Macroeconomics

Macroeconomics
A European Text
SEVENTH EDITION
Michael Burda
and
Charles Wyplosz
Great Clarendon Street, Oxford,Ox2 6dp,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Michael Burda and Charles Wyplosz 2017
The moral rights of the authors have been asserted
Fourth edition published 2005
Fifth edition published 2009
Sixth edition published 2013
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2016963302
ISBN 978–0–19–873751–3
Printed in Great Britain by
Bell & Bain Ltd., Glasgow
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
The cover image shows the face of Dionysus.

Preface to the Seventh Edition
The seventh edition of Macroeconomics: A European Text will be published more than a quarter-century after we first set out on our long journey. In 1990, two
colleagues at INSEAD in Fontainebleau, a Frenchman and an American with Polish family names, decided to write a textbook that dropped the usual US-
American frame of reference, consistent with our chosen workplaces and the probing questions our students posed to us over the years. We think we have been
somewhat successful in identifying some of the big macroeconomic questions for Europe—as opposed to the United States, where the institutional, political, and
social setting is quite different. These questions have inspired young people that we have taught to appreciate why macro is such a vibrant, interesting and
relevant area of economics. More than ever, we have tried to accentuate the positive and common ground in our profession, despite the eternal squabbling—in a
quest to make the material more credible and digestible for our students.
Since the last edition appeared in 2013, much has happened in Europe. The book was written in the wake of the Great Recession’s aftershocks, and Europe was
still dealing with—or muddling through—their implications for trade, monetary union, and fiscal policy. Add civil war in the Ukraine, the refugee crisis, and
terrorism, all of which had implications for trade, migration policy and the further steps of European integration. And then came Brexit, in which the UK’s citizens
decided—perhaps not entirely consciously, but decisively all the same—against further participation in the European integration project, at least as it was
envisioned by Monnet, Werner, de Gaulle, Churchill, Kohl and many prominent figures of the last century. It was an outcome that all democrats must accept, yet
one with deep implications for all of Europe. It remains to be seen how and when the UK separates from the European Union.
And yet, a lot has not happened. The common currency project remains a construction site. Objectively, there is little chance of Europe agreeing to further
integration steps in the years to come. Fiscal policy coordination remains in the prohibitive rather than proactive mode, even though the collapse of aggregate
demand in the southern Euro-periphery was far deeper than was foreseen in 2011—and might well have been prevented. The Great Recession left a deep scar on
the world economy, and even the US and China have grown more slowly since.
Europe may be paying the price for its haste to adopt the common currency without considering all the ramifications of a monetary union without a nation-
state. Many consider the lack of a fiscal union to be crucial, but an even more ominous error was to overlook that banks are the most important component of the
monetary plumbing of an integrated economy. The introduction of the euro had focused too much on the currency and not enough on the integration of the
financial system and the robust ability of banks to provide the means of payment across national boundaries. A decade afterwards this mistake is still being
corrected, and the banking union has yet to be implemented. If anything, national financial systems are at high risk of becoming re-nationalized or even
Balkanized. The vision of a unified European capital market seems even more distant.
Understanding the macroeconomics disturbances of the dimensions that struck the world in 2008 is an important part of macroeconomics. We remain
convinced that the crisis was a failure of policy, not of macroeconomics. We have enough tools to analyse and understand the events of the past decade. The
fault lies with the users of those tools. Policymakers either used them incorrectly or took their eye off one of the many balls of sound policy. The principles
developed in the current and previous editions of this textbook can help organize our thoughts on a very complex situation, and allow us to evaluate the
adequacy and appropriateness of current policy responses. Because the study of economics, especially macroeconomics, tends to be countercyclical, a
continuing objective of the seventh edition is to give our readers the ability to assess the current European crisis in a structured way.
Like much of Europe, our book needed a structural overhaul, and we have added a number of new features to the present edition:
We moved asset markets to the front of the book as part of the building block strategy for mastering the determinants of aggregate demand;
We brought the banking system to the fore, both in our description of the macroeconomic system and in the characterization of monetary policy;
We added a new chapter on monetary policy which stresses the importance of financial stability in macroeconomic health. We discuss the zero lower bound problem and
nonstandard policy measures meant to counter it, including quantitative easing, negative deposit rates and more;
We adopted the new IMF standards for measuring balance of payments;
We revised and rewrote the core IS–TR chapters in the closed and open economy, as well as the central AS-AD chapters;
We have added much material on the ongoing Euro crisis, including boxes on Greece, Ireland, the TARGET-2 debate, the Swiss franc episode, and secular stagnation.
As with every new version of our book, we reviewed our book word for word to improve readability and stimulate student interest. This time we were able to
engage a large team of students to comb the text for typos, misplaced references and other issues. At the technical level, we have further simplified the math and
notation.
This textbook can be used in second or third-year bachelor-level courses, as well as macroeconomic modules in business schools, public policy and political
science programmes. While every chapter contains macro content and insights into the European context, some instructors may prefer to use only those parts
that fit into their teaching plans. While designed for a full year, two-semester course, Macroeconomics: A European Text can certainly be used for shorter
courses as well. Here are three suggested options:
Full course track Medium-length track Short track
All 20 chapters 15 chapters 10 chapters
General introduction General introduction General introduction
Chapters 1 and 2 Chapters 1 and 2 Chapters 1 and 2
The long run The long run and the real economy The long run and the real economy
Chapters 3 to 5 Chapters 3, 4, and 5 Chapters 3, 4, and 5
Building blocks The short run The short run
Chapters 6 to 10 Chapters 8 to 14 Chapters 11 to 14
Equilibrium The exchange rate The exchange rate
Chapters 11 to 15 Chapter 15 Chapter 15
Policies and debates Policies and debates
Chapters 16 to 19 Chapters 16 to 18
History of macro thought
Chapter 20
Oxford University Press continues to provide valuable support at various stages of the editorial process of the seventh edition. The reviews by current users
and non-users were extremely helpful, and gave us a finger on the pulse of the teaching profession. We are indebted to the Publishing Editors, Becci Curtis and
Kat Rylance, for expert guidance in keeping the book close to schedule. The production process sailed smoothly thanks to Sal Moore. Caroline Quinnell provided
excellent copy editing, Angela Foskett produced a superb index of words, and Dr Giancarlo Ianulardo conducted a thorough accuracy check of the final book.
We are also grateful for our many adopters, colleagues and students for offering invaluable feedback on adapting and restructuring our book to the new
challenges posed by the current crisis. In particular, Stan Standaert and Falk Mazelis provided very useful discussions and feedback. Thanks to Gian Maria
Milesi-Feretti for his recent updates of international investment position data.
Our textbook has always been a data-intensive enterprise, and this edition would not have been possible without the help of an army of student research
assistants. We are especially grateful to Niklas Flamang who worked for almost six months meticulously gathering data for the project. Thomas Dengler, Daniel
Jacob, Jacob Meyer, and Max Reinhardt worked hard and long in the final weeks, and were invaluable for their careful proofreading and research assistance.
Michael Burda and Charles Wyplosz
2017
Brief Contents
List of Tables
List of Figures
List of Boxes
PART I
Introduction to Macroeconomics
1 What is Macroeconomics?
2 Macroeconomic Accounts
PART II
The Macroeconomy in the Long Run
3 The Fundamentals of Economic Growth
4 Labour Markets and Unemployment
5 Money, Prices, and Exchange Rates in
the Long Run
PART III
The Building Blocks of Macroeconomics
6 Borrowing, Lending, and Budget Constraints
7 Asset Markets 168
8 Private Sector Demand: Consumption and Investment
9 Money and Interest Rates
10 Monetary Policy, Banks, and Financial Stability
PART IV
Macroeconomic Equilibrium
11 Macroeconomic Equilibrium in the Short Run
12 International Capital Flows and Macroeconomic Equilibrium
13 Output, Employment, and Inflation
14 Aggregate Demand and Aggregate Supply
15 The Exchange Rate
PART V
Macroeconomic Policy in a Global Economy
16 Demand Management Policies
17 Fiscal Policy, Debt, and Seigniorage
18 Policies for the Long Run
19 The Architecture of the International
Monetary System
20 Epilogue
References
Glossary
Index
Detailed Contents
List of Tables
List of Figures
List of Boxes
PART I
1.1 Overview of Macroeconomics
1.2 Macroeconomic Concepts
1.3 Macroeconomics in the Long Run: Economic Growth
1.4 Macroeconomics in the Short Run: Business Cycles
1.5 Macroeconomics as a Science
1.6 The Methodology of Macroeconomics
1.7 Preview of the Book
2.1 Overview
2.2 Gross Domestic Product
2.3 Flows of Incomes and Expenditures
2.4 Balance of Payments
Summary
PART II
3.1 Overview
3.2 Thinking about Economic Growth: Facts and Stylized Facts
3.3 Capital Accumulation and Economic Growth
3.4 Population Growth and Economic Growth
3.5 Technological Progress and Economic Growth
3.6 Growth Accounting
3.7 Why are Some Countries Rich and Others Poor?
Summary
4.1 Overview
4.2 Demand and Supply in the Labour Market
4.3 A Static Interpretation of Unemployment
4.4 A Dynamic Interpretation of Unemployment
4.5 The Equilibrium Rate of Unemployment
Summary
5 Money, Prices, and Exchange Rates in the Long Run
5.1 Overview
5.2 Money and the Neutrality Principle
5.3 Nominal and Real Exchange Rates
5.4 The Exchange Rate in the Long Run:Purchasing Power Parity
Summary
PART III
6.1 Overview
6.2 Thinking About the Future
6.3 The Household’s Intertemporal Budget Constraint
6.4 The Firm and the Private Sector’s Intertemporal Budget Constraint
6.5 Public and Private Budget Constraints
6.6 The Current Account and the Budget Constraint of the Nation
Summary
7 Asset Markets
7.1 Overview
7.2 How Asset Markets Work
7.3 Functions of Asset Markets
7.4 Asset Prices and Yields
7.5 Information and Market Efficiency
7.6 Asset Markets and the Macroeconomy
Summary
8 Private Sector Demand:Consumption and Investment
8.1 Overview
8.2 Consumption
8.3 Investment
Summary
9.1 Overview
9.2 Money: What is it and Where Does it Come From?
9.3 The Demand for Money and the Market for Money
9.4 Money: Past, Present, and Future
Summary
10.1 Overview
10.2 The Instruments of Monetary Policy: Nuts and Bolts
10.3 Objectives, Targets, and Instruments of Monetary Policy
10.4 The Channels of Monetary Policy
10.5 Financial Stability as a Prerequisite for Monetary Policy
Summary
PART IV
11.1 Overview
11.2 Aggregate Demand and the Goods Market 277
11.3 The Goods Market and the IS Curve
11.4 The Money Market, Monetary Policy, and the TR Curve
11.5 Macroeconomic Equilibrium
Summary
12.1 Overview
12.2 The Implications of Being Small
12.3 International Financial Flows
12.4 Output and Interest Rate Determination under Fixed Exchange Rates
12.5 Output and Interest Rate Determination under Flexible Exchange Rates
12.6 Fixed or Flexible Rates?
Summary
13.1 Overview
13.2 General Equilibrium with Flexible Prices: The Neoclassical Case
13.3 The Phillips Curve: Chimera or a Stylized Fact?
13.4 Accounting for Inflation: The Battle of the Mark-ups
13.5 Inflation, Unemployment, and Output
Summary
14.1 Overview
14.2 Aggregate Demand and Supply under Fixed Exchange Rates
14.3 Aggregate Demand and Supply under Flexible Exchange Rates
14.4 How to Use the AS–AD Framework
Summary
15.1 Overview
15.2 The Foreign Exchange Markets
15.3 The Interest Parity Conditions
15.4 Exchange Rate Determination in the Short Run
15.5 The Exchange Rate in the Long Run
15.6 From the Long to the Short Run
15.7 Exchange Rate Volatility and Currency Crises
Summary
PART V
Macroeconomic Policy in a Global Economy
16.1 Overview
16.2 Demand Management: What are the Issues?
16.3 Feasible Demand Management Policy
16.4 Beyond Controversies: The Synthesis
16.5 The Great Recession and Demand Management: New Challenges or Old Dilemmas?
Summary
17.1 Overview
17.2 Fiscal Policy and Economic Welfare
17.3 Macroeconomic Stabilization
17.4 Deficit Finance: Public Debt and Seigniorage
17.5 How to Stabilize the Public Debt
Summary
18.1 Overview
18.2 Market Efficiency and the Theory of Supply-Side Policy
18.3 Product Market Policies
18.4 Taxation as the Price of Intervention
18.5 Labour Market Policy
18.6 Supply-Side Policy in Practice
Summary
19 The Architecture of the International Monetary System
19.1 Overview
19.2 History of Monetary Arrangements
19.3 The International Monetary Fund
19.4 Currency Crises
19.5 The Choice of an Exchange Rate Regime
Summary
20 Epilogue
20.1 The Keynesian Revolution
20.2 The Monetarist Revolution
20.3 The Rational Expectations Revolution
20.4 The Microfoundations of Macroeconomics
20.5 New Keynesian Macroeconomics: The Latest Synthesis
20.6 Institutional and Political Economics
20.7 Labour Markets
20.8 Search and Matching
20.9 Growth and Development
20.10 Demographics, Low Productivity Growth, and Secular Stagnation
20.11 Conclusions
References
Glossary
Index
List of Tables
1.1 Real Income per Capita (GDP in euros, 2000 prices)

1.2 Openness (ratio of average of exports and imports to GDP, %)
1.3 GDP Growth Forecasts, 2009 and 2010 (% per annum)
2.1 Growth Rates of Nominal GDP, Real GDP, and GDP Deflator: Euro Area 2005–2015 (% per annum)
2.2 Estimates of 2008 German Nominal GDP
2.3 Components of GDP by Expenditure, 1999–2015 (% of GDP)
2.4 GDP and Household Disposable Income, 2014
2.5 The Accounting Identity in 2010 (% of GDP)
2.6 The Balance of Payments
2.7 Balance of Payments: Some Examples
2.8 Balance of Payments, Various Countries, 2014 (US$ billion)
3.1 The Growth Phenomenon

3.2 Capital–Output Ratios (K/Y), 1913–2009
3.3 Average Annual Growth Rates of GDP per Capita since 1960
3.4 Growth of Real Gross Fixed Capital Stock, 1913–2010 (% per annum)
3.5 Population, Employment, and Hours Worked, 1913–2010
3.6 The Solow Decomposition (average annual growth rates)
4.1 Annual Total Hours Worked and Average Wages, 1870–2014

4.2 Weekly Hours Worked per Person of Working Age
4.3 Female Labour Force Participation Rates
4.4 European Trade Unions: Membership and Coverage, 1950–2013
4.5 Standardized Unemployment Rates 1960–2016 (% of labour force)
4.6 Minimum Wages and Youth Labour Market Experience, 2013–2014
4.7 Unemployment Stocks and Flows, 2015
4.8 Inflows into Unemployment and Unemployment Rates in the UK, March 2004
4.9 Unemployment Compensation: Conditions for Eligibility and Benefit Levels, 2015
4.10 Estimates of the Equilibrium Unemployment Rate (% of the labour force)
5.1 Inflation and Money Growth in the Long Run: A Rule of Thumb (assuming that real money demand grows at 3% per annum)
6.1 Interest Rates for Government and Corporate Bonds, 29 February 2016 (% per annum)
7.1 Stock Market Capitalization and Trading, February 2016
7.2 Bond Prices and Yields: Some Examples
9.1 Money in Five Places, 2015
9.2 Reserve Ratio Requirements in Selected Countries, 2015
12.1 Measures of Openness and Economic Size, 2014 (% of GDP)
12.2 Sterilized and Unsterilized Foreign Exchange Market Interventions
12.3 The Mundell–Fleming Model: A Summary of the Importance of the Exchange Rate Regime
13.1 Wage Share of Value Added by Country and Selected Industries, 2014
13.2 Equilibrium Unemployment Rates
14.1 Tracking Movements in Figure 14.6
14.2 Iceland and Ireland: Key Economic Indicators, 2004–2015
15.1 Shares of Currencies in Foreign Exchange Market Transactions
15.2 Shares of Currency Pairs in Foreign Exchange Market Transactions
15.3 Average Daily Foreign Exchange Transaction Volume (US$ billion) 390
15.4 GDP per Capita and Price Levels, 2014 (USA = 100)
16.1 Inflation Targets Set by Inflation-Targeting Central Banks in 2016

16.2 Persistence of Inflation
16.3 Changes in Inflation and Growth in the Great Recession, 2008–2009
17.1 General Government Spending and Finances: Eurozone, USA, UK, and Japan, 2016
17.2 Government Transfers, Various Countries, 1960 and 2010
17.3 Government Budget Balances, Various Countries, 1975–2015 (% of GDP)
17.4 Fiscal Implication of German Reunification (% of GDP)
17.5 Expected and Realized Government Budgets in 2010 (% of GDP)
17.6 Gross Public Debt, Various Countries, 1970–2015 (% of GDP)
17.7 Net Government Indebtedness and Primary Budget Balances, 2015 (% of GDP) 461
17.8 Economic Growth in Southern Europe, 1980–2015 (% per annum, average)
17.9 Primary Budget Balances for Selected European Countries, 2000–2015
17.10 Public Finances in Italy, 1918–1928
18.1 Ease of Doing Business Rankings, 2015
18.2 Subsidies in Various Countries (% of GDP)
18.3 Expenditures on Labour Market Programmes, 2013 (% of GDP)
18.4 Measures of Employment Protection, 2013
18.5 Labour Taxation in 2015 (in %)
19.1 Inflation Rates in Five Countries, 1900–1913 (annual average rate of increase in GDP deflator, %) 512
19.2 Beggar-thy-Neighbour Depreciations, Various Countries, 1931–1938 (value of currencies as percentage of 1929 gold parity)
19.3 IMF Quotas and Voting Rights (January 2016)
19.4 Impact of a 0.1% Tobin Tax and the Holding Period of Investments
19.5 Exchange Rate Regimes in European Union Countries, 2016
List of Figures
1.1 GDP per Capita and Life Satisfaction in 2014

1.2 Unemployment Rates in the Eurozone, Switzerland, and the USA, 1970–2015
1.3 Labour Share of Income in Manufacturing and Stock Prices, Four Countries, 1950–2016
1.4 Capacity Utilization Rates and Changes in Inflation Rates, USA, 1967–2015
1.5 Gross Domestic Product (GDP), France, Germany, and the UK, 1870–2015
1.6 Quarterly Gross Domestic Product, UK, 1956:1–2015:3
1.7 Burns–Mitchell, Now: Eight Countries, 1970–2015
1.8 Burns–Mitchell, Back Then: US Monthly Data, 1914–1938
1.9 Price Levels and Inflation Rates, France and the UK, 1870–2015
1.10 Endogenous and Exogenous Variables
2.1 Inflation Rates, GDP Deflator, and Consumer Price Index: Italy, 1985–2014
2.2 Estimates of the Size of the Underground Economy (% of GDP)
2.3 The Circular Flow Diagram
2.4 From Domestic Product to Disposable National Income, 2013
3.1 GDP per capita in the UK 1920–2015

3.2 The Production Function
3.3 The Production Function in Intensive Form
3.4 The Output–Labour and Capital–Labour Ratios in Three Countries (1820–2015)
3.5 The Steady State
3.6 Catching Up
3.7 Investment, GDP per Capita, and Real GDP Growth
3.8 An Increase in the Savings Rate
3.9 The Golden Rule
3.10 Raising Steady-State Consumption
3.11 Population and Employment in the Eurozone and the USA, 1960–2016
3.12 The Steady State with Population Growth
3.13 Population Growth and GDP per Capita, 1960–2009
3.14 The Steady State with Population Growth and Technological Progress
3.15 Growth Rates along the Steady State
3.16 Growth Rate of Total Factor Productivity in the USA (% per annum)
3.17 The Convergence Hypothesis in Reality
3.18 Conditional Convergence
3.19 Gross Fixed Capital Formation in the Public Sector, 1995–2015 (% of GDP)
3.20 Index of Economic Freedom Around the World in 2016
4.1 Household Preference

4.2 The Household Budget Line and Optimal Choice
4.3 Reaction of the Household to a Wage Increase: Labour Supply
4.4 Individual and Aggregate Labour Supply
4.5 The Production Function and the Labour Demand Curve
4.6 An Increase in Labour Productivity
4.7 Equilibrium in the Labour Market
4.8 Shifting Labour Demand and Supply
4.9 Involuntary Unemployment
4.10 Trade Unions’ Indifference Curves
4.11 The Collective Labour Supply Curve
4.12 Labour Market Equilibrium with a Trade Union
4.13 Minimum Wages
4.14 A Map of Labour Markets
4.15 Employment and Real Wages in the Euro Area and the USA, 1970–2015
4.16 Unemployment in France and Germany, 1955–2014
4.17 Actual and Equilibrium Employment
5.1 Money Growth, Inflation, and Exchange Rate Depreciation, 1975–2006

5.2 Annual Inflation in Zimbabwe, 2001–2015
5.3 Zimbabwean Banknotes, Mid-Year 2008
5.4 Money Market Equilibrium
5.5 The Swedish Krona’s Nominal and Real Exchange Rates, 1970–2015
6.1 Endowment, Wealth, and Consumption
6.2 Inheriting Wealth or Indebtedness
6.3 The Production Function
6.4 Productive Technology
6.5 Unproductive Technology
6.6 Investment Increases Wealth
6.7 The Government Budget Line
6.8 Primary Consolidated Government Budget Surpluses, Four Countries, 1970–2015
6.9 Ricardian Equivalence
6.10 Public and Private Borrowing Rates in Italy, 2003:1–2016:1
6.11 Borrowing Constraints
6.12 Ricardian Equivalence in the UK, 1970–2016
6.13 Net Asset Positions, 2014 (% of GDP)
7.1 Trading Hours of Stock Markets Around the World, Greenwich Mean Time
7.2 Yield Curves
7.3 The Risk–Return Curve
7.4 Triangular Arbitrage
7.5 Ostmark–DM Rate and Bid–Ask Spreads, August 1989–June 1990
7.6 The SP500 Index in Early 2016
7.7 Possible Stock Price Paths
7.8 Tulipmania, 1637
7.9 Housing Prices in Ireland, 1978:1–2015:3
7.10 The Rise and Fall of NASDAQ Stocks, 1992–2016
8.1 Indifference Curves
8.2 Optimal Consumption
8.3 Temporary and Permanent Income Changes
8.4 Consumption Smoothing in Greece and Germany, 1995–2016
8.5 Life-Cycle Consumption
8.6 The Effect of an Increase in the Interest Rate
8.7 Consumption, Disposable Income, and Wealth in France, 1978–2014
8.8 Credit Constraints
8.9 GDP, Domestic Demand, and the Current Account: Poland and East Germany
8.10 The Optimal Capital Stock
8.11 Technological Progress
8.12 The q-Theory of Investment
8.13 Investment and Tobin’s q, Weimar and Modern Germany
8.14 Tobin’s q
9.1 Currency in Circulation (in per cent of nominal GDP)

9.2 Balance Sheet of a Typical Commercial Bank, Abstract and Concrete
9.3 The Money Multiplier
9.4 The Reserves–Money Stock Link
9.5 Interest Rates in the Euro Area, 2007–2015
9.6 The Composition of Money Demand and Interest Rates, Euro Area, 2000–2016
9.7 The Money Market
10.1 Balance Sheet of Typical Central Banks, Abstract and Concrete
10.2 Balance Sheet of a Central Bank Before and After an Open Market Purchase
10.3 Setting the Interest Rate
10.4 ECB Interest Rates (January 1999–June 2016)
10.5 Choice of the Monetary Policy Strategy
10.6 Inflation in OECD Countries, 1970–2015
10.7 Money Multipliers (M1/M0) Before and After the Crisis
10.8 Inflation Forecasts of Central Banks and Outcomes
10.9 Examples of Taylor Rules
10.10 Euro Area Yield Curves
10.11 Quantitative Easing in the Euro Area
10.12 Mistrust in the Interbank Market (3-month interest rates)
10.13 The Effect of Loan Losses on a Bank’s Balance Sheet
10.14 The Effect of a Government Recapitalization on a Bank’s Balance Sheet
11.1 Cyclical Fluctuations

11.2 General Macroeconomic Equilibrium
11.3 The ISCurve
11.4 An Exogenous Increase in Aggregate Demand
11.5 Consumption and House Prices in the USA, 1975–2015
11.6 The TR Curve
11.7 The TR Curve and Money Market Equilibrium
11.8 The Slope of the TR Curve
11.9 The LM Curve
11.10 Shifts of the TR Curve
11.12 Macroeconomic Disturbances
11.13 Short-term Interest Rates in the UK, USA, and Germany (% per annum), 1990–2016
11.14 The Zero Lower Bound and the Great Recession
12.1 International Financial Market Equilibrium
12.2 The Evolution of Financial Account Liberalization 1970–2013
12.3 The Money Market Under Fixed Exchange Rates
12.4 Demand Shocks Under Fixed Exchange Rates
12.5 An International Financial Shock
12.6 A Devaluation
12.7 China: Foreign Exchange Reserves (US$ billion)
12.8 Monetary Policy Disturbances Under Flexible Exchange Rates
12.9 A Demand Disturbance Under Flexible Exchange Rates
12.10 Long-Term Interest Rates, 1970–2015
12.11 Swiss Franc/Euro Exchange Rate and the Swiss Money Supply, 1999–2015
12.12 An Interpretation of the Swiss Case
13.1 From the Short to the Long Run
13.2 Output and the Labour Market in the Long Run
13.3 The Phillips Curve in Theory
13.4 The Phillips Curve in Reality
13.5 The Output Gap and Unemployment in Germany, 1970–2016
13.6 Okun’s Law in Theory
13.7 Okun’s Law in Reality
13.8 The Aggregate Supply Curve
13.9 The Long Run
13.10 Phillips Curves: Recent Experiences in the Euro Area and the UK
13.11 The Price of Oil, 1950–2015
13.12 The Expectations-Augmented Phillips and Aggregate Supply Curves
13.13 From the Short to the Long Run
14.1 Aggregate Demand and Aggregate Supply, Short Run and Long Run
14.2 Inflation in Denmark and the Euro Area, 1992–2016
14.3 The Aggregate Demand Curve Under Fixed Exchange Rates
14.4 Shifts in the Aggregate Demand Curve
14.5 Aggregate Demand and Supply Under Fixed Exchange Rates
14.6 Fiscal Policy Under Fixed Exchange Rates
14.7 Devaluation
14.8 Expansionary Monetary Policy Under a Fixed Exchange Rate Regime
14.9 The Real Exchange Rate, France vs. Germany, 1975–2011
14.10 Nominal and Real Interest Rates on French Government Debt, 1998–2011
14.11 The Aggregate Demand Curve Under Flexible Exchange Rates
14.12 Aggregate Demand and Supply Under Flexible Exchange Rates
14.13 Monetary Policy Under Flexible Exchange Rates
14.14 An Adverse Supply Shock
14.15 An Adverse Demand Disturbance
14.16 Monetary Policies 2007–2015
14.17 Disinflation
15.1 Fluctuations of the Euro/Sterling Pound Exchange Rate
15.2 The UIRP Condition Repeated Over and Over Again
15.3 An Increase in the Domestic Interest Rate
15.4 Monetary Policy Under Floating Exchange Rates
15.5 The Primary Current Account Function
15.6 The Equilibrium Real Exchange Rate
15.7 The Equilibrium Exchange Rate and Non-Price Competitiveness
15.8 Real Exchange Rates, 1995–2015
15.9 Primary Current Accounts and Real Exchange Rates, Greece and Ireland 1999–2015
15.10 Korea’s Exchange Rate (index 100 = 2005 average)
16.1 Real GDP Levels (2005 = 100)
16.2 The Neoclassical Case
16.3 The Keynesian Case
16.4 Exchange Rate Anchors in Argentina and Bulgaria, 1992–2010
16.5 The Swiss Franc and Prices for Traded and Non-traded Goods, 2007–2016
16.6 Variability of Inflation, Real Wages, and Unemployment: OECD Countries, 1960–2010
16.7 The Impulse-Propagation Mechanism
16.8 Impulses and Propagations: An Example
16.9 The Propagation Framework in the AS–AD Model
16.10 Lags and Demand Management Policy
16.11 Burns–Mitchell Diagrams, Macroeconomic Indicators in Eight OECD Economies, 1970–2016
16.12 Political Business Cycles
16.13 Partisan Politics
16.14 Size of the Balance Sheet of Central Banks
17.1 Public Debt in Four Countries (% of GDP)
17.2 Stabilization Policies
17.3 Cyclical Behaviour of the Primary Budget in the Netherlands, 1980–2010
17.4 Endogenous and Exogenous Components of Budgets
17.5 Changes from 2008 to 2010 in Actual and Cyclically Adjusted Budget Balances, 20 OECD Countries (% of GDP)
17.6 Yields on Greek, Irish, Portuguese, Spanish, and Italian 10-year Bonds, Relative to Germany, 2009–2011 (% per annum)
18.1 The Macroeconomics of Supply-Side Policies
18.2 The Supply-Side Economics of Immigration
18.3 The Impact of Supply-Side Policies
18.4 The Effect of Taxation
18.5 The Laffer Curve in Theory
18.6 Laffer Curves in Reality?
18.7 Beveridge Curves in the UK, Germany, and the USA, 1960–2010
18.8 Incentives and the Social Safety Net
18.9 Two Successful Supply-Side Reforms: the Netherlands and Ireland
19.1 Monetary Gold Stock and Cumulative Gold Production, 1840–1980
19.2 The Decline of World Trade During the Great Depression
19.3 The Three Layers of Bretton Woods
19.4 US Official Liabilities and Gold Reserves, 1950–1970
19.5 Inflation Rates: USA, UK, Germany, and Italy, 1960–1976
19.6 Fundamental Crisis in Iceland in 2008
19.7 Currency in Crisis (index 100 ∙ January 1997)
19.8 The N–1 Problem
19.9 TARGET-2 Imbalances; 2001–2016 (€ bn)
19.10 Cumulative Price Level and Unit Labour Cost Changes Relative to Eurozone-12 Average, 1999–2010
List of Boxes
1.1 Burns–Mitchell Diagrams, Now and Then

1.2 All Kinds of Cycles
1.3 Forecasting the Crisis Years
1.4 Macroeconomic Schools of Thought: A Primer
2.1 Value Added and Value Subtracted: Two Examples

2.2 What GDP Measures
2.3 Price Deflators and Price Indices
2.4 The Underground Economy and Unpaid Work
2.5 How National Accounts Estimates Can Vary over Time
2.6 Examples of Balance of Payments Accounts
2.7 Links Between National Income Accounts and the Balance of Payments
2.8 China’s Foreign Exchange Reserves
3.1 China and the Chinese Puzzle of Economic Growth

3.2 For the Mathematically Minded: The Cobb–Douglas Production Function
3.3 Growth Miracles Eventually Come to an End
3.4 Population Growth and GDP per capita
3.5 Techno-pessimists vs. Techno-optimists
3.6 A Snail’s Pace: The 2% Convergence Rule
4.1 Technical Change and Unemployment

4.2 Market Hours at Work: Europe versus the USA
4.3 The European Unemployment Problem
4.4 Minimum Wages and Youth Unemployment
4.5 European Labour Market Policy and the German ‘Miracle’
5.1 Evidence for Monetary Neutrality: The Introduction of the Euro

5.2 Zimbabwe: The Top of the Inflation Heap
5.3 Arithmetic of Rates of Change
5.4 Computing and Comparing Effective Exchange Rates
6.1 Neither a Borrower nor a Lender Be: Economics and the Sociology of Credit
6.2 Discounting and Bond Prices
6.3 Gross Investment, Depreciation, and the Capital Stock
6.4 The European Debt Crisis
6.5 Global Imbalances
6.6 Pyramids: Is it Possible to Beat the Budget Constraint?
7.1 The Term Structure of Interest Rates
7.2 Risk Diversification
7.3 The Price of Risk
7.4 Mortgage Securitization and the US Financial Crisis of 2008–2009
7.5 The Short-Lived Market for Ostmarks
7.6 Tulipmania
8.1 Indifference Curves, Intertemporal Substitution, and Optimal Consumption

8.2 Germany and Greece During the Crisis
8.3 Does China Defy the Permanent Income Hypothesis?
8.4 Current Income and Spending in East Germany and Poland
8.5 Optimal Capital Investment: Looking Beyond Two Periods
8.6 User Cost of Capital, Tobin’s q, and the Price of Investment Goods
9.1 Wicksell’s Cashless Society

9.2 Money, the Payments System, and the Role of the Banks
9.3 The Money Multiplier
9.4 How Can Interest Rates be Negative?
10.1 How the ECB Does It

10.2 Hyperinflation and Central Banks: Historical and Current Episodes
10.3 The Taylor Rule
10.4 Bankers’ Mistrust, Eurozone Edition
10.5 Basel Regulation and Capital Adequacy
10.6 Bagehot Rules!
11.1 The Multiplier’s Algebra
11.2 The American Dream, An American Nightmare
11.3 When Central Banks Target the Money Supply: The LM Curve
11.4 Monetary Policy under Duress: The Zero Bound
11.5 Secular Stagnation Ahead?
12.1 What is Openness?
12.2 Monetary Side Effects of Foreign Exchange Market Interventions
12.3 Fixed Exchange Rates in China
12.4 The Challenges of the Swiss Franc
13.1 Where Does the Money Growth Come From?
13.2 When and How Firms Change Prices
13.3 The Wage Mark-up and the Labour Share
14.1 The Real Exchange Rate and Money Growth Under a Fixed Exchange Rate Regime
14.2 Conflict and Coexistence with Different Inflation Rates: France and Germany
14.3 Ireland v. Iceland: Vulnerable Islands in a Global Financial Tsunami
14.4 Wage Negotiations: The Time Dimension
15.1 The Real Interest Rate Parity Condition

15.2 Mussa’s Stylized Facts and the Asset Behaviour of Exchange Rates
15.3 Iterating Forward the UIRP Condition
15.4 Interest Rate Parity and Short-Run Open Economy Macroeconomics
15.5 The Norwegian Cure for the Dutch Disease
15.6 What Goes Down Must Come Up, and Conversely
15.7 The Balassa–Samuelson Effect
16.1 The Exchange Rate Anchor: Argentina and Bulgaria
16.2 Switzerland: The Safe Haven
16.3 Optimal Inflation
16.4 Computers, Scientists, and Business Cycles
16.5 Impulse-Propagation Mechanisms in the AS–AD Model
16.6 Leading and Lagging Indicators
16.7 This Time It’s Different? Unconventional Policies for Unconventional Times
17.1 Tax Smoothing after German Reunification
17.2 The Budgetary Process
17.3 Debt–Deficit Arithmetic
17.4 Euro Area and the Stability and Growth Pact
17.5 Mussolini and the Public Debt
17.6 Interest Rates Unhinged: The Great European Sovereign Debt Crisis
18.1 EU Competition Policy and National Preferences
18.2 The Supply-Side Economics of Immigration
18.3 Telecommunications: Successful Supply-Side Policy
18.4 The Deadweight Loss from Taxation
18.5 Crusoe Caught in the Safety Net
18.6 Taxes and the Labour Market in Europe
18.7 Supply-Side Reforms: A Solution to Greece’s Problems?
19.1 Bimetallism and Gresham’s Law
19.2 How the IMF is Managed
19.3 Icelandic Aftershocks following the Financial Earthquake
19.4 Two South-East Asian Crises and the Aftermath: 1997–1998 and 2007–2008
19.5 The Tobin Tax (a.k.a. Financial Transactions Tax)
19.6 Exchange Rate Regimes in EU Member Countries
19.7 The N–1 Problem
19.8 The TARGET-2 Controversy
19.9 Charlemagne’s European Monetary Union
19.10 The Euro-Crisis: Fundamental or Non-Fundamental?
PART I
Part I of this textbook sets the stage for macroeconomics as a subject. Chapter 1 explains the objectives and methods of
macroeconomics, its history and usefulness, as well as its controversies and open questions. It provides a number of
essential definitions and offers a preview of what will follow.
Macroeconomics is about measurement as well as analysis. Chapter 2 introduces the methods of measurement used
in macroeconomics. In particular, we will explore the national income accounts, the language economists use to describe
and communicate the economic activities of a region or a nation, and the balance of payments, which summarizes its
transactions with the rest of the world.
What is
Macroeconomics? 1
1.2.1 Income and Output
1.2.2 Unemployment
1.2.3 Factors of Production and Income Distribution
1.2.4 Inflation
1.2.5 Financial Markets and the Real Economy
1.2.6 Openness
1.5.1 The Genesis of Macroeconomics
1.5.2 Macroeconomics and Microeconomics
1.5.3 Demand and Supply
1.6.1 What is to be Explained?
1.6.2 Theory and Realism
1.6.3 Positive and Normative Analysis
1.6.4 Testing Theories: The Role of Data
1.6.5 Macroeconomic Modelling and Forecasting
1.7.1 Structure
1.7.2 Controversies and Consensus
1.7.3 Rigour and Intuition
1.7.4 Data and Institutions
1.7.5 Europe
The Theory of Economics does not furnish a body of settled conclusions immediately applicable to policy. It is a method rather than a doctrine, an apparatus of the mind, a technique
of thinking, which helps its possessor to draw correct conclusions.
J. M. Keynes

Unemployment, inflation, booms and busts, financial markets, interest rates, and exchange rates are everyday fare in the news. All these phenomena affect our well-
being. This why macroeconomics is so exciting. Macroeconomics is more than just headlines, however: it is a fascinating intellectual adventure. The breadth of
issues it covers is evidence enough of its inherent complexity. Yet, a few simple ideas can go a long way seeing through complex situations. Macroeconomics is also
enormously useful. People, rich or poor, are affected as jobs appear or vanish and as prices change. Businesses can gain or lose large amounts of money when their
environment changes, regardless of how well they are managed. Citizens want to understand what their governments are up to. Dramatic events like depressions,
when overall economic activity is very far below average and unemployment soars, or hyperinflations, when prices are increasing at monthly rates of 50% or more, can
tear at a society’s social fabric, and yet can be prevented when policy-makers apply sound economic principles.
But are ‘sound economic principles’ valid? The events that started in 2007 and led to the Great Recession and the Eurozone crises have led some observers to
conclude that macroeconomics needs to be overhauled. Yet it is also common to blame medical doctors for epidemics, and or messengers for bringing bad news, and
that doesn’t sound fair. Macroeconomists are in a similar situation today. We will argue that macroeconomics as a field has proved its mettle. To be sure, our
understanding is very far from perfect. In particular, macroeconomists are not good at making forecasts and the profession has failed to see the crisis coming. Yet, the
principles underlying our field made it possible to understand these dramatic events surprisingly well as they unfolded. They also contained prescriptions for what
needed to be done. When governments heeded this advice, the economic situation in fact improved. In any case, we did not experience the Great Depression of the
1930s, and we know why. Besides helping diagnose such situations, the material in this book will teach you humility: Some events, both economic and political,
cannot be predicted. This includes banking and financial crises, which contain strong behavioural elements.
A key feature of macroeconomics is that it is geared to finding solutions. Its principles may look abstract at first, but they readily lead to policy prescriptions. These
prescriptions are directly related to the observation of economic events with undesirable consequences, which in turn presumes agreement on what is to be measured
and how measurements are made. Before the arrival of macroeconomics, concepts such the gross domestic product (GDP), the consumer price index, or the
unemployment rate simply did not exist. Some of these concepts are presented in Section 1.2, many more will be introduced as we move through the book. The next
step is to formulate a diagnosis. A key challenge at this stage is to distinguished between business cycles, or short-run fluctuations of output and employment, and
long-run trends. This distinction, which is presented in Sections 1.3 and 1.4, shapes the structure of the book; both represent the subject of macroeconomics and the
objective of macroeconomic policies. Equipped with these preliminary essentials, we can then examine in what sense macroeconomics is a science (Section 1.5) and
how it helps us to reason about the world around us (Section 1.6). The chapter concludes with a preview of the book’s contents (Section 1.7).
1.2.1 Income and Output

The most important indicator of a nation’s economic well-being is a measure of its output and income during a given year, which is called the gross domestic
product (GDP). The concept will be discussed in greater detail in Chapter 2. GDPs are very large numbers, and it is good to develop a feel for their magnitudes. In
2015, the GDP of the UK in 2015 amounted to about 1.89 trillion pounds; in Germany, 3.02 trillion euros; in the Netherlands 0.68 trillion euros; in Denmark 1.981 trillion
kroner.1 Obviously, these numbers are related to country size, which is why we often look at GDP per capita, which range from 25,000 to 40,000 euros in the same year
across the countries of Europe. We should be careful when comparing these data over time because GDPs can change for two reasons:(1) more is produced, and (2)
prices of those goods change. We will learn in Chapter 2 how to deal with this issue using so-called real GDP numbers—at constant prices , so to speak. Table 1.1,
which corrects for this using prices in the year 2000, displays the course of GDP over the past century.
A quick look at the table reveals that (1) most countries are characterized by steady growth of GDP over time, but that (2) there are significant and recurring
fluctuations of GDP around its trend, called business cycles. We return to the distinction between trends and cycles in separate sections later. Another important
observation is that small differences in average annual growth rates cumulate over time to huge differences in levels. A striking example: GDP per capita in Argentina
was about 15% larger than that of Sweden in 1900; 110 years later, it had fallen to half the latter’s value. The reason? While both grew over the period, Sweden’s
economy expanded at a 2.1% rate compared with Argentina’s 1.3%. 2 Similarly, it is striking to note how much poorer the Asian countries were in comparison to
Western Europe and the US in 1900. Some of them caught up (Japan), or started to catch up recently (China after the late 1970s) while others still have not
(Bangladesh).
A natural concern is whether GDP, or GDP per capita, is a good indicator of individual well-being. GDP says nothing about the distribution of income in the
economy, which may also matter to its residents. Many researchers are now examining issues related to happiness and quality of life, not just economic output. One of
the robust findings on research in this area is that while GDP is not everything, it is certainly one important determinant of economic and social well-being. Some
evidence is provided in Figure 1.1, which displays on the vertical axis the results of coordinated public opinion polls. People were asked how satisfied they were with
their lives. The horizontal axis corresponds to that year’s GDP per capita in each country. The figures show that deep unhappiness is only found in poor countries.
Yet, people in some countries seem to be poor and happy (e.g. Bhutan) while the French and the Germans seem less happy than the Danes and the Finns, who have
similar standards of living. A number of explanations are currently being explored, but this is an entirely different subject!
1.2.2 Unemployment
One important phenomenon associated with cyclical fluctuations is unemployment, the fact that people seeking paid work cannot or do not find it, sometimes even
when the economy is growing rapidly. The unemployment rate is the proportion of unemployed workers in the labour force. The labour force consists of those
who are either working or are actively looking for a job. In comparison with the total population, the labour force leaves out young people who are not yet working,
the old who are retired, and those who cannot or do not wish to work—or have given up hope of working.
Table 1.1 Real Income per Capita (GDP in euros, 2000 prices)
1900 1913 1929 1950 1987 1992 1999 2002 2007 2010 2015 Av. growth rate
Austria 2,462 2,961 3,160 3,167 13,085 14,937 17,145 17,920 20,289 20,280 20,824 1.9%
Belgium 3,188 3,606 4,319 4,667 13,280 15,078 17,010 17,987 20,069 20,071 20,104 1.6%
Denmark 2,578 3,343 4,337 5,933 15,401 16,192 19,017 19,717 21,414 20,346 20,552 1.8%
Finland 1,426 1,804 2,322 3,634 13,144 12,837 15,931 17,152 21,051 20,008 19,582 2.3%
France 2,457 2,978 4,025 4,504 14,144 15,774 17,549 18,610 19,040 18,562 18,904 1.8%
Germany 2,550 3,117 3,462 3,316 13,417 14,433 15,737 16,399 18,066 18,018 19,420 1.8%
Italy 1,526 2,191 2,643 2,992 12,771 14,216 15,612 16,339 17,229 16,329 15,330 2.0%
Netherlands 2,925 3,459 4,861 5,124 13,447 15,165 17,966 18,493 20,854 20,489 20,715 1.7%
Norway 1,604 2,091 2,895 4,639 15,521 16,715 21,019 22,093 24,399 24,096 24,714 2.4%
Sweden 2,188 2,646 3,306 5,759 14,483 14,509 16,962 18,144 21,688 21,192 22,285 2.1%
Switzerland 3,275 3,645 5,410 7,745 16,912 17,800 18,590 19,179 21,175 21,534 21,944 1.7%
UnitedKingdom 3,838 4,205 4,703 5,930 13,154 13,785 16,650 17,817 20,183 19,158 20,311 1.5%
Canada 2,488 3,800 4,328 6,231 15,678 15,511 18,347 19,654 21,619 21,292 22,446 1.9%
USA 3,496 4,529 5,895 8,170 18,618 19,908 23,669 24,383 27,052 26,348 28,065 1.8%
Argentina 2,355 3,245 3,732 4,261 6,237 6,406 7,443 6,083 8,803 9.310 10,069 1.4%
Bangladesh 417 443 445 461 515 574 708 778 960 1,088 1,396 1.1%
China 466 472 481 383 1,484 1,822 2,702 3,586 4,857 6,299 8,948 2.6%
India 512 575 622 529 961 1,146 1,568 1,719 2,402 2,847 3,639 1.7%
Japan 1,008 1,185 1,731 1,641 13,887 16,648 17,597 17,918 19,505 19,005 19,699 2.6%
Source: www.eco.rug.nl/~Maddison/and The Conference Board and Groningen Growth and Development Centre, Total Economy Database, January 2011,
http://www.conference-board.org/economics/The Maddison-Project, http://www.ggdc.net/maddison/maddison-project/home.htm, 2013 version. IMF, World Economic Outlook
Database April 2016.
There are many reasons to be concerned about unemployment. A first reaction to the image of idle workers is one of lost production and income. At the same time,
we need to ask whether the unemployed are able to find offers of work, whether they are turning down job offers, and if so, for what reason. Are the jobs that workers
are searching at all, or are workers really unavailable for jobs on offer, perhaps because their expectations are unreasonable? Or are they simply frustrated not to find a
job and have given up looking?
Unemployment is generally not a pleasant affair. Even with well-developed and efficient unemployment assistance programmes, long-term jobless workers can
experience emotional stress and their skills may deteriorate. Even if they are not measurable, the social and psychological costs of unemployment are high for the
affected individuals and for society as a whole. By that criterion, Europe has not done well over the last decades, as Figure 1.2 shows. The average rate of
unemployment has grown inexorably to reach double-digit numbers. In the USA, in contrast, unemployment has closely followed the business cycle, rising in periods
of slowdown, declining when growth returned. At the same time, not all European countries have shared this misery, as the case of Switzerland shows. Chapters 4 and
18 will help explain these different outcomes.
Fig. 1.1 GDP per Capita and Life Satisfaction in 2014

Money is not everything in life. While public opinion polls show a clear link between life satisfaction—a measure of happiness—and GDP per capita, that link is far from tight.
Sources: GDP per capita from World Economic Outlook, IMF; Index of satisfaction with life: http://en.wikipedia.org/wiki/Satisfaction_with_Life_Index
Fig. 1.2 Unemployment Rates in the Eurozone, Switzerland, and the USA, 1970–2015
The unemployment rate, measured as the proportion of workers who do not have a job but are looking for one, varies considerably across countries. In the USA, the unemployment
rate moves tightly with the business cycle. In those European countries that use the euro today, unemployment rose markedly in the 1970s and 1980s and stayed there for a long time.
In contrast, Switzerland avoided high unemployment for the entire period, but has also suffered a significant increase over the past two decades.
Sources: OECD
1.2.3 Factors of Production and Income Distribution

The output of an economy, its GDP, is by and large the result of work effort by men and women combined with equipment—‘machines’, but also buildings and other
structures. Labour and capital are the technical names given to the two main inputs, or factors of production.3 The distribution of total income between these
two factors of production is often a political matter, even if it is largely determined by economic forces. The employees are paid in the form of salaries, benefits, fees,
and bonuses. Governments take their share in the form of various taxes. What is left are profits, or the capital share of income. These profits go to the owners. In stock
markets, ownership of companies, or shares, are traded in open markets and valued on the basis of the firms’ profitability. Figure 1.3 shows the fraction of total income
in manufacturing that goes to labour, the labour share. It also plots the evolution of the stock market index over the same period, which tracks the value of shares
in companies traded on the stock exchange or bourse. An index is an indicator of the evolution of some phenomenon over time, designed to take a standardized value
(e.g. 1 or 100) on a particular date to highlight relative changes. The figure reveals a clear, but not perfect, inverse relationship between the labour share and average
stock prices. When the share of income going to labour is high, less is available for the firms’ owners, and stock prices tend to be lower. While it would be premature
to assert that one causes the other, it is certainly plausible that both are driven by common economic phenomena. In Chapter 8, we will see that depressed stock prices
may adversely affect the accumulation of productive equipment and, ultimately, the growth and size of the economic pie itself.
1.2.4 Inflation
The consumer price index measures the evolution of the cost of a basket of goods meant to represent average spending by consumers. Its per cent increase is the rate
of inflation. The inflation rate is usually stated in terms of percentage change per year, even when it is measured more frequently, such as every quarter or every
month. Most of the time, inflation is low or moderate at rates ranging from just above 0% to 4%. In the 1970s, many European countries experienced double-digit
inflation, with rates rising to 10%, 20%, or even more. In the 2010s, prices declined in a number of countries leading to negative inflation rates. In Latin America or in
the transition countries of Eastern Europe, inflation rates of several hundred per cent were quite common in the 1980s. When inflation is very high it is usually
measured on a monthly basis; the term hyperinflation describes situations when this monthly inflation rate exceeds 50%. A sign of exceptional economic distress,
hyperinflation has been observed in Central Europe in the early 1920s, in Latin America in the 1980s, and in many countries which emerged from the collapse of the
Soviet Union in the early 1990s.
Fig. 1.3 Labour Share of Income in Manufacturing and Stock Prices, Four Countries, 1950–2016
Labour and capital share the fruits of the economic activity. The labour share is the fraction of economic output which is paid to workers in wages and other forms of compensation.
The valuation of companies, reflected in stock prices, is negatively associated with the labour share. Real stock prices are computed as a share price index divided by the GDP
deflator, a measure of the price level.
Sources: Labour share: AMECO database, European Commission; Stock prices: Economic Outlook, OECD.
In normal times, inflation is related to the business cycle. Figure 1.4 shows how the rate of inflation changes when the rate of capacity utilization varies. The rate
of capacity utilization measures the degree to which companies are truly employing their available plant and equipment, and it serves as a good indicator of cyclical
conditions. The inflation rate is generally procyclical: it tends to rise when activity is high and declines in periods of slack. In contrast, the unemployment rate is
countercyclical: it moves against the cyclical behaviour of output, falling when output is growing rapidly and rising when output is growing more slowly or falling.
The behaviour of inflation is investigated in Parts III and IV of this book.
1.2.5 Financial Markets and the Real Economy

Financial markets play a central role in modern economies. Either literally or with the help of sophisticated communications technologies, they represent arenas where
buyers and sellers of financial assets such as bonds, stocks, currencies, and other financial instruments meet to trade. Together with banks and other financial
institutions, financial markets gather resources from households and firms in the form of savings and lend them out to others who will spend them. One specific
feature of these markets is the extreme day-to-day variability of prices at which financial instruments are traded.
Fig. 1.4 Capacity Utilization Rates and Changes in Inflation Rates, USA, 1967–2015
When measures of the utilization of capacity indicate a high level of activity in factories, the rate of inflation tends to rise, i.e. prices rise at an increasing rate. Conversely, low levels
of activity are accompanied by declining rates of inflation.
Sources: Bureau of Labor Statistics (CPI) and Federal Reserve Bank of Saint Louis (capacity utilization).
Physical investment, the accumulation of productive capital by firms, is intimately related to financial conditions. It is one channel through which financial markets
affect the real economy. The other channel is consumption spending by households. Stocks—shares in corporations—represent one form of private wealth. When
share prices rise, people feel richer and consume more. The real economy is contrasted with the financial or monetary economy: the former concerns the
production and consumption of goods and services, and the incomes associated with productive activities. The latter deals with trade in assets, i.e. monetary and
financial instruments. Chapter 11 brings the real and the monetary spheres of the economy together to understand how output and interest rates are determined from
year to year. Chapter 12 explores these short-run linkages in more detail when the economy is open to international trade in financial assets. Chapters 13 and 14 bring
together the issues of inflation, output, and exchange rate determination.
1.2.6 Openness
In the modern world, all countries engage in international trade by exporting and importing goods and services to and from each other. In addition, an increasing
number of countries are also connected through trade in financial assets. One measure of a country’s openness, or exposure to the various economic influences
coming from the rest of the world, is the ratio of the average of its exports and imports to its GDP. Table 1.2 shows that openness has considerably increased over the
past decades. This process of increasing trade and trade integration in goods, services, and financial assets over time is frequently called globalization. Smaller
countries tend to be more open than larger countries, and indeed the USA and Japan are fairly closed by international standards. This is also the case of the European
Union vis-à-vis the rest of the world, even though considerable trade integration has taken place among its member countries and with the rest of the world. Since
mid-2004, the European Union has expanded to 28 nations to become an economic region of more than 500 million inhabitants. These substantial enlargements have
further accentuated the importance of international trade, financial, and policy links between EU member countries. The planned departure of the UK (known as Brexit)
represents a historical reversal in this integration process.
Table 1.2 Openness (ratio of average of exports and imports to GDP, %)
1960 2015
Belgium 38.2 83.6
China 4.4 20.6
Denmark 33.3 50.1
Germany 19.0 43.0
Greece 12.5 30.0
Hungary — 85.0
Ireland 32.4 111.0
Netherlands 47.9 77.2
Poland — 48.0
Portugal 16.7 39.9
Spain 7.6 31.9
Sweden 23.2 43.1
Russian Federation — 25.4
Switzerland 27.3 57.3
Ukraine — 53.8
United Kingdom 20.9 28.4
United States 4.6 14.0
Japan 10.5 18.4
European Union 6.1 12.0

Sources: World Bank, OECD.
As a consequence, no country’s fate is truly independent of events that occur elsewhere, sometimes very far away. A good example is the financial crisis that
began in the summer of 2007 in the USA yet spread in the aftermath not only to Europe and Asia but also to remote parts of Africa and South America. Chapter 19
looks more closely at these problems and potential solutions.
Figure 1.5(a) displays estimates of GDP for France, Germany, and the UK since 1870. A positive long-run general tendency, or trend, clearly dominates shorter-run
fluctuations. The trend rate of growth has been fairly stable, perhaps with a slight increase after the Second World War. Another way of seeing this is to plot the
natural logarithm of GDP against time, as in panel (b) of Figure 1.5. With this so-called logarithmic scale the slope of the curve is a direct measure of the annual
growth rate: a constant growth rate would yield a straight line.4 In the long run, on average, we seem very close to a rather robust and steady trend.
Fig. 1.5 Gross Domestic Product (GDP), France, Germany, and the UK, 1870–2015
National output and income, as captured by the gross domestic product, exhibits a robust growth trend. Growth tends to be exponential; that is, annual percentage increases are
reasonably stable in the long run. This does not preclude significant year-to-year variations (panel a). When the data are displayed on a logarithmic scale instead (panel b), the slope
of the curve measures the annual rate of growth.
Sources: 1870–1949: Maddison (1991) (http://www.ggdc.net/maddison/oriindex.htm); 1950–2015: Conference Board’s Total Economy Database (https://www.conference-
board.org/data/economydatabase/index.cfm?id =27762).
This trend growth in total output implies remarkable increases in living standards. Let’s take another look at Table 1.1. Note that per capita, or average, income
increased to more than sixfold in Belgium since 1900, ninefold in Sweden, and 18-fold in Japan. Yet the growth phenomenon is not shared by all countries at all times.
In Bangladesh, real income per capita rose by only 120% over the same period. Some countries have faced serious setbacks, such as wars and famines, while others
have expanded rapidly. Some, like China and India, stagnated for many decades before suddenly entering a period of rapid increase in living standards. China poses a
particularly interesting case because its explosive take-off was so recent, and because it was the world’s most advanced economy 700 years ago.
Because of these momentous increases in standards of living, economic growth is one of the most exciting issues in macroeconomics. Chapter 3 explores in
detail the reasons why economies grow. One reason is an increase in population, since more people can work and produce more output. Another is the accumulation
of means of production: plant and equipment, roads, communication networks, and other forms of infrastructure make workers more productive. Most important is the
development and harnessing of knowledge and technology to economic ends. The sharp acceleration of scientific discoveries towards the end of the eighteenth
century is thought to have triggered the industrial revolution, and some believe we are now witnessing the onset of a new wave of advances related to information
and telecommunications technology.
While output and income have increased by staggering amounts over many decades, growth is not constant or even steady. Figure 1.5 shows that real output tends
to fluctuate around its trend. This is even more apparent in Figure 1.6, which plots the quarterly rate of change in GDP for the UK. Quarterly data tend to accentuate
the relative importance of short-run fluctuations. These sustained periods of ups and downs are called business cycles. One important challenge of macroeconomics
is to explain such deviations of GDP from its underlying trend, referred to as the output gap. Why do these fluctuations occur and persist over periods ranging from
three to 10 years, and what can be done, if anything, to avoid the disruptions that are associated with them? This is the common theme of Parts III and IV of this book.
While business cycles are hardly identical across countries and across time, they have a number of common features. These features are represented in Burns–
Mitchell diagrams. The idea is simple. Imagine cutting up a curve like those depicted in Figures 1.5 or 1.6 into a set of cycles and superposing all these cycles on
top of each other. To do this, it is necessary to identify calendar dates for cyclical turning points of output (GDP)—here we will be concerned with peaks, but the
procedure can be used for troughs as well. Having identified those cyclical peaks, simple numerical averages of other macroeconomic variables of interest can be
calculated around the calendar date of the output’s peak. The behaviour of those variables around the turning points of output tells us something about whether the
variables are leading—meaning that they can help forecast future turning points in output. More importantly, they can help inform the formulation of theories which
will help us better understand how business cycles arise and develop over time. Box 1.1 provides more details on the Burns–Mitchell methodology.
Fig. 1.6 Quarterly Gross Domestic Product, UK, 1956:1–2015:3

With growth rates, fluctuations of economic activity become more apparent.
Sources: Office of National Statistics.
Box 1.1 Burns–Mitchell Diagrams, Now and Then
The fluctuations of economic activity in the rapidly industrializing economies of Europe and the USA attracted much interest in the early part of the twentieth
century. In the 1920s and 1930s, the National Bureau of Economic Research (NBER) in New York was a centre of such research, associated with Gottfried
Haberler, Simon Kuznets, Wassily Leontief, Allyn Young, and many other economic luminaries of the day. A common view of these researchers was that the
emergence of powerful statistical methods made it possible to study economic phenomena in general, and the business cycle in particular, in a more
scientific fashion.
Two NBER researchers, Arthur Burns and Wesley Mitchell, were somewhat sceptical of the statistical approach but committed to a data-driven, descriptive
assessment of business cycle regularities. Mitchell had already written a book in 1927 that more or less laid out the research programme, but together with
Burns the project ultimately came to fruition after the Second World War. The idea was to reduce the time series of data to a sequence of cycles and then study
the average behaviour of other important variables a number of periods before and after the peak of the average cycle. This highly data-intensive empirical
approach was considered modern and useful, even if it was criticized in some quarters as ‘measurement without theory’.
The identification of business cycles is always a tricky procedure with some element of arbitrariness. We employ a particular method which has gained
acceptance in recent years and apply it to eight advanced economies over about 35 years of quarterly data.5 The result is a reference cycle which appears in
the first panel of Figure 1.7. The cycles identified by this procedure are used to produce the other panels of Figure 1.7, which give the behaviour of variables 10
quarters before and 10 quarters after the cyclical peak, usually stated in proportion to the average value over that period. In Figure 1.8, we present the original
findings of Burns and Mitchell, who worked on primarily monthly US data from the late nineteenth and especially the early twentieth century.
1.5.1 The Genesis of Macroeconomics

Why do we observe cyclical fluctuations—ups and downs—in the level of GDP around its trend? Why is unemployment generally countercyclical, while changes in
inflation appear procyclical? For a long time, economists paid little attention to such phenomena. In fact, it was believed that properly functioning markets would
deliver the best possible collective outcome, to a good approximation at least, and that there was no point in looking into aggregate behaviour. This principle was
called ‘laissez-faire’. Opponents of laissez-faire endorsed some form of interventionism—government support for particular markets and industries, including
subsidies and protection from foreign competition.
This does not mean that business cycles were ignored completely. In fact, cycles of varying lengths were identified and studied, ranging from inventory cycles of one
or two years’ duration to long-wave cycles lasting half a century. Box 1.2 provides details on these cyclical movements. Such cycles were seen as the cumulative
outcomes of disturbances such as discoveries, inventions, exceptionally good or bad crops, wrong bets by firms on goods that customers want to buy, or even
changing tastes of consumers at home and abroad. Inflation was seen as the consequence of rapidly growing money stocks, first because of gold discoveries in the
nineteenth century, afterwards because of reckless paper money creation by central banks. As will be seen in Chapter 16, much of this wisdom remains valid today.
Yet the Great Depression of the 1930s, which spread throughout the world and sent millions into unemployment and misery, seemed too severe to be simply bad luck.
Reflecting upon the Great Depression in 1936, British economist John Maynard Keynes published The General Theory of Employment, Interest and Money, a book
that is often said to have started the field of macroeconomics. Keynes stressed the role of aggregate demand in macroeconomic fluctuations. His followers later
persuaded policy-makers to engage in aggregate demand management, that is, to manipulate government demand in order to smooth out fluctuations, mainly to
avoid protracted recessions.
Fig. 1.7 Burns–Mitchell, Now: Eight Countries, 1970–2015

These six figures display the average behaviour of variables around cyclical peaks, where the cycle is measured using a procedure described in the text. The vertical line around
zero shows the quarter in which real GDP reaches its peak (upper left panel). The peak is followed on average by a sharp drop in the growth rate of output (upper right panel).
Unemployment is countercyclical, rising most after the peak in GDP has been passed, but is rising across cycles, indicating that average unemployment rates in the sample were
increasing over time. The inflation rate is procyclical but lagging, peaking in the quarter after output. Short-term interest rates and employment are strongly procyclical.
Sources: OECD; authors’ calculations.
Fig. 1.8 Burns–Mitchell, Back Then: US Monthly Data, 1914–1938
These original figures document the work of Burns and Mitchell on US monthly pre-World War II data for industrial and agricultural production as well as durable and non-durable
goods, interest rates, employment, payrolls, and food prices (GDP data did not exist at the time these diagrams were constructed). As in Figure 1.7, the cyclical peak is identified using
a well-defined procedure, and averages of the cycles around that peak (denoted ‘P’) to trough (denoted ‘T’). Production, employment, wages, wholesale prices, and interest rates are
procyclical and coincident, while retail prices seem to lag slightly. The procyclical behaviour of interest rates became significantly less pronounced after 1914, which was the year
after the US central bank, the Federal Reserve System, was founded.
Sources: Mitchell (1951).
An evaluation of the success of demand management policies is presented in Chapters 16 and 17. Since the Second World War, the amplitude of the business cycle
appears to have diminished considerably, as can be seen in Figure 1.5. Of course, the global recession of 2008–2009, sometimes called the ‘Great Recession’, is a major
setback. For centuries, earlier generations assumed that favourable periods of growth were inevitably followed by periods of declining activity. Following Keynes,
concern initially centred on fluctuating demand. The attention then turned toward the supply side—meaning the productive capacity of an economy—and the
efficient utilization of labour and capital resources. This applies especially to unemployment, which remains a big problem in much of Europe. These topics are the
subject of Chapter 18.
Another remarkable change in the behaviour of the post-war economy concerns the general price level, or the cost of goods in terms of money. Until the First
World War, prices were as likely to rise as they were to fall, as can be seen from Figure 1.9. Apart from war periods, the price level was trendless; over long periods of
20 to 50 years, the consumer price index—a measure of the average price level—was remarkably stable. One interpretation of the post-war era—a controversial one, as
we shall see—is that macroeconomics has led to more steady output growth at the cost of inflation. In the mid-1980s, concern with high inflation triggered a change of
heart. In particular, most central banks have toned down demand management policies and refocused their energy on keeping inflation low. The crisis that started in
2007 is now forcing central banks to recognize that they have a broader duty than price stability; in particular, they have to be concerned with financial stability and
the health of the banking system.
1.5.2 Macroeconomics and Microeconomics

The macroeconomy is just the sum of hundreds or thousands of markets, each of which is explained by microeconomic principles. Microeconomics is devoted to the
study of prices of individual goods and of the markets where these goods are produced and sold. Why do we need two separate disciplines? To a great extent they
are linked. Microeconomics is dedicated to the analysis of market behaviour of individuals. Macroeconomics is concerned with collective behaviour, the outcome of
individual decisions taken without full knowledge of what others do. Thus macroeconomics should be built on microeconomic principles. This is how the field has
evolved over the past three decades. However, seeking to bring together all markets into one framework is a daunting challenge. In this book, we go half the way. We
recall the microeconomic principles that are needed to understand decisions like consumption and productive investment, as well as the overriding importance of
budget constraints. Yet, we do not insist on grounding the overall picture in these principles. We do so for two reasons.
Box 1.2 All Kinds of Cycles
Business cycles, like comets, bear the names of their discoverers. Simon Kuznets (1901–1985) was a Russian-born US economist who received a Nobel
Prize for his work on growth. Russian economist Nikolai Kondratieff (1892–1938) developed his theory of long-wave cycles in the 1920s before he was
arrested and disappeared; the official Soviet Encyclopaedia then wrote about his work: ‘this theory is wrong and reactionary’. It was also in the 1920s that
Joseph Kitchin (1861–1932), a South African statistician and gold trader, uncovered his own more rapid cycles of 2–4 years’ periodicity, which are associated
with inventory movements, bank clearings, and wholesale prices. Clement Juglar (1819–1905), a nineteenth-century French physician, first studied cycles in
human births, deaths, and marriages before turning his skills to interest rates and credit conditions. These Juglar cycles—which involve fluctuations of
investment spending, GNP, inflation, and unemployment—are perhaps the closest thing to the business cycle that we will study in this book.
Interestingly, one of the most robust and regular cycles in economic activity is the cycle that coincides with the seasons of the year. Movements of output in
agriculture, manufacturing, construction, and tourism have obvious seasonal components which sometimes swamp business cycle fluctuations in magnitude,
as do patterns in overall output associated with bank holidays, summer and winter weather, and harvests.
Fig. 1.9 Price Levels and Inflation Rates, France and the UK, 1870–2015
Until the outbreak of the First World War, the price level was stable, and inflation was close to zero on average. Since the Second World War, the price level has risen secularly,
average inflation has been positive, high in the late 1970s and much of the 1980s, declining over the 1990s and becoming very low, close to zero, in the 2010s.
Sources: Maddison (1991); OECD.
First, bringing together microeconomic principles leads to considerable complexity. We believe that simplicity and clarity are an essential ingredient of a textbook.
Of course, simplicity should not come at the expense of accuracy. It turns out that, at this stage of the development of the discipline, the complexity of the effort has
led to a large number of arbitrary simplifications that do not ensure accuracy. Second, in creating macroeconomics, Keynes stressed the notion of coordination
failures, which arise in decentralized markets as illustrated in the following example. A consumer wants to purchase a car, but her income is insufficient for her to do
so. A car manufacturer could actually hire her to build cars, and with her salary she would then be able to buy one. That one sale, however, would not suffice to pay
her salary, so other buyers would need to be found. In order to generate sufficient demand for her employment, several other individuals would need to be hired,
perhaps in different industries. For this scheme to work, a considerable amount of coordination among producers and consumers would be required.
The laissez-faire principle is that prices and markets automatically and perfectly perform this coordinating role. Keynes’ critique of markets was that sometimes they
fail to produce the desired result as quickly as we would like to see. There may be many consumers wishing to buy goods and willing to work to produce them, and
many firms that would benefit from hiring them if only they could be persuaded that their sales would increase. But this potential may not be realized and we have
both recession (fewer sales) and unemployment (fewer jobs). Even if market forces tend to correct this imbalance—which they eventually do—the period of time
necessary may be long enough to involve significant social costs. Macroeconomics started with the idea that prices and markets do not continuously resolve all the
coordination requirements of a modern economy.
1.5.3 Demand and Supply
In its most concentrated form, macroeconomics boils down to separating events into two categories: (1) those that affect the demand for goods and services, and (2)
those that affect the supply of those goods and services. The demand side relates to spending decisions by economic agents—households, firms, and
government agencies—both at home and abroad. The principle of aggregate demand management policies is that the government can take actions to offset or smooth
out those of private agents—firms and households—in order to dampen or eliminate fluctuations in total spending. The idea is to take the edge off recessions as well
as booms. Two traditional demand management instruments are fiscal and monetary policy. Fiscal policy manipulates government expenditures or taxes in an
attempt to affect the volume of national spending. This subject is studied in detail in Chapter 17. Monetary policy is directed at influencing interest and exchange
rates, and more generally conditions in financial markets. This in turn affects spending on goods and services. Chapters 9 and 10 provides an in-depth analysis of
money and monetary policy.
The supply side relates to the productive potential of the economy. The choice of hours worked by households, the productivity of their labour, and in general the
efficiency with which resources are allocated in generating output, all affect an economy’s aggregate supply. Accordingly, supply-side policies represent
government’s effort to increase an economy’s long-run capacity as well as its overall efficiency. Frequently, this effort is about reducing or eliminating government-
induced inefficiencies, which were introduced before the importance of the supply side was understood, or as the result of successful lobbying by interest groups. It
is also about bringing idle or underutilized resources into productive uses. Unemployment policy—designed to fight the scourge of market economies—occupies a
key role in the supply side. Chapter 18 explores these issues and shows how the government can improve or worsen the economic climate.
1.6.1 What is to be Explained?

Macroeconomics is concerned with aggregate activity, the level of unemployment, interest rates, inflation, wages, the exchange rate, and the trade in goods, services
and assets with other countries. Before beginning to think about these questions, it is essential to be clear about what we want to explain and what we take as given,
or outside the realm of analysis. The variables to be explained using economic principles are called endogenous variables. The other variables—those we do not try
to explain—are called exogenous variables. This distinction is represented in Figure 1.10. Examples of variables often considered exogenous are policy instruments
(the tools of fiscal and monetary policies), economic conditions abroad (foreign levels of activity and interest rates), the price of oil, and sometimes even domestic
social conditions such as business optimism or trade union militancy.
The distinction between endogenous and exogenous variables is necessarily arbitrary. Many exogenous variables are not strictly independent of the endogenous
variables. In fact, the book progresses in steps. Many variables initially are considered exogenous and then progressively made endogenous, or endogenized. For
example, fiscal and monetary policy decisions are often responses to the course of inflation or unemployment. While it is convenient to regard policy variables as
exogenous, it is sometimes interesting and useful to understand how these exogenous variables themselves are determined. For example, we will often treat monetary
policy as a systematic response to economic conditions.
Fig. 1.10 Endogenous and exogenous Variables

Endogenous variables are the object of analysis in an economic model. Exogenous variables are determined outside the economic model. The weather and political decisions are
examples of variables generally considered to be exogenous.
1.6.2 Theory and Realism

Macroeconomics proceeds by making simplifying assumptions. We never literally believe in our assumptions, but we need them in order to see through the vast
complexity of an economy. This is why the distinction between endogenous and exogenous variables is artificial. Truly exogenous variables are rare. Earthquakes are
truly exogenous but what about climatic conditions or scientific discoveries and inventions? The task of systematically linking the behaviour of endogenous variables
to changes in exogenous variables is accomplished by specifying relationships between all the variables of interest. You might say that economics—and in particular
macroeconomics—is in the business of establishing relationships involving causality.
All these relationships, when brought together, constitute a theory. Almost by definition, in social sciences, theory must be an abstraction, an intentional departure
from realism. If the real world could be understood without simplifying assumptions, theories would be unnecessary. The problem is not with economics, but rather
with the world’s inherent complexity. Karl Marx, who was no friend of conventional political economy, seemed to hit it on the head:
The body in its entirety is easier to study than are the cells of that body. In the analysis of economic forms, moreover, neither microscopes nor chemical reagents are of use. The
force of abstraction must replace both. ((1867) Foreword to Volume I, Das Kapital)
Progress is made by weeding out those assumptions and theories that lead us to false conclusions. As time passes, some theories prove to be unfounded, while
others gain acceptability. This process is long and complex, and far from complete. Because macroeconomics is a young discipline, a number of controversies
continue, and this aspect is discussed in Section 1.7.
1.6.3 Positive and Normative Analysis

Macroeconomic analysis and policy are closely linked. Because a number of exogenous variables are under the control of government, it makes sense to ask what is
good and what is bad policy. At its best, macroeconomics can explain the economy. For example, it can link particular events to exogenous events or policy decisions.
This is positive economics: it refrains from value judgements. Normative economics takes a further step and passes judgement or makes policy
recommendations. In so doing, it must specify what criteria are used in arriving at particular conclusions. This inevitably implies a value judgement. Economists
generally like to make policy recommendations. As long as they are truthful about their own preferences and reveal their criteria, this is part of their professional
activity. In this textbook, we will generally refrain from normative economic analysis. 6 At the same time, we believe and hope that many readers will make use of their
newly acquired knowledge to indulge in the normative side of macroeconomics: this is what makes it fun.
1.6.4 Testing Theories: The Role of Data

The generally accepted way of evaluating theories is to subject them to scientific testing. In macroeconomics, this means looking at the facts, i.e. at data. This is easier
said than done, and there are a number of unusual difficulties. First, data often correspond to elusive concepts, as Chapter 2 will show. Second, constructing
aggregate data implies enquiring into the behaviour of millions of individuals, who sometimes have good or bad reasons to misrepresent the truth. Third, economics
shares a predicament common to other social sciences: experimentation is not really possible—when observed, people often change their behaviour. Not only is it
possibly immoral—no macroeconomist would wish to start a hyperinflation just to test a theory—but more crucially, many important variables simply are not
observable. This is the case of people’s expectations of the future, for example. Macroeconomists are forced to conduct empirical tests with the data that they have.
They develop statistical techniques, often sophisticated ones, to deal with observation and measurement errors. They refine their techniques for gathering and
analysing data. This allows the elimination of inadequate theories and the modification of others. The surviving theories will be those that withstand the test of time in
this scientific process.
1.6.5 Macroeconomic Modelling and Forecasting

Economists are frequently asked to make forecasts. Governments, international organizations, and large financial institutions employ large teams of economists to
prepare them. If macroeconomics were to be judged by the performance of forecasts, the verdict would not be unkind. However, the respectable track record of
forecasters has been sullied by some large historical errors. Box 1.3 illustrates this fact by examining the accuracy of forecasts, after the fact, for the disastrous crisis
years 2009 and 2010.
There are several reasons why economic forecasting is inherently difficult. First, even an excellent understanding of an economy’s structure—how its endogenous
variables interact—can fall victim to unexpected changes in exogenous variables. Good examples of this are the oil price increases of 1973 and 2000, their fall in 2015–
2016, the high-tech crash of 2001 and, of course, the latest crisis. Second, expectations—which are volatile in nature—wield an important influence over the economy.
Governments sometimes react to their own forecasts by implementing policies designed to prevent those forecasts from happening. Political changes occur quickly
and can disrupt the economic environment. Finally, it takes time—often several months—to know what has really happened at any given point, so forecasts are
always based on provisional information which becomes more precise only with time.
Box 1.3 Forecasting the Crisis Years
Economic forecasts can be wrong, and often spectacularly so. Table 1.3 presents a few examples about the crisis years. The collapse of the investment bank
Lehman Brothers in September 2008 triggered banking crises in the US and Europe. The result was a deep, unforeseeable recession in 2009. Determined
action by governments and central banks often limited the recession to just one year. The table shows how forecasts of GDP growth have changed over time.
We use the forecasts published every six months by the Organisation for Economic Co-operation and Development (OECD), an organization of industrialized
countries. These forecasts usually reflect the consensus of professional public and private forecasters. The eventual outcome is also shown.
The Year 2009
The financial crisis began in mid-2007, but was hardly noticed and downplayed by most analysts at the time. This explains why the initial GDP growth
forecasts, produced in December 2007, were fairly optimistic. Six months later, in June 2008, concern started to grow for the US and Europe, but it was
perceived that the rest of the world (here Japan and Korea) would not be much affected. The December 2008 forecast, after the Lehman collapse, became
alarmist but, with hindsight, still much too optimistic.
The Year 2010
The initial forecasts, in December 2008, correctly expected that the recession would only last one year; they projected relatively modest but positive growth
rates for 2010. However, as 2009 turned out much worse than expected, in mid-2009 forecasters became overly pessimistic and revised downward their
numbers for 2010. The last forecasts, from December 2009, show that pessimism had declined, yet they remained too pessimistic.
Most forecasts are generated by computer-based models. These models resemble those that we present in this book. They consist of hundreds, sometimes
thousands, of equations. Constructing these equations is a long and difficult task. The exogenous variables must be guessed by forecasters before they can ask their
computers for an answer. This introduces many margins of error. The models can never be fully reliable, and the exogenous variables may be difficult to pinpoint. For
these reasons, the forecasters themselves take their results with a grain of salt, and often, when the outcome is not completely satisfactory, ‘drop in’ their own
subjective factor to the results.
Table 1.3 GDP Growth Forecasts, 2009 and 2010 (% per annum)
2009 France Germany Japan Korea UK USA

Forecast (Dec 2007) 2.0 1.6 1.8 5.1 2.4 2.2
Forecast (June 2008) 1.5 1.1 1.5 5.0 1.4 1.1
Forecast (Dec 2008) -0.4 -0.8 -0.1 2.7 -1.1 -0.9
Actual Outcome -2.7 -4.7 -6.3 0.3 -4.9 -2.6
2010 France Germany Japan Korea UK USA
Forecast (Dec 2008) 1.5 1.2 0.6 4.2 0.9 1.6
Forecast (June 2009) 0.2 0.2 0.7 3.5 0.0 0.9
Forecast (Dec 2009) 1.4 1.4 1.8 4.4 1.2 2.5
Actual Outcome 1.4 3.5 4.0 6.2 1.3 2.9
Source: OECD, Economic Outlook.
1.7.1 Structure
The book proceeds in steps. Parts I–III build up an understanding of the measurement and the behaviour of the underlying economy. Part I is concerned mostly with
defining terms and constructing a macroeconomic vocabulary. Part II studies the behaviour of the economy in the long run: growth and output, labour markets, and
prices and exchange rates. Part III develops our understanding of macroeconomy in the short run, that is, from quarter to quarter or from year to year. This part spans
many subjects, ranging from the demand of households and firms for goods and services, the financial system, the short-run determination of output, interest rates,
and the exchange rate. It also extends the analysis to include the analysis of inflation, output, asset prices, and exchange rates over a longer horizon. It introduces a
framework for thinking about inflation and the business cycle. Part IV then uses this framework to explore policy issues facing governments: demand management,
fiscal policy, and macroeconomic policies designed to enhance long-run performance. The book concludes with an extensive look at the world international financial
system, and the current state of economic thought.
1.7.2 Controversies and Consensus

Economists often make a bad name for themselves by quarrelling in public. Visible disagreements among economists frequently have to do with finer points, if not
outright hair-splitting. This discourse is intellectually healthy, but misleading to outside observers, whose opinions are often based on accounts in the popular press
and who are more apt to recall sensational talk-show appearances rather than sober analysis of theory and data. It is unfortunate that many disagreements have
important policy implications. Perhaps as a result, politicians often see economics as a sort of debating event, with economists acting as advocates for one particular
ideology or another, and may even abuse economists’ opinions to get a stamp of approval for a particular policy.
In this textbook we do not shy away from presenting some of the most important disagreements among economists, leaving the reader free to judge. Yet we do not
dwell upon these controversies either, choosing to focus rather on the common ground. Because there is so much that is not controversial, it is best first to
understand the broad areas of consensus. Box 1.4 provides more details.
1.7.3 Rigour and Intuition

The only possible scientific approach to the complexities of the real world is to employ the rigour of reasoning. However, to be useful, macroeconomics must be
versatile and easily put to work when we want to understand particular events. This is why a great deal of macroeconomics amounts to the organized accumulation of
intuition about particular phenomena. Our objective is, therefore, to leave readers with a natural understanding of how the economy functions. We do this by trying to
draw robust yet simple conclusions from the various and often intricate principles presented. Such intuition is never completely rigorous, but can be useful in practice.
Rigour plays a crucial role in telling us when our intuition is correct, and when it is leading us astray.
1.7.4 Data and Institutions

Macroeconomics is fascinating because it tells us a great deal about the world in which we live. It is not merely a set of abstract principles with interesting logical
properties. Some theories will look odd at first sight, yet they capture key aspects of the real world. This is why at each important step we pause to look at facts. Facts
can be data or particular episodes. Studying them carefully shows how theories work and shape our understanding of macroeconomic phenomena. It broadens our
knowledge of important events that have shaped the lives of millions of people.
Box 1.4 Macroeconomic Schools of Thought: A Primer
Almost from its beginning, macroeconomics has been divided into two main schools of thought. Keynesians (and their neo-Keynesian heirs) and monetarists
(and neo-monetarists) continue to pursue the old debate about the role of the marketplace and government in society. Keynesians are more likely to believe
that markets function imperfectly and that governments can and should use economic policy actively to combat recessions. Monetarists7 tend to reject this
view, seeing politics and the power of bureaucracies as barriers to government efforts to steer the economy away from business cycles and more generally
market failures, which they see as either unavoidable or of lesser importance. Given these premises, each school uses theories and data to build and support
its case.
These labels are not exclusive. In the USA reference is sometimes made to saltwater versus freshwater macroeconomists. Saltwater economists come
from universities located on the two US seaboards (Harvard, MIT, Yale, Stanford, Berkeley) and tend to defend the Keynesian legacy. Freshwater economists
are more frequently associated with monetarism and laissez-faire; they hail from universities located near the Great Lakes, e.g. Chicago, Rochester, or
Minnesota. In Europe, similar controversies characterize national, and increasingly European, debates. National traditions tend to make British and French
economists more Keynesian, while German or Swedish economists are more monetarist. Dutch, Italian, and Spanish economists are hard to classify, having
as many exceptions as examples for any rule. In recent years, the differences have tended to diminish, both among researchers and policy-makers and a
synthesis was in sight. When the Great Recession came, a predictable new polarization emerged of views on theory and policy, as many economists mixed
normative and positive aspects in their public pronouncements.
On the other hand, a graph or a table is no substitute for more rigorous analysis of the data. Merely demonstrating that two economic variables move closely
together is a far cry from proving that one causes the other. Our motive in using data to illustrate economic phenomena is to give readers a feel for economics itself.
On our website we offer a list of suggested reading for each chapter—which is by no means meant to be exhaustive—for those who want to learn more about the
theory and practice of macroeconomics.
Finally, good economic theories must be valid under different conditions. At the same time, the response of different countries to economic shocks is often shaped
by their particular economic and political institutions. These include their form of government, the roles of labour unions and employers’ associations, and
regulations. The interplay of macroeconomic principles and institutions is an essential part of a proper understanding of the field, and this is why we spend a lot of
time reviewing them. The economics of these institutions is, however, far beyond the level of this textbook.
1.7.5 Europe
Our textbook bears the subtitle ‘A European Text’. Does this mean that we think that macroeconomics in Europe is fundamentally different from macroeconomics in
the USA, Asia, or Latin America? Most certainly not! To the contrary, we take the view that macroeconomics is sufficiently global in scope to apply to economies
around the world. On the other hand, we do wish to send a more subtle signal: we believe strongly that European economies have important distinguishing features
that make them hard to study through the lens of, say, the leading textbooks from North America.
There is much in Europe that warrants such a European emphasis. Rather than a collection of states under a federal government, Europe is a mosaic of nation-
states, each with a sovereign macroeconomic policy-maker, but also with distinct preferences and endowments. The completion of the Single European Market, the
creation of a monetary union, and the significant enlargement of the European Union increase the pressure towards integration, raising specific new challenges along
the way. To varying degrees, European countries also share a common view of the relationship between market forces and social justice. The attachment to fairness
and economic solidarity is deeply ingrained in Europe’s traditions and history, which explain why her labour markets differ so much from those in the USA. This
observation alone warrants a markedly different perspective, even if the underlying theory is the same.
Key Concepts
exchange rate
macroeconomics
business cycle
trend
gross domestic product (GDP)
unemployment, unemployment rate
labour force
labour
capital
factors of production
labour share
index
inflation
hyperinflation
capacity utilization
procyclical and countercyclical
real economy, monetary economy
openness
globalization
logarithmic scale
economic growth
output gap
Burns–Mitchell diagram
laissez-faire versus interventionism
aggregate demand management
supply side
price level
coordination failures
demand side
economic agents
fiscal policy
monetary policy
endogenous
exogenous
positive and normative economics
Media
Students can greatly benefit from reading daily the economic section of their newspaper. Some publications with high-quality analyses (but not free of prejudices) are
(in English): the Financial Times and The Economist. There is also a wealth of information on the internet, but don’t believe everything you read in Wikipedia!
Data are produced by national statistical institutes and central banks. Some international institutions produce comparable data and are of easy access: the IMF’s
International Financial Statistics and its biannual survey World Economic Outlook, the OECD’s biannual Economic Outlook, the World Bank’s World Development
Report and Global Economic Prospects, the European Commission’s European Economy, and the European Bank for Reconstruction and Development’s annual
Transition Report. All maintain websites, more or less generous in allowing access to their data and publications.
1 A trillion (1 000 000 000 000) (US usage) is 1 followed by 12 zeros (1012).
2 A useful rule of thumb is that a country growing at rate g% per annum needs about 70/g years to double in size.
3 Land, energy, intellectual property, and many other inputs also matter, but are quantitatively less important in macroeconomics and will be ignored to make matters simpler.
4 For mathematically inclined readers, if x(t ), the value of x at time t , grows at constant rate g, defined as (1/x)dx/dt , then x(t ) = x(0)egt and ln x(t ) = ln x(0) + gt .
5 The method is due to Harding and Pagan (2001) and can be summarized as follows. For a four-quarter moving average of the original series, define a peak as the quarter for which GDP is
higher than the two preceding and the two successive quarters. In case two or more consecutive peaks are found, only the highest is retained. Next examine the highest and lowest points in a
neighbourhood of the peaks. Now repeat the procedure using an unweighted short-term moving average of the original series. In the neighbourhood of these intermediate turning points,
troughs and peaks are determined in the unsmoothed time series. If these pass a set of additional restrictions on the magnitude of the fluctuation, they are selected as the final cyclical
turning points.
6 Many are motivated by ‘social conscience’ to study economics. Much like idealistic health professionals who want to cure the sick, economists can also be interested in ways to provide
relief to the disadvantaged and suffering.
7 The term ‘monetarist’ derives from the Latin ‘moneta’ signalling their original emphasis on excess money growth as the sole cause of inflation. Now the term is sometimes applied to
those who advocate unregulated markets and criticize government intervention at both the microeconomic and the macroeconomic level.
Macroeconomic
Accounts 2
2.1 Overview
2.2.1 Three Definitions of Gross Domestic Product
2.2.2 Real versus Nominal Quantities, Deflators versus Price Indices
2.2.3 Measuring and Interpreting GDP
2.3.1 The Circular Flow Diagram
2.3.2 Summary of the Flow Diagram
2.3.3 More Detail
2.3.4 A Key Accounting Identity
2.3.5 Identities versus Economics
2.4.1 The Current Account and its Components
2.4.2 The Capital Account
2.4.3 Net Borrowing or Lending
2.4.4 The Financial Account and its Components
2.4.5 Errors and Omissions
2.4.6 The Meaning of the Accounts
Summary
Facts and theories meet in analysis. The combination of the two is essential if economics is to progress, since it is neither a pure subject, like mathematics, of which one does not ask
that the theories should be applicable to actual phenomena, nor is it a collection of facts, like the objects on a junk heap, of which one does not ask how they are related.
Richard Stone1
2.1 Overview
Every science has its own special language; not necessarily to exclude non-experts, but to make discussion more meaningful and precise. In this chapter we will start
by learning the language of macroeconomics. We first provide a quantitative description of the economy and definitions of more frequently used concepts. As a
natural point of departure, the chapter begins with a discussion of the national income accounts and accounting identities—how magnitudes we are interested in
relate to each other, by construction. The national income accounts play a central role throughout the study of macroeconomics.
Knowing the facts and how these facts are measured are essential for our understanding of the macroeconomy. Chapter 2 presents the most important indicators
of an economy’s health that we have: the national income and product accounts and the international trade and financial accounts. Moreover, the distinction
between description (this chapter) and analysis (the rest of the book) is similar to that found in biology. Describing living organisms as a collection of different cells
is only a first step. The second step is to analyse and understand how cells function and affect each other. In a similar way, decomposing the gross domestic
product into its components and examining the external accounts describe interactions and relationships without explaining how or why. That is the job of
subsequent analytic chapters. It is thus unavoidable to spend time with these definitions. As we shall frequently see, there is much more to them than first meets the
eye.
2.2.1 Three Definitions of Gross Domestic Product

The gross domestic product (GDP) is a measure of productive activity. It is defined for a particular geographic area—usually a country, but possibly a region or a
city, or a group of countries such as the European Union (EU) or the euro area. It is also defined over a time interval, usually a year or a quarter. This is because the
GDP is a flow variable, much like the amount of water flowing down a river. Flow variables differ from stock variables, which are always defined with reference
to a particular point in time, such as the quantity of water held back by a dam.2
It turns out that there are three ways to measure GDP, and a different definition of GDP corresponds to each measurement. Our first definition of GDP is the sum of
all final sales of goods and services sold during the measurement period.
Definition 1
GDP = the sum of final sales within a geographic location during a period of time, usually a year.
This definition refers specifically to final sales, i.e. goods and services sold to the consumer or firm that will ultimately use them. For example, the
purchase of a loaf of bread or a motor car by a household is a final sale. In contrast, a car sold to a dealer which is subsequently resold during the
measurement period, or a loaf of bread purchased by a grocery store which is later sold to a household are not final but intermediate sales. Inter‐
mediate sales are excluded from GDP to avoid double counting. For this reason, GDP should never be confused with total sales, or turnover. Consistent
with this approach, exports are always counted as final sales regardless of how the foreigners use them, because they leave the geographic borders of the
national economy.
Our second definition of GDP recognizes that each final sale of a good or service represents the ultimate step that validates all the efforts that have gone into
producing and making it available to the buyer. It encompasses a chain of economic activities which are each seen as value added.
Definition 2
GDP = the sum of value added occurring within a given geographic location during a period of time.
A firm creates value added by transforming purchased input goods and raw materials into products it can sell in the marketplace. Value added is thus the
difference between sales (turnover) and the costs of raw materials, unfinished goods, and imports from abroad. If the firm produces intermediate goods,
its sales are costs to its customers, who themselves are producers. This value added should not be counted twice, and it is deducted from those
customers’ own sales in computing its own value added. When the final consumer purchases a good or a service in the market, the price includes all the
value added created at each stage in the production process; hence the consistency between Definitions 1 and 2. Box 2.1 uses a concrete example to show
how various productive activities contribute to an economy’s total value added.
GDP also measures all incomes earned within a country’s borders—by residents and non-residents alike. Because one person’s final spending must be someone
else’s income, the third definition of GDP is also consistent with the first.
Definition 3
GDP = the sum of incomes earned from economic activities within a geographic location during a period of time.
GDP statistics are mentioned constantly in the financial and political press. The GDP, and in particular, its rate of growth, are generally considered to be
the most important indicators of an economy’s health, and their evolution is closely watched by managers, economists, and politicians. They allow us to
study the performance of a single economy over time as well as to compare different countries. Three important points are worth mentioning:
(1) For comparison over time, we want to distinguish two reasons why GDP can increase: (1) more real economic activity and (2) higher prices for the same economic
activity. This aspect is taken up in Section 2.2.2.
(2) For comparison across countries, we need to convert all GDP measures into a common currency. But, as we will see later, exchange rates are quite volatile and
may give a faulty picture. For that reason, we usually use the concept of purchasing power parity, which is presented in Chapter 5.
(3) Small countries tend to have small GDPs, and yet they may still be well-off. This is why we often look at GDP per capita, dividing the GDP measures by the
population.
The definition of GDP contains a fair amount of arbitrariness, and it is open to debate whether every positive movement in GDP constitutes an
improvement in national well-being. All the same, it is the best indicator we have. More details on this controversial issue are provided in Box 2.2.
Box 2.1 Value Added and Value Subtracted: Two Examples
Consider the following example of value added. A keg of beer is produced and sold for final use for€100. It is useful to break up this final sale into the steps of
value added which were involved in its production. First, a brewery bought barley from a farmer, paying€10, used and paid for energy in the brewing process
with a value of€20, and bought a keg from a keg manufacturer at a cost of€5. (For simplicity, the intermediate inputs of the farmer, energy producer, and keg
manufacturer are assumed to be zero.) The beer is sold to a wholesaler for€80, so the brewery’s own contribution to value added per keg is€45, given by his
sale price (€80) less costs of inputs (€10 + €20 + €5 = €35). Next, the wholesaler sells the filled keg for€90 to a retailer, contributing value added of€10. The
retailer sells the keg for€100 to some consumer (not a pub, which would add another layer to the value added chain!), generating€10 of value added.
Summing up, the final price can be broken down into value added at each stage of production and delivery of the final good:
Value added contributed by the:
Farmer €10
Energy producer €20
Keg manufacturer €5
Brewery €45
Wholesaler €10
Retailer €10
Sum €100
Each step in the value added chain represents a source of income for factors of production involved. Suppose for example that the brewer had labour costs of
€35 (wages and salaries as well as social security contributions) and €5 in beer taxes. Then the brewery’s activity led to profits of €5, which are the income to
the owners—assuming there were no further costs such as interest on loans, royalties for brands or trademarks, or rent. Similarly, if the wholesaler had no
costs (employees, rent, or interest), the €10 of value added would represent his income, which can also be thought of as the profit he receives as owner of
the business. The example shows how the division of the value added is arbitrary and potentially separable from the issue of whether value added is
generated at all.
Because value added is the source of income for labour, capital, and other factors of production, few economic activities could survive very long if they
subtracted value, i.e. if sales did not even cover material input costs. Not only would labour and capital not receive income for their efforts, but also someone
would have to pay for the operating loss on each unit of output sold. Yet there are many examples of value subtraction. Often firms sell goods ‘below cost’
simply to clear inventories for new products. In the centrally planned economies before the fall of communism, many industries were forced to sell their
output cheaply. As a result, firms produced too few goods of poor quality, exemplified by the Trabant car, for which the citizens of the German Democratic
Republic had to wait for years. In capitalist countries, some public services are thought to be worth less than they cost, and some government enterprises
show chronic losses which are ultimately paid for by taxpayers. The non-market value of these services might be high enough to justify those losses; if they
can’t, then the public sector is also engaged in value subtraction.
2.2.2 Real versus Nominal Quantities, Deflators versus Price Indices

Real and nominal GDP
Now that we know what GDP is and how GDP data are constructed, we can immediately see how the national income statisticians have solved the problem of adding
up apples and oranges: the solution is to use prices to convert volumes (the numbers of apples and oranges) into values (final sales of apples and oranges).
Suppose an economy produces only these two goods and requires no imports. Final sales of apples and oranges are obtained by multiplying the quantities of apples
and oranges sold, Qa and Qo, by their respective prices, Pa and Po, yielding nominal GDP, or GDP at current prices:
(2.1)
Yet there is a slight problem: if the price of oranges increases from one year to the next, nominal GDP rises even while the volume of final sales has not changed at
all! An increase in nominal GDP can result from either higher prices or more output. To separate the effects of output and price movements, national income
accountants distinguish between nominal and real GDP. Increases in real GDP correspond to increases in physical output, the number of apples and oranges
produced and sold. Whereas nominal GDP is computed as in (2.1), using the actual selling prices, real GDP is computed by using prices observed in some agreed
base year. 4 In our example, suppose that in the base year 0 the prices of apples and oranges are and final sales of apples and oranges are and
respectively. Then real GDP in year t is given by:
(2.2)
Box 2.2 What GDP Measures
The GDP is a measure of recorded market transactions as well as non-market production by the public sector. This leaves out many activities which are not
carried out through legal channels or do not reach the marketplace, like growing vegetables in a garden at home. Furthermore, since the value of goods and
services is measured using their transaction prices, two identical goods may enter the GDP differently if one of them is sold at a discount. Finally, it is not a
measure of happiness: painful expenses (having a tooth removed, for example) enter the GDP in the same way as pleasurable ones. When someone dies,
GDP rises: the funeral service, the hospital expenses, and the execution of the will by lawyers and bankers all represent additional final sales of goods and
services. Despite the fact that they are costs, pollution and other forms of environmental damage do not count toward GDP, since they are not traded in
markets.
Services enter the GDP exactly like goods. Services include medical doctors’ fees or an estate agent’s commission when an existing house is sold. The
GDP also excludes many forms of income. In the case of real estate, if the house’s value has increased since it was purchased, the previous owner enjoys a
capital gain, but capital gains are not counted in GDP. Used-goods sales, such as cars or antique furniture, do not enter GDP either. Such transactions
represent a transfer of ownership rather than production.
Public services are part of GDP, even if they are not really sold. Their price is simply measured by their cost of production. For example, public education
enters GDP as the sum of teachers’ salaries, operating costs such as electricity or heating costs, and equipment including rents. Similarly, the national
defence enters the GDP as total expenditure on armed forces.
The digital revolution represents a new challenge to the measurement of GDP. It is claimed that GDP may have been growing faster by as much as 0.3 to
0.7% per year than actually measured.3 One example is phone calls with VoIP (Voice over Internet Protocol) which allows users to use free internet
connections. Another is music streaming, which has led to both a massive increase of music listening and an abrupt fall in sales of CDs and other
conventional means of music storage. While the sales of recordings have declined significantly, sales of music services have increased—a radically different
good, yet one that is a close substitute. To date, conventional income and product accountants are challenged to find adequate ways of capturing the impact
of digital goods on GDP.
This distinction is very general and applies to all macroeconomic variables: nominal variables represent values at current prices; real variables represent volumes at
constant prices. As an example, consider Table 2.1, which reports growth rates of nominal and real GDP for the euro area.5
Table 2.1 Growth Rates of Nominal GDP, Real GDP, and GDP Deflator: Euro Area 2005–2015 (% per annum)
Nominal Real GDP

GDP GDP deflator
2005 3.6 1.7 1.9
2006 5.2 3.2 1.9
2007 5.5 3.1 2.4
2008 2.4 0.5 1.9
2009 −3.6 −4.5 1.0
2010 2.8 2.1 0.7
2011 2.7 1.6 1.1
2012 0.4 −0.9 1.2
2013 1.0 −0.3 1.3
2014 1.8 0.9 0.9
2015 2.8 1.6 1.2
Source: AMECO on line, European Commission.
Price deflators and indices

The distinction between nominal and real GDP can be used as a measure of the general price level, or the price of the broadest possible basket of goods in terms of
money. The GDP deflator, one way of measuring the price level, is simply the ratio of nominal to real GDP:
(2.3)
In the base year, nominal and real GDP coincide and the GDP deflator equals 1.0. Often it is multiplied by 100 for ease of comparison over time. The GDP deflator can
be thought of as an average of all prices of final goods in terms of money, where each price is implicitly weighted by the share of the corresponding good in the GDP.
As these shares change over the years, so do the weights.
The inflation rate can be measured by the rate of increase in the GDP deflator, which in turn can be approximated by the following formula:6
(2.4)
For example, Table 2.1 shows that in 2015 the nominal GDP of the euro area rose by 2.8% while the real GDP increased only by 1.6%. On average, therefore, prices
rose by roughly 1.2%.7
An alternative measure of inflation is based on an average of prices with fixed weights, called a price index. A basket of goods is selected and the amount of each
good, or category of goods, in the basket is used to weight the corresponding prices. An example is the consumer price index (CPI). This is based on a basket
of goods consumed by a representative individual.
Figure 2.1 shows the growth rates of the GDP deflator and of the CPI in Italy. Differences between the two measures of inflation are usually not very large, but
they can become significant when import prices—which matter for the CPI but not for the GDP deflator—behave differently from domestically produced good prices.
For example, in the late 1980s the price of crude oil increased less than prices of goods and services produced in Italy. In the early 2000s, domestic wage increases
put pressure on domestic production costs and the GDP deflator rose faster than the CPI, an evolution that was abruptly reversed in the financial crisis after 2009.
Other price indices can be tailored to track prices of certain types of goods, consumers, or sectors of the economy. Along with price deflators, there is a large
menu to choose from, with each price index or deflator having its own special emphasis. Box 2.3 presents some frequently used deflators and indices.
2.2.3 Measuring and Interpreting GDP

The GDP, which represents the economic performance of an entire economy, is not easy to measure. The task is generally carried out by official statistical offices
which draw on various sources of information. One natural source is the tax authorities. Firms report sales (first definition of GDP), individuals report incomes (third
definition), and in most countries (all EU countries, but not the USA) value added taxes (VAT) are collected by intermediate and final sellers who then report their
value added when they pay the tax (second definition).
Fig. 2.1 Inflation Rates, GDP Deflator, and Consumer Price Index: Italy, 1985–2014
Both the GDP deflator and the consumer price index (CPI) measure the price level, or the price of goods in terms of money. The inflation rate is simply the rate of growth of one of
these measures. The figure shows that both GDP deflator and CPI measures of inflation tend to move together over time, with occasional exceptions when the difference in the
underlying ‘baskets’ matters. In the late 1980s and in 2015, world oil prices declined sharply. Since gas and heating oil are part of household consumption, inflation measured by the
CPI declined. Since oil is imported, it does not contribute value added directly in Italy, and has only a small impact on the GDP deflator. The opposite occurred between 2009 and 2014.
Sources: World Development Indicators, the World Bank.
Box 2.3 Price Deflators and Price Indices
The price index closest to the GDP deflator is the producer price index (PPI), with fixed weights corresponding to a basket representative of national
production. Similarly, the CPI is closely tracked by the consumption deflator, the ratio of nominal and real aggregate consumption expenditures by
households. A price index like the CPI or the PPI is an example of a fixed-weight, or Laspeyres index. The consumption deflator, which is based on the actual
share of goods in the corresponding year’s consumption, is called a variable weight or Paasche index. The CPI and the consumption deflator include goods
and services produced abroad and imported, while the PPI and the GDP deflator do not, but these latter measures include goods and services locally
produced and exported. Figure 2.1 suggests a growing divergence between the PPI and the CPI in Italy in the late 1980s. The reason is that imported goods
prices increased by less than those of domestically produced goods.
Other frequently used deflators are related to exports, imports, investment goods, and government purchases. The wholesale price index (WPI) measures
the average price of goods at the wholesale stage, and various commodity price indices track the evolution of raw materials prices. The dizzying diversity of
indices and deflators reflects the fact that a perfect price index simply does not exist. Different price measures are used for different purposes. For example,
wage-earners would like to tie their wages to their cost of living; in this case, the relevant index is the CPI or the consumption deflator. In the case of Italy,
linking wages to the CPI rather than to the PPI resulted in higher profits for firms whose sales are better tracked by the PPI. Because the CPI and other
Laspeyres indices are easier to compute, they are used most often in practice.
Box 2.4 The Underground Economy and Unpaid Work
Who hasn’t had an offer from a carpenter, a car mechanic, or painter to do some work ‘without a receipt’? Agents engage in the underground, or informal,
economy for straightforward reasons. First, they want to avoid taxes (the value added tax, employment and social security charges, profit taxes). Second, while
significant, criminal activities, such as drug-dealing, prostitution, or racketeering, are intentionally concealed and kept underground by market participants. By
definition, the size of the underground economy is unknown, but national income statisticians often attempt to guess its importance. They use various
approaches such as monitoring household electricity use, which tends to be higher in economies where unreported market activity is more significant, or
looking at the amount of large-denomination currency in circulation, since underground transactions do not use bank accounts and profits are conveniently
held in large bills. The sale of intermediate inputs related to final production often indicates underground economy activities. For example, a large
discrepancy exists between the purchase of construction materials and reported construction activity. Figure 2.2 shows the extent of the underground
economy in a number of countries.
The fact that GDP figures are collected through tax returns immediately raises the suspicion that individuals and firms may misrepresent their finances to the fiscal
authorities. Such unreported income and output is frequently referred to as the underground economy. Box 2.4 presents estimates of how large it could be. It
also alerts us to the importance of work that is not paid for in the marketplace.
Another shortcoming associated with the magnitude of the task is the time it takes to get reasonably accurate numbers. Data from tax returns are processed with
some delay. Usually at the end of the first month of each quarter, figures for the preceding quarter are released. Box 2.5 explains how such flash estimates are
produced and updated several times over the following years. The inaccuracy of these estimates is unsettling because they are frequently used by governments
when deciding on economic policies, by investors when valuing their assets, and by firms deciding on hiring or firing workers and on acquiring new plant and
equipment. This is why other indicators are often used to supplement the GDP figures.8 It is also why analysts tend to concentrate on growth rates rather than
levels. As long as the distortions do not change much over time, measured GDP growth rates offer a good picture of average economy performance.
Box 2.5 How National Accounts Estimates Can Vary over Time
Because governments, firms, and investors require timely information about the economy, national statistical institutes in advanced economies have devised
ways of quickly producing preliminary estimates of GDP. The procedure is based on the knowledge that the value added of, for example, the 100 largest
corporations represents a given proportion of GDP. If the proportion were 10%, as these firms fill in VAT tax reports or respond to specially designed
questionnaires, multiplying by 10 their combined value added provides a rough early estimate of GDP. A few months later, revised estimates can be based
on data provided by a larger sample of firms. Waiting still longer will allow the incorporation of estimates based on an early and partial analysis of tax returns.
Detailed analysis of all tax returns data—using procedures to reconcile differences between measures based on the three definitions—leads to a final figure.
Table 2.2 shows successive estimates of German GDP in 2008. The first estimate, published in January 2009, exceeded the final figure by €8 billion, or by
about 0.3% of the initial estimate.
Fig. 2.2 Estimates of the Size of the Underground Economy (%of GDP)
Another serious drawback of GDP as a measure of economic activity is unpaid work. Minor repairs around the house, caring for children, cooking for the family, and cleaning up
take up much time and effort. Wealthier people hire help for these chores, in which case it becomes part of GDP (if reported to the tax authorities). Most people do it themselves, and
it is unrecorded.
Sources: Schneider (2015), AMECO, own calculations.
It is tempting to compare GDPs across countries. Because countries have different populations, it is natural to look at GDP per capita, or the average income
earned within a country’s boundaries. Such data must be regarded with caution, however. First, GDP is a measure of income, not wealth. Income is a flow, while
wealth is the stock of assets accumulated over longer periods of time. For example, the average income earned in the UK is lower than that of Abu Dhabi. Yet
average British wealth is likely to be much higher because Britain has been accumulating wealth for centuries, in the form of private assets (e.g. houses, factories,
jewels, stocks) and national assets (e.g. the London Bridge, paintings in the British Museum, railroads, highways and telecommunication networks, and much more).
Table 2.2 Estimates of 2008 German Nominal GDP
Date of publication GDP(€bn, 2000 prices) %difference from previous estimate %difference from Jan 2009
Jan 2009 2489.4 — —
Feb 2009 2489.4 0.00% 0.00%
May 2009 2492.0 0.10% 0.10%
Aug 2009 2491.4 –0.02% 0.08%
Nov 2009 2495.8 0.18% 0.26%
May 2010 2495.8 0.00% 0.26%
Nov 2010 2481.2 –0.58% –0.33%
Feb 2011 2481.2 0.00% –0.33%
Source: Estimates as published in the monthly bulletin of the Deutsche Bundesbank, various issues.
Second, a large number of transactions are not recorded, especially in developing countries. They belong to what is sometimes called the informal economy. For
example, much food can be produced within the extended family (a non-market activity), or exchanged for other food (a non-reported market activity). Very low
reported per capita income levels in developing economies are believed to underestimate true value added and income, even though efforts are being developed to
reduce the gap. Finally, GDPs are measured in the country’s local monetary unit, or currency, and are then converted into a common currency using the exchange
rate. But local costs are often much lower in poor countries, for reasons discussed in Chapter 15. To correct for this effect, economists often use GDP figures that
have been adjusted for differences in purchasing power, as already mentioned above.
2.3.1 The Circular Flow Diagram

From final expenditures to net taxes and factor income
Each individual’s expenditure necessarily contributes to some other individual’s income. The simplified circular flowdiagram represented in Figure 2.3 is based on
this simple truth and goes a long way in tracking the functioning of an economy. Based on the first and third definitions of GDP, it shows how GDP arises as final
sales, and how it is paid out to households, the owners of the factors of production. In addition, it shows how firms—around which market activity is organized—
households, and the government interact to make GDP possible.
The GDP appears in the left part of the figure. It represents the final net sales of firms. Since firms are owned by households, the GDP represents the gross income
of factors of production employed in the country or geographic region under consideration. To see what firms do with revenues coming from final sales, we move
clockwise. The government, shown as the circle inside the flow diagram, takes (in the form of taxes) and gives (in the form of various transfers) to both households
and firms. It also purchases goods and services directly. Because it needs to pay for them, the government takes in more than it gives away in transfers. The
difference between taxes and transfers is called net taxesand is represented by T. These taxes are taken in at several points of the value added chain—as indirect
taxes such as value added taxes, or as direct taxes on different types of income. Here we consolidate them for simplicity. What is left of GDP after these taxes and
transfers are subtracted is called private income, Y − T.
Fig. 2.3 The Circular Flow Diagram
The lower left part of the wheel represents final sales of goods and services which are domestically produced. It is the sum of consumption spending (C), investment spending (I),
government purchases (G), and exports (X) less imports (Z). In the upper left part of the wheel this is interpreted as income to residents. This income is taxed by the government,
but is also supplemented by transfers to households and firms, resulting in net taxes (T). The remainder is private income, which may be saved (S) or spent (C). The private sector
invests in productive equipment (I), which it finances in part by savings (S); the balance S − I is the private sector’s net saving behaviour. Similarly, the public sector’s net saving
behaviour is reflected in its budget surplus T − G. The balance X − Z represents net exports of goods and services.
From private income to absorption plus net exports—GDP

Private income is ultimately earned by those households which own the factors of production involved in creating the value added. The largest part of private
income is wages and salaries paid to workers plus payments for rent and royalties, as well as net interest on loans from banks. The residual which remains after firms
pay all these other factors is known as gross profit. This profit can either be saved by the firm or redistributed back to firm owners as income.9 Since all factors of
production are ultimately owned by some households, households receive income as employees, as bondholders, as shareholders, as owners of land, and as holders
of patents. Households can either save this income, or spend it on consumption.
The private sector represents the consolidation of households with the firms that they own. The flow diagram shows how the aggregate savings of the private
sector (S) are deposited with the financial sector. The financial sector includes banks, financial institutions, and stock markets whose function is to collect savings
and channel them to firms seeking to invest, that is, to purchase productive equipment. This activity, called financial intermediation, is represented by the
lower-right circle. In the aggregate, the private sector uses its savings—the income that it does not consume—to finance, or pay for, the acquisition of new
productive equipment by firms. The stock of existing productive equipment, including structures, is referred to as physical capital, while the purchase of new
equipment is called investment. The excess of private saving over investment (S − I) is called net private saving. Net private saving can be positive or negative.
Firms and households spend their income—part of it borrowed—to consume (C) and to invest (I).10
To private sector expenditures on goods and services (C + I) the government adds its own demand (G). Governments purchase goods (e.g. roads, military
equipment, newly built buildings, or stationery for the bureaucracy) and services (of civil servants and other employees). While governments transfer lots of income,
distribute various subsidies to firms and households, and pay interest on the public debt, these are not purchases of goods and services and not included here.
Total domestic spending, sometimes called absorption, is the sum (C + I + G) of private and public spending on all goods and services. Part of absorption includes
the purchase of imported goods and services (Z). This is shown as the branch going into the leftmost circle, which represents the rest of the world. Similarly, while
some domestic income thus leaks abroad, foreigners buy domestically produced goods and services. Those purchases are the country’s exports ( X). Netting these
two flows with the rest of the world yields net exports (X − Z).11 When positive, net exports increase demand for domestic production above that originating with
domestic residents; when negative, demand for domestic production is less than total domestic demand.
The sum of absorption and net exports represents the total final sales that occur within the geographic area, i.e. the GDP. The circular flow of income is closed.
This circularity is the essence of economic activity: we (collectively) earn to (collectively) spend.
2.3.2 Summary of the Flow Diagram

The flow diagram can be summarized using the first and third definitions of GDP (Y). As final sales, the GDP is broken down into four main categories: (1) final sales
of consumption goods and services (C), (2) final sales of investment goods and changes in inventory stocks (I), (3) final sales to the government (G), and (4) sales to
the rest of the world (X). Since part of domestic income leaks abroad to pay for imported goods, imports (Z) must be subtracted, which gives the first decomposition
of GDP by final expenditures:
Table 2.3 Components of GDP by Expenditure, 1999–2015 (% of GDP)
Consumption (C) Investment (I) Government Purchases (G)

Australia 56.5 26.9 17.6
Canada 55.4 22.2 20.4
France 55.2 21.8 23.1
Germany 56.3 20.3 18.7
Italy 60.3 19.8 19.2
Japan 58.6 22.4 18.9
Switzerland 56.0 24.1 11.0
United Kingdom 64.5 17.6 19.9
united States 67.6 20.8 15.3
Euro area 56.1 21.5 20.3
Source: AMECO, European Commission.
(2.5)
The flow diagram also shows that GDP can be viewed as net incomes earned by the owners of production factors. What do they do with this income? The three
possibilities are given on the right-hand side of the flow diagram: they pay taxes net of transfers (T), they save (S), and they consume (C). Hence the second
decomposition by uses of income:
(2.6)
Table 2.3 displays the components of the first decomposition as a percentage of GDP for a few countries. Consumption typically represents about 60% of GDP in
Europe, but it is much higher in the US. The investment rate—the ratio of investment expenditures to GDP—amounts to some 20%, with few differences among the
developed countries, but it is about twice as much in China. Because investment corresponds to the accumulation of productive equipment, it matters for future
economic growth. The table shows that public spending varies quite a bit, but comparisons are not always easy. Investment in infrastructure equipment (roads,
bridges, public utilities) may be undertaken privately in some countries and publicly in others. Many goods and services are privately produced in some countries
while they are delivered freely as public goods in others: these include medical services, schools, child care, and public transport. Finally, the ‘size of government’ is
considerably greater than the share of government purchases of goods and services: transfers to firms and households, not reported in Table 2.3, may be as large as
direct expenditures or even larger. When total spending is considered, which adds transfers to direct purchases, the government often ‘handles’ more than half of
GDP.
The flows of incomes and spending captured by Figure 2.3 constitute the real, as opposed to financial, side of an economy. Parts of these flows leak out to the
financial side in the form of corporate and household savings; others leak out to the government in the form of tax payments or social security contributions; others
to foreigners through imports. To the extent that withdrawals of resources from the circular flow due to a particular sector are not matched exactly by inflows in the
form of spending, then that sector’s net asset position must be changing, by definition. 12 If savings of the private sector exceed investment spending, for example,
this means that the private sector is accumulating assets. The same holds for the general government if net taxes exceed government purchases of goods and
services, or for a nation if net export of goods and services, broadly defined, is positive. Asset accumulation or decumulation has economic consequences. How the
financial side of the economy functions, and how the real and financial sides are linked, is studied in Part III of this book.
2.3.3 More Detail

Domestic vs national
Economic boundaries are important for the measurement of national output and income. As noted earlier, the GDP defines a country by the people and firms that
operate within its borders, quite independently of their nationality or residence. An alternative is to define the economic activity of a country by its residents, both
people and the firms they own, wherever they produce or earn income. This leads to an alternative concept, called gross national income (GNI). The GNI is
obtained by adding to GDP those incomes earned abroad by resident entities and subtracting incomes generated by non-resident entities within the country. The
net of these two measures is the balance on primary international income. These are called primary incomes because they are directly associated with
production.
The balance on secondary international incomeadds up various forms of payment ‘without consideration’ (taxes paid to the home authorities by non-
residents or paid to foreign authorities,.pngts and transfers sent to or received from abroad, etc.). Adding the secondary income balance to the GNI provides the
gross disposable national income. This is summarized in Figure 2.4, which explains these concepts using the case of Belgium and Germany. In 2013, Belgian GDP was
about half a billion euros greater than GNI. This means that Belgium had a deficit on its primary income balance; in total, more payments were made by Belgian
residents for the use of foreigners’ labour, capital, or other contributions to value added which occurred in Belgium in 2013. Similarly, the net flow of.pngts, transfers,
taxes, and related payments meant that more was paid to the rest of the world than Belgian residents received. As a result, Belgian gross disposable national income
was a whopping €7.3 billion, or about 5% less. In contrast, Germany received more primary income from abroad than it paid out, while it was a net payer on
secondary income.
Fig. 2.4 From Domestic Product to Disposable National Income, 2013
Many foreigners reside in both Belgium and Germany, but GDP deviates from GNI in qualitatively different ways. In Belgium, interest income of foreign firms and wages of workers
who live outside of the country dominate, so that GNI and GDNI are less than GDP. In Germany, foreign investment income of German residents outweighs that of non-residents. On
the other hand, German residents transferred more money abroad than they received ( €51 bn in 2013), which partially offsets the inflow of foreign primary income.
Sources: knoema.com.
Table 2.4 GDP and Household Disposable Income, 2014
GDP (billions of €) Households Disposable Income

in € %of GDP
Germany 2916 1710 58.7
France 2132 1307 61.3
Sweden 431 216 50.1
Switzerland 516 315 61.1
united States 13058 9399 72.0
united Kingdom 2253 1352 60.0
Note: Data for Switzerland: 2013.
Sources: AMECO, European Commission.
Household incomes
It is also important to note that a big share of GDP does not reach individual households. It either goes to the government (as net taxes) or is saved by firms (as
retained earnings). This is illustrated in Table 2.4.
Gross and net
All of the concepts presented so far are ‘gross’. What is ‘net’, then? In the process of producing output, productive equipment is subjected to wear and tear and
obsolescence. Properly measured,
this depreciation should be subtracted to give a clearer picture of the output that is really available as income if we are to preserve the value of our productive
capacities. Subtracting depreciation from GDP gives us the net domestic product (NDP), GNIbecomes NNI, etc.13
2.3.4 A Key Accounting Identity

The two decompositions of GDP, (2.5) and (2.6), are accounting identities: they hold by definition. Therefore it is always the case that:
Consumption C appears on both sides of this equality and can be eliminated. When this is done and terms are rearranged, the two accounting identities yield a third
one:
(2.7)
The last term, X − Z, is the balance of exports over imports of goods and services. Parentheses highlight the fact that the corresponding expressions appear in Figure
2.3 as net flows of the private sector (household and business), government, and the rest of the world, respectively. Each of the three net flows can be thought of as
a form of saving or withdrawal from the circular flow income and expenditure: a leakage if positive, or an injection if negative. If S > I, the private sector in the
aggregate is a net saver. If S < I, the private sector is a net borrower. Similarly if T > G, the government is saving, and if G > T it is borrowing by issuing debt to
domestic or foreign residents. The identity (2.7) shows how these leakages are linked, by definition. Table 2.5 provides some examples in the year 2010, when the
European debt crisis was in full swing.
The table shows that, in 2010, both Italian private and public sectors were spending more than they took in; the country as a whole, therefore, had to borrow
abroad, which explains why they were running sizeable external deficits, as will become clear in Section 2.4. The Eurozone as a whole was running balanced net
exports as budget deficits were about matched by private sector savings. Japan, the USA, and the UK were running huge budget deficits but Japan’s net exports
were in surplus, due to even larger net private savings, while the US and the UK had to borrow abroad.
Table 2.5 The Accounting Identity in 2010 (% of GDP)

S–I T–G X–Z
USA 5.6 –8.8 –3.2
Japan 10.3 –6.7 3.6
Belgium 3.9 –2.6 1.3
Denmark 4.7 0.8 5.5
France 2.6 –4.8 –2.2
Germany 8.1 –2.5 5.6
Italy –1.3 –2.2 –3.5
Netherlands 11.5 –3.8 7.7
Spain 0.7 –5.2 –4.5
Sweden 4.7 1.6 6.3
UK 5.8 –8.3 –2.5
Euro area 4.1 –3.9 0.2
Note:Data for net exports include incomes that have been ignored so far: primary incomes that remunerate the services of factors of production employed
outside national borders and secondary incomes that include various taxes and transfers. Section 2.4.1 explains these additions and identifies extended
net exports as the current account balance, which is shown in this table.
Source: OECD.
2.3.5 Identities versus Economics

Identity (2.7) is a particular way of stating that all goods and services produced (Definition 3 of GDP) must be purchased (Definition 1). For example, if private
savings exceed private investment (S > I), either net exports must be positive or the government budget must be in deficit, or both. This is an accounting fact, true by
definition.
It is important to remember that identities tell us little or nothing about causation. Without more information, it is impossible to know whether (1) the government
deficit is at the origin of positive net private savings, (2) high exports are generating income that is simply saved by residents, or (3) low domestic investment
spending is coinciding with a domestic recession, in which both imports and tax revenues are low, bringing either current account or government budget or both into
deficit. This highlights the difference between simply collecting data and interpreting them, that is, the difference between accounting and economics. The identity
(2.7) is not only a requirement that accounts be correctly measured; we will see later that, as a market equilibrium condition, it also tells us a lot about how adjustment
mechanisms work.
The balance of paymentsaccounts record all economic transactions between a geographical entity and the rest of the world. Usually this entity is a country, and
almost all countries have adopted a classification of payments designed by the International Monetary Fund. Following that system, the presentation in Table 2.6
separates out international transactions in goods and services in a broad sense (Parts I and II) from financial transactions (the lower Part III). Some useful rules of
thumb on how transactions are recorded in the balance of payments are presented in Box 2.6.
Simply put, the current account captures commercial transactions as explained in Section 2.4.1. The capital accountfurther broadens the notion of commercial
transactions to include purchases and sales of rights to exploit natural resources or buildings, and copyrights. It also includes transactions such as nationalizations,
private and public debt forgiveness, and ownership transfers to residents abroad. The current account, or net national saving in the form of acquisition of claims on
foreigners, is found by combining I and II. It can also be thought of as net national lending if positive, and net borrowing if negative.14 What is being lent or
borrowed is described in the financial account, which is Part III of the balance of payments. Financial transactions imply the purchase and sale of assets such as
stocks and bonds, loans and credit, bank deposits and currency, as well as real estate and other forms of wealth. Purchases of assets are recorded as a positive
entry, sales with a negative entry.
Table 2.6 The Balance of Payments
I. Current account
A. Goods and services
1. Goods
2. Services
B. Income account
1. Primary income
2. Secondary income
II. Capital account
III. Financial account
1. Direct investment
2. Portfolio investment
3. Other investment
4. Reserve assets
IV. Errors and omissions
2.4.1 The Current Account and its Components

The first account to consider in Table 2.6 records exports and imports of goods and services. Imports are entered with a minus sign (called debits), while exports
contribute positively to the balance (called credits). The net result is line IA, the balance of goods and services. The balance of trade in goods (line IA1)
includes merchandise trade as well as trade in intermediate inputs, goods repair, goods held in ports, and non-monetary gold. The balance of trade in services (line
IA2) incorporates a wide and growing variety of intangibles (so-called ‘invisibles’) such as transport and travel, communication, insurance, financial, and other
services. It also includes royalties and licence fees.
A second account to consider is the balance of international income (line IB). This account summarizes the net income of a nation which originates from
abroad, as already noted in Section 2.3.3. It captures those transactions that do not represent direct sales of goods and services but can be considered as commercial
in a general sense. Following that same logic, the international income account is broken down into two components. Primary incomeconsists of non-resident
employee incomes, interest earned abroad and repatriated profits of foreign-owned corporations. Secondary incomeincludes remittances of employees who
reside in a foreign country and send money home, as well as.pngts and transfers. For example, when a Polish plumber living in London sends money to relatives
living in Warsaw, this counts as a credit to the Polish secondary income account and as a debit to the UK’s. The secondary income account also includes tax
payments by foreign firms or persons, insurance claims, as well as private or public.pngt or transfers. An example of the latter is foreign aid granted at the national or
multinational level. It may be helpful to think of the primary international income account as a summary of transactions involving the sale (‘export’) and purchase
(‘import’) of factors of production (capital and labour services), while the secondary international income account captures sale and purchases of ‘good will’ broadly
defined.
The sum of the balances of trade in goods, services, and of international income is called the current accountbalance.
2.4.2 The Capital Account

In the IMF classification the capital account captures various commercial transactions that involve ‘intangible’, non-produced assets like the right to use land or the
acquisition (or loss) of good will through debt cancellation (or the expropriation of foreigners).15 Because the capital account is usually small and captures rare
transactions not easily classified elsewhere, we will ignore it throughout the rest of the book.
2.4.3 Net Borrowing or Lending

The financial account is a country’s net lending or borrowing. Under the convention that borrowing is negative lending, we will call it net lending for short. But why
is it called net lending? The answer is provided by the national income decomposition (2.5) which can rewritten as:
where X − Z is the balance on goods and services, Y is GDP, and A = C + I + G is referred to as absorption, or total domestic spending on goods and services by
domestic and foreign households, firms, and government agencies. Net lending is defined as the sum of the balance on goods and services (X − Z) and the income
account balance (IAB). Again ignoring the capital account, it follows that:
(2.8)
where YD = Y + IAB is the Gross National Disposable Income, an extended definition of income further explained in Box 2.7. This makes it clear that Net Lending is
the excess of income (YD) over spending (A) and thus indicates indeed whether the country is a net borrower or a net lender. When a country earns more than it
spends (i.e. CA = YD − A > 0), it is a net lender vis-à-vis the rest of the world. Conversely, a country running a current account deficit spends more than it earns (CA
< 0 and A > YD) and must match the difference by borrowing abroad.
2.4.4 The Financial Account and its Components

Mechanically, when a country saves—equivalently, when it is a net lender—it acquires foreign assets or gets rid of some of its foreign liabilities by repaying some
of its debt. Conversely, a net borrower increases its liabilities or disposes of some its foreign assets. These financial transactions, collected in the financial account,
are the counterpart of the current account (plus the capital account, which is ignored) and they are shown in Part III of Table 2.6. Somehow, they must be equal. The
remaining question is: who actually performs the balancing act, and how?
The answer is to be found in the financial account, which aims at fully describing all lending and borrowing activities by the country’s private and public
sector. The IMF defines four categories of financial transactions which affect this account:
(1) Direct investment (line III.1)—purchases and sales of shares (equity investment) in business enterprises in excess of 10% of the total value; reinvested profits;
real estate purchases.
(2) Portfolio investment (line III.2)—purchases and sales of shares (equity investment) in business enterprises of less than 10% of total value; purchases and issues
of bonds, money market instruments, and financial derivatives.
(3) Other investments (line III.3)—trade credits; loans by banks; deposits held at banks; currency.
(4) Reserve asset transactions (line III.4)—transactions involving monetary gold, special drawing rights and reserve positions with the IMF, and foreign exchange
reserves held by the monetary authorities.
Box 2.6 Examples of Balance of Payments Accounts
Two simple rules help understand balance of payments. The first one concerns the current account: international transactions are recorded with a minus
sign when they involve an outgoing payment (e.g. for imports) and positive signs represent receipts in the domestic currency (e.g. for exports). If incoming
payments exceed outgoing payments—the current account is in surplus—the country earns more than it spends, which means that it is a net saver. If the
current account balance is negative, it is a net borrower.
The second rule applies to the financial account, and matches the saving/borrowing distinction. Buying foreign assets represents saving and is registered
with a positive sign, even though money if leaving the country. Selling assets represents borrowing and is registered with a negative sign. In Table 2.7 we
give some examples of transactions and how they are recorded, using these two rules.
Table 2.7 Balance of Payments: Some Examples

Transaction Credit (+) or debit Country Account
(–)
UK exports chemicals to France to the amount of £1 million + £1 m UK Goods and services
–£1 m France Goods and services
French school trains German cyclists for €500,000 +€500,000 France Goods and services
–€500,000 Germany Goods and services
German construction company is paid SF 5 million to build a Swiss bridge +SF5 m Germany Goods and services
–SF5 m Switzerland Goods and services
Swiss ski instructor is paid salary of €80,000 for work performed in Austria +€80,000 Switzerland Primary income
–€80,000 Austria Primary income
UK fast food franchises remit £1 million in profits to headquarters in the USA +£1 m USA Primary income
–£1 m UK Primary income
Austrian government gives €3 million in relief aid to tsunami victims in Thailand +€3 m Thailand Secondary income
–€3 m Austria Secondary income
Estonian worker in Denmark sends DK 100,000 to family in Tallinn +DK 100,000 Estonia Secondary income
–DK100,000 Denmark Secondary income
Spanish government forgives debt of €10 million owed by Peru +€10m Peru Capital account
–€10m Peru Capital account
Swedish investor purchases a factory in Germany for €100 million –€100 m Germany Financial account/direct
investment
+€10 m Sweden Financial account/direct

investment
Portuguese bank buys €20 million of stock in German company from French bank –€20 m France Financial account/portfolio
based in France investment
+€20 m Portugal Financial account/portfolio

investment
UK bank based in London lends £50 million to subsidiary in Ireland –£50 m Ireland Financial account/other
investment
–£50 m UK Financial account/other

investment
Slovenian resident transfers €100,000 from home account to a bank account in Italy –€100,000 Italy Financial account/other
investment
+€100,000 Slovenia Financial account/other

investment
Box 2.7 Links Between National Income Accounts and the Balance of Payments
In this section and in Section 2.2.2, we face three different definitions of income. The differences are all related to international transactions and are therefore
linked to the balance of payments. This box makes all this more precise. As before, assume a zero capital account balance. As explained in Box 2.2, the
Gross National Income (GNI) concerns earnings by domestic residents while the Gross Domestic Product (GDP) refers to incomes earned within the
country’s borders. The link with the balance of payments is:
GNI = GDP + primary income account balance,
since the primary income account captures payments received from nationals living abroad (a credit) and payments from foreigners who live domestically (a
debit).
The Gross National Disposable Income (GNDI) builds upon the GNI by adding net secondary incomes:
GNDI = GNI + secondary income account balance.
Combining these two definitions, we have:
GNDI = GDP + income account balance.
Thus, to the extended definition of income, GNDI, relative to GDP, corresponds the extended definition of trade as we move from the balance on goods and
services (X − Z) to the current account balance CA, as shown by a comparison between (2.5) and (2.8) rewritten respectively as:
An overall positive financial account indicates net lendingto the rest of the world, a deficit corresponds to net borrowing. In other words, a positive entry in the
financial account indicates that money flows out of the country to purchase foreign assets, it is a credit item, while inflows are recorded as deficits as domestic
assets are sold to foreigners. While the first three types of financial account transactions can be conducted by the private or public sector, the fourth item is
performed exclusively by the monetary authorities, usually the central bank. This action is called a foreign exchange market intervention.
Summarizing, net lending is measured by the current account but also by the financial account plus changes in official reserves, or the foreign reserves
account balance. Ignoring the capital account, we have:
(2.9)
2.4.5 Errors and Omissions
Accounting principles require that net lending be the same whether calculated as (2.8) or as (2.9). Trade data originate with customs authorities. Financial data come
from the banking system, since international transactions are mediated by financial institutions. Given the sheer number of observations that must be gathered, it
should not be surprising that, in practice, discrepancies emerge. While there are genuine mistakes—the sheer volume of data to be treated is an invitation for errors
—there may also be omissions which are less than innocent such as tax evasion, illicit trade (drugs, arms, and counterfeiting come to mind) or money laundering.16
Table 2.8 Balance of Payments, Various Countries, 2014 (US$ billion)
Eurozone US Sweden Turkey Brazil China Russia UK

Current account 320 –390 33 –47 –104 220 58 –152
Balance on goods 332 –741 18 –64 –7 435 190 –203
Balance on services 94 233 9 25 –48 –151 –55 146
Primary income balance 79 238 15 –9 –52 –34 –68 –54
Secondary income balance –186 –119 –10 1 3 –30 –8 –41
Capital account 27 0 –1 0 0 0 –42 –2
Net lending 346 –390 32 –47 –104 220 16 –154
Financial account balance 403 –240 13 –45 –100 79 23 –166
Direct investment, net 62 489 4 –7 –71 –209 34 –134
Portfolio investment, net 97 –167 21 –20 –39 –82 40 –189
Other investment, net 183 –240 –7 –17 –3 253 51 170
Reserve assets 6 –4 0 0 11 117 –108 12
Net errors and omissions 56 150 –17 2 3 –140 6 –12
Note: By construction, 'net errors and omissions' are equal to the financial account balance less the sum of current account and capital account balances.
Deviations are due to rounding error.
Source: OECD.
This is why an additional account called ‘errors and omissions’ or ‘balancing items’ is needed, item IV in Table 2.6. Again suppressing the capital account, the
errors and omissions are computed as the amount that must be added to (2.9) to make it ‘add up’:
There is no presumption where these errors come from. Table 2.8 shows that errors and omissions can at times be embarrassingly large.
Table 2.8 also illustrates the balance of payments accounts for a number of different countries with widely different experiences. The enormous current account
deficit in the USA can be linked to a deficit in trade on goods, only partially offset by surpluses in services and income balance. The primary income balances of the
Eurozone and the US are positive, partly reflecting income from subsidiaries abroad. Their negative secondary income balances are partly driven by money sent
home by foreign employees. The financial account of the Eurozone is in surplus as local firms and investors acquire shares in foreign firms. The US financial account
is in overall deficit, but it combines a large surplus in direct investment, i.e. the acquisition of foreign firms, and the acquisition by foreigners of minority shares in US
firms (portfolio investment) and of US government debt, some of which is held by foreign central banks to use as reserves. Indeed, we can observe a large increase
of reserve by the Chinese central bank, while the central bank of Russia has been active limiting the depreciation of its currency by selling foreign exchange reserves.
Finally, note that errors and omissions can be very large, as is the case in the US and China.
Box 2.8 China’s Foreign Exchange Reserves
In 2008, just as the developed world plunged into a massive financial crisis, China posted a large current account surplus, making it a net lender of some
$420 billion. Its non-official financial account was slightly negative as foreign corporations were rushing to escape the financial crisis and invest in the world’s
fastest growing country. In order to avoid a currency appreciation, the central bank increased its reserves by nearly $500 billion. By end 2014, China was
holding foreign exchange reserves of $4,000 billion, more than one third of the world total and more than four times as much as the whole European Union.
Then the tide reversed. In 2015, its current account surplus shrank and the financial account turned positive as Chinese residents sought to park their money
abroad and foreigners retreated. This time the central bank sold about $500 billion worth of its reserves to avoid a sharp currency depreciation.
2.4.6 The Meaning of the Accounts

Equation (2.9) implies that net lending corresponds to the current account, which also equals the financial account balance. While this is obvious from an
accounting viewpoint, it is remarkable that it occurs in practice since it sums up millions of commercial and financial transactions. Of course, accounting mistakes
occur, and they are captured as errors and omissions, but there is much more to it. To see that, consider (2.9) again, breaking out the change in reserve assets from
the financial account:
(2.10)
where NOFA is defined as the financial account excluding change in foreign exchange reserves held by the central bank. Changes in foreign exchange reserves—
sometimes called ‘official interventions by monetary authorities’—occur when the central bank also gets involved in the financing of international transactions. If
the central bank has been a net buyer of foreign currency or other financial assets, for example, this would be registered as net lending and mean that ∆ R > 0.
Why should the monetary authorities accumulate (∆ R > 0) or decumulate (∆ R < 0) reserves? Consider the case when the sum of the current and the capital
accounts is in surplus but investment abroad NOFA is smaller (CA + KA > NOFA). This means that, collectively, the residents want to save and yet they do not
acquire enough foreign assets. In chapters to follow, we will see that foreign currency is also a foreign asset. In the event that CA + KA > NOFA, this implies that
domestic residents will have received through CA + KA more foreign currency than they voluntarily acquired through NOFA. They will want to get rid of this
unwanted money, which means selling it to acquire domestic money. This will strengthen the value of the domestic currency. 17 At this stage the monetary
authorities will decide whether they want the domestic currency to become stronger, that is to appreciate. If not, they will step in and absorb the unwanted foreign
currency, which means that they increase their stock of foreign exchange reserves (∆ R >0) and restore the equality in (2.10). Later chapters will examine this decision
in detail. At this stage, we just note that item (4) in the financial account (line III.4 in Table 2.6) is an important variable of economic policy. Box 2.8 provides an
example.
Summary
1 The gross domestic product (GDP) can be defined in three equivalent ways: (1) as the sum of final sales, (2) the sum of value added, or (3) the sum of factor
incomes. GDP is a flow variable measured over a well-defined time interval, usually a year.
2 ecause nominal GDP measures final sales at market prices, an increase in the price level leads to an increase in GDP even if quantities sold are constant. Real GDP
is computed by pricing current output with constant prices, corresponding to a chosen base year.
3 The GDP deflator is the ratio of nominal to real GDP. It is one measure of the price level. Inflation is approximately equal to the difference between the growth rates
of nominal and real GDP. Price indices, also used to compute inflation rates, use constant-weight baskets of goods and services.
4 Measurement of GDP is imperfect, costly, and time-consuming. A large amount of economic activity is unmeasured, such as household services and the
underground economy. Yet year-on-year comparisons, such as annual growth rates, are less affected by measurement problems.
5 GDP is equal to the sum of consumption, investment, government spending, and net exports (Y = C + I + G + X − Z). At the same time, GDP is equal to
consumption, plus private sector savings, plus net taxes (gross taxes less public transfers received by the private sector) (Y = C + S + T). It follows as an identity
that the current account surplus is equal to the surplus of the government plus the surplus of the private sector (X − Z) = (T − G) + (S − I).
6 The balance of payments is a record of current account transactions and their financial counterparts, the financial account. The current account is the sum of the
merchandise, invisibles, and income accounts. Net lending is measured either by the sum of the current and capital accounts or by the balance of the financial
accounts. Net errors and omissions are a balancing item to account for any discrepancy between the two.
7 If the monetary authorities want to maintain the value of their country’s exchange rate, they must intervene on exchange rate markets to match any possible
balance of payments imbalance. Conversely, the exchange rate floats freely when the monetary authorities refrain from intervening; then all adjustment for
balance of payments equilibrium occurs within the private sector, as a result of changes in the market-determined exchange rate.
Key Concepts
accounting identities
flow versus stock variables
final versus intermediate sales
value added
GDP per capita
nominal and real GDP
GDP deflator
consumer price index (CPI)
underground economy
circular flow
net taxes
private income
consumption
financial intermediation
savings
physical capital
investment
absorption
gross national income (GNI)
net national income (NNI)
primary international income
secondary international income
gross disposable national income
net domestic product (NDP)
balance of payments
capital account
balance of goods and services
balance on international income
primary and secondary incomes
current account
net borrowing or lending
financial account
foreign exchange market intervention
foreign reserves account balance
currency depreciation
Exercises
1 You are given the following data:

GDP 2,500
Depreciation 250
Before-tax corporate profits 500
Social security contributions 350
Transfers to households and firms 500
Net interest to foreigners 100
Proprietary income 35
Net corporate saving 300
Indirect taxes 500
Subsidies to enterprises 200
Fines and fees 50
Net remittances to rest of world 250
Corporate taxes 50
Consolidated government deficit 50
Personal taxes 750
Household savings 100
Investment expenditure 600
Compute: NDP, national income, personal disposable income, consumption, government purchases, GDP, the current account balance. State your assumptions
clearly.
2 What happens to GDP when a music teacher marries his student whom he was tutoring previous to the marriage, but stops billing her for her private lessons? What happens when
a housewife becomes self-employed at her own day-care centre?
3 ‘Services do not contribute to GDP as much as industry because industry produces things—tangible goods.’ Comment on this using what you have learned about the national
income accounts.
4 I bought my house for €100,000. I have just sold it for €200,000, and the estate agent received a 10% commission from the buyer. What is the effect on GDP?
5 Suppose you have the following data on prices and quantities transacted:
Prices (€)
Apples Pears Petrol
2006 1.0 2.0 5.0
2007 1.0 3.0 6.0
Quantities
Apples Pears Petrol
2006 300 100 50
2007 400 150 40
(a) If the economy produced all three (and only these three) goods, compute the nominal GDP in both periods, and real GDP at 2006 prices. What is the rate of
inflation in 2007, as measured by the change in the GDP deflator?
(b) Suppose a CPI is constructed using weights corresponding to quantities produced in 2006. What is the rate of inflation measured by the CPI?
6 Give the three definitions of GDP. Explain whether the following transactions contribute to UK GDP, and, if so, explain how all three definitions apply.
(a) A resident of York purchases a bag of sweets produced in Manchester.
(b) A tourist visiting York purchases a bag of sweets produced in Manchester.
(c) An operator of a Slovakian food chain purchases a large order of sweets produced in Manchester which are sent in the measurement period to Bratislava.
(d) A Manchester businessman purchases a machine to mould the sweets, which is manufactured in the Czech Republic.
(e) A store manager purchases several boxes of sweets from the Manchester sweet company, but stores them in the stock room, where they sit until the beginning
of next year.
(f) The city government of York purchases several boxes of sweets from the Manchester sweet company for its reception for the Lord Mayor.
7 Over the past five years, taxes were about 60% of GDP in Sweden. Yet disposable income over the past five years also amounted to 60% of GDP. How can these numbers be
reconciled?
8 ‘Commuters increase GDP because they send home a large fraction of their earnings.’ Comment.
9 How would the following transactions be recorded in the French balance of payments?
A French resident buys an Austin Mini produced in the United Kingdom.
A French resident purchases stocks in a German corporation from a German bank based in Germany.
A French national living in Switzerland buys stock in a French company from a bank based in Switzerland.
A French resident builds a house in Italy, paying Italian residents to do the job.
A French resident gives money to Greenpeace located in Hamburg, Germany.
A French banker sends a wire transfer of euros to his daughter at the Humboldt University of Berlin.
The same French banker wires euros from his bank account in Berlin to his account in Paris.
A Tunisian worker in Marseilles sends money to his family in Tunis.
Peugeot SA, a French concern, pays dividends to a resident of Finland.
Profits of Owen Corning, a US company, are reinvested in capacity expansion of a factory in Fontainebleau, France.
The Banque de France (a part of the European Central Bank) purchases Danish kroner to prevent the exchange rate (in euro) from falling in Copenhagen.
A French resident of Colmar, a town in Alsace near the German border, smuggles home a stereo purchased in Freiburg (Germany).
10 By definition, we do not know what is in the Errors and Omissions account. Beyond genuine data collection mistakes, we can only speculate what is in there. Table 2.8 shows
large values in the cases of the US, China and, to a lesser extent, the Eurozone. Can you guess what items these entries might include?
Essay Questions
1 Evaluate the following statement: ‘Bringing the underground economy above ground would increase the GDP, but worsen tax receipts and the balance of payments.’
2 In recent years firms in Europe as well as in the USA have begun to ‘outsource’ or divest themselves of many traditional service functions, purchasing them on the market instead.
Companies are increasingly obtaining computer, catering, legal, consulting, and other business services by ordering them from outside, independent companies. What is the effect of
outsourcing on GDP? (Think of the example of a firm which turns its cafeteria into an independent contractor.) How does your answer depend on what the new firm does with its
independence?
3 GDP mixes up everything. It includes ‘goods’ such as apples we eat and theatre shows we enjoy, but also ‘bads’ such as petrol burnt in traffic congestion and burial costs. It
ignores many ‘goods’ as well: the value of a good neighbour or the free time that we can spend watching a sunset. Comment and explore how you would compute gross domestic
happiness?
4 Countries which generate most economic activity from the exploitation of natural resources often experience large fluctuations in their national income and product account results.
Explain why this might be the case, carefully defining the terms value added and costs of production. Why might the national income and product accounts lose their relevance for
these countries? How might the problem be solved?
5 ‘The EU member states don’t need balance of payments statistics. The individual states of the USA get along fine without them.’ Discuss.
1 Sir Richard Stone (1913–1991) of Cambridge University received the Nobel Prize in Economics in 1984 and is generally regarded as the father of national income accounting.
2 Another example of a stock variable is a company’s balance sheet, which measures its financial state at a single point in time, say 31 December of each year. A corresponding example
of a flow variable is an income statement, which records the profit or loss attributed to the firm over a time period, say 1 January to 31 December.
3 As estimated by Charles Bean in ‘Time to rethink the way we measure economic activity’, see the fun-to-read article on http://www.voxeu.org/article/rethinking-measurement-
economic-activity.
4 Problems arise when new goods are introduced (e-readers, tablets, or drones), or existing goods improve in quality (personal computers). National income accountants have devised
procedures to deal with such effects.
5 In previous decades, it was common to update the base year every five or more years. Rapidly changing product mixes and globalization have increased the importance of redefining
base years more frequently. It is now EU national income accounting practice to redefine the base year every year, effectively rendering real GDP a so-called chain standard index. With
this change, growth rates in real GDP are more reliably measured. As a result, some national statistically authorities have even stopped publishing real GDP levels, using an index number
without units instead.
6 Box 5.3 in Chapter 5 provides the derivation of for this formula. If we denote nominal GDP as Y N, real GDP as Y and the GDP deflator as P, we have Y N = PY, which means P = Y
N/Y. Then, using the results from Box 5.3, we have
Δ P/P = Δ Y N/Y N−Δ Y/Y.
7 To see why this formula is an approximation, suppose real GDP increased at rate g and inflation at rateπ. If these rates are expressed as fractions (per cent change divided by 100), then
the rate of nominal growth is given by (1 + g)(1 + π) − 1 = g + π + g π. For g andπsmall, g π ≈ 0, so the rate of growth of nominal GDP is approximately g + π.
8 Chapter 16 discusses some of the most frequently used indicators.
9 The other production costs are mainly land and buildings, financial costs (borrowing from banks and bondholders), and raw materials and intermediate goods which, for the country as a
whole, are frequently imported.
10 There is an important difference between this terminology and that used in the business or popular press, in which ‘investment’ includes the acquisition of existing assets or financial
instruments. Although stocks and bonds are often issued by firms to finance purchases of productive equipment, their simple acquisition or sale does not necessarily imply ‘investment’ in
economics, i.e. the creation of new productive capacity.
11 As we will see in Section 2.4, net exports (X − Z) do not take into account some international income movements that go beyond sales of goods and services.
12 If the net outflow of goods and services is positive, net assets owned by the sector must be increasing; if the net outflow is negative, net assets are decreasing. Technically, increases in
net assets can occur by increases in gross financial assets holding liabilities constant, or by reducing gross liabilities (i.e. paying off existing debt to other sectors) holding assets constant.
13 In practice, tax laws determine financial accounting practices as regards depreciation of machines and other forms of physical capital. Firms are generally allowed to subtract part of the
book value of equipment from their revenues each period when computing taxable profits. It may under- or overstate actual economic depreciation by a wide margin.
14 Net lending thus includes the purchase of stocks, real estate, and other forms of asset purchase.
15 The capital account also accounts for changes in ownership or transfers of certain types of non-financial assets.
16 By definition, the sum of the current accounts of all countries in the world should equal zero. In fact, it is systematically negative, as receipts are ‘omitted’ more often than
expenditures.
17 Later chapters will explore this process in much greater detail. A currency appreciates when its value in terms of other currencies increases. Conversely, if its value decreases, we speak
of a depreciation.
PART II

5 Money, Prices, and Exchange Rates in the Long Run
Part II studies the long run. The long run is what economists mean when they talk about the behaviour of an economy over
decades, rather than over short time spans of quarters or a few years. The long run describes attainable and sustainable
aspects of the national economy, and goes far beyond the short-term perspective of the business cycle fluctuations
described in Chapter 1. The most important of these aspects is the material standard of living enjoyed by present and
future generations.
We begin with economic growth, the most fundamental of all long-term macroeconomic phenomena. Economic growth
is the rate at which the real output of a nation or a region increases over time. As the ultimate determinant of the poverty
or wealth of nations, sustained economic growth is a central aspect of the long run. Next, we look at the labour market,
one of the most important markets in modern economies. In the labour market, households trade time at work for the
ability to purchase goods and services in the goods market. We will see how labour is allocated: where it comes from,
who demands it, and how to think about unemployment.
The last chapter in Part II introduces key nominal variables—money, interest rates, and the exchange rate—which are
generally denoted in nominal terms, pounds or euros or dollars or renminbi. It shows that, in the long run, these nominal
variables are linked to each other in a very particular way. Indeed, the monetary neutrality principle asserts that prices,
nominal exchange rates, and all nominal variables change at the same rate as money is growing, leaving the real side of
the economy unaffected.
The Fundamentals of Economic Growth
3
3.1 Overview
3.2.1 The Economic Growth Phenomenon
3.2.2 The Sources of Growth: The Aggregate Production Function
3.2.3 Kaldor’s Five Stylized Facts of Economic Growth
3.2.4 The Steady State
3.3.1 Savings, Investment, and Capital Accumulation
3.3.2 Capital Accumulation and Depreciation
3.3.3 Characterizing the Steady State
3.3.4 The Role of Savings for Growth
3.3.5 The Golden Rule
3.6.1 Solow’s Decomposition
3.6.2 Capital Accumulation
3.6.3 Employment Growth
3.6.4 Technological Change
3.7.1 The Convergence Hypothesis
3.7.2 Conditional Convergence and Missing Inputs
Summary
The consequences for human welfare involved in questions like these are simply staggering: once one starts to think about them, it is hard to think about anything else.
R. E. Lucas, Jr1
3.1 Overview
Output, as measured by the gross domestic product at constant prices, tends to grow in most countries over time. Is economic growth a universal phenomenon?
Why are national growth rates of the richest economies so similar? Why do some countries exhibit periods of spectacular growth, such as Japan in 1950–1973,
the USA in 1820–1870, Europe after the Second World War, or China and India more recently? Why do others sometimes experience long periods of stagnation,
as China did until the last two decades of the twentieth century? Do growth rates tend to converge, so that periods of above-average growth compensate for
periods of below-average growth? What does this imply for levels of GDP per capita? These questions are among the most important ones in economics, for
sustained growth determines the wealth and poverty of nations.
This chapter will teach us how to think systematically about growth and its determinants. It turns out that the explanation of this fundamental process is rather
simple. Figure 3.1 shows the per capita GDP of the UK since 1920. It is plotted in logarithms because, as explained in Chapter 1, the slope of the curve directly
shows the growth rate. The curve reminds us of important historical events: the build-up of a war economy and the massive destruction and disruption
associated with and following the Second World War. It also reveals the excessive growth of the early 2000s that led to the 2008–2009 crisis. The figure also plots
the trend. The striking observation is that the UK economy has always returned to an overall trend growth rate of about 2%, more or less independent of
historical circumstance, while these circumstances have varied widely over the past two centuries. Fundamental forces seem to be at work that, if left free to
operate, keep an economy growing indefinitely.
Fig. 3.1 GDP per capita in the UK 1920–2015
Using a logarithmic scale, we show the evolution of real, per capita GDP along with a trend that corresponds to a 1.8% growth rate over the period. We see important events:
the Great Depression in the early 1930s, the build-up of a war economy and the sharp fall as a result of the Second World War, the overheated ‘bubble economy’ of the 2000s.
Remarkably, the GDP returns over time to a seemingly immutable growth trend.
Source: Maddison (1995), World Bank.
Indeed, in this chapter we will focus on three main forces or engines of economic growth. First, the capital stock can grow, and a greater capital stock
enables workers to produce more. The capital stock grows when an economy dedicates resources to expanding its productive capacity—we called this
investment in the previous chapter. Second, the working population or labour force can grow, which means that more workers are potentially available for market
production. This growth can arise for many reasons—increases in births two or three decades ago, immigration now, or increased labour force participation by
people of all ages, especially by women. The third reason is the productivity of factors in place, or technological progress. As knowledge accumulates and
techniques improve, workers and the machines they work with become more productive. For both theoretical and empirical reasons, technological progress turns
out to be the ultimate driver of economic growth.
3.2.1 The Economic Growth Phenomenon

Despite setbacks arising from wars, natural disasters, or epidemics, economic growth seems like an immutable economic law of nature. Over the centuries, it has
been responsible for significant, long-run material improvements in the way the world lives. Table 3.1 displays the annual rate of increase in real GDP—the
standard measure of economic output of a geographic entity—for various periods in a number of currently wealthy countries since 1820. (The early data are
clearly rough estimates.) Over almost two centuries, GDP has increased 30 to 60-fold or more, while per capita GDP has increased 12- to 30-fold. Our grandparents
are right to say that we are much better off than they were.
Table 3.1 The Growth Phenomenon
Average rates of growth in GDP (%per annum) Av. Growth GDP per capita 1820–2010 (%per
annum)
1820– 1870– 1913– 1950– 1973– 2001– 1820–
70 1913 50 73 2001 10 2010
Austria 1.4 2.4 0.2 5.2 2.5 2.2 2.0 1.6
Belgium 2.2 2.0 1.0 4.0 2.1 2.0 2.1 1.5
Denmark 1.9 2.6 2.5 3.7 2.0 1.7 2.4 1.5
Finland 1.6 2.7 2.7 4.8 2.4 2.3 2.6 1.8
France 1.4 1.6 1.1 4.9 2.3 1.9 1.9 1.6
Germany 2.0 2.8 0.3 5.5 1.8 1.7 2.2 1.6
Italy 1.2 1.9 1.5 5.5 2.3 1.7 2.1 1.5
Netherlands 1.7 2.1 2.4 4.6 2.4 2.2 2.4 1.4
Norway 2.2 2.2 2.9 4.0 3.4 2.9 2.7 1.9
Sweden 1.6 2.1 2.7 3.7 1.9 1.9 2.3 1.6
Switzerland 1.9 2.5 2.6 4.4 1.2 1.3 2.4 1.7
United 2.0 1.9 1.2 2.9 2.1 2.0 1.9 1.4
Kingdom
Japan 0.4 2.4 2.2 8.9 2.7 2.2 2.6 1.8
United States 4.1 3.9 2.8 3.9 3.0 2.7 3.5 1.7
Source: Maddison (2007) Historical Statistics for the World Economy: 1–2003 AD, OECD.
Box 3.1 China and the Chinese Puzzle of Economic Growth
At the end of the fourteenth century, China probably was the world’s most advanced economy. While Europe was just beginning to recover from centuries
of inward-looking backwardness and relative decline, Chinese society had reached a high degree of administrative, scientific, and economic
sophistication. Innovations such as accounting, gunpowder, the maritime compass, moveable type, and porcelain manufacture are just a few attributable
to the Middle Empire. Marco Polo was one of many famous European traders who tried to break into the Chinese market. According to crude estimates by
economic historian Angus Maddison, fourteenth-century Western Europe and China were on roughly equal footing in terms of market output—and many
experts claim the Chinese were technically more advanced.2 Yet over the next six centuries, standards of living increased 25-fold in Western Europe
compared with only sevenfold in China.
Most of that sevenfold increase in GDP per capita has occurred in China over the last three decades. After adopting far-reaching market economy
reforms in the 1980s, economic growth has averaged a phenomenal 10% per annum since 1990. At this rate, the economy doubles in size every seven
years. If this growth continues, China will easily reach the standard of living of poorer EU countries by 2025.
The Chinese growth phenomenon raises a host of intriguing questions. Why did China stagnate for centuries, while Europe flourished? Why did China
literally explode in the 1990s? It is widely agreed that the Chinese success story would have been impossible without China’s shift from isolationism to a
policy of openness to international trade and foreign direct investment. Almost as a converse proposition, some historians associate the economic
stagnation of China after the fifteenth century with the grounding of 3,500 great sailing ships of the Ming dynasty in 1433, the world’s largest naval
expeditionary fleet under the command of Admiral Zheng He. A policy of ‘inward perfection’, fear of Mongol threats, lack of government funding, and a deep
mistrust of merchant classes which benefited most from the international excursions of the Imperial ‘Treasure Fleet’, all led China to close itself off from
foreign influences, with disastrous consequences. For many economists, this is a warning shot about potential risks of an unquestioning anti-
globalization movement.
The table also reveals that the growth process is not necessarily smooth. We will see that this variation reflects the effect of wars, colonial expansion and
annexation, and dramatic changes in population as well as political, cultural, and scientific revolutions. It also reflects the business cycle. Despite these swings, it
is striking that the overall average growth of GDP per capita is remarkably similar across these countries and over time. Indeed, the last column shows that per
capita growth since 1820 has been 1.5–1.9% per annum.
Small average annual changes displayed in Table 3.1 cumulate surprisingly fast. The advanced economies of the world grow by roughly 2–4% per year. A
growth rate difference of 2% per annum increases a gap by 49% in 20 years, and by 170% in half a century. The recent phenomenal growth successes of China
and India and the troubling slowdowns in Western Europe and Japan show that growth is by no means an automatic birthright. Moreover, fortunes can change:
Box 3.1 reminds us that China was a leading world economy in the fourteenth century, only to fall into half a millennium of decline and stagnation.
3.2.2 The Sources of Growth: The Aggregate Production Function

Growth theory asks how sustained economic growth across nations and over time is possible. Do we produce more because we employ more inputs, or because
the inputs themselves become more productive over time, or both? What is the contribution of each factor? The most important tool we will use to think about
these questions is the production function. The production function relates the output of an economy—its real GDP—to productive inputs. The two most
important productive inputs are the physical capital stock (equipment and structures), represented by K, and labour employed, represented by L. The capital
stock includes factories, buildings, and machinery as well as roads and railroads, electricity, and telephone networks. Employment or labour is the total number of
hours worked in a given period of time. The labour measure L is the product of the average number of workers employed (N) during a period (usually a year) and
the average hours (h) that they work during that period (L = Nh). We speak of person-hours of labour input.3 Symbolically, the production function is written:
(3.1)
The plus (‘+’) signs beneath the two inputs signify that output rises with either more capital or more labour.4
The production function is a useful, powerful, and widely used short-cut. It reduces many and complex types of physical capital and labour input to two. In
microeconomics, the production function helps economists study the output of individual firms. In macroeconomics, it is used to think about the output of an
entire economy. Box 3.2 presents and discusses the characteristics of a widely-used production function, the Cobb–Douglas production function.
Box 3.2 For the Mathematically Minded: The Cobb–Douglas Production Function
The use of mathematics in economics can bring clarity and precision to the discussion of economic relationships. An illustration of this is the notion of a
production function, which formalizes the relationship between inputs (capital and labour) and output (GDP). One particularly well-known and widely-used
example is the Cobb–Douglas production function:
(3.2)
where α is a parameter which lies between 0 and 1, and is called the elasticity of output with respect to capital: a 1% increase in the capital input results in
an α % increase in output.5 Similarly 1 − α is the elasticity of output with respect to labour input. It is easy to see that the Cobb–Douglas production
function possesses all the properties described in the text.
Diminishing marginal productivity
The marginal productivity of capital is given by the derivative of output (Y) with respect to capital (K): ∂Y/∂K = αKα − 1L1 − α = α(L / K)1 − α. Since α < 1, the
marginal product of capital is a decreasing function of K and an increasing function of L. Similarly, the marginal productivity of labour is given by ∂Y / ∂L = (1
− α)(K / L)α, which is increasing in K and decreasing in L.
Constant returns to scale
The Cobb–Douglas function has constant returns to scale: for a positive number t, which can be thought of as a scaling factor:
(3.3)
Intensive form
The intensive form of the Cobb–Douglas production function is obtained by dividing both sides of (3.2) by L, which is the same as setting t = 1/L in
equation (3.3), to obtain:
(3.4)
where k = K/L and y = Y/L are the intensive form measures of input and output defined in the text. Since α < 1, the intensive form production is indeed well-
represented by Figure 3.2.
The production function is a technological relationship. Taken alone, it does not reflect the profitability of production, and it has nothing to do with the quality
of life or the desirability of work. It is meant to capture the fact that goods and services are produced using factors of production: here, equipment and hours of
labour. In the following, we describe some basic properties that are typically assumed for production functions.
Marginal productivity
Consider an economy producing output with workers and a stock of capital equipment. Then imagine that a new unit of capital—a new machine—is added to the
capital stock, raising it by the amount ∆K, while holding labour input constant.6 Output will also rise, by ∆Y. The ratio ∆Y/∆K, the amount of new output per unit
of incremental capital, is called the economy’s marginal productivity of capital. Now imagine repeating the experiment, adding capital again and again to the
existing stock, always holding labour constant. Should we expect output to increase by the same amount for each additional increment of capital?
Generally, the answer is no. As more and more capital is brought into the production process and the number of hours worked remains the same, each new
piece of equipment is allocated less of workers’ time. As one might expect, the increases in output become smaller and smaller. This is called the principle of
diminishing marginal productivity.7 It is represented in Figure 3.2, which shows how output rises with capital, holding labour unchanged. The flattening of
the curve illustrates the principle since the slope of the curve is equal to the economy’s marginal productivity.
It turns out that the principle of diminishing marginal productivity also applies to the labour input. Increasing the employment of person-hours will raise
output; but output from additional person-hours declines as more and more labour is being applied to a fixed stock of capital.
Fig. 3.2 The Production Function
Holding labour input L (the number of hours worked) unchanged, adding to the capital stock K (available productive equipment) enables an economy to produce more, but with
smaller and smaller increments.
Returns to scale
Output increases when either inputs of capital or labour increase. But what happens if both capital and labour increase in the same proportion? Suppose, for
example, that the inputs of capital and labour were both doubled—increased by 100%. If output doubles as a result, the production function is said to have
constant returns to scale. If a doubling of inputs leads to more than a doubling of output, we observe increasing returns to scale. Decreasing returns
is the case when output increases by less than 100%. It is believed that decreasing returns to scale are unlikely. Increasing returns, in contrast, cannot be ruled
out, but we will ignore this possibility for the moment. In fact, the bulk of the evidence points in the direction of constant returns to scale.
With constant returns we can think of the link between inputs and output—the production function—as a zoom lens: as long as we scale up the inputs, so
does the output. In this case, an attractive property of constant returns production functions emerges: output per hour of work—the output–labour ratio (Y/L)
—depends only on capital per hour of work—the capital–labour ratio (K/L). This simplification allows us to write the production function in the following
intensive form:8
(3.5)
where y = Y/L and k = K/L. The output–labour ratio Y/L is also called the average productivity of labour: it says how much, on average, is being produced with
one unit (one hour) of work.9 The capital–labour ratio K/L measures the capital intensity of production.
The intensive-form production function is depicted in Figure 3.3. Because of diminishing marginal productivity, the curve becomes flatter as the capital–labour
ratio increases. The intensive-form representation of the production function is convenient because it expresses the average productivity of labour in an
economy as a function of the average stock of capital with which that labour is employed. In a world of constant returns, the intensive form production function
says that standards of living do not depend on the absolute size of an economy. Indeed, Ireland, Singapore, and Switzerland have matched or exceeded the per
capita GDP of the USA, the UK, or Germany.
Fig. 3.3 The Production Function in Intensive Form

The production function shows that the output–labour ratio y grows with the capital–labour ratio k. Its slope is the marginal productivity of labour since with constant returns to
scale ∆Y/∆K = ∆y/∆k. The principle of declining marginal productivity implies that the curve becomes flatter as k increases.
3.2.3 Kaldor’s Five Stylized Facts of Economic Growth

At this point it will prove helpful to look at the data: How have inputs and outputs in real-world economies changed over time? In 1961, the British economist
Nicholas Kaldor (1908–1986) studied economic growth in many countries over long periods of time and isolated several stylized facts about economic growth
which remain valid to this day. Stylized facts are empirical regularities found in the data. Kaldor’s stylized facts will organize our discussion of economic growth
and restrict our attention to theories which help us to think about it, just as a police detective uses clues to limit the number of possible suspects in a criminal
investigation.
The first of Kaldor’s stylized facts concerns the behaviour of output per person-hour and capital per person-hour.
Stylized Fact No. 1: both output per capita and capital intensity keep increasing
The most remarkable aspect of the growth phenomenon is that real GDP seems to grow without bound. Yet labour input, measured in person-hours of work ( L),
grows much more slowly than both capital (K) and output (Y). Put differently, both average productivity ( Y/L) and capital intensity (K/L) keep rising. Because
income per capita (Y/N) is closely related to average productivity or output per hour of work (Y/L), economic growth implies a continuing increase in material
standards of living.10 Figure 3.4 presents the evolution of the output–labour and capital–labour ratios in three important industrial economies.
Fig. 3.4 The Output–Labour and Capital–Labour Ratios in Three Countries (1820–2015)
Output–labour and capital–labour ratios are continuously increasing. Growth accelerated in the USA in the early twentieth century, and after 1950 in Japan and the UK.
Sources: Maddison (1995); Penn-World Tables 8.1, AMECO, chained.
Stylized Fact No. 2: the capital–output ratio exhibits little or no trend over time
As they grow in a seemingly unbounded fashion, the capital stock and output tend to track each other. As a consequence, the ratio of capital to output ( K/Y)
shows little or no systematic trend. This is apparent from Figure 3.4, while Table 3.2 shows that it is only approximately true. For example, while output per hour in
the USA has increased by roughly 600% since 1913, the ratio of capital to output actually fell slightly over the same period. At any rate, the capital–output ratio
may not be exactly constant, but it is far from exhibiting the steady, unrelenting increases in average productivity and capital intensity described in Stylized Fact
No. 1.
Table 3.2 Capital–Output Ratios (K/Y), 1913–2009
1913 1950 1973 1992 2009

France 1.6 1.7 1.8 2.3 2.6
Germany 2.3 2.1 2.4 2.3 2.6
Japan 1.0 1.8 1.7 3.0 3.9
UK 1.0 1.1 1.7 1.8 1.9
USA 2.9 2.3 2.1 2.4 2.3
Source: Maddison (1995); 2009 data from Economic Outlook, OECD.
Stylized Fact No. 3: hourly wages keep rising

The long-run increases in the ratios of output and capital to labour (Y/L and K/L) mean that, over time, an hour of work produces ever more output. Simply put,
workers become more productive. It stands to reason that their wages per hour should also rise (the theoretical reasoning will be outlined in Chapter 4), and this
is indeed the case. Growth evidently delivers ever-increasing living standards for workers.
Stylized Fact No. 4: the rate of profit is trendless

Note that the capital–output ratio (K/Y) is just the inverse of the average productivity of capital (Y/K). The absence of a clear trend for the capital–output ratio
(K/Y) implies that average capital productivity too is trendless: over time, the same amount of equipment delivers about the same amount of output. It is to be
expected therefore that the rate of profit does not exhibit a trend either. This stands in sharp contrast with labour productivity, whose secular increase allows a
continuing rise in real wages. Yet income flowing to owners of capital has increased, but only because the stock of capital itself has increased. Indeed, with a
stable rate of profit, income from capital increases proportionately to the capital stock.
Stylized Fact No. 5: the relative income shares of GDP paid to labour and capital are trendless
We just saw that incomes from labour and capital increase secularly. Surprisingly perhaps, it turns out that they also tend to increase at about the same rate, so
that the distribution of total income (GDP) between capital and labour has been relatively stable. In other words, the labour and capital shares have no long-run
trend. We will have to work a little harder to explain this remarkable fact.
3.2.4 The Steady State

Stylized facts are not meant to be literally true at all times, certainly not from one year to the other. Instead, they highlight central tendencies in the data. As we
study growth, we are tracking moving targets, variables that keep increasing all the time, apparently without any upper limits. Thinking about moving targets is
easier if we can identify stable relationships among them. This is why Kaldor’s stylized facts will prove helpful. Another example of this approach is given by the
evolution of GDP: it seems to be growing without bounds, but could its growth rate be roughly constant? The answer is yes, but only on average, over five or 10
years or more. In Chapter 1, we noted the important phenomenon of business cycles, periods of fast growth followed by periods of slow growth or even declining
output. As we look at secular economic growth, we are not interested in business cycles. We ignore shorter-term fluctuations—compare Figures 1.5 and 1.6 in
Chapter 1—and focus on the long run.
This is why it is useful to try and imagine how things would look if there were no business cycles at all. Such a situation is called a steady state. Its central
feature is that certain variables, such as GDP growth rates, or ratios of key variables, such as the capital–output ratio or the labour income share—are constant.
Just as the stylized facts are not to be taken literally, think of the steady state as the long-run average behaviour that we never reach, but fluctuate about. From
the perspective of 10 years ago, we thought of today as the long run, but now we can see all the details that were unknown back then. Given that modern GDPs
double every 10–30 years, a temporary boom or recession which shifts today’s GDP by one or two percentage points amounts to little in the greater order of
things, the powerful phenomenon of continuous long-run growth. Steady states—and stylized facts—are not just convenient ways of making our lives simpler;
they are essential tools for distinguishing the forest from the trees.
This section presents the Solow growth model, first proposed by Nobel Prize Laureate Robert Solow of the Massachusetts Institute of Technology.
3.3.1 Savings, Investment, and Capital Accumulation

Kaldor’s first stylized fact highlights a relationship between output per hour and capital per hour. This link is in fact predicted by the production function in its
intensive form. It suggests that a good place to start if we want to explain economic growth is to understand why and how the capital stock rises over time. We
will thus first study how the savings by households—foregone consumption—is transformed into investment in capital goods, which causes the capital stock to
grow.
The central insight is delivered by the circular flow diagram presented in Chapter 2 (Figure 2.3). GDP represents income to households, either to workers as
wages and salaries, or to owners of firms as distributed profits. Households and firms save part of their income. These savings flow into the financial system—
banks, stock markets, pension funds, etc. The financial system channels these resources to borrowers: firms, households, and the government. In particular, firms
borrow—including from their own savings—to purchase capital goods used in production. This expansion of productive capacity, in turn, raises output, which
then raises future savings and investment, and so on. This is the machinery behind the growth phenomenon.
To keep things simple, let us first assume that the size of the population, the labour force, and the numbers of hours worked are all constant. At this stage, we
ask some fundamental questions: can capital accumulation proceed without bound? Does more saving mean faster growth? And since saving means
postponing consumption, is it always a good idea to save more?
3.3.2 Capital Accumulation and Depreciation

Let us start from the national accounts of Chapter 2. Identity (2.7) shows that investment (I) can be financed either by private savings by firms or households (S),
by net savings of the government (the consolidated budget surplus, or T − G), or the net savings of foreigners (which is roughly net exports, or Z − X):
(3.6)
As a description of the long-run or a steady state, suppose that the government budget is in balance (T = G), and the current account surplus equals zero (Z = X).
In this case, the economy’s capital stock is ultimately financed by savings of resident households. 11 More precisely, we reach the conclusion that, in the steady
state, I = S. Investment expenditures are financed entirely by domestic savings. This is a first explanation of the growth phenomenon: we save, we invest, we
grow. As a first approximation, let s be the fraction of GDP which households save to finance investment. Equality of investment and saving implies I = sY, so
dividing by L we have:
(3.7)
This relationship is shown in Figure 3.5 as the saving schedule. It expresses national savings as a function of national output and income. The saving
schedule lies everywhere below the production function because saving is just a fraction of GDP.
Fig. 3.5 The Steady State

With labour assumed constant, the capital–labour ratio stops changing when investment is equal to depreciation. This occurs at point A, the intersection between the saving
schedule sf (k) and the depreciation line δk. The corresponding output–labour ratio is determined by the production function f (k) at point B. When away from point A, the economy
moves towards its steady state. Starting at k1 below the steady state, investment (point C) exceeds depreciation (point D) and the capital–labour ratio will increase until it
reaches its steady-state level . Starting at k2 above the steady state, the process works in reverse and capital per worker declines.
We next distinguish between gross investment, the amount of money spent on new capital, and net investment, the increase in the capital stock. Gross
investment represents new additions to the physical capital stock, but it does not represent the net change of the capital stock because, over time, previously
installed equipment depreciates—it wears out, loses some of its economic value, or becomes obsolescent. The fraction of the capital stock that is routinely lost
each period is called depreciation, and the proportion of the total stock lost each period δ is called the depreciation rate. The depreciation rate for the
overall economy is fairly stable and will be taken as constant. Logically, this implies that the more capital is in place, the more depreciation will occur. Because it is
proportional to the capital stock, depreciation is represented in Figure 3.5 by a ray from the origin, the depreciation line, with slope δ .
Whenever gross investment exceeds depreciation, net investment is positive and the capital stock rises. If gross investment is less than depreciation, the
capital stock falls. While it may seem odd to imagine a shrinking capital stock, it is not uncommon to observe this in declining industries or regions. Net
investment is therefore:
(3.8)
or equivalently, written in intensive form:
We see that the net accumulation of capital per unit of labour is positively related to the savings rate s and negatively related to the depreciation rate δ . How
about the link between capital accumulation ∆k and capital intensity k? It is ambiguous. On the one hand, a larger capital stock means a higher income level (y =
f(k)), and therefore more savings and more gross investment. On the other hand, more capital means more depreciation and less net investment. This ambiguity is
a central issue in the study of economic growth and will be addressed in the following sections.
3.3.3 Characterizing the Steady State

Let us summarize what we have done up to now. The production function (3.5) relates an economy’s output to inputs of capital and labour. Its intensive form,
presented in Figures 3.4 and 3.5, relates the output–labour ratio to the capital–labour ratio. According to equation (3.8), capital accumulation is also driven by the
output–labour ratio. Putting all these pieces together, we find that capital accumulation (∆ k) is determined by stock of capital already accumulated up to that
point (k):
(3.9)
In Figure 3.5, ∆k is the vertical distance between the savings schedule sf(k) and the depreciation line δk. It represents the net change in the capital stock per unit
of labour input in the economy. The sign of ∆ k tells us where the economy is heading. When ∆k > 0, the capital stock per capita is rising and the economy is
growing, since more output can be produced. When ∆k < 0, the capital stock per capita and output per capita are both declining. At the intersection of the saving
schedule and the depreciation line (point A), gross investment and depreciation are equal, so the capital–labour ratio (point B) no longer changes. The capital
stock is thus similar to the level of water in a bathtub when the drain is slightly open: gross investment is like the water running through the tap, while
depreciation represents the loss of water through the drain. The newly accumulated capital exactly compensates that lost to depreciation—the water flows into
the bathtub at the same speed as it leaks out. This is the steady state.
Regardless of where it starts, the economy gravitates to the steady state and stays there. Suppose, for instance, that the economy is to the left of the steady-
state capital–labour ratio say at the level k1.12 Figure 3.5 shows that gross investment sf(k1) at point C exceeds depreciation δk1 at point D. According to (3.9),
the distance CD represents net investment, the increase in the capital–labour ratio k, which rises towards its steady-state level .
Can the capital stock proceed beyond going all the way to say, k2? It turns out that it cannot. As the economy gets closer to point A, net investment
becomes smaller and smaller as the steady state is approached; it equals nil precisely when k = . This process, called catch-up, is represented in Figure 3.6 and
illustrated in Box 3.3. To see how the economy behaves when capital is above its steady state, consider k2 > . Gross investment sf(k2) is less than depreciation
δk2, so the capital–labour ratio declines, and the economy moves leftward towards , its stable resting point.
Fig. 3.6 Catching Up

Catching up occurs when a country starts below its steady state GDP (lower curve) and embarks on a faster growth path, approaching the steady state (the upper curve) from
below.
Box 3.3 Growth Miracles Eventually Come to an End
In each period, we see ‘growth miracles’, countries that go through a period of rapid growth. The USA in the late nineteenth century and Australia in the
early twentieth century went through this catch-up phase. For historical reasons, a country with a low capital–labour ratio unleashes the process
described in Figure 3.6. The USA and Australia were initially poor colonies. Table 3.3 shows some more recent examples. Europe recovered in the 1960s
from wartime destruction. The next wave included Japan, also recovering from the war and economic backwardness. Then came China, a very poor
country after centuries of economic and political repression. Catch-up started once it adopted fundamental economic reforms and opened to international
trade in the late 1970s under the leadership of Deng Xiaoping. More recently, India and Vietnam have joined in. Much of Latin America has also started to
catch up over the last decade or two. Africa, finally, seems about to extricate itself from poverty as well.
These are remarkable episodes. A 10% annual growth rate means that standards of living double every seven years. Why do these growth explosions
take off, and how long do they last? Growth theory tells us that catch-up occurs when people save and firms borrow these savings to invest and raise the
capital stock, and when firms can adopt state-of-the-art technology. This requires adequate financial systems and economic openness. Importantly, it
requires a reliable system of property rights, which in turn requires effective legal systems and the absence of armed conflict. As for the latter question,
Figure 3.6 confirms the answer given by theory: until the economy reaches the steady state. This happened to Europe in the 1970s, to Japan a decade
later. China is still relatively poor and should keep growing for a decade or two, albeit at a decreasing rate.
Table 3.3 Average Annual Growth Rates of GDP per Capita since 1960
1961–1970 1971–1980 1981–1990 1991–2000 2001–2010

France 4.6 3.1 1.8 1.5 0.5
Japan 8.5 3.3 4.1 0.9 0.8
China 2.4 4.4 7.8 9.3 9.8
India 4.1 0.8 3.4 3.6 6.3
Vietnam 2.3 5.9 6.1
Source: World Development Indicators, The World Bank.
3.3.4 The Role of Savings for Growth

We now show that the more a country saves, the more it invests; the more it invests, the higher is its steady-state capital–output ratio; and the larger its capital–
output ratio, the higher its output–labour ratio will be in the steady state. This long-run proposition implies that countries that save and invest a lot have high per
capita incomes. Is this true? Figure 3.7(a) looks at investment rates and GDP per capita for the whole world and is suggestive of such a relationship. For example,
the poor countries of Africa typically invest little, in contrast to the richer countries of Europe and Asia. Yet, the link is not strong enough to explain everything.
Similarly, Figure 3.7(b) shows that the investment rate can account for differences in economic growth between countries, but only weakly. Our story is too
simple and we will need more flesh on the bare theoretical bones that we have just assembled.
Fig. 3.7 Investment, GDP per Capita, and Real GDP Growth
For the available data from a sample of 165 countries in the Penn World Table dataset, the correlation coefficient between the average investment rate (the ratio of investment to
GDP) and the average per capita GDP is positive but modest (0.44). The correlation of the investment rate in the countries with real GDP growth over the period is also positive
(0.41). High investment rates are clearly correlated with per capita GDP and growth, but other influences play an important role as well.
Source: Penn World Table Version 8.1: Feenstra, Inklaar and Timmer (2015) (http://www.ggdc.net/pwt); authors’ calculations.
Still, at this stage, we reach an important result: savings and investment are more closely linked to steady-state levels of output rather than their growth rates .
This means that nations which save more should have higher standards of living in the steady state, but they do not grow faster, indefinitely. This is an
important result which may seem counterintuitive. Yet it is implied by the simple Solow model.
To see this, consider Figure 3.8, which illustrates the effect of an increase in the savings rate from s to s ´. The savings–investment schedule shifts upwards
while the production function schedule remains unchanged. The new steady-state output–labour and capital–labour ratios are both higher at point B than they
were at point A beforehand. It will take time, however, for the economy to reach that new steady state. At the initial steady-state position (point A), an increase in
the saving state causes gross investment to rise. Since depreciation is unchanged, net investment must be positive. The capital–labour ratio starts rising, which
raises the output–labour ratio. This will go on until the new steady state is reached at point B. During this interim period, therefore, growth is higher, which can
give the impression that higher investment rates cause higher economic growth. The boost is only temporary; once the steady state has been reached, no further
growth effect can be expected from a higher savings rate. We still need a story to explain growth in output per capita. This is the story told in Sections 3.4 and
3.5.
Fig. 3.8 An Increase in the Savings Rate

An increase in the savings rate raises capital intensity (k) and the output–labour ratio (y).
It may be surprising that increased savings do not affect long-run growth. That result is an implication of diminishing returns. An increase in savings causes
the capital stock to rise, but this also means that more capital depreciates and thus needs to be replaced. Increasing amounts of gross investment are needed just
to keep the capital stock constant at its higher level. Yet the resources for that increased investment are not sufficiently forthcoming because the marginal
productivity of capital decreases. Further additions to the capital–labour ratio yield smaller and smaller increases in income, and therefore in savings and
investment, while depreciation rises proportionately with the capital stock. Put simply, decreasing marginal productivity implies that, at some point, saving more
is simply not worth it.
3.3.5 The Golden Rule

Figure 3.8 contains an important message: to become richer, you need to save and invest more. But is being richer—in the narrow sense of accumulating capital
goods—always necessarily better? Saving requires the sacrifice of giving up some consumption today against the promise of higher income tomorrow, but does
saving more today always mean more consumption tomorrow? The answer is not necessarily positive. To see why, note that in the steady state, when the capital
stock per capita is k, savings equal depreciation and the part of income that is not saved—steady state consumption—is given by:
(3.10)
In Figure 3.9, consumption per capita in the steady state is the vertical distance between the production function and the depreciation line.13 If we could choose
any saving rate we wanted, we could effectively pick any point of intersection of the savings schedule with the depreciation line, and therefore any level of
consumption we so desired. Figure 3.9 shows that steady-state consumption is highest at the capital stock for which the slope of the production function is
parallel to the depreciation line.14 The corresponding steady-state capital–labour ratio is indicated as . Now remember that the slope of the production function
is the marginal productivity of capital (MPK) while the slope of the depreciation schedule is the rate of depreciation δ . We have just shown that the maximal level
of consumption is achieved when the slope of the production function, the marginal product of capital, is equal to the slope of the depreciation line:
(3.11)
Fig. 3.9 The Golden Rule

Steady-state consumption (expressed in per capita terms) is the vertical distance between the production function and the depreciation line. It is at a maximum at point A
corresponding to the situation where the slope of the production function, the marginal productivity of capital, is equal to δ, the slope of the depreciation line.
This condition is called the golden rule, and can be thought of as a recipe for achieving the highest consumption per capita, given existing technological
capabilities. In this case, with no population growth and no technical progress, the golden rule states that the economy maximizes steady-state consumption
when the marginal gain from an additional unit of GDP saved and invested in capital (MPK) equals the depreciation rate.
What are the consequences of ‘disobeying’ the golden rule? If the capital–labour ratio exceeds , too much capital has been accumulated, and the MPK is
lower than the depreciation rate δ . By reducing savings today, an economy can actually increase per capita consumption, both today and in the future. This
looks like a free lunch, and indeed, it is one. We say that the economy suffers from dynamic inefficiency. Dynamically inefficient economies simply save and
invest too much—in order to maintain a capital stock which is simply too large—and consume too little as a result.
A different situation arises if the economy is to the left of . Here, steady-state income and consumption per capita may be raised by saving more, but not
immediately; consumption can be increased only in the long run, after the adjustment has occurred. No free lunch is immediately available. More consumption in
the future—in the steady state—must be ‘earned’ by increased saving and reduced consumption today. Moving towards from a position on the left requires
current generations to sacrifice so future generations can enjoy more consumption resulting from more capital and income in the steady state. An economy in
such a situation is called dynamically efficient because it is not possible to do better without paying the price for it.
The difference between dynamically efficient and inefficient savings rates is illustrated in Figure 3.10, which shows how we move from one steady state to
another one with higher consumption. In the dynamically inefficient case (a), it is possible to permanently raise consumption by consuming more now and during
the transition to the new steady state. In the dynamically efficient case (b), a higher steady-state level consumption is not free and implies a transitory period of
sacrifice.
Fig. 3.10 Raising Steady-State Consumption

In a dynamically inefficient economy (a), it is possible to permanently raise consumption by reducing saving. In a dynamically efficient economy (b), higher future consumption
requires early sacrifices.
Dynamic inefficiency—in short, too much capital—may have characterized some of the centrally planned economies of Central and Eastern Europe. It may also
characterize today’s China after years of savings in excess of 50% of GDP. We say ‘may’ because the proof that an economy is dynamically inefficient lies in
showing that the marginal productivity of its capital lies below the depreciation rate, and neither of these is easily measurable. What we do know is that leaders of
planned economies of the last century often boasted about their economies’ high investment rates, which were in fact considerably higher than in the capitalist
West. Yet overall standards of living were considerably lower than in market economies, and consumer goods were in notoriously short supply.
In dynamically efficient economies, future generations would benefit from raising saving today, but those currently alive would lose. Should governments do
something about it? Since it would represent a transfer of resources from current to future generations, there is no simple answer. It is truly a deep political choice
with no solution since future generations don’t vote today. A number of factors influence savings, such as taxation, health and retirement systems, cultural
norms, and social custom. Importantly, too, saving and investment are influenced by political conditions. Political instability and especially wars, civil or
otherwise, can lead to destruction and theft of capital, and hardly encourage thrifty behaviour. As we discuss in Section 3.7, in many of the world’s poorest
countries, property rights are under constant threat or simply non-existent.

A major shortcoming of the previous section is that it does not explain permanent, sustained growth, our very first stylized fact. Capital accumulation, we saw,
can explain high living standards and growth during the transition to the steady state but the law of diminishing returns ultimately kicks in. Clearly, some crucial
ingredients are missing. One of them is population growth, more precisely, growth in the employed labour force. This section shows that, once we introduce
population growth, sustainable long-run growth of both output and the capital stock is possible.
Recall that labour input (person-hours) grows either if the number of people at work increases, or if workers work more hours on average. Later on in this
chapter, we will see that the number of hours worked per person has declined steadily over the past century and a half. Figure 3.11 shows that, despite this fact,
employment has been rising, either because of natural demographic forces (the balance between births and deaths), increasing labour force participation
(especially by women) or immigration. Overall, more people are at work but they work shorter hours, so the balance of effects is ambiguous. Because the number
of hours worked per person cannot and does not rise without bound, we will treat it as constant. Then any change in person-hours is due to exogenous changes
in the population and employment, and output per person-hour changes at the same rate as output per capita.
Even though population and employment are growing, the fundamental reasoning of Section 3.3 remains valid: the economy gravitates to a steady state at
which the capital–labour and output–labour ratios (k = K/L and y = Y/L) stabilize. If L grows at an exogenous rate n, y and k will remain constant if output Y and
capital K also grow at rate n. The relentless increase in the labour input will be the driver of growth in this case. Quite simply, if income per capita is to remain
unchanged in the steady state, income must grow at the same rate as the number of people.
Fig. 3.11 Population and Employment in the Eurozone and the USA, 1960–2016
Population of working age (between 15 and 64) has been growing both in the USA and the countries of the Eurozone, the part of the European Union that uses the euro.
Employment has also been growing, albeit less fast in Europe. Note the jump in the euro area in 1991, the year after German unification.
The role of saving and capital accumulation remains the same as in the previous section, with only a small change of detail. The capital accumulation condition
(3.9) now becomes:15
(3.12)
The difference is that, for the capital–labour ratio to increase, gross investment must not only compensate for depreciation, it must also provide new workers with
the same equipment as those already employed. This process is called capital-widening and it explains the presence of n in the last term of the equation.
The situation is presented in Figure 3.12. The only difference with Figure 3.5 is that the depreciation line δ k has been replaced by the steeper capital-
widening line (δ + n)k. The fact that the capital-widening line is steeper than the depreciation line captures the greater need to save when more workers are
being equipped with productive capital. The steady state occurs at point A1, the intersection of the saving schedule and the capital-widening line. At this
intersection , savings are just enough to cover the depreciation and the needs of new workers, so ∆k = 0.
Fig. 3.12 The Steady State with Population Growth
The capital–labour ratio remains unchanged when saving and investment are equal to (δ + n1)k. This occurs at point A1, the intersection between the saving schedule sf (k) and
the capital-widening line (δ + n1)k. An increase in the rate of growth of the population from n1 to n2 is shown as a counter-clockwise rotation of the capital-widening line. The
steady-state capital–labour ratio declines from 1 to 2.
The role of population growth can be seen by studying the effect of an increase in the rate of population growth, from n1 to n2. In Figure 3.12 the capital-
widening line becomes steeper and the new steady state at point A2 is characterized by a lower capital–labour ratio . This result makes sense, since we assume
that savings behaviour is unchanged and yet we need more gross investment to equip the constant inflow of new workers. The solution is to provide each
worker with less capital. Of course, a lower implies a lower output–labour ratio f( ). The Solow model with population growth implies that, all other things being
equal, countries with a rapidly growing population will tend to be poorer than countries with lower population growth. Box 3.4 examines whether it is indeed the
case that high population growth lowers GDP per capita.
At what level of investment does an economy with population growth maximize consumption per capita? Because the number of people who are able to
consume and save is growing continuously, the golden rule must be modified accordingly. Following the same reasoning as in Section 3.3, we note that steady-
state investment per capita is (δ + n) , consumption per capita is given by f( ) – (δ + n) . Proceeding as before, it is easy to see that consumption is at a
maximum when:
Box 3.4 Population Growth and GDP per capita
Figure 3.13 plots GDP per capita in 2008 and the average rate of population growth over the period 1960–2009. The figure could be seen as confirming
the negative relationship predicted by the Solow growth model—and might even be interpreted as support for the hypothesis that population growth
impoverishes nations. Thomas Malthus, a famous nineteenth-century English economist and philosopher, claimed exactly this: population growth causes
poverty. He argued that a fixed supply of arable land could not feed a constantly increasing population and that population growth would ultimately result in
starvation. He ignored technological change, in this case the green revolution which significantly raised agricultural output in the last half of the twentieth
century. As we confirm in Section 3.5, technological change can radically alter the outlook for growth and prosperity.
Yet the pseudo-Malthusian view has been taken seriously in a number of less-developed countries, which have attempted to limit demographic growth.
The most spectacular example is China, which pursued a one-child-only policy for decades. At the same time, we need to be careful with simple
diagrams depicting relationships between two variables. Not only do other factors besides population growth influence economic growth, but it may well
be that population growth is not exogenous. Figure 3.13 could also be read as saying that as people become richer, they have fewer children. There exists
a great deal of evidence in favour of this alternative interpretation.
Fig. 3.13 Population Growth and GDP per Capita, 1960–2009

The figure reports data on real GDP per capita and average population growth for 190 countries over almost a half century. The plot indicates a discernible negative association
between GDP per capita and population growth, especially when the rich oil-producing countries (United Arab Emirates, Qatar, Kuwait, Bahrain, Oman, Brunei, and Saudi Arabia)
are excluded. The sharp population growth observed in these countries is largely due to immigration.
Source: Alan Heston, Robert Summers and Bettina Aten, Penn World Table Version 7.0, Center for International Comparisons of Production, Income and Prices at the University of
Pennsylvania, May 2011.
(3.13)
The ‘modified’ golden rule equates the marginal productivity of capital with the sum of the depreciation rate δ and the population growth rate n. The intuition
developed here continues to apply: the marginal product of an additional unit of capital (per capita) is set to its marginal cost, which now includes not only
depreciation, but also the capital-widening investment necessary to equip future generations with the same capital per head as the current generation. A growing
population will necessitate a higher marginal product of capital at the steady state. The principle of diminishing marginal productivity implies that the capital–
labour ratio must be lower. Consequently, output per head will also be lower.
Taking population growth into account gives one good reason why output and the capital stock can grow permanently, and at the same rate. While this satisfies
Kaldor’s second stylized fact, the picture remains incomplete: so far, we conclude that the capital–labour and output–labour ratios are constant in the steady
state. That standards of living are not rising is inconsistent with Kaldor’s first stylized fact and the data reported in Table 3.1. We still need to account for the fact
that per capita income and capital stock grow, and grow at the same rate.
As it turns out, we have ignored technological or technical progress. Over time, increased knowledge and better, more sophisticated techniques make workers
and the equipment they work with more productive. With a slight alteration, our framework readily shows how technological progress can explain ever-rising
living standards. To do so, once more, we reformulate the aggregate production function introduced in (3.1). Technological progress means that more
output can be produced with the same quantity of equipment and labour. The most convenient way to think of technology is as an additional factor of
production. In the production function, we add a measure of the state of technology, A, which raises output at given levels of capital stock and employment:
(3.14)
When A increases, Y rises, even if K and L remain unchanged. For this reason, A is frequently called total factor productivity. Yet A is not really a factor of
production, but rather a method or technique of production. No firm pays for knowledge; each firm just benefits from it. It is best thought of as ‘best practice’,
which is assumed to be available freely to all. Innovations of all sorts continuously raise productivity. At this point, it will be convenient to assume that A
increases at a constant rate a, without trying to explain how and why. Technological progress, which is the increase in A, is therefore considered as exogenous.
It turns out that it is possible to relate our analysis to previous results in this chapter in a straightforward way. First, we modify (3.14) to incorporate technical
progress in the following particular way:
(3.15)
In this formulation, technological progress acts directly on the effectiveness of labour. For this reason it is sometimes called labour-augmenting technical
progress.16 An increase in A of, say, 10% has the same impact as a 10% increase in employment, even though the number of hours worked hasn’t changed. The
term AL is known as effective labour (or labour in efficiency units) to capture the idea that, with the same equipment, one hour of work today produces more
output than before because A is higher. Effective labour AL grows for two reasons: (1) more labour L, and (2) greater effectiveness A. For this reason, the rate of
growth of AL is now given by a + n.
Now we change the notation a little bit. We redefine y and k as ratios of output and capital relative to effective labour: y = Y/AL, k = K/AL. Once this is done, it
is only a simple step to recover the now-familiar production function in intensive form, y = f(k).17 Not surprisingly, the ratio of capital to effective labour evolves
almost exactly as before:
(3.16)
The reasoning is the same as when we introduced population growth. There we noted that, to keep the capital–labour ratio K/L constant, the capital stock K must
rise to compensate for both depreciation (δ ) and population growth (n). Now we find that, to keep the capital–effective labour ratio k = K/AL constant, the capital
stock K must also rise to keep up with workers’ enhanced effectiveness ( a). So k will increase if saving sf(k), and hence gross investment, exceeds the capital
accumulation needed to make up for depreciation δ , population growth n, and increased effectiveness a. From there on, it is a simple matter of modifying Figure
3.12 to draw Figure 3.14. The steady state is now characterized by constant ratios of capital and output to effective labour (y = Y/AL and k = K/AL).
Constancy of these ratios in the steady state is a very important result. Indeed, if Y/AL is constant, it means that Y/L grows at the same rate as A. If the
average number of hours remains unchanged, then income per capita must grow at the rate of technological progress, a. In other words, we have finally
uncovered the explanation of Kaldor’s first stylized fact: the continuous increase in standards of living is due to technological progress. Since K/AL is also
constant, the capital stock per capita also grows in the steady state at the same rate as A, which is Kaldor’s second stylized fact.
Fig. 3.14 The Steady State with Population Growth and Technological Progress
In an economy with both population growth and technological progress, inputs and output are measured in units per effective labour input. The intensive form production function
inherits this property. The slope of the capital accumulation line is now δ + a + n, where a is the rate of technological progress. The steady state occurs when investment is
equal to (δ + a + n)k (point A), which is the intersection of the saving schedule sf (k) with the capital-widening line (δ + a + n)k. At the steady-state A, output and capital increase
at the rate a + n, while GDP per capita increases at the rate a.
Figure 3.15 summarizes these results. Because of diminishing marginal productivity, capital accumulation alone cannot sustain growth. Population growth
explains GDP growth, but not the sustained increase of standards of living over the centuries. Technological progress is essential for explaining economic growth
in the long run. Rather than creating misery in the world, it turns out to be central to improvements in standards of living.
Note that an increase in the rate of technological progress, a, makes the capital-widening line steeper than before. In Figure 3.14 this would imply lower steady-
state ratios of capital and output to effective labour. This does not mean that more rapid technological progress is a bad thing, because what counts is income
per person (Y/L), not income per unit of effective labour (Y/AL). The latter is just an accounting device to help describe a dynamic economy analytically and
graphically. In the steady state, the higher a is, the faster Y/L and standards of living will grow.
Fig. 3.15 Growth Rates along the Steady State
While output and capital measured in effective labour units (Y/AL and K/AL) are constant in the steady state, output–labour and capital–labour ratios (Y/L and K/L) grow at the
rate of technological progress a, and output and the capital stock (Y and K) grow at the rate a + n, the sum of the rates of population growth and technological progress.
The discussion can be extended in a natural way to address the issue of the golden rule. Redefining c as the ratio of aggregate consumption (C) to effective
labour (AL), the following modified version of (3.10) will hold in the steady state:
The modified golden rule now requires that the marginal productivity of capital be the sum of the rates of depreciation, of population growth, and of
technological change:
(3.17)
Maximizing consumption per capita is equivalent to making consumption per unit of effective labour as large as possible. To do this, an economy now needs to
invest capital per effective unit of labour to the point at which its marginal product ‘covers’ the investment requirements given by technical progress ( a),
population growth (n), and capital depreciation (δ ).
3.6.1 Solow’s Decomposition

We have now identified three sources of GDP growth: (1) capital accumulation, (2) population/labour growth, and (3) technological progress. It is natural to ask
how large the contributions of these factors are to the total growth of a nation or a region. Unfortunately, it is difficult to measure technological progress.
Computers, for instance, probably raise standards of living and growth, but by how much? Some people believe that the ‘new economy’, brought on by the
information technology revolution, will push standards of living faster than ever. Others are less optimistic that the effect is any larger than other great
discoveries which mark economic history. Box 3.5 provides some details on this exciting debate.
Robert Solow, who developed the theory presented in the previous sections, devised an ingenious method of quantifying the extent to which technological
progress accounts for growth. His idea was to start with the things we can measure: GDP growth, capital accumulation, and hours worked. Going back to the
general form of the production function (3.14), we can measure output Y and two inputs, capital K and labour L. Once we know how much GDP has increased,
and how much of this increase is explained by capital and hours worked, we can interpret what is left, called the Solow residual, as due to the increase in A, i.e.
a = ∆A/A:
Solow residual output growth due to growth in capital and hours worked.18
This method, called the Solow decomposition, is used to estimate total factor productivity as presented in Figure 3.16.
3.6.2 Capital Accumulation

Table 3.4 shows that, typically, capital has been growing at about 3–5% per year over most of the twentieth century in the developed countries. Capital
accumulation accelerated sharply in the 1950s and 1960s as part of the post-war reconstruction. Many European countries accumulated capital considerably
faster than the USA and the UK up until the mid-1970s, the reason being that continental Europe was poorer at the end of the Second World War. These
sustained periods of rapid capital accumulation fit well the description of catch-up, when the capital stock is below its steady-state level.
Fig. 3.16 Growth Rate of Total Factor Productivity in the USA (%per annum)
The average annual increase in A (total factor productivity) accelerated sharply from the 1930s to the 1970s and has slowed down sharply ever since.
Source: Gordon (2016).
Table 3.4 Growth of Real Gross Fixed Capital Stock, 1913–2010 (% per annum)
1913–50 1950–73 1973–87 1987–2010

France 1.2 6.4 3.7 3.3
Germany 1.1 7.7 2.7 1.9*
Netherlands 2.4 6.9 2.2 2.5
UK 1.6 5.7 2.3 3.6
USA 1.7 3.8 2.6 2.8
* 1991–2008.
Sources: Maddison (1991); OECD, Economic Outlook.
3.6.3 Employment Growth

The most appropriate measure of labour input is total number of hours worked. For several reasons, however, growth in population or the number of employees
does not necessarily translate into increased person-hours. To understand this, we can rewrite the total number of hours in the following way:
Total hours worked = (hours/employee) × (employee/population) × population.
It follows that the total number of hours worked can increase for three reasons:
Population growth. Everything else remaining unchanged, more people provide more working hours. But many things can and do change.
The proportion of people who work. Some working-age people are unemployed and others voluntarily stay out of work for various reasons. In addition,
people live longer, study longer, and retire earlier. Furthermore, women have increased their labour force participation over the past 30 years.19
Hours worked per person. Over time, those who work tend to work fewer hours per day and fewer days per year.
Table 3.5 shows that these effects have roughly cancelled each other out so that, in the end, employment and population size have increased by similar amounts
in our sample of developed countries.
Table 3.5 also documents the sharp secular decline in the number of hours worked per person in the developed world. The long-run trend is a consequence of
shorter days, shorter workweeks, fewer weeks per year, and fewer years worked per person. This is why the number of person-hours has declined across the
industrial world. The dramatic decline in hours worked per person is a central feature of the growth process; an average annual reduction of 0.5% per year means
a total decline of 40% over a century. As societies become richer, demand for leisure increases. The last two columns of the table reveal a massive jump in leisure
time available, which is as important a source of improvement of human welfare as increases in material wealth.
Table 3.5 Population, Employment, and Hours Worked, 1913–2010
Population growth (% Employment growth (% Growth in hours worked per Hours worked per Hours worked per
per annum) per annum) person (%per annum) person in 1913 person in 2010
France 0.4 0.3 − 0.5 2,588 1,561
Germany 0.5 0.7 − 0.5 2,584 1,418
Netherlands 1.1 1.3 − 0.5 2,605 1,372
UK 0.5 0.5 − 0.4 2,624 1,647
USA 1.5 1.6 − 0.4 2,605 1,690
Japan 0.9 0.9 − 0.4 2,588 1,713
Sources: Maddison (2006); Groningen Growth and Development Centre and the Conference Board, Total Economy Database, January 2011.
Table 3.6 The Solow Decomposition (average annual growth rates)
(a) 1913–1987*
Country GDP Contribution of inputs Residual

France 2.6 1.1 1.0
Germany 2.8 1.4 0.8
Netherlands 3.0 2.0 0.4
UK 1.9 1.2 0.5
USA 3.0 2.0 0.7
Japan 4.7 3.0 0.5
* An adjustment is made to account for the modernization of productive capital.
Source: Maddison (1991: 158).
(b) 1990–2014
GDP Contribution of inputs Residual

France 1.5 1.5 0.1
Germany 1.5 0.6 0.9
UK 1.9 1.9 0.0
USA 2.4 1.8 0.6
Japan 1.1 0.7 0.4
Source: Conference Board Total Economy Database™, http://www.conference-board.org/data/economydatabase/ September 2015.
3.6.4 Technological Change

Table 3.6 presents the Solow decomposition of GDP growth between the contribution of inputs (capital and labour) and the residual that is meant to measure
technological change. The first table shows the situation over much of the twentieth century; the second examines the same countries over the last 25 years. A
general feature is that GDP growth has markedly declined, but for different reasons. The largest decline concerns Japan, mostly due to fewer inputs, in this case
capital accumulation after its post-war catch-up. To a smaller extent, the same applies to Germany, the Netherlands, and the USA. There is no growth decline in
the UK, where labour and capital growth made up for an interruption in technological change. A similar combination explains the French decline in GDP growth.
This generalized slowdown has led to a lively debate between techno-pessimists and techno-optimists about the possibility that we have entered a period of
secular stagnation.20 Box 3.5 provides a summary of the debate.
In the Solow model of Section 3.5, the steady state depends on five factors: (1) the production function; (2) the saving rate; (3) the rate of capital depreciation; (4)
the rate of population growth; (5) the rate of technological progress. It follows that all countries should eventually converge to the level of income per capita if
these five parameters are the same everywhere. This is a big ‘if’ and indeed, across the world, we can see huge differences in incomes per capita.
The Solow model also implies that countries with a poor endowment capital will grow faster as they move to the steady state. In this view, different growth
rates are a transitory phenomenon on the way to steady-state convergence. Once the steady state is reached, per capita growth does not depend on the savings
rate, nor on the depreciation rate, but only on the rate of technological change a. Different rates of technological progress can explain different growth rates.
Box 3.5 Techno-pessimists vs. Techno-optimists
The striking changes brought about by the ICT (information and communications technologies) revolution, which include the internet, wireless
telecommunications, the conspicuous use of electronic equipment, robots, or 3D-printings have led many observers to conclude that a new industrial
revolution is upon us. Robert Gordon, from Northwestern University, describes three industrial revolutions. The first one, which occurred over 1770–1840,
involved the steam engine leading to steamships and railroads, and the replacement of wood by steel. During 1870–1920, the second industrial
revolution was based on the discovery of electricity, the internal combustion engine (petrol motors), industrial chemicals (plastics, antibiotics, and modern
medicine), running water, TV, and telephonic communication. ICT is therefore the third industrial revolution. Gordon’s careful estimates of the evolution of
total factor productivity growth (a = ΔA/A) in the USA are displayed in Figure 3.16 as annual averages over 10-year periods. He argues that new
technologies takes decades to affect total factor productivity and attributes its rapid growth from the 1930s through the 1970s to the second industrial
revolution. A techno-pessimist, he considers that the third revolution, which started in the 1960s, is much smaller than the second one and that most of its
effects have peaked. He also claims that ‘headwinds’, such as the declining impact of education, higher inequality, and population ageing will further dent
per capita GDP growth rates over the decades to come.
Over some 150 years, per capita GDP has grown by 2% in the USA, what he calls the ‘2% rule’. Combining the weakness of the third industrial
revolution with headwinds, he foresees that the long-run growth rate will be cut by half. A difference of 1% per year cumulates to 28% after 25 years and
65% after 50 years. Thus, he sees average real GDP per capita, currently at about $50,000, reaching $86,000 by 2077 instead of $200,000 with the 2%
rule. This is a huge difference. Naturally, the techno-optimists vigorously disagree. They argue that most of the benefits of ICT are yet to be reaped, that
more innovations are under way (e.g. DNA engineering), and that it is impossible to foresee what can be discovered over the next decades. The debate is
as vigorous as the issue is important.21
While it may be too restrictive, the convergence hypothesis is a good starting point to explain differences in income levels and in growth rates. Indeed, it
attracts attention to a few precise reasons behind economic success or backwardness and behind rapid or slow growth. This section reviews the evidence and
uses the Solow growth model to examine various interpretations.
3.7.1 The Convergence Hypothesis

Panel (a) of Figure 3.17 plots the average growth rate of 34 OECD (advanced) countries over the period 1960–2011 against their real per capita GDP in 1960. The
convergence hypothesis predicts that countries with low initial per capita GDP will grow faster than those that started richer—that is what catching up is all
about. Put differently, we expect to observe a negative relationship between initial conditions (per capita GDP in 1960) and the growth which followed. This does
indeed appear to be the case, but less clearly so in Panel (b) that looks at 135 non-OECD countries. A number of these less advanced countries have converged,
being now called emerging-market countries. Others have regressed, becoming absolutely poorer. Conflicts explain some, but not all of these cases. Box 3.6
explains, among the OECD countries, it takes about 35 years to eliminate half of the gap between any given nation and the richest ones.
Fig. 3.17 The Convergence Hypothesis in Reality
All things being equal, growth theory predicts that poorer countries should grow faster than richer ones. Panel (a), which plots the average growth rate for 34 advanced OECD
countries over 1960–2011 against initial GDP per capita in 1960, lends strong support to the convergence hypothesis. Panel (b), which displays the experience of 135 non-OECD
countries over the same period, does not.
Source: Penn World Data.
While Panel (a) of Figure 3.17 shows that the convergence hypothesis does well among the club of richer nations, Panel (b) shows that, for the world as a
whole, there is no general evidence of convergence. Many poorer countries seem to be ‘stuck’ in growth traps with low per capita GDP and low or even
negative growth. How can we explain this apparent contradiction?
One possibility is that countries with low income per capita suffer from insufficient capital accumulation because of inadequate saving or destruction by war or
natural disaster. Yet low domestic savings alone cannot explain the situation. It is very possible for foreigners—either individuals or corporations—to invest in
poor countries and thus finance investment. In fact, if capital is free to move, we should expect multinational firms, investors, and financiers to invest in countries
where capital–labour ratios are low and, therefore, returns on investment ought to be high. Over time, capital–labour ratios (k) should be equalized across
countries. This process may take some time, of course, but a process of convergence towards similar capital–labour ratios, and therefore per capita GDPs should
occur.
Box 3.6 A Snail’s Pace: The 2% Convergence Rule
The convergence hypothesis can be expressed in the following way:22

(3.18)
which asserts that the growth rate of an economy in period , will be higher than its steady-state growth rate (a) when current GDP (Yt) is less than
the steady-state level (Ȳt) which is determined by the saving rate, the depreciation rate, and technology. The parameter β captures the speed of
convergence. The greater the distance between the current GDP and its steady-state value, the faster the rate of growth. For the countries shown in Figure
3.17, Robert Barro of Harvard and Xavier Sala-i-Martin of Pompeu Fabra University23 estimated β to be about 0.02, which means that about 2% of the gap
in per capita income is closed per year. This implies that it takes roughly 35 years to eliminate 50% of the difference between a given region and the
leading one, assuming they have the same underlying production and other attributes.24 As an example, consider the convergence of regions within the
USA. In 1880, the US South had a GDP per capita of about one-third of the richer, north-eastern New England region. This very low initial condition
reflected the destruction of capital and infrastructure during the Civil War.
A second possibility is that chronically poor countries suffer from technological backwardness or a lack of a properly functioning economy, which prevents
them from reaching a steady-state level attainable by rich countries—point B in Figure 3.5. In that case, when and if they eliminate the barriers that prevent them
from moving towards their steady states, they should catch up and therefore grow faster than the wealthy economies, which can only grow at the speed of
technological progress. A prominent example is China, which has grown by some 10% per year ever since it abandoned central planning in the 1980s. India is
following, although not quite as fast. Plainly, this process hasn’t occurred in many countries. 25 Poorer countries’ inability to close the income gap is suggestive
of a systematic factor that we have not taken into account.
3.7.2 Conditional Convergence and Missing Inputs

So far, we have implicitly assumed that all countries have access to the same technology, represented by the production function. What if they were to differ?
Figure 3.18 gives some intuition by depicting two different production functions. Two countries with dissimilar production functions with the same saving rates
will converge to different steady states. The notion that convergence occurs, but to steady states that depend on the individual attributes of an economy, nation,
or region, is called conditional convergence. Furthermore, if they start from the same initial capital per capita, the country with the more productive
production function will accumulate more capital in each period and grow faster.
Fig. 3.18 Conditional Convergence

The economies depicted in the two panels are identical in all respects except that the available production technology is more productive at all levels of capital per capita in (a)
compared with (b). In the steady state, the more productive economy is richer because it has a higher capital–effective labour ratio. For two countries with identical initial capital
stock per effective labour (k0), conditional convergence will occur to fundamentally different steady states. The richer economy at the outset will also grow faster along its path
to the steady state, as predicted by equation (4.1). Conditional convergence means taking not only an economy’s initial condition into account, but also its steady state.
But why should production functions differ across countries? International differences are unlikely to be due to technology in the strict sense. It is hard to see
why knowledge or the mere availability of state-of-the-art methods can differ across countries for any period of time. Information is freely available, so producing
clothing or electronic chips in an internationally competitive market should involve the same production techniques whether in Taiwan, Germany, India, or
Zimbabwe. Or should it? An important reason why production functions are different is the existence of other inputs to the production function which have been
overlooked so far. Research has identified a long list of such influences. They can be thought of as ‘technology’ in the broader sense, those features that
increase the productivity of both labour and capital in use. We examine a number of possibilities.
Human capital: education and training

Just as firms acquire physical capital for producing goods and services, individuals expend time, energy, and money to acquire knowledge and learning how to
use that knowledge. These activities range from going to school, learning a skill, or taking a training course. Acquisition of knowledge is an investment. It
represents a sacrifice today—foregoing a paid job, or the costs of education—for gains in the future. Such sacrifices are generally made because future gains
exceed the initial costs.26 It is for this reason that one speaks of investment in human capital. Better-trained and educated workers are more productive, and
more productive workers can earn higher wages. It makes sense to think of production as combining not only physical capital K and hours of work L, but also a
third input, human capital H. The economy’s production function becomes:
(3.19)
Much of the formal reasoning earlier in this chapter can be repeated using this extended version of the production function. Like physical capital, human capital
is accumulated over time and is subject to depreciation as knowledge progresses and people age. This implies that countries that invest more in education and
training tend to be better off in the long run—they would tend to have a higher production function, as in panel (a) in Figure 3.18. A country that invests more in
human capital will have a higher level of per capita income. On the other hand, as before, long-run growth rates will still be the same, determined by the rate of
technological progress. The reason is that the marginal product of human capital is likely also to be decreasing.27
Human capital: health

Human capital is not just education. It includes anything that raises labour productivity for a given capital stock. In particular, human capital also includes the
state of health of workers. Freedom from disease and chronic illness as well as access to adequate medical care are considered by many to be fundamental human
rights, but they also represent a critically important determinant of economic prosperity. Workers who are ill or must care for relatives cannot be available for
productive activities. Furthermore, in countries where health services are poor or access is limited to the richest segment of the population, life expectancy will be
low. Reduced life expectancy reduces incentives for individuals to invest in education, and can lead to emigration of wealthy elites. Poor health services,
therefore, are a plausible further interpretation of growth traps.
Public infrastructure
A second important factor which is missing from the production function is public infrastructure. This includes streets, public transport, telecommunication,
postal services, airports, systems of water distribution, electricity provision, and sewage treatment, etc. Like the private capital stock, public infrastructure
consists of goods which contribute to general productivity and are widely available to all users, often at low or no cost, and are most often provided by the
government. The aggregate production function can be further augmented to include the stock of public capital, which we will call KG:
(3.20)
The importance of public infrastructure raises a number of questions. Public infrastructure is often free to use, but it is definitely not costless. Like private
physical capital, it must be financed. Governments are generally responsible for providing infrastructure, and use taxation or user fees to finance these
expenditures. However, it is difficult, if not impossible, for a government to evaluate the benefits that accrue in millions of ways to millions of users. Furthermore,
politicians rarely reap the benefits of infrastructure spending directly. At best, infrastructure enhances tax revenues in the future, and politicians may not be in
power to claim credit for those future growth effects. Figure 3.19 shows the average amounts that European governments spent on public capital accumulation
over the period 1995–2005.
Social infrastructure
Even accounting for ‘hard factors’ like education, health, and infrastructure, it remains difficult to account for the enormous differences in total factor
productivity between the richest and the poorest nations of the world. For instance, the level of GDP per capita in the USA is roughly 35 times as high as in Niger
in Africa. This means that an average US worker works 10 days to produce the annual output of a worker in Niger. Naturally, part of this difference is linked to the
fact that US workers have better capital equipment, education and skills, and public infrastructure to work with. Yet even if these differences are taken into
account, US workers are still almost eight times more productive in per capita terms. Economists attribute this wide gap to ‘soft’ factors which improve both
directly and indirectly the effectiveness of workers.
Fig. 3.19 Gross Fixed Capital Formation in the Public Sector, 1995–2015 (%of GDP)
On average, European countries spend 2–3% of GDP on public infrastructure. Yet the variation across countries is high.
The first of these is property rights. We have seen that investment in physical and human capital matters a great deal. The common feature of an investment,
spelled out in more detail in Chapters 7 and 8, is that it implies spending now for uncertain benefits in the future. It is often taken for granted that ownership of
capital will be unquestioned both today and in the future, and that individuals will be able to use the skills that they acquire. In both cases, we can speak of
clearly defined rights of ownership, or property rights, which are respected by others and systematically enforced by the state. While property rights are well
established in the richest countries of the world, they cannot be taken for granted universally.
Property rights require clear, credible, and enforceable legislation or constitutional provisions which guarantee that individuals and firms cannot be
dispossessed of their belongings, except if they violate the law and even then, only after due process. The concept of property rights is not restricted to merely
retaining one’s belongings. It also guarantees that one’s possessions can always be used as intended and disposed of, under all circumstances. Property rights
can be denied in many ways. One instance is nationalizations of firms, which occurred in France as recently as the early 1980s. They are an example of retroactive
legislation, enacted after the original investment was carried out. Milder forms of violations of property rights include government ‘orders’, corruption, or simply
political pressure.
At the individual level, property rights should be extended to human rights. For instance, human capital is harmed when people are arbitrarily sent to jail,
barred from jobs, or prevented from performing (non-harmful) economic activities. Mere threats of imprisonment or assassination also deny property and human
rights. As long as individuals do not have basic freedoms of association, expression, and protection from violence—and these irrespective of sex, race, political
opinions, or religious beliefs—their property rights are not established.
It is easy to see why property rights are a precondition for long-term economic growth. If investors cannot be sure that they will own their investments
tomorrow, why bother to invest today? This would also explain why capital does not always seem to flow from rich to poor countries. Even if the rate of return on
capital in poor countries is much higher, the risk of expropriation or arbitrary restriction of property rights may convince investors to keep their money in richer,
more stable places. And if investment is held back, future growth will be too. This elementary proposition is surprisingly often ignored. This was the case under
communism, a doctrine which explicitly rejected private ownership of means of production. Nowadays, property rights are routinely denied by arbitrary,
undemocratic regimes and by wars, both civil and international. Figure 3.20 displays the Economic Freedom Index produced by the Heritage Foundation.
Fig. 3.20 Index of Economic Freedom Around the World in 2016

This index averages the performance of countries (evaluated from 0 to 100) on 10 dimensions, including property rights. The top five countries are Hong Kong, Singapore, New
Zealand, Switzerland, and Australia, in that order. The bottom five are, from the worst upwards: North Korea, Cuba, Venezuela, Zimbabwe, and Turkmenistan.
Source: The Heritage Foundation (http://www.heritage.org/index).
The relationship between property rights, broadly defined, and growth is more complicated than meets the eye. One view is that property rights are a
prerequisite for sustained economic growth. Another view is that affluence makes basic freedoms and property rights more desirable. There are cases of
countries which embarked on a stable, often fast growth path while enjoying limited property and human rights: the communist countries, Chile under Pinochet,
or some countries of South-East Asia. Conversely, it is often argued that some countries visibly fail to grow because property rights are impaired or non-existent.
Summary
1 Economic growth refers to the steady expansion of GDP over a period of a decade or longer. Growth theory is concerned with the study of economic growth in the steady
state, a situation in which output and capital grow at the same pace and remain in constant proportion to labour in effective terms. This approach reflects key stylized facts.
2 The aggregate production function shows that output grows when more inputs (capital and labour) are used, and when technology improves the effectiveness of those inputs.
3 The capital stock, the sum of productive equipment and structures, is accumulated through investment, and investment is financed by savings. When savings are a stable
proportion of output, the steady-state capital stock is determined by savings and capital depreciation.
4 The assumption that the marginal productivity of capital is declining implies that output, and therefore savings, grow less proportionately to the stock of capital, in contrast
to depreciation, which is proportional to capital. The steady state is reached when capital accumulation exhausts the productive potential of additional savings and comes to
an end.
5 In the absence of population growth and technological change, the steady state is characterized by zero output and capital growth. Adding population growth provides a first
explanation of secular output growth, but standards of living—measured as output per capita—still do not increase. It is only when we allow for technological progress that
permanent growth in per capita output and capital is possible.
6 The golden rule describes a steady state in which consumption is as high as possible. It occurs where the marginal productivity of capital equals the rate of depreciation (so as
to replace lost or worn out capital) plus the rate of population growth (to provide new workers with adequate equipment) plus the rate of technological change (adjusting
capital to enhanced labour effectiveness): MPK = δ + n + a.
7 An economy is dynamically efficient when steady-state consumption can be raised tomorrow only at the expense of lower consumption today. An economy is dynamically
inefficient when both current and future steady-state consumption can be increased. In the former case, the capital stock is lower than the golden-rule level. In the latter case,
the capital stock is above the golden-rule level.
8 Saving does not affect the steady-state growth rate, but only the level of output per capita.
9 The Solow decomposition is a method of accounting for the sources of economic growth. It breaks down growth in GDP into the sum of growth attributable to changes in the
factors of production and growth due to improved production. The latter is called the Solow residual, and is usually interpreted as technological change.
10 While the Solow model does a good job of explaining the behaviour of developed economies, it has difficulties accounting for the poor growth performance of poor and
middle income countries. One reason is because many factors which affect total factor productivity, such as infrastructure, education and training, human rights, and law and
order, are absent. When these elements are considered, the Solow model is better able to predict low growth of poor countries as an artifact of poor institutions, poor
education and training, and property rights.
Key Concepts
economic growth
technological progress
production function
marginal productivity
diminishing marginal productivity
returns to scale (constant, increasing, decreasing)
decreasing returns
output–labour ratio
capital–labour ratio
stylized facts
steady state
Solow growth model
capital accumulation
saving schedule
gross investment
net investment
capital depreciation
depreciation line
golden rule
dynamic inefficiency, dynamically efficient
capital widening, capital-widening line
aggregate production function
total factor productivity
effective labour
Solow residual
Solow decomposition
convergence hypothesis
growth traps
conditional convergence
human capital
public infrastructure
property rights
human rights
Exercises
1 Draw intensive-form production functions f(k) with decreasing, constant, and increasing returns to scale.
2 Use graph paper (or a computer spreadsheet) to plot the following intensive-form production functions (expressing y as a function of k) from the interval [0, 100]:
(a) f(k) = 2k;
(b) f(k) = 10 + k0.5;
(c) f(k) = k1.1;
(d) f(k) = 10 + 2k − 0.5k2;
(e) f(k) = max(0, − 10 + 2k).
Which of these functions has constant returns? Decreasing or increasing returns? Are your answers always unambiguous, i.e. do they hold for all k∈[0,
100]?
3 Define the steady state. How is a steady state important in the context of the Solow growth model? Explain why stylized facts are helpful to organize the discussion of
economic growth.
4 Suppose K/Y is constant at 2. (a) Assume first that there is no population growth and no technological progress. What is the steady-state saving–output ratio consistent with
a rate of depreciation of 5%? (b) Now allow for positive economic growth, due to either population growth or technological progress. What is the steady-state saving–output
ratio consistent with a rate of depreciation of 5% and 3% real growth?
5 Consider a country with zero technological progress and K/L = 3. Its population grows at the rate of 2% per year. What is the steady-state rate of growth of GDP per capita
if the saving rate is 20%? If it is 30%? How do your answers change if depreciation occurs at a rate of 6% per year?
6 Suppose the aggregate production function is given by . Does it have increasing, decreasing, or constant returns to scale? Show that the marginal products of
capital and labour are declining. Show that they are increasing in the input of the other factor.
7 Define the golden rule. Explain why it is achieved at in Figure 3.9. To establish this result, imagine that you start to the left of and explain why moving to the right
increases consumption. Similarly show that consumption decreases when moving rightwards from a position to the right of .
8 The golden rule (3.17) is MPK = δ + a + n. In comparison with the no-technological change case, we now require a higher marginal productivity of capital. With diminishing
marginal productivity, this means a lower capital stock per effective unit of labour. Is that not surprising? How can you explain this apparent paradox?
9 Draw a graph showing the evolution of Y/L in the catch-up phase and then in the ensuing steady state when the economy starts from a capital–effective labour ratio below its
steady-state level. Draw a diagram showing investment (not the capital stock).
10 Explain, formally or informally, the difference between average labour productivity and output per capita.
11 In the steady state, output per effective labour Y/AL is constant. What happens to output per capita when the average number of hours worked declines, holding all else
constant?
12 Suppose two regions in the same country have identical production functions, savings rates, population growth rates, and depreciation rates. Suppose that per capita GDP
differs by 20% between them, but that this gap is closed by 2% per annum. How many years will it take to close the gap to 5%?
13 Suppose output Y is produced with physical capital K, human capital H, and labour L using the production function Y = KαHβL1−α−β. Assume that households save a
fraction sH of their income in human capital and sK in physical capital and that sH = sK = s and that both forms of capital depreciate at the rate δ. Solve for the steady-state
level of output and output per capita. Why isn’t this economy growing?
Essay Questions
1 Japan in the 1960s, Korea in the 1980s, China in the 1990s, and India in the current decade have experienced periods of rapid growth, in effect catching up with the richer and
more developed countries. How can you explain this phenomenon, and why did it not happen earlier?
2 In earlier centuries, colonial expeditions were launched to increase a country’s land and population. How might such activities make a country richer?
3 ‘Globalization is bad for growth since it means that countries invest abroad and expand the capital stocks of other countries instead of their own. The Solow model would
consider this a decrease in the savings rate, thus leading to lower GDP per capita.’ Comment.
4 Examining the panels of Figure 3.7, one immediately notices that African countries are bunched in the lower left-hand part of the diagram, while the European countries are
clustered in the upper right-hand region. What explanation of this fact is offered by the Solow growth model? Do you think this is a sufficient explanation? Explain.
5 It is often claimed that the defeated nations after the Second World War grew faster than the victor nations. Is this hypothesis consistent with the Solow growth model?
6 Are you a techno-optimist or a techno-pessimist? Justify your position.
7 Why is the acquisition of human capital called an investment, like the acquisition of physical capital? Explain why knowledge and human capital depreciate, like physical
capital. What is the link between property rights and investment in both physical and human capital?
1 Robert E. Lucas, Jr (1937–), Chicago economist and Nobel Prize Laureate in 1995, is generally regarded as one of the most influential contemporary macroeconomists.
Among his many fundamental contributions to the field, he has researched extensively the determinants of economic growth.
2 Maddison (1991: 10).
3 Since output and labour inputs are flows, they could also be measured per quarter or per month, but should be measured over the same time interval. Note that capital is a stock,
usually measured at the beginning of the current or end of the last period. We discussed the important distinction between stocks and flows in Chapter 1.
4 Formally, the first partial derivatives FK(K, L) ≡ ∂F(K, L)/∂K and FL(K, L) ≡ ∂F(K, L)/∂L are positive.
5 The elasticity of output with respect to capital is defined as (∂Y/∂K)(K/Y) and is (αKα − 1L1 − α) (K1 − αLα − 1) = α. Similarly, 1−α is the elasticity of output with respect to the labour
input.
6 Throughout this book, the symbol ‘∆’ (Greek delta) stands for a step change in a variable, holding all else constant. Here, ‘all else’ means the labour input.
7 Following footnote 4, this feature of the production function implies mathematically that the second partial derivatives FKK(K, L) ≡ ∂2F(K, L)/∂K2 and FLL(K, L) ≡ ∂F2(K, L)/∂L2
are negative.
8 The constant returns property implies that if we scale up K and L by a factor t , Y is scaled up by the exactly same factor—for all positive numbers t , it is true that tY = F(tK, tL).
In the text we use the case t = 2; we double all inputs and produce twice as much. If we choose t = 1/L, we have y = Y/L = F(k, 1). Rename this f(k) because F(k, 1) depends on k only.
The intensive production function f(k) expresses output produced per unit of labour (y) as a function of the capital intensity of production (k).
9 It is important to recall the distinction between average productivity (Y/L) and marginal productivity (∆Y/∆L).
10 If population size N and total hours worked L change in the same proportion, the growth rates of Y/N and Y/L are identical. In developed (mature) economies, people tend to work
fewer hours and fewer years, so Y/L grows more slowly than Y/N. Formally, ∆(Y/L)/(Y/L) = ∆Y/Y – ∆L/L and ∆(Y/N)/(Y/N) = ∆Y/Y – ∆N/N, so the difference between the growth rates
of per capita output Y/N and average productivity Y/L is ∆N/N – ∆L/L. Section 3.6.3 examines this difference.
11 This need not be true for a region within a nation: the capital stock of southern Italy, eastern Germany, or Northern Ireland may well be financed by residents of other parts of their
countries. Yet even these financing imbalances are unlikely to be sustainable for the indefinite future.
12 In general, steady-state values of variables will be indicated here with an upper bar, e.g. , ȳ , etc.
13 Note that everything, including consumption and saving, is measured as a ratio to the labour input, person-hours. As already noted, if the number of hours worked does not change,
the ratios move exactly as per capita consumption, savings, output, etc.
14 An exercise asks you to prove this assertion.
15 The proof requires some calculus based on the principles presented later on in Box 5.3. The change in capital per capita obeys ∆k/k = (∆K/K) − (∆L/L). After substituting ∆K = I − δK
and ∆L/L = n and setting I = sY, the equation simplifies to ∆k = sy − δk − nk = sf(k) − (δ + n)k.
16 Technological innovation sometimes also makes capital more productive. It is much simpler to assume that technological progress is labour-augmenting and it is enough to provide a
reasonable—if incomplete—interpretation of the growth phenomenon. It is consistent with important growth facts to a greater extent that the capital-augmenting alternative.
17 Constant returns to scale imply that y = F(K, AL)/AL = F(K/AL, 1) = f(k).
18 Technically, the Solow residual is where sL is the labour share, defined as the share of national income paid to labour in the form of wages and non-wage
compensation, and 1 − sL is the income share of capital. The WebAppendix shows how this formula can be derived from the production function.
19 Chapter 4 explores these and related issues in more detail.
20 An up-to-date review of the debate can be found in Coen Teulings and Richard Baldwin, Secular Stagnation: Facts, Causes and Cures, VoxEU Ebook (CEPR, 2014).
21 For an entertaining introduction, see the articles published on VoxEU: Robert Gordon, ‘Is US economic growth over? Faltering innovation confront the six’, September 2012
(http://www.voxeu.org/article/us-economic-growth-over) and Joel Mokyr, ‘Is technological progress a thing of the past?’, September 2013 (http://www.voxeu.org/article/technological-
progress-thing-past).
22 Interested readers can see in the WebAppendix how this formulation is derived.
23 See Barro and Sala-i-Martin (1991).
24 This is the useful rule-of-thumb for computing doubling time or half-lives. Suppose a variable x grows at a constant positive rate of g% per annum, then it will double in roughly 70/g
years. If it is shrinking at rate g, it will shrink to half its size in 70/g years. See the WebAppendix for a derivation of this rule.
25 Nobel Laureate Robert E. Lucas, Jr of Chicago estimated that if India and the United States had the same production function, the marginal product of capital in the former would be
about 58 times that of the latter! In his words, ‘If this model were anywhere close to being accurate, and if world capital markets were anywhere close to being free and complete, it is
clear that, in the face of return differentials of this magnitude, investment goods would flow rapidly from the United States and other wealthy countries to India and other poor
countries. Indeed, one would expect no investment to occur in the wealthy countries in the face of return differentials of this magnitude’ (Lucas 1990: 92).
26 In Chapters 7 and 8 the notion of an investment and the assessment of the desirability of investment will be made much more precise.
27 In general, if the production function is subject to constant returns to scale in all inputs, it must be the case that the marginal productivity of each input taken individually is
decreasing.
Labour Markets and
Unemployment 4
4.1 Overview
4.2.1 Labour Supply and the Consumption–Leisure Trade-off
4.2.2 Labour Demand, Productivity, and Real Wages
4.2.3 Labour Market Equilibrium
4.2.4 The Interpretation of Unemployment
4.3.1 Involuntary Unemployment and Real Wage Adjustment
4.3.2 Collective Bargaining and Real Wage Rigidity
4.3.3 Social Minima and Real Wage Rigidity
4.3.4 Efficiency Wages and Real Wage Rigidity
4.4.1 Labour Market States and Transitions
4.4.2 Stocks, Flows, and Equilibrium Unemployment
4.4.3 Job Separation and the Incidence of Unemployment
4.4.4 Job Finding and the Duration of Unemployment
4.5.1 The Concept
4.5.2 The European Experience
4.5.3 Actual and Equilibrium Unemployment
Summary
In our present-day complicated economic life we are likely to be confused by the many industrial operations and money transactions. But net income remains exactly what it was
to primitive Robinson Crusoe on his island—the enjoyment from eating the berries we pick, so to speak, less the discomfort or the labour of picking them.
Irving Fisher1
Labour is the source of all value.
Karl Marx2
4.1 Overview
In the last chapter, we showed how output is a function of the endowment of factors of production—capital and labour—as well as technical sophistication.
We examined the evolution of the capital stock and of technology in some detail, but we took the supply of labour as given, regardless of the wage or other
variables. Even in storybooks, life is not so simple. In the famous novel by Daniel Defoe, the shipwrecked castaway Robinson Crusoe was endowed with a stock
of capital (coconut trees) to produce output (coconuts), but needed to expend time and effort to gather and transport the fruits that he would eventually
consume. Like most people, Crusoe had to choose whether to work or not, and how much effort he would put into it. Presumably, the rewards of working played a
role in his decision. And the decision of how much to work determined how many coconuts would be harvested. This chapter is all about the labour market,
which determines how much people work and earn.
To introduce the labour market, we need to think hard about the decisions of households, which supply hours of work to the market, and of firms, which
demand them. Households work so that they can consume, but they also want to spend some of their time not working and enjoying leisure or free time. The
supply of labour is seen as a trade-off between consumption and leisure. At the same time, someone must be willing to employ and pay for the hours of work that
workers want to supply. If firms demand labour, it must have some value to them in production. As we study the interaction of the demand and supply of labour,
we will learn how unemployment emerges.
We begin by looking at a household that supplies hours of work to the labour market. Then we turn to the demand for labour by firms. This leads naturally to
the standard confrontation of demand and supply, which is sorted out by the wage, the price of labour. Yet we will see that labour is not a standard ‘commodity’.
Workers are not identical, and the quality of labour services is difficult to ascertain and harder to monitor. Unlike machines or raw materials, workers can decide
whether they would like to work for a particular employer and under which conditions to sell their labour services. Frequently, the employment relationship
involves explicit and implicit arrangements specific to both firm and employee. The functioning of labour markets is also influenced by country-specific
institutions, such as labour law or collective bargaining, and is subject to complex legal rules and customs. Finally, the labour market is dynamic, with suppliers
and demanders of labour entering into and exiting employment relationships at a remarkable rate. We show how these interactions help us to understand the
concept of equilibrium unemployment, which may differ from actual unemployment observed at a particular point in time.
While Karl Marx may be out of favour these days, he was certainly right to see labour as the most important factor of production. Virtually everything stems
directly or indirectly from labour. Raw materials are brought forth from the earth by human hands. Equipment used in this and other forms of economic activity is
made using labour and previously manufactured equipment, itself the output of labourers and capital in a more distant past. Even the knowledge embodied in
people—human capital—comes from our own efforts at mastering skills and techniques, as well as the time our teachers spent trying to educate us.
4.2.1 Labour Supply and the Consumption–Leisure Trade-off

In modern societies, consuming goods and services requires income. Earning income most often means working for firms for a wage or salary. 3 But labour has a
cost, too: every hour of work means an hour less of free time. Because households value both consumption and leisure, they balance the two the best they can,
given the possibilities available to them. This trade-off is known as the consumption–leisure trade-off. The consumption–leisure trade-off is well-illustrated
by the choices faced by Robinson Crusoe.4
The Crusoe we consider consumes all the fruits of his labour, which are the coconuts that he picks up on the beach, shucks, and cracks open—hard work by
anyone’s measure. (We postpone issues related to saving and investment to Chapters 7 and 8.) Crusoe values both the consumption (coconuts) and the free
time he can enjoy not working to obtain them (resting, contemplating, watching the stars at night). His preferences or tastes with respect to consumption and
leisure can be summarized with indifference curves. Each indifference curve presented in Figure 4.1 shows combinations of consumption and leisure which
make Crusoe equally happy, or to have the same utility—the word economists use for a consumer’s state of satisfaction.
Indifference curves are designed to tell us something about the preferences of households. Higher indifference curves correspond to higher levels of utility or
happiness. The negative slope of the curve shows that a trade-off is involved: taking a unit of consumption from Crusoe requires compensation in the form of
more leisure. The (absolute) slope or steepness of each curve at a given point shows how readily Crusoe substitutes consumption for leisure, holding his level of
satisfaction constant. The bowed shape of Crusoe’s indifference curves tells us that the greater his consumption is relative to its leisure, the more of that
consumption he is willing to give up for an additional unit of relatively scarcer leisure. The rate at which a household is willing to give up consumption for leisure,
holding satisfaction constant, is called the marginal rate of substitution of consumption for leisure. An important principle is that as a good becomes
increasingly scarce, the marginal rate of substitution of other goods for that particular good increases.
Fig. 4.1 Household Preference

An indifference curve shows how readily a household is willing to substitute consumption C for leisure ℓ holding its level of satisfaction or utility constant. Curves further out
from the origin correspond to higher levels of utility.
On Crusoe’s island there is no money, but there is time, in fact lots of it. Yet even Crusoe’s time is in limited supply over a certain period (a day, a month, or
year), after sleep and other important functions are accounted for. All scarce resources have a price, and time is no exception. The price of an hour of leisure is its
opportunity cost: how much consumption could one otherwise earn in that hour by working in the market? For the representative household with access to a
labour market, the price of leisure is the real (consumption) wage. In practice, it is measured as a ratio of an average nominal (money) wage (W) to the
consumer price index (P), the price of goods. If total available time for work is denoted as and the hourly real wage is w = W/P, the value of Crusoe’s total time
endowment in terms of consumption is w. This endowment can be allocated between C units of consumption and ℓ hours of leisure, with value ℓw, according
to the following equation:5
(4.1)
Equation (4.1) is Crusoe’s budget constraint; it states all possible combinations of leisure and consumption that he can afford. It is depicted in Figure 4.2 as the
negatively sloped line AB. The horizontal distance OA is equal to the number of hours at his disposal. The distance OB measures the value of that endowment
in terms of consumption goods. It is the total amount of consumption attainable when leisure is zero—i.e. if Crusoe were to work all the time.6
Fig. 4.2 The Household Budget Line and Optimal Choice
The household has hours at its disposal (measured by the distance OA) for either leisure or work. For every unit of leisure that it gives up, it can earn w consumption goods.
The real wage w determines the slope of the budget line AB. Given the budget line, the highest possible utility is achieved at point R, where an indifference curve is tangent to
the budget line.
The negative slope of the budget line (−w) measures the trade-off of consumption for leisure offered by the market: how much consumption must be given
up to get an additional unit of leisure, or how much consumption can be ‘purchased’ with an additional hour of foregone leisure. In both cases it is the hourly
wage rate w. It explains why the real wage w is often referred to as the relative price of leisure in terms of consumption. If the real wage changes, the budget
line rotates around point A, which is the fixed time endowment . It represents the amount of leisure possible when consumption is zero.
Optimal choice and the individual labour supply schedule

Crusoe maximizes his utility by choosing the highest possible indifference curve without moving to the right or above his budget line. This is achieved at point R
in Figure 4.2, where the indifference curve is tangent to the budget line. At this point, the marginal rate of substitution of consumption for leisure is equal to the
real wage w. Given the terms of trade offered by the market, he cannot make himself any better off (reach a higher indifference curve) by choosing any other point
on the budget line, i.e. by trading leisure against consumption.
An important issue is the influence of the real wage on household behaviour. How does the household react to an increase in the real wage depicted in the first
panel of Figure 4.3? Graphically, OB increases and the budget line rotates clockwise. Crusoe’s optimal choice is now represented by point R′. This point is above
R—Crusoe’s consumption increases—but it is not clear whether it is to the left—Crusoe works less—or to the right—he works more. This ambiguity is a
consequence of two different effects. First, the relative attractiveness of leisure declines because its relative price has risen. Taken alone, this would encourage
Crusoe to take less leisure, work harder, and consume more. This is called the substitution effect. Second, and at the same time, each hour of Crusoe’s labour is
paid better. Holding leisure constant means earning more income overall. A normal reaction to an increase in income should be to enjoy both more consumption
and more leisure, which means working less. This incentive to work less in response to a wage increase is called the income effect.
So, if the wage rises, will Crusoe work more or less? This question cannot be answered unambiguously without knowing more about preferences. In Figure 4.3,
the substitution effect dominates, so the net effect is positive: an increase in the wage leads to a decline in leisure and an increase in labour supply. This is
depicted in the second panel of Figure 4.3 as an upward-sloping household labour supply curve.
In practice, the response to rising wages varies widely across individuals, depending on tastes, family circumstances, age, etc. It also depends on the time
horizon under consideration.7 In the short run, most individuals do not seem to react much to changes in the real wage. Over the very long run, increasing real
wages have led to decreasing labour supply as the income effect dominates. Table 4.1 shows that, over the last century and a half, real wages have increased by
eight- to more than 20-fold, while working hours have declined by one-half, with interesting differences across countries, to be discussed shortly. Labour supply
behaviour also varies according to gender. For men, the average work-week, the retirement age, and the labour force participation rate have fallen secularly
since 1900.8 In contrast, labour force participation and hours per week of women have risen in the past half-century. One possible interpretation is that the
income effect of higher wages dominates for men, whereas the substitution effect dominates for women. Another interpretation is that customs and sociological
factors change. Services such as child care and schooling, as well as improvements in the technology of home production, have made it possible for more people,
especially women, to take up paid jobs. Yet another interpretation is that the nature of work has changed since the relative decline of industry and the rise of the
service economy, for which women are comparably and possibly absolutely better equipped than men.
Fig. 4.3 Reaction of the Household to a Wage Increase: Labour Supply
When the real wage increases, the budget line rotates around point A (the endowment of time remains unchanged) and becomes steeper, because a unit of leisure is
exchanged for more units of consumption. This allows both consumption and leisure to increase at the same time (income effect). Because leisure is more expensive, however,
some is given up (substitution effect). In the case depicted here, the substitution effect dominates.
Table 4.1 Annual Total Hours Worked and Average Wages, 1870–2014
1870 1913 1938 1973 1992 2000 2014

Annual hours worked per person
France 2945 2588 1848 1771 1542 1517 1473
Germany 2941 2584 2316 1804 1563 1469 1371
UK 2984 2624 2267 1688 1491 1491 1677
USA 2964 2605 2062 1717 1589 1660 1789
Sweden 2945 2588 2204 1571 1515 1588 1609
Real wage (index: 1870 = 100)
France 100 205 335 1048 1417 1434 1668
Germany 100 185 285 944 1178 1222 1318
UK 100 157 256 439 640 733 790
USA 100 189 325 596 659 737 826
Sweden 100 270 521 1228 1493 1727 2135
Source: Hours worked are from Maddison (1991) for years 1870–1938 and from Groningen Growth and Development Centre and The Conference Board, Total Economy
Database for the years 1973 to 2000; wages are from Mitchell (1978, 1983). German wage data from 1913–38 are approximated using average labour productivity growth.
Data for 2014 are from the OECD, Main Economic Indicators.
The aggregate labour supply curve

We have studied a representative household’s decision to work. The next step is to add up the supplies of individual households across the economy, while
noting the special aspects of labour and labour markets. In many instances, individuals cannot vary the hours of work that they supply. At best, they can choose
between working or not working at all. Most labour contracts specify a standard working time (length of the work-week, days of holiday leave per year). Workers
frequently receive and accept ‘take it or leave it’ job offers. Sometimes workers are better off not working at all. In cases such as these, small wage increases may
not be sufficient to motivate households to take up jobs, although large ones might.9 In any case, the aggregate labour supply remains the sum of many
individual decisions (to work or not work, and how many hours to work). While individual labour supply is measured in hours during some period of time (e.g. per
year as in Table 4.1), aggregate supply is measured in person-hours, the total amount of hours supplied by all workers during that same period.10 As in Chapter
3, total hours are the product of the number of persons in work times the hours worked per person. When wages rise, even if those who already work do not
modify their supply of labour (the benchmark case), others who had preferred not to work may now decide to join the labour force. Figure 4.4 shows how it is
then possible for a steep or even near-vertical (inelastic) labour supply curve at the individual level to be consistent with a flatter aggregate supply curve.
Fig. 4.4 Individual and Aggregate Labour Supply

The aggregate labour supply curve is less steep than that of individual households for two important reasons. First it represents the summation of a great number of upwardly
sloped individual supply curves. Second, new workers choose to enter the labour force as wages rise.
4.2.2 Labour Demand, Productivity, and Real Wages

The labour demand curve and its slope
Firms use both capital and labour to produce goods and services. As discussed in Chapter 3, the capital stock at any particular point in time is best thought of as
given; from month to month firms vary their output by adjusting employment of labour (person-hours). Holding capital constant, we can express the link between
output Y and employment L using the production function, as shown in Figure 4.5. The slope of the production function measures the marginal productivity
of labour (MPL), the quantity of additional output which results from one more unit of labour input (an hour). The shape of the curve reflects the principle of
decreasing marginal productivity. With all else held constant, the MPL in a representative firm declines as the amount of labour employed increases.
Fig. 4.5 The Production Function and the Labour Demand Curve
When labour input increases, output increases, but at a declining rate. This additional output is the marginal productivity of labour (MPL). In Panel (a), the ray OR represents the
labour cost of producing when the hourly real wage is w . The vertical distance between the production function and the cost line represents the firm’s profit. The firm maximizes
its profit at point A, where the curve is parallel to OR, i.e. where MPL = w . Its demand for labour is given in Panel (b) by the declining marginal product of labour (MPL) curve.
In deciding how much labour to employ, the representative firm seeks the highest possible profit given the cost of labour, the real hourly wage w. The line OR
represents the total cost of labour to the firm at different levels of employment. Because L hours of work cost wL, its slope is w. For each level of employment,
profit is measured as the vertical distance between the curve depicting the production function and the labour cost line OR. Profit is at a maximum at point A,
where the production function is parallel to OR. At this point, the MPL, the slope of the production function, is equal to the real wage, the slope of OR. If the
MPL exceeds the real wage, hiring one more hour of work raises revenues by the amount MPL, while raising costs by only w. This implies an increase in profits.
A firm interested in maximizing profits would hire the extra hour, and continue to do so as long as the MPL exceeds the wage. But the MPL will decline as more
employment is utilized, so the expansion of employment has a natural end: point A where MPL equals the real wage. Suppose, taking the opposite case, that the
real wage exceeds the MPL. Now the firm can increase its profit by reducing its demand for labour. Because it is optimal to set labour such that MPL = w, the
MPL schedule in Panel (b) of Figure 4.5 is also the firm’s labour demand curve.
The slope of the labour demand curve depends on the rate at which the MPL falls. If MPL is relatively insensitive to labour and falls only slightly as labour
inputs are increased, then the demand curve will be relatively flat. Small changes in the wage will lead to large responses in the demand for labour. The labour
demand curve is elastic. In contrast, if the MPL drops sharply as labour inputs are increased, the labour demand curve is steep and the demand for labour is
unresponsive to changes in the wage. We say that the demand for labour is inelastic.
Shifts in the demand for labour

After studying the slope of the labour demand curve, we consider reasons for the labour demand curve to shift—to change its position. One possible reason is
an increase in the capital stock K, which we took as given, i.e. exogenous. At the aggregate level, an increase in the capital stock tends to make additional labour
more productive. In Panel (a) of Figure 4.6 this is shown as an increase in MPL—the production function becomes steeper at every level of production. In Panel
(b), this implies that the labour demand curve shifts out. A technological improvement that makes labour more productive at any given level of capital and labour
input will have a similar effect on the MPL. This is identical to an increase in total factor productivity discussed in Chapter 3. Increasing total factor productivity
can help account for the fact that wages have grown secularly over time.
Conversely, it is also possible for labour demand to shift back, i.e. to decline holding the wage constant. A decline in the capital stock brought about by war,
natural disaster, or technical obsolescence causes the labour demand curve to shift inwards, i.e. down and to the left. In the case of labour-saving technical
change, it is possible for labour demand to decline, even as total output is increasing. Box 4.1 explains the important phenomenon of technical change in more
detail.
Fig. 4.6 An Increase in Labour Productivity

Labour can become more productive either because more capital is put in place, or because technological progress makes labour more productive using the existing stock of
equipment. In panel (a), at any level of labour input, more output is produced and the production function is steeper everywhere. The MPL increases and the demand for labour
schedule shifts up in Panel (b).
Box 4.1 Technical Change and Unemployment
What is the effect of technological progress on jobs? A common perception is that technology destroys jobs by making people redundant, i.e.
unnecessary for the production of goods and services. Aren’t cars mostly produced by robots? Don’t computers reduce the need for secretaries? Isn’t the
internet replacing mail and jobs in the post office? It would be hard to deny that technological advances destroy some jobs, occupations, or even whole
sectors of economic activity. At the dawning of the industrial revolution in the early nineteenth century, the Luddites in Britain or the soyeux in France
agitated against mechanized weaving machines and mobilized mobs to destroy them in large numbers. And indeed, weaving cloth by hand has
disappeared as a trade, as have many other technically backward activities. Nevertheless, overall productivity of labour has increased roughly tenfold in
the nations of Western Europe in the past century and a half, and if anything, unemployment is somewhat lower now than it was a century and a half ago,
so the story must be more complicated than simply technology killing jobs.
Indeed, the story is more complicated. Technological progress does mean that the same output can be produced with fewer person-hours, but this
does not mean that overall employment must fall. The key question is: how are technology-induced increases in labour productivity (Y/L) split between
growth in output (Y) and employment (L)? Because both employment and output are endogenous, a correct answer requires much more knowledge in
order to follow all the mechanisms that influence the final outcome. Sorting this out is the task of macroeconomics. But what do the facts tell us? Kaldor’s
fourth stylized fact discussed in Chapter 3 says that the wage share is stable long term. The wage share is WL/PY, which can be rewritten as w/(Y/L). For
the wage share to remain constant when productivity rises, real wages w must have to keep up with productivity (Y/L).
We also know that, over the decades, the number of hours that people work has been secularly falling, but the number of workers employed in the USA
and Europe has increased dramatically—a consequence of increasing labour force participation. Higher real wages, more leisure and more employment
—it seems that technological progress has made a positive contribution to our well-being.
The Luddites and their modern-day equivalents are guilty of the ‘fallacy of composition’: the disappearance of some jobs or the decline of some regions
need not apply to the economy as a whole. The demand for secretaries may have fallen, but who builds, programmes, and maintains the computers that
make secretaries redundant? Who designs, markets, and services the cars built by robots? New jobs replace old jobs. Of course, these are not the same
people. At the aggregate level, old skills are replaced by new skills, but some individuals must be retrained, which takes time, and some are sidelined.
That some people become unemployed does not mean that total employment declines. Economic progress may be painful for some, but it is by no
means a systematic job-killer. And from Chapter 3 we have learned that it is the most important engine of sustainable increases in the standards of living.
4.2.3 Labour Market Equilibrium

We now have the building blocks needed to understand the labour market: a supply curve derived from household behaviour, and a demand curve derived from
firm behaviour. The interaction of supply and demand for labour is depicted in Figure 4.7. Equilibrium occurs at the intersection of the two curves (point A). At
wage w the market clears (there is no excess demand or supply): L is the number of hours firms want to hire and households want to work. Both the real wage rate
and employment are endogenously determined in the labour market.
This basic characterization of the labour market is an important part of an economist’s toolkit, and its predictions serve as the benchmark for the rest of the
chapter. Figure 4.8 provides two examples of its usefulness. Panel (a) shows the outcome of an increase in labour productivity resulting from capital accumulation
or technological advances, already examined in Box 4.1. The labour demand curve shifts outwards; the supply curve remains unaffected. The result is an
unambiguous increase in real wages. In the figure, employment increases, but if the labour supply curve is vertical, employment remains unchanged and labour
income wL rises proportionally with the real wage. It turns out that if the income effects of wage increases dominate the substitution effect, the supply curve can
even be backward-bending: employment (person-hours) would even decline. Table 4.1 suggests that this has been the case at least over the past 150 years.
Fig. 4.7 Equilibrium in the Labour Market
Labour market equilibrium occurs at point A where demand and supply are equal. The real wage w clears the market at employment level L. If N is working age population, total
labour endowment is given by N , and voluntary unemployment is given by the distance N – L.
The second panel of Figure 4.8 shows that an exogenous increase in the supply of labour leads to an increase in employment, but also to a reduction in real
wages. Yet, this is a short-run effect. One of the stylized facts presented in Chapter 3 states that, in the long run, real wages grow with productivity. To the extent
equilibrium employment increases, an increase in labour supply will initially lower the capital–labour ratio K/L. A lower capital–labour ratio means that the
productivity of capital has risen. By the logic of the previous chapter, capital accumulation will rise. As the capital–labour ratio returns to its previous growth
path, the labour demand curve shifts outward (not shown). In the long run, an increase in the supply of labour is eventually matched by an increase in the stock
of productive capital, and, ceteris paribus , no reduction in real wages.
Fig. 4.8 Shifting Labour Demand and Supply

When labour demand increases (Panel (a)), the real wage and the employment level both increase. When labour supply increases instead (Panel (b))—because of new entries
into the labour force, for example—employment rises but the real wage declines.
4.2.4 The Interpretation of Unemployment

While the supply-and-demand apparatus allows us to evaluate the effect of various changes on equilibrium employment and real wages, it is disappointing in one
crucial respect. At point A in Figure 4.7, labour supplied is equal to labour demanded and denoted by L. Any unemployed labour—literally, labour not employed
—reflects the voluntary decisions of households and firms. Indeed, represents total potential labour supply, the maximum amount of hours of work that all
the people of working age in the economy (N) can contribute. If L is actual employment, corresponding to point A, unemployment (measured as hours of work
not employed) is simply Since point A lies on the labour supply curve, unemployment reflects the choices of households. Why then do we see
unemployment? The reasoning so far leads us to conclude that unemployment in Figure 4.7 is voluntary. The real wage w is simply too low to persuade workers
to give up all their leisure: some may wish to work only part-time, others may not want to work at all.
It might be disturbing to think that unemployment could be chosen freely. Yet voluntary unemployment—or, more precisely, the decision not to work—is
an important phenomenon in modern economies. It is not only the very wealthy who can afford not to work. Those who receive an income from other sources
(from a spouse or from the state, for example) may also find that the net wage they can earn does not compensate for foregone leisure or non-market activities,
including working at home or raising children. Voluntary unemployment is likely among low-skilled people who cannot earn much, or in countries where taxes are
so high that working yields little net gain. The most obvious costs to working are faced by families with children. The high cost of child care—or simply the
unavailability of such services—explains why two-earner families are not as common in some countries as in others. Box 4.2 presents more general evidence on
how very different patterns of household labour supply can emerge, suggesting that labour market institutions have a more important role for explaining
differences across countries than preferences themselves.
Box 4.2 Market Hours at Work: Europe versus the USA

International comparisons should always be taken with a pinch of salt, because data from different countries—especially labour market data—are
frequently difficult to compare. Yet often differences are just too large to attribute to measurement error. Table 4.2 compares hours worked per week, on
average, per adult of working age (in general, aged 15–64). In the mid-1990s, continental Europeans spent considerably less time in market work than
North Americans, Scandinavians, and Britons, 20–25% less in some cases. Two decades later, differences among countries within these two groupings
have narrowed somewhat, yet fundamental differences persist.
Table 4.2 Weekly Hours Worked per Person of Working Age
1995 2014
Austria 17.9 17.9
Canada 19.9 19.9
Denmark 17.5 16.3
France 15.2 14.3
Germany 15.8 15.1
Italy 15.0 14.2
Sweden 18.1 20.5
Sources: International Labor Office, OECD, and authors' calculations.
The decision for women to enter and remain in the labour force is of central importance for an economy’s productive potential. Table 4.3 displays the female
labour force participation rate (the proportion of women of working age in the labour force, whether employed or not). It shows a very substantial increase in
almost every country. The exceptions are the Scandinavian countries, which already had very high participation rates in 1984 and where these rates remain
highest. The variation across countries points to differences in both cultures and institutions. For example, one reason why so many women work in Scandinavia
is the existence of highly developed and subsidized child care systems. Participation remains lower in societies in which women have traditionally played a
greater role at home, but also where taxation may discourage a second earner in the family.
Table 4.3 Female Labour Force Participation Rates
Country 1984 2014

Belgium 45.0 63.0
Canada 61.6 74.2
Denmark 73.3 75.0
Finland 72.3 73.8
France 56.7 67.4
Germany 51.0 72.9
Ireland 36.2 62.5
Japan 54.5 66.0
Norway 65.8 75.9
Portugal 55.1 70.0
Spain 34.3 69.8
Sweden 78.6 79.3
Source: OECD.

Our first attempt at defining unemployment in the last section is somewhat unsatisfactory. Surely, unemployment is more than simply labour withheld voluntarily
from the market. The International Labour Organization (ILO) and the Organisation for Economic Co-operation and Development (OECD) define an individual as
unemployed if he or she does not have a job during the reference period, is actively looking for one, and is ready to work. Let us start by defining the labour force
as those members of the adult population who are either working (L) or unemployed (U). The labour force mainly excludes young people in school, the retired,
and those who are not voluntarily looking for work. It corresponds closely to the amount of labour supplied to the market, given current conditions, including the
level of real wages. Denoting the labour force as LS, we can write:
(4.2)
Labour force = employment + unemployment
The unemployment rate u is the fraction of the labour force which is out of work, or u = U/LS. Perhaps it is now clear why the picture of
unemployment in Figure 4.7 is incomplete: according to the ILO definition, it would have to be zero! In the rest of this chapter, we examine alternative
reasons for unemployment.
4.3.1 Involuntary Unemployment and Real Wage Adjustment

One interpretation of unemployment in the ILO definition is the excess of labour supply over labour demand. In equilibrium depicted in Figure 4.7 this outcome
was ruled out by assumption, because the labour market was assumed to clear. Figure 4.9 considers the important case where the real wage is fixed at , which is
higher than the level which equates supply and demand, w. At , firms are willing to hire labour, while workers supply LS. Since firms cannot be forced to hire
more than they wish, actual employment is and is labour supplied but not demanded by the market. The labour market does not clear and there is
involuntary unemployment. Involuntary unemployment occurs when labour is willing and able to work at the wage but cannot find employment. At point
B, the marginal product of labour (MPL) exceeds the valuation of leisure time by households that are out of work. Put differently, at point B, firms are not willing
to hire all labour that workers are willing to supply at wage .
Fig. 4.9 Involuntary Unemployment

At the real wage rate the quantity which is supplied by households but not demanded by firms, represents involuntary unemployment.
The solution to the unemployment problem in Figure 4.9 appears rather straightforward. If the real wage were to fall from to w (point A), demand would
increase, supply would decrease, and involuntary unemployment would be eliminated. It is the failure of the wage to decline that perpetuates unemployment.
This is a key result: the existence of involuntary unemployment must be explained by real wage rigidity, which we examine next.
4.3.2 Collective Bargaining and Real Wage Rigidity

For sustained real wage rigidity to occur, involuntarily unemployed workers must be unable, on average, to supply their labour services at wages below or
firms must be unwilling or not allowed to take up such offers.11 There are many reasons why wages may not adjust, or do so only slowly. The most important
features of labour markets which have been neglected until now are labour market institutions. Labour market institutions are distinguishing national
features of labour markets based on legal, cultural, operational, or organizational characteristics, which are designed to interfere with or modify the functioning of
labour markets.
Labour (or trade) unions are one of the most fundamental and universal institutions that operate in modern labour markets today. Unions are organizations
of employees which advocate interests of labour in a number of dimensions, including workers’ rights, working conditions, and most importantly, wages. They
exist because workers have considerably less ability to influence these conditions individually than collectively. Labour unions are often matched by equally
powerful employers’ associations. Bargaining between employers and unions contrasts sharply with the perfect-competition description of labour markets
described in Section 4.2. In what follows, we study the motives of negotiators, and how they impact on wage determination. In doing so, we discover how
unemployment can be voluntary from the perspective of trade unions, and nevertheless be involuntary from the viewpoint of the individual household.12 We will
also see that unions try to smooth out the ultimate fundamentals of the labour market as determined by supply and demand in Figure 4.7.
The rationale for labour unions

The employer–employee relationship has inherently conflictual aspects. One is the distribution of income. While economic principles assert that income should
be split according to the marginal productivities of capital and labour, in practice it is difficult, if not impossible, to measure these concepts. Another subtler
reason is that firms cannot observe work effort, a key element of productivity, which is under the control of each individual employee. Individual workers facing a
large employer are in a poor bargaining position. They have little influence over their own wage rate and may not even feel safe discussing working conditions,
fearing reprisals in the form of a salary cut or dismissal. They may even feel pressure to accept conditions that would not be acceptable under competitive
conditions.
Historically, workers have organized themselves into unions to help counteract such pressures, in particular, to achieve higher pay levels and a voice in the
day-to-day operation of the workplace. Table 4.4 gives some details on union organizations and their organizational strength (membership) and impact on
workers’ contracts (coverage). Clearly, labour market institutions vary considerably from country to country. Scandinavian countries have a tradition of
centralized unionization. German workers are organized by industry unions; in Britain unions tend to be associated with members’ craft or occupation. In France,
Italy, and Spain, unions are frequently tied to political parties. In the Nordic economies, union membership is generally greater than three-quarters of the
workforce and coverage is often as high as 90%, while in the USA unions have more or less disappeared from the private (non-governmental) workplace.
Table 4.4 European Trade Unions: Membership and Coverage, 1950–2013
Country Structure Union Membership (%) Collective Bargaining Coverage (%)

1960 1970 1990 2013 2013
Sweden Umbrella (ILO, TCO, SAGO) 72.1 67.7 78.0 67.7 89
Finland Umbrella (SAK, STTK, AKAVA) 31.9 51.3 69.4 69.0 93
Denmark Umbrella (LO, HK, FTF) 56.9 60.3 78.6 66.8 84
Norway Umbrella (LO, AF, YS) 60.0 56.8 58.3 53.5 70
Belgium Party, religious (ACV/CSC, ABVV/FGTB, ACLVB/CGSLB) 39.3 39.9 51.3 55.1 96
Ireland Mostly crafts in ICTU, fragmented 43.1 50.6 54.3 29.6 32
Austria Umbrella/industrial (OGB) 67.9 62.8 56.7 27.4 98
Italy Party, religious (CGIL, CISL, UIL) 24.7 37.0 49.6 36.9 80
United Kingdom Mostly crafts in ICTU, fragmented 38.8 43.0 49.7 25.4 30
Germany Umbrella/industrial (DGB, IG Metall, Ver.di) 34.7 32.0 34.9 17.7 58
Portugal Party, religious/umbrella (CGTP, UGT) 54.8 18.1 67.0
Netherlands Party, religious (FNV, CNV, MHP) 41.7 36.5 34.8 17.6 85
Greece Occupational/sectoral (GSEE, ADEDY) 39.0 20.8 40
Spain Industrial, company level (CC.OO, UGT, ELA, CIGA) 13.5 17.2 79
France Party, religious (CGT, CFDT, CGT-FO. CFTC. CFE-CGC) 19.6 21.7 18.3 7.7 98
United States Mostly local plant-level (AFL-CIO) 30.9 27.4 22.1 10.8 12
Source: International Labor Organization and Labor Force Statistics, OECD.
These vast differences reflect social history as well as the costs and benefits associated with union membership. The costs are dues (fees) that members must
pay. The benefits vary, ranging from higher wages and protection from arbitrary employer decisions to more specific advantages, including priority for certain
jobs and income supplements when unions are on strike. In some countries, many advantages accrue to all workers, so there is little point in paying union dues.
This is the case in France, for example.13 In other countries, such as Belgium and in Scandinavia, unions help manage some social benefits systems, including
unemployment insurance. In the USA, some unions even issue credit cards and provide other financial services to their members. In the end, the influence of
unions is usually much stronger than simple membership statistics would suggest, and is partly related to labour laws which institutionalize their role. This is
evident from Table 4.4, which shows the fraction of all workers working under contracts negotiated by unions, whether or not they are union members.
The economics of labour unions

Simplifying somewhat, we can think of two primary economic objectives which motivate unions: higher real wages, and more jobs.14 It is helpful to think of union
preferences in terms of indifference curves shown in Figure 4.10. The slope of the indifference curve represents the willingness of the union to trade off
employment for wages. It is flat for ‘hard-line’ unions which are unwilling to trade wages against employment, steeper for unions which care relatively more about
jobs than wages.
Fig. 4.10 Trade Unions’ Indifference Curves

When a trade union values both higher wages and more employment, its preferences are described by indifference curves. A ‘hard-line’ union is not willing to give up much in
lower wages to raise employment and would have relatively flat indifference curves. For a union preoccupied with employment, in contrast, the curves would be steep.
In contrast to Section 4.2, we now describe a labour market in which union indifference curves replace those of the representative individual as the relevant
preferences. This captures the fact that the active agent in the labour market is not the individual, but the labour union, or the collective bargaining process more
generally. The ‘budget line’ faced by the union is simply the labour demand curve. From Section 4.2.2 we know that labour demand is given by the MPL (either a
firm’s MPL, or that of an industry or the entire economy: if firms are all alike, their individual demand does not differ from the collective one represented by an
employers’ association).15
Given the demand for labour that it faces, the optimal choice for the union is the tangency point of the highest indifference curve with the current demand for
labour. This outcome is the best the union can do, given the economic environment it faces, which is the behaviour of firms. Should the demand for labour shift,
the union’s menu of options will change too. Capital accumulation or technological progress which shifts the labour demand curve outward, for example, will
increase the wage that firms are willing to pay at a given employment level, or increase the sustainable employment that can be hired, holding the wage constant.
Inward shifts of labour demand are also possible. These are attributable to destruction or obsolescence of the capital stock (e.g. due to wars, earthquakes, the
lack of new investment), or new inventions which make some labour unnecessary. They have the opposite effect, shrinking available options to the union. As the
labour demand schedules shifts, the set of tangency points map out the collective labour supply curve shown in Figure 4.11. The curve describes the most
desired joint evolution of real wages and employment from the union’s perspective.
Fig. 4.11 The Collective Labour Supply Curve

The collective labour supply curve is obtained by connecting the points of tangency between the indifference curves and a shifting labour demand schedule.
The slope of the collective labour supply curve thus reflects the preferences of the union for employment and wages as well as its economic environment. It
will also reflect whether demand is shifting outward or inward.16 Union members who are currently employed, or who enjoy seniority or job protection, tend to
fight for a hard-line union. If their influence is high, the slope of the collective labour supply curve will be steep. In contrast, a ‘jobs-first’ union accepts moderate
real wage increases and supplies more labour. It will tend to exhibit a flat collective labour supply curve.17
Employment effects of collective bargaining

The collective labour supply curve resembles the individual supply curve of Figure 4.3, but has different origins. Collectively, through their unions, workers
increase their bargaining power and, accordingly, aim at better outcomes. In particular, for a given amount of labour supplied, they ask for higher real wages: for
this reason, it should be clear that the union-driven collective labour supply curve can only lie above the individual labour supply curves. Without the union’s
influence on wage setting, equilibrium would occur in Figure 4.12 at point A: individuals would be willing to work up to L at wage w. They cannot, however,
because the wage is set in negotiations between the firms and the trade union, and individuals cannot simply underbid their employed colleagues. The outcome
is a and the resulting unemployment is involuntary for affected individuals, but voluntary from the union’s point of view.
Fig. 4.12 Labour Market Equilibrium with a Trade Union

When a labour union represents workers at wage negotiations, labour market equilibrium occurs at point B. If the union collective labour supply curve is above the individual
labour supply curve, the real wage is higher and employment (hours or number of workers) lower than at point A, which would be the outcome if individuals were
negotiating individually. The result is the existence of union-voluntary, individual-involuntary unemployment (LS – ) which is the difference between actual employment and
the amount of work LS that workers are willing to supply individually at the real wage rate (point C).
How can unions enforce wage rigidity, apparently against the will of unemployed individual workers? One reason is that the leadership is typically elected by
the employed, sometimes called the insiders. The employed are always a dominant majority of the membership, even at record high unemployment rates of 10%
or 20%. Unemployed workers often give up their membership or lose interest in union affairs. They are called the outsiders. Unions end up representing the
insiders, who have jobs, rather than the outsiders, who are an unemployed minority. Employed workers lobby for high real wages (for themselves) at the cost of
some unemployment (for others). Box 4.3 explains how this effect can explain the relentless rise of unemployment in Europe after the two oil price shocks of the
1970s.
The split between unions and unemployed workers cannot go too far, though. After unemployment increased to high levels in Europe in the 1970s and 1980s,
unions became more employment-conscious and real wage growth moderated significantly. There are several reasons for this wage moderation. First, members
became worried that they too might become unemployed. Second, there was criticism from the non-unionized workers. Third, the loss in membership revealed in
Table 4.4 has meant lower income from union dues as well as less overall influence.
Box 4.3 The European Unemployment Problem

In the 1960s, unemployment rates were generally very low, much lower than in the United States (Table 4.5). They increased over the next two decades,
exceeding the US rate in the 1990s. A major reason for the increase was the sudden oil price increases of the mid-1970s and early 1980s, the so-called
oil shocks. These shocks made many production processes obsolete, because they had been developed on the assumption that oil, and more generally
energy, would remain cheap. The consequence was a sharp drop in productivity, which can be represented by an inward shift of the aggregate labour
demand curve. The effect of the oil price increase on employment was limited in the US because wages fell. In contrast, militant labour unions in Europe
successfully fought wage declines for more than a decade.
As unemployment rises, pressure to do something grows on the insiders and their unions. In a number of countries, measures have been taken to help
these countries turn the corner on the European unemployment problem. This was the case in the UK in the 1980s and 1990s and in Germany in the
2000s; the unemployment rate eventually decreased in both countries, but they take time to move the supply curve effectively down. Measures taken in the
2010s in Italy and Spain will eventually produce similar effects.
Table 4.5 Standardized Unemployment Rates 1960–2016 (% of labour force)
1960–1969 1970–1979 1980–1989 1990–1999 2000–2009 2010–2014 2016

France 1.7 3.1 7.6 9.7 8.4 9.5 10.4
Germany 0.7 2.0 5.8 7.8 8.9 5.7 4.9
Italy 4.8 5.9 8.4 10.2 7.9 10.4 11.8
Spain 2.5 4.5 16.3 18.0 11.2 23.3 20.5
Sweden 1.7 2.1 2.6 7.2 6.7 8.1 7.7
United Kingdom 1.6 3.5 9.5 8.0 5.4 7.5 5.4
United States 4.8 6.2 7.3 5.8 5.5 8.0 4.7
Note: For Germany, we use data for West Germany until 1990 and for reunited Germany afterwards.
It would be unfair to assert that unions are solely responsible for real wage rigidity. Employers’ associations can also contribute to real wage rigidity. These
associations represent the collective interests of firms. They represent an additional mechanism for policing collective bargaining agreements reached with
unions. In the end, employers’ associations do not control the demand for labour: this is the prerogative of the individual companies. So, while it is in firms’
interest to keep wages low, it is also in their interest to ‘take the wage out of competition’. This means keeping the wages of their competitors high, or at least
preventing them from hiring cheap labour.
4.3.3 Social Minima and Real Wage Rigidity

Beyond trade unions and employers’ associations, several other institutional and economic factors can contribute to wage rigidity, and therefore involuntary
unemployment. Frequently mentioned in the European context are social minima, or minimum standards for income and earnings mandated by the government for
reasons of social equity or protection. Social minima may stem from social or unemployment benefits, since few will work for a net wage which falls below the
level available for no work at all. They can also be set at a higher level by minimum wages, which are legal limits on how low wages can be.
Minimum wages have been enacted for a variety of reasons. One was to prevent employers with excessive market power from abusing it by depressing wages.
Another reason was to protect young people from exploitation. With schooling rudimentary and poverty endemic, for many youngsters on-the-job training was
the only way to get started. Unscrupulous employers would offer very low wages, sometimes below minimal survival needs. Social protection was and often still
is justified. Even in countries without statutory or legal minimum wages, it is frequently the case that collective agreements are extended to uncovered workers,
and contract wages assume the characteristics of a legal minimum wage. With occasional exceptions, this is generally the case in continental Europe.
The primary economic effect of minimum wages is to discourage firms from hiring workers with low MPL, which tend to be the young, the unskilled, those with
little training, or those with the wrong skills. Figure 4.13 illustrates the effect of minimum wages. To serve any purpose at all, the minimum wage wmin must be
higher than the wage that would be obtained otherwise and which is itself higher than what individuals would accept with market clearing (w). The result is
employment equal to Lmin; unemployment is even higher than the level implied by the wage set in collective bargaining .
Some evidence on the effect of minimum wages is discussed in Box 4.4. Those most likely to be hurt by the existence of minimum wage legislation are poorly
educated young people with no job experience and older workers with obsolete skills; their MPL is simply too low to justify the going minimum wage. The
minimum wage can also affect jobs paying more than the floor, because once in place, it tends to push up into the higher echelons of the wage pyramid, just like a
spring which is compressed. This is because market forces tend to re-establish wage differences between different skill and productivity levels.
Fig. 4.13 Minimum Wages

Minimum wages reduce the demand for labour below the level that would result with either union-negotiated wages or individual-supplied labour.
Box 4.4 Minimum Wages and Youth Unemployment
It is striking that teenagers in the USA often work during the summer when their European counterparts go on vacation. One reason might be that wages
that must be paid for young, unskilled labour are too high in Europe. The minimum wage amounts to about one-third of the average manufacturing wage
in the USA, while in many EU countries it well exceeds 50%. This is one reason why filling station attendants and grocery shop assistants have all but
disappeared in most European countries. Table 4.6 shows non-employment and unemployment rates for young people in a number of countries, as well
as the average minimum wage, as a fraction of the average overall wage. In interpreting the table, it is important to note that Danish youth under 18 years
receive an exemption from collectively bargained minimum wages, as they do in the USA, while in France, Belgium, and the Netherlands, the minimum
wage as a fraction of the median wage is high, meaning that a great many jobs are paid the minimum wage.
In France, the minimum wage, called the SMIC (Salaire minimum interprofessionnel de croissance), is an important element of the collective
bargaining system. It is set by a council on which both the government and unions are represented. Many government employees receive the SMIC. In
recent years, 10–15% of workers in industry, commerce, and services earned the SMIC or near it, a much higher proportion than in the USA (about 3–4%).
While the minimum wage in general has a negative effect on youth employment, it may lead to a substitution of adults for youths, even increasing the
employment of the former. Recent evidence from Portugal’s experience with the minimum wage supports this hypothesis.
We have emphasized the economic effects of minimum wages. The public debate, in contrast, is frequently guided by considerations such as social
equity and social norms. A minimum wage is supposed to help the fight against ‘working poverty’. Yet it is rarely mentioned that the minimum wage
destroys jobs as long as the labour demand curve is elastic. Thus, those who might have held those jobs are implicitly ‘financing’ the minimum wage—
those whose productivity is just below the level necessary to justify the job’s existence. The public purse might be a better approach to solving the problem
of the working poor.
Table 4.6 Minimum Wages and Youth Labour Market Experience, 2013–2014
2014 Minimum wage as % %affected (at or near Youth minimum as %of Non-employment ratio for Unemployment rate for
of median wage minimum) 2013 adult level (relevant age) youth age 15–24 (%) youth aged 15–24 (%)
Belgium 51 0.3 0 (under 16); rising to 94 (22) 69.8 23.2
Denmark Collective 40(<18) 38.5 12.6

agreements
France 61 8.2 (16); 90(17); 30–75 (trainees) 63.4 23.2
Germany Collective Industry agreements 50 7.8

agreements
Greece 46 0.8 89 72 52.4
Italy Collective Industry agreements 70 42.7

agreements
Netherlands 48 6 30 (15); rising to 85 (22) 32.6 12.7
Spain 41 0.2 60.4 53.2
UK 48 8.3 61.7 (16–17); 85 (18–21) 38.8 16.3
USA 37 4.3 58 (<20) 45 13.4
EU 28 56.5 21.9
Source: OECD.
4.3.4 Efficiency Wages and Real Wage Rigidity

Another reason why real wages may not decline in the presence of involuntary unemployment is that firms themselves may not wish to lower them for various
reasons. The phenomenon is often called efficiency wages, and it is related to another special aspect of labour. In contrast to other factors of production,
work effort is not easily observed by firms, and yet it matters a lot. By paying higher wages, firms may attempt to elicit more work of better quality. If the wage is
relatively high, a worker who is dismissed for lack of effort is unlikely to obtain such a good deal elsewhere, especially if dismissals are interpreted as a sign of
poor work effort. Firms may also pay higher wages to obtain a better selection of applicants and to keep workers from quitting too often. If all firms behave this
way, the outcome will be a wage level which exceeds the market-clearing value w in Figure 4.13.
In capital-intensive industries, where shirking could seriously disrupt the production process and where a high-quality workforce is of primary importance,
firms may have a strong incentive to pay efficiency wages. In this case, the function of real wages goes beyond simply equilibrating demand and supply in labour
markets. Generally, wages will not be able to satisfy both functions. They will tend to be rigid and lie above the market-clearing level, as in Figure 4.12.
4.4.1 Labour Market States and Transitions

Any person in the working-age population is either employed, unemployed, or out of the labour force. Figure 4.14 displays these three states and how flows
occur between them. A striking aspect of labour markets in developed economies is the sheer size of these flows. Table 4.7 shows that the flow of individuals
moving into and out of unemployment over a year represents several times the stock of unemployment at any particular point in time. In contrast to the static
impression conveyed by Section 4.3, labour markets are remarkably dynamic, even when unemployment seems stuck at high levels.
There are three ways of becoming unemployed:
First, new entrants to the labour market join the labour force before they have found work, but are initially unsuccessful.
Second, voluntary separations of workers from jobs can lead to unemployment. In general, separations occur due to quits, layoffs (redundancies), plant
closures, or expiration of fixed-term contracts. Quits represent voluntary separations from the employee’s viewpoint. They account for roughly 50% to 66% of
all separations from employment in the UK, and up to 70% in the USA. Most workers who quit one job take up another immediately—a transition from
employment to employment. Of those who quit but do not start a new job, most leave the labour force, usually for family reasons (e.g. maternity leave), return
to school, or retire. Relatively few workers quit into unemployment.
Finally, involuntary separations, or job losers, tend to flow into unemployment (although some may just decide not to look for work). Job loss may result from
termination of fixed-term contracts (common in France and Spain), unanticipated redundancies or layoffs (more common in Denmark, the UK, and the USA),
and closure or relocation of factories (which occurs everywhere).
Fig. 4.14 A Map of Labour Markets

Every individual is in one of three states: employed, unemployed, or out of the labour force. Over any period of time, large numbers of workers are flowing from one state to
another.
Table 4.7 Unemployment Stocks and Flows, 2015
Average unemployment Unemployment flows (millions per year)

(level, in millions of persons)
inflows outflows
Austria 0.35 1.01 1.13
Germany 2.79 7.52 7.60
UK 1.77 3.10 3.35
USA 8.30 47.50 48.26
Sources: AMS (Austria); Bundesagentur für Arbeit (Germany); Office of National Statistics (UK); Bureau of Labor Statistics (USA).
4.4.2 Stocks, Flows, and Equilibrium Unemployment

In earlier sections, we have ignored the obvious fact that no two positions and no two persons are the same. This has led us to overlook that pairing a worker
and an unfilled job opening—also called a vacancy—is not always easy and may take time. The matching of skills, occupation, industry, and geographical
location requires a large amount of information. The more efficient the labour markets are, the faster the match is achieved. From this perspective, unemployment
emerges as an unavoidable result of the dynamics of labour force movements, and the normal process of job creation and destruction as firms grow or shrink, are
born or die.
In addition to the efficiency of the job-matching process, equilibrium unemployment depends on the number of job separations and the number of vacancies. If
we ignore the flows from and to ‘Not in the labour force’ in Figure 4.14, the number of workers who become unemployed (per month or per year) represents a
fraction s , called the separation rate, of existing employment relationships (L). While sL workers flow into unemployment each period, a number of
unemployed workers find jobs and flow out of unemployment. If we use f to denote the job finding rate, i.e. the fraction of the unemployed (U) who find
employment during the period, the change in unemployment in a given period is given by:
(4.3)
In the dynamic interpretation of labour markets, equilibrium unemployment is the stock that results when flows into and out of unemployment are equal, or when
∆U = 0 in (4.3). Expressing unemployment as a proportion of the labour force LS = L + U and neglecting transitions in and out of the labour force allows us to use
the flow perspective of labour markets to write the equilibrium unemployment rate u as:18
(4.4)
We can think of the proportion s of workers separated from their jobs as the probability of losing a job if currently employed and, similarly, the proportion f of
unemployed workers represents the probability of finding a job if unemployed.19 Equilibrium unemployment is higher, the higher the job separation rate and the
lower the job finding rate. The next two sections will examine more closely the determinants of these two indicators—the job separation rate s and the job finding
rate f.
4.4.3 Job Separation and the Incidence of Unemployment

The separation rate s is a measure of the incidence of unemployment. It has both structural and cyclical components. The structural aspect is linked to the ease
with which firms can dismiss workers, regardless of the business cycle. It is lower in countries where legal and social restrictions on layoffs exist (as in most
continental European countries) than in countries where redundancies are more acceptable (e.g. Denmark, the UK, and the USA).20 The cyclical aspect simply
refers to the fact that during recessions the probability of losing a job rises and so, therefore, does equilibrium unemployment.
Table 4.8 shows that in one particular country (the UK) in a given year (2004), job separation rates vary considerably across various characteristics of labour
market participants. Just as equation (4.4) implies, those specific labour force groups that exhibit higher separation rates of inflow into unemployment tend to
have higher unemployment rates.
Table 4.8 Inflows into Unemployment and Unemployment Rates in the UK, March 2004
Inflow rate into unemployment (monthly, as %of employment) Unemployment rate (%of relevant labour force)
By region: Britain
East Midlands 0.7 2.9
Eastern 0.5 2.3
London 0.8 3.6
North-East 1.0 4.6
North-West 0.8 3.3
Northern Ireland 0.7 4.0
Scotland 0.9 3.9
South-East 0.5 1.8
South-West 0.5 1.9
Wales 0.8 3.5
West Midlands 0.8 3.6
Yorkshire and the Humber 0.8 3.4
Total 0.7 3.1
By demographic group: UK
Aged 16–17 1.0 21.1a
Aged 18–24 2.2 9.9a
Aged 25–49 0.6 3.9a
Aged 50 and over 0.4 2.5a
Male 0.9 4.3
Female 0.4 1.7
a Average from Dec 2003 to Feb 2004.
Source: UK Labour Market Trends, March 2004.
4.4.4 Job Finding and the Duration of Unemployment

Like the separation rate, the job finding rate f has cyclical and structural components. It is highly procyclical, increasing when output and employment are rising
and declining when the economy is contracting. On the structural side, the job finding rate depends on a number of long-run factors, including labour market
institutions, the effectiveness of the worker–job matching process, and the suitability of the unemployed for work in the vacancies which firms are posting. It
depends on how hard the unemployed look for jobs, how many job openings are available, and how easy it is to spot an opportunity (and how many
opportunities are available). It depends on incentives to remain unemployed, and unemployment insurance may therefore slow down the exit rate out of
unemployment. Many of these aspects will be discussed in Chapter 18.
Unemployment benefits are intended to assist the jobless while they search for work. Table 4.9 shows that unemployment benefit systems vary considerably
from country to country, with respect to eligibility criteria, income replacement, and the period over which they are paid. While unemployment benefits reflect
a widely perceived need for income insurance, solidarity, and social conscience, they may encourage unemployed workers in declining industries to wait for an
unlikely recovery rather than to retrain and change sectors, possibly at a lower wage. They also act as a disincentive for looking for a job, or as an incentive for
being ‘choosier’.21 If the benefits are generous, and particularly if they are long-lasting, unemployed workers may take more time to find an acceptable job, a time
in which their skills and re-employability may deteriorate. This ‘unemployment trap’ often applies to unemployed low wage-earners who face loss of benefit upon
taking on a new job. The last column of Table 4.9 shows a tendency for people to remain unemployed longer in countries where unemployment benefits are more
generous, paying more income over longer periods. As the finding rate declines, equilibrium unemployment rises. This confirms an uncomfortable trade-off
between social concern and economic efficiency.
4.5.1 The Concept

If all unemployment were voluntary, it would hardly attract any attention. The existence of high and evidently involuntary unemployment means that labour
markets do not function like other markets. The existence of market imperfections, arising from both economic and institutional factors, makes it necessary to
modify the market-clearing paradigm of Section 4.2. We should also consider alternative concepts of equilibrium besides the equality of demand for labour by
firms and the supply of labour by households.
The unemployment rates shown in Figure 1.2 clearly fluctuate in the short run over 2–3-year intervals, but also move slowly over a longer horizon. It is these
longer-run movements in unemployment which are of interest to us now. Long-run labour market equilibrium can be summarized by an unemployment rate that
would occur in the absence of cyclical disturbances. Because of imperfections, labour markets may be in equilibrium and yet unemployment may not be limited to
voluntary unemployment. Equilibrium unemployment can be thought of as the sum of frictional and structural unemployment:
(4.5)
Table 4.9 Unemployment Compensation: Conditions for Eligibility and Benefit Levels, 2015
Country Eligibility conditions: Maximum benefit Replacement rate of net Restrictions Long term unemployment (>12
duration (in weeks) earnings/Benefit levelsa months) as %of all unemployed, 2015
Employment Period
Austria 28 weeks 12 52 55% according to 27.3
months wage class
Belgium 312 days 21 36 months 60% of max 52.3
months earnings
Canadab from 420 to last 1 45 55% average 12.9 (2014)
700 hours year covered
earnings
Denmark 52 weeks 3 years 104 90% of average 27.3
earnings
Finland 26 weeks 28 72 €32.66 per day 23.1
months
France 4 months 28 104 57.4% (75%) of average 43.1
months daily wage
Germany 12 months 2 years 104 60% of net 43.2
earnings
Greece 200 days 2 years 51 55% (70%) daily wages 71.6
(monthly
salary)
Ireland 39 weeks last 1 45 €188.00 per week 60.7
year
Italy 52 weeks 2 years 61 75%, 60%, 45% for every 6 57.7
months of gross average daily
wage
Japan 12 months 24 47 50–80% of average 37.6 (2014)
months daily wage
Netherlands 26 weeks 36 38 months 70% of average 43.5
weeks salary
New resident for indefinite NZ$173.92 per week 13.6 (2014)
Zealand 24 months
Norway registered 104 0.24% of annual 29.0
employment income per
day
Spain 12 months last 6 102 70% (60%) of average 52.3
years earnings
Sweden 480 hours or 12 43 320 kr per day 19.3
6 months months
Switzerland 12 months 2 years 74 80% of last 35.7
earnings
UK c 26 £65.45 per week 32.7
16-20
USAb 1 year 26 50% taxable 23.0 (2014)
weeksd
a Single household without children.
b Data from 2013.
c Contributions must have been paid on earnings equal to at least 26 times the lower earnings limit (£111 from April 2014) in one of two tax years.
d Different across US states.
Sources: Social Security Programs Throughout the World http://www.ssa.gov.
Frictional unemployment can be thought of the unemployment which is unavoidable, even under the best of conditions, because it takes time for a match to
occur between a worker seeking a job and a vacancy needing to be filled. It depends on the efficiency of the labour market, including the eagerness of both
parties to find a match quickly. The frictional unemployment rate may well vary over time, but only slowly as the market’s efficiency changes and policies make it
more or less likely for people to find jobs or to become unemployed in general. Is frictional unemployment voluntary or involuntary? Probably a bit of both. Some
laid-off workers do find it genuinely difficult to quickly find a new job. Others may be choosy.
Structural unemployment describes the unemployment caused by inappropriate wage levels that are not allowed to adjust to their market-clearing levels. It
has many causes. The common theme is that the demand and supply of labour are influenced by a number of institutions and regulations. Collective labour
supply, which is brought into balance with labour demand in equilibrium, is generally inconsistent with individual supply behaviour. Some workers are
involuntarily unemployed even when real wages equate the collective supply of labour with the demand of firms. Another possibility is that an economy’s
industry has suffered a loss of competitiveness, meaning that at any level of wages employment is lower, meaning that the inflow rate ( s ) and equilibrium
unemployment rate (u) are higher.
Estimates of equilibrium rates of unemployment are provided in Table 4.10. In many respects, they track down the averages presented in Table 4.5. This is
normal as actual unemployment rates fluctuate around their equilibrium over business cycles. Like in Table 4.5, the contrast between Europe and North America
is striking. The equilibrium unemployment rate was generally very low in Europe in the 1960s. Since then it has risen considerably while remaining stable in the
USA. In Germany and the UK, however, they have come down to levels not much above those in the 1960s. To understand this dramatic evolution, we return to
the two components of the equilibrium rate of unemployment—demand for labour and the supply of labour.
4.5.2 The European Experience

The evidence suggests that European unemployment rose when large numbers of workers lost their jobs at the time of the great oil price increases. The expected
subsequent return to pre-oil-shock levels was thwarted in many European countries by a fall in the finding rate, so exit from unemployment became increasingly
difficult. Is the development of the social safety net to be blamed for having provided workers with the incentive to wait out their unemployment rather than to
accept reductions in their wages? Simple comparisons between Europe and the USA, such as the one in Figure 4.15, seem to confirm this suspicion. Naturally,
Europe was at a different point on its growth path than the US, given the destruction of capital and productive labour which occurred in the Second World War,
yet such a growth rate is hardly consistent with GDP growth registered during the period. Not surprisingly, employment grew less in Europe than in the US.
Table 4.10 Estimates of the Equilibrium Unemployment Rate (% of the labour force)
1970–1979 1980–1989 1990–1999 2000–2009 2010–2015

Germany N/A N/A 7.3 8.4 5.6
Italy 5.5 7.4 9.3 8.4 8.6
Japan 1.7 2.4 2.9 4.1 4.0
Spain 5.8 11.0 15.0 13.9 17.9
United Kingdom 5.3 9.3 8.0 5.8 6.4
United States 5.9 6.6 5.7 5.7 5.5
Source: OECD, Economic Outlook No 98.
Fig. 4.15 Employment and Real Wages in the Euro Area and the USA, 1970–2015
Real wages are defined as total compensation per employee deflated by the GDP deflator. Over the period 1970–1990, real wages doubled in the euro area, while employment
rose by about 10%. During the same period, real wages in the USA increased by only about 50%, while employment rose more than 70%. Since 1990, real wage growth in the
euro area has moderated sharply, and growth of employment has increased. The distribution of this growth across Europe is not even, however.
Source: OECD.
Yet there is also striking evidence that Europeans can combine high employment, low unemployment, and social protection. The social safety net is even more
developed in Denmark, Sweden, and Norway, where long-term unemployment has remained lower. This implies that what really matters is not the safety net itself,
but the disincentives that it can generate. Unemployment benefits, for example, provide an alternative to finding a new job, and help transform temporary
unemployment into permanent—that is, structural—unemployment. Long-term unemployment has become increasingly widespread, and as workers gradually
lose their human capital and contact with the active labour force, they become unsuitable for any vacancy.
The strikingly different evolution of the equilibrium unemployment rate across countries also points to the importance of wage-setting institutions. When the
labour market cannot adjust and wage flexibility is obstructed, equilibrium unemployment is bound to respond adversely to abrupt changes to labour demand
such as energy crises, globalization, or technological change. This applies to labour supply shocks as well, such as immigration, increasing female participation,
etc. The sharp contrast between the US or the UK and Europe suggests this. This is also the reason why reforming labour market institutions to react to these
changes can make a big difference. Box 4.5 shows that the reforms adopted in Germany in the mid-2000s—which had political costs for the politicians who
supported them—helped Germany deal better with the Great Recession than other economies. We return to these issues in Chapter 18.
Box 4.5 European Labour Market Policy and the German ‘Miracle’
Most European countries have experienced chronically high rates of unemployment since the mid-1970s. In many places, high and rising unemployment
has become an accepted fact of life, with average unemployment rising after every economic downturn. Yet it may be surprising that in the 1960s and early
1970s, Europe had lower joblessness than the USA. Economists widely agree that the ratcheting upward of unemployment in Figure 1.2 was the
culmination of bad private and public responses to a string of adverse economic events: sharply higher energy prices in the 1970s, technical change in
the 1980s, and globalization and increasing international trade in the 1990s. All of these acted like large negative productivity shocks and shifted labour
demand to the left, only to be reinforced by a slowdown in investment in response to the bad times. The late 1960s and early 1970s were a particularly
militant period for European trade unions, with a number of general strikes and labour unrest across the continent and the UK were able to push for real
wage rigidity. Passive support for the unemployed was an easier choice than promoting structural change. Facing a rising tide of unemployment, many
politicians simply accepted unemployment as inevitable and as beyond the influence of policy, focusing instead on ways to redistribute income from those
who have jobs to those who don’t. The term ‘Eurosclerosis’ was coined by German economist Herbert Giersch of the Kiel Institute of World Economics to
characterize this dark period.
Yet all was not lost in Europe. Unemployment today in Germany has fallen to half of what it was in 2003, the year the Economist described it as the ‘sick
man of Europe’. In achieving this, it stands out, together with a handful of other European countries that have turned the tables on Eurosclerosis. Figure
4.16 provides the details.
How did Germany do it? The following factors were all crucial:
(1) Unions took a less confrontational stance, effectively recognizing that the labour demand curve was flatter than originally thought, the result of Europe’s
opening up to trade with Eastern Europe and Asia;
(2) Starting in the mid-1990s, firms and unions began to bargain in a more decentralized fashion, allowing individual firm profitability to influence wage
determination;
(3) Regulatory burdens on flexible forms of employment, including part-time work, temporary help agency, and minijobs (low paid jobs …) were
significantly reduced;
(4) Administration of unemployment insurance and assistance was tightened and benefit levels were reduced both in amount and duration. Pressure was
increased on workers to accept offers of employment or accept retraining, or lose some or all benefit.
Figure 4.16 compares the trajectory of unemployment in Germany with that of France since the late 1950s. Both countries are of comparable size and
experienced very similar macroeconomic disturbances: sharp oil price increases in the 1970s, the introduction of cheap mass computing,
telecommunications, and the internet; globalization and the opening to trade with Eastern Europe and Asia; and the introduction of the euro in 1999.22
Despite different institutions, until 2005, the experiences of the three countries were largely similar; every recession was associated with a long-lasting
increase in unemployment rates in both countries. Only after the tough labour market reforms of 2003–5 was Germany able to turn the corner on
unemployment.
Fig. 4.16 Unemployment in France and Germany, 1955–2014

Until the mid-2000s, French and German unemployment rates moved in tandem. After 2005, when major labour market reforms were undertaken in Germany, they parted
company.
Source: FRED Database, Federal Reserve Bank of Saint-Louis.
4.5.3 Actual and Equilibrium Unemployment

It can take many years before real wages adjust to their long-run values in Figures 4.7 and 4.12. In the meantime, actual unemployment can deviate from
equilibrium unemployment. Actual employment is below, and actual unemployment above, equilibrium when the real wage is above the equilibrium level, as at
point A1 in Figure 4.17. When the real wage is low, firms may be able temporarily to move away from the union-set collective labour supply curve towards the
individual labour supply curve (point A2), for example by using agencies specializing in temporary jobs or overtime work. Workers may have overestimated the
real wage by underestimating the rise in the price level. Firms may be willing to hire more workers at the going wage. In such situations, employment is above, and
unemployment below, the equilibrium level. These deviations, while short-term in nature, tend to be associated with fluctuations of the economy that we know as
the business cycle. We revisit this important topic in Chapter 12.
Fig. 4.17 Actual and Equilibrium Employment
When unions negotiate on behalf of workers, market equilibrium occurs at point A, and equilibrium unemployment is . Actual employment and unemployment may differ if the
real wage is slow to move to its equilibrium level w . If it is above the market equilibrium level (w 1 > w ), firms reduce employment to L1 and actual unemployment exceeds
equilibrium unemployment. Conversely, below-equilibrium real wages (w 2 < w ) enable firms to connect with structurally unemployed workers willing to work at lower wages
than the collectively bargained level. The resulting unemployment rate is lower than the equilibrium level.
Summary
1 Households trade off leisure against consumption. An increase in wages can induce more labour supply if the substitution effect dominates, i.e. labour supply is relatively
elastic. It will supply less labour if the income effect dominates. When the two effects offset each other, labour supply is inelastic and does not change.
2 Individual labour supply seems to be inelastic in the short run. But in the long run it is more likely to be backward-bending, as the higher incomes afforded by higher real
wages allow households to enjoy both more leisure and more consumption. Aggregate labour supply is more responsive to real wage changes than that of households in the
short run, as real wage increases draw new individuals into the labour force.
3 The demand for labour by firms depends on its (marginal) productivity, which is determined by the available technology and the capital stock. Firms hire labour to the point
where the marginal productivity of labour is equal to the real wage. The labour demand schedule is shifted outwards by an improvement in technology or an increase in the
capital stock.
4 Equilibrium employment and the wage level are given by the intersection of labour demand and labour supply. Improvements in technology or increases in capital will be
reflected in higher wages if labour supply is inelastic, and in higher employment if labour supply is elastic.
5 Involuntary unemployment arises when real wages do not decline to clear the market so that not all labour supplied by households is hired.
6 Labour unions care about real wages and employment. In determining their target wage, given the demand for labour by firms, they ask for higher real wages than if the labour
market were perfectly competitive. While the resulting unemployment rate is voluntary for unions, it may be involuntary for individuals.
7 Because firms cannot easily monitor work effort or wish to elicit lower turnover or improve worker quality, they may pay efficiency wages. This is yet another reason why
real wages may be set above market-clearing levels.
8 Labour markets are characterized by widespread government interventions. Minimum wages, designed to protect workers, can actually cause unemployment. Despite this,
governments may see minimum wages as an effective means of guaranteeing a socially acceptable minimum income for those who find work.
9 The labour market is characterized by large flows between states of employment, unemployment, and being out of the labour force. These flows are an important dimension
of labour markets and determine equilibrium unemployment.
10 The efficiency of job search can vary across individuals and countries, and is affected by government labour market policies. Unemployment benefits, designed to make
unemployment more bearable, provide disincentives to finding a new job quickly, thereby increasing equilibrium unemployment. Other programmes, such as training and
relocation subsidies, can reduce equilibrium unemployment.
11 Because there are labour market institutions as well as natural market imperfections, unemployment is never zero or entirely voluntary, even under the best of conditions.
Frictional unemployment is the result.
12 Equilibrium unemployment is the sum of frictional and structural unemployment. Structural unemployment is due to fundamental problems related to the demand and
supply of labour. Individuals may be willing to work at lower wages than those prevailing in equilibrium, but may not be able to underbid in the market. This is the sense in
which real wages are downwardly rigid.
13 Real wages are slow to adjust to disequilibria, if only because they fulfil many other roles. This speed of adjustment will depend on labour market institutions, among other
things. As a result, actual and equilibrium unemployment may differ for some time.
Key Concepts
endowment
leisure
unemployment
consumption–leisure trade-off
indifference curves
utility
marginal rate of substitution
real (consumption) wage
relative price
substitution effect
income effect
household labour supply curve
labour force, labour force participation rate
person-hours
marginal productivity of labour
labour demand curve
elastic, inelastic
voluntary unemployment
involuntary unemployment
real wage rigidity
labour market institutions
labour (or trade) unions
employers’ associations
collective labour supply curve
insiders and outsiders
minimum wages
efficiency wages
separations, separation rate
finding rate
equilibrium unemployment rate
unemployment benefits
equilibrium unemployment
frictional unemployment
structural unemployment
Exercises
1 Suppose that the household in Figure 4.2 receives an inheritance. Show the effect on its decision to work and to consume. According to this result, do rich people work more
or less than poor people? Explain your answer.
2 Suppose Robinson Crusoe is paid a higher wage (‘overtime pay’) if he works more than 8 hours a day, but only has 16 hours at his disposal.
(a) Draw his budget line in this case.
(b) Does the existence of overtime necessarily make him better off?
(c) Show Crusoe’s optimal behaviour for ‘normal’ indifference curves.
Under which conditions will he choose to work overtime? Under which conditions will he refuse?
3 Suppose Crusoe must pay for his work tools (coconut husks?) and must spend an hour a day climbing the trees to get to the coconuts. Think of these as fixed costs which
must be paid as long as he is working at all in the labour market.
(a) Draw his budget constraint in this case.
(b) Show Crusoe’s optimal choice for the case that he goes to work, grumbling about the fixed costs, but voluntarily. What do you conclude about his utility level
compared with staying at home?
(c) Now sketch the case in which Crusoe decides to stay at home. What can you say about his utility level and preferences in this case? Show that if the wage rises
enough, Crusoe can be enticed to leave his tree house.
(d) Now apply this to a modern-day worker who faces commuting cost as well as a monetary/resource cost of working, regardless of how high the wage is or how many
hours are worked. Show both cases—the worker who works and the worker who stays at home. How could an ‘in-work’ grant to those who work any positive
amount of time entice the worker to accept a job? Could it fail?
4 It is frequently claimed that Europe’s unemployment problem is due to high labour taxes in the form of employer and employee contributions to social security, which are
usually proportional to wages. Assume that individuals care about after-tax wages.
(a) Using the machinery of Figure 4.3, show the effect of a high tax versus low tax environment. What is the effect on labour supply and equilibrium wages? State your
assumptions carefully.
(b) What is the effect on equilibrium employment levels resulting from the increased tax for the case of the utility function U = cαℓ1−α (ignore labour demand)?
(c) What changes in your answer to (b) if the tax revenues are rebated to the households as a lump-sum payment, i.e. unrelated to their income tax payments? Given
that high labour taxation (social security charges and income taxes) in France, Germany, and Italy coincides with high rates of transfers to households (as opposed to
Denmark and Sweden) how could you explain the observed pattern of market labour supply in Table 4.2?
5 Derive the collective labour supply curve graphically. Show how different union preferences can interact with the same shift of labour demand and lead to: ( a) rigid wages
around some ‘target level’ with flexible employment; (b) flexible wages around some target level of employment; (c) flexible wages when employment is rising, but rigid when
employment is declining.
6 In the 1980s, many European countries tried to tackle the unemployment problem by encouraging early retirement, with arguments that only a fixed amount of jobs are
available for job-seekers. Use the supply and demand framework to show how the advocates of such a policy might have argued. Why do you think that most policy-makers
have now given up on early retirement as a way of fighting unemploy-ment?
7 It is well-known in Europe that the unemployment rate among well-educated workers is considerably lower than for those who left school early or without training. Why
might this be the case? How might an ‘education offensive’ help solve the unemployment problem, even if wages for the low-skilled are rigid?
8 In Japan the bonus system is widespread. Workers receive as much as 30% of their pay in the form of a profit-contingent payment, which can go up or down depending on
the fortunes of the enterprise in which they work. What are the implications of such a system for real wage rigidity and equilibrium employment?
9 It is sometimes claimed that the influx of foreign migrants into the labour force is a cause of unemployment. Show the effect of immigration on the labour market as described
by Figure 4.12. What does it mean for real wages and employment?
10 Show the effect of a technological innovation which raises the MPL in a market with an upward-sloping supply curve for labour when: (a) wages are flexible; (b) wages are
rigid. Now show the effect of technical change which reduces the MPL. Given these differences, evaluate how unions might regard technical change differently, depending on
their preferences. Given your answer, assess why unions are frequently in disagreement about the merits of a minimum wage.
Essay Questions
1 How would you determine whether unemployment is voluntary or involuntary? How do you respond to the criticism that it is virtually impossible to distinguish between
the two?
2 Critically evaluate the statement that labour unions should assume responsibility for the administration, financing, and disbursement of unemployment benefits.
3 Severance regulations—government interference with the separation of workers from firms which frequently occurs in modern economies—are universally thought to
influence labour markets, but it is hotly disputed whether they increase or decrease unemployment. Discuss. Do you think there are other costs involved in the regulation of
redundancies?
4 Discuss both sides of the minimum wage debate. What kind of hard facts might help resolve the argument?
5 From equation (4.4) it follows that all policies which reduce unemployment can be boiled down to two types of measures: reducing the inflow rate s and increasing the
outflow rate f. Discuss policies which could affect these two rates. Are they necessarily good for the economy—in the sense of increasing output of the economy or utility
of members of society?
1 Irving Fisher (1867–1947) was an American who is often described as one of the greatest mathematical economists of all time. Among other subjects, he contributed to the
theory of investment, capital, and interest rates; monetary economics and the theory of inflation; and most notably to the theory and practice of price index numbers. He was
founder and first president of the Econometric Society.
2 Karl Marx (1818–1883) was a German economist and political philosopher whose theories predicted the immiseration of labour and the ultimate crisis and collapse of market
(capitalist) economies. Although he clearly got all that wrong, his empirical observations on the plight of the working class at the time and his focus on issues relating to product and
labour markets—and his influence on the lives of millions of people who lived and worked under communism and socialism—earn him a central place in the history of economic
thought.
3 ‘Firms’ are understood to include public as well as private enterprises, national, regional, and local governments. It can even mean working for oneself, in the case of self-
employment. In what follows we will focus on firms which strive to maximize profits.
4 The use of this character, based on the classic novel by Daniel Defoe (1660–1731), is traditional in economics and will be seen again in Chapters 7 and 8, in which goods are
distinguished by the point in time that they are consumed.
5 The nominal budget constraint is To write it in real terms of consumption goods in (4.1), we simply divide both sides of the equation by P, the price of
consumption goods.
6 If Crusoe had some initial wealth to begin with, the budget line would be shifted upwards (vertically) by that amount. This would then represent consumption attainable without
having to work at all.
7 For details, the reader is invited to look at the WebAppendix.
8 The labour force participation rate is defined as the proportion of working-age people which is either working or registered as unemployed.
9 These important cases are taken up again in detail in a box in Chapter 18.
10 Aggregate employment is sometimes measured as the number of people who have a job. Generally, we will use the first definition (person-hours), and make explicit mention when
referring to the number of employed workers. Under any definition, when more workers enter the labour force, the labour supply curve shifts to the right independently of the wage
level.
11 It is important to stress ‘on average’, since workers are different, varying by occupation, skill level, and industrial affiliation. Thus, real wage rigidity may affect workers with low
productivity more than those with high skill levels.
12 It should be stressed that we limit ourselves strictly to the economic significance of trade unions. As the history of the labour movement amply demonstrates, unions have had an
enormous influence on modern society beyond their economic impact.
13 This is an illustration of the so-called free-rider problem. If no workers pay dues, the union disappears and no one is protected. So some workers must pay the dues for all to have a
union. Often workers tend to join unions for reasons related to peer pressure, social norms, or ideology.
14 As noted already, unions care about many other aspects of labour markets, such as safety at work, working time, employees’ rights, and influence over working conditions and
organization. They also care about the size of their membership. To simplify the analysis, these aspects are not considered here.
15 Here we assume that firms are behaving competitively—each firm on its own. More likely than not, associations of employers will arise in response and exert their own power in the
bargaining process, presuming they can organize and discipline enough firms. A more general treatment of bargaining when employers are non-competitive is possible but beyond the
level of this textbook.
16 Note that the collective labour supply curve depends on the successive shifts in the demand for labour that occur. Different shifts can change both the position and the slope of the
labour demand curve. A number of outcomes are possible. These are described in more detail graphically in the WebAppendix to Chapter 4.
17 More details on different possible shapes of the collective labour supply curve are discussed in the WebAppendix.
18 If ∆U = 0, it follows that sL = fU. Substituting LS − U for L and dividing by LS yields (4.4).
19 These figures conceal a large degree of heterogeneity in the labour market: some individuals find a job easily after becoming unemployed, whereas others may have very low
probabilities of exiting unemployment.
20 It is, however, usually the case that outflow rates in countries with employment protection are also lower, because firms are more reluctant to hire new workers. For that reason, the
effect of employment protection on the unemployment rate is ambiguous.
21 Strictly speaking, this applies only to those who already qualify for benefits. Prior work experience is often required before one can receive unemployment insurance benefits. In this
case, individuals will be more willing to accept the first job. This is called the ‘entitlement effect’.
22 The data refer to West Germany before 1990, to unified Germany afterwards.
Money, Prices, and
Exchange Rates
in the Long Run
5
5.1 Overview
5.2.1 Money
5.2.2 Money and Prices
5.2.3 Money, Prices, and Output
5.2.4 Nominal and Real Interest Rates
5.3.1 Nominal Exchange Rates
5.3.2 Real Exchange Rates
5.3.3 Movements in Nominal and Real Exchange Rates
5.3.4 Measuring the Real Exchange Rate in Practice
5.4 The Exchange Rate in the Long Run: Purchasing Power Parity
Summary
It is indeed evident that money is nothing but the representation of labour and commodities, and serves only as a method of rating or estimating them. Where coin is in greater plenty, as a greater quantity of it is required to represent the same quantity of goods, it can have no effect, either
good or bad… any more than it would make an alteration on a merchant’s books, if, instead of the Arabian method of notation, which requires few characters, he should make use of the Roman, which requires a great many.
David Hume1
One must know the value of money; the spendthrifts know it not, and the greedy even less.
Montesquieu
5.1 Overview
In this chapter we bring a new actor to the macroeconomic stage: money. Money is a central theme in macroeconomics—in fact, so central that macroeconomics is sometimes even called monetary economics. It may be surprising that we waited until now to
introduce this concept! But there is a good reason to do so. The key message of this chapter is the principle of monetary neutrality: in the long run, the supply of money does not matter for the real aspects of the economy—called the real economy—
considered up to now: long-run economic growth, standards of living, employment, and unemployment. This is why we did not need to bring money into the picture, but the neutrality principle must be established. This chapter therefore represents a critical
juncture: it is the last one dealing with the long run and it is the first one dealing with money. This chapter only offers a cursory introduction to the topic; the central role of money in our economy and in macroeconomic policy is the subject of Chapters 9 and 10.
In addition to money, this chapter will open up the economy to the rest of the world in an important set of ways: we will see that different economies can trade goods, services, and financial asserts with each other. Throughout this book we will be concerned
with such open economies. To this end, we need to introduce the exchange rate, the price of our money in terms of other monies. The exchange rate, appropriately adjusted, can give us insight into the meaning of external competitiveness, or how well our
goods compete with those of other countries. It will also provide another implication of money neutrality: in the long run, there is not much that monetary policy can do about competitiveness. Short-run changes in the price of one money in terms of another
cannot increase our long-term welfare.
We start by presenting the logic behind the neutrality principle. This leads us to derive its central implication—that inflation is a monetary phenomenon. If you wonder why a country undergoes persistently high inflation, you have nowhere else to look but the
central bank, which has let money grow too fast in the past. In addition, the overwhelming evidence also shows that high-inflation countries see their currency depreciate at about the same rate.
In order to understand this second result, we will need to define two exchange rates: a nominal and a real one. The nominal exchange rate is familiar; it is the price of one money in terms of another that is quoted in newspapers and by banks. The real
exchange rate is a more subtle concept which tries to address the notion of what money can actually buy in the two countries being compared. It involves adjusting the nominal exchange rate for price levels at home and abroad. While the real exchange rate is
characterized by both short-run fluctuations and wider medium-term swings, the neutrality of money implies that the real exchange rate is constant in the long run, and independent of monetary factors.
One difficulty with quoted exchange rates in practice is that they always link two monies, very often the home currency and the US dollar. But what about other bilateral exchange rates that relate the home currency to important neighbours and trading partners?
In order to tackle this issue, we discuss the construction of effective exchange rates, both nominal and real, which represent averages of bilateral nominal and real rates.
In the end, this chapter serves two purposes. First, it presents the fundamental neutrality result and some of its implications. This will serve as a bridge between the long run discussed up to now and the short-run evolution of the economy that is the subject of
the next eight chapters. Second, it introduces key variables—money and the exchange rate—which are the bread and butter of macroeconomics.
5.2.1 Money
We all think we know what money is, don’t we? After all, it is rare to spend a day not having to deal with money in some form or another. Well, it is considerably trickier than it looks and Chapter 9 will provide a detailed definition, or rather several definitions, of
money. To make things even more confusing, people tend to use the term ‘money’ when they mean income (wages, interest, or pension benefits) or wealth. For economists, money is what you use when you go shopping: an asset which is readily accepted in
exchange by others. It is a medium of exchange.
A related question is where that money comes from. Again, a full explanation will have to wait until Chapter 9. While civilizations have seen money in the form of gold, silver, salt, exotic seashells, cigarettes, or private IOUs 2 of banks or other companies, money
in modern times has been ultimately provided by a government or by a government-sanctioned institution. While this need not necessarily be the case, Chapter 10 will explain that the control of the supply of money is exerted by the monetary authority, the central
bank; this is all that we need to know here.
A fundamental principle of economics is that money ultimately only affects the price level, leaving the real side of the economy untouched. This is what David Hume meant in the epigram to this chapter. Think of money as a numbering system that allows us to
express prices of all the goods and services we buy and sell in common terms. The exchange rate can be seen as a way of converting this numbering into a different measurement unit—going from euros to pounds sterling, for example. This suggests that prices
and the nominal exchange rate should grow at the same rate as money, in the long run at least. This common-sense idea is called the monetary neutrality principle. It is readily illustrated in the left-hand side chart of Figure 5.1, which shows average annual
inflation and money growth over about 30 years, a suitably long time period. Later chapters will provide more precision about what we mean by ‘the long run’. The chart shows that, indeed, those countries where money has been growing fast have had high
inflation rates. Over a period of decades, high monetary growth is no accident or unavoidable fate, but is subject to the control of the central bank. At the same time, the figure also shows that the neutrality principle suffers from many exceptions. We will first look
at why the principle makes sense and then at why it is only an approximation of reality.
Fig. 5.1 Money Growth, Inflation, and Exchange Rate Depreciation, 1975–2006
The left-hand side chart shows the average annual rates of inflation rate and money growth in 155 countries. The average annual depreciation vis-à-vis the US dollar is related to the average annual difference of inflation in the same countries (excluding the USA) and in the USA. It is
displayed in the right-hand side chart. We have excluded the pathological countries where inflation rates have been in excess of 100% per year on average during the entire period. For most countries, the annual averages are computed over the period 1975–2006. For some, the period is
shorter, because data are not available for all years (this concerns the formerly planned economies until about 1995 and the euro area countries after 1999 since the latter have abandoned their own monies).
Source: International Financial Statistics, IMF.
5.2.2 Money and Prices

Money is held by households and firms to carry out transactions. For this reason, money is often called a medium of exchange. As a first approximation, let us start by assuming that people simply hold cash as a proportion k of their income in each period,
measured in money units.3 In Chapter 9, we will refine this analysis. For a country as a whole, total income is measured by GDP. Denoting the real GDP as Y and the price level as P, the nominal GDP is PY, we state that the total demand for money, denoted M,
is simply proportional to the nominal GDP:
(5.1)
This equation is known as the Cambridge equation, and is the simplest and most readily understood formulation of the demand for money. 4 The neutrality principle is easily understood with the following simple thought experiment: imagine that prices of all
goods and services were to double overnight. As a result, a €20 banknote today only buys today what a €10 banknote could buy yesterday. The value of money—the purchasing power of money—has fallen by 50%. In fact, its value is inversely proportional to
the price level, which has doubled.
Box 5.1 Evidence for Monetary Neutrality: The Introduction of the Euro
An excellent example of monetary neutrality was the introduction of the euro, the money currently used by almost 340 million Europeans. On 1 January 1999, 11 members of the European Union gave up their national currencies ‘irrevocabl’ and adopted
the euro as their new money. On 1 January 2002, euro banknotes became legal tender for transactions in Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain. (Later over the next 15 years, Cyprus,
Estonia, Greece, Latvia, Lithuania, Malta, Slovakia, and Slovenia would join.) The adoption of irrevocable exchange rates three years earlier had already rendered these national currencies ‘non-decimal units’ of the euro. As monetary neutrality would
predict, the introduction of the euro and the abandonment of national monies represented nothing but a mere rescaling of prices, with no lasting consequences for the real economy. The short-term impact of the event on the inflation rate has been
estimated at 0.2–0.5% and this was a one-off effect, well within the margin of measurement error.
What followed the introduction of the euro is a different story, however. It lies at the heart of the interaction between real and nominal variables. The introduction of a new money means new institutions and agencies—the European Central Bank—as
well as new responsibilities for the creation of money. In addition, the introduction of the euro has affected the behaviour of many economic agents in the euro area. This is the exciting subject of the rest of the textbook.
Money is the ultimate nominal variable. All nominal variables are measured using the monetary unit: euros, pounds, dollars, roubles, kroner, etc. We will even sometimes use the word ‘monetary’ as a substitute for ‘nominal’. Box 5.1 uses the introduction of the
euro in continental European countries as a plausible illustration of monetary neutrality.
Since the purchasing power of money is inversely proportional to the price level, it can be written as the ratio M/P. This is also called the real value of money, or real money for short. M/P measures the purchasing power of the money stock M. Dividing both
sides of the equation by P, we can rewrite (5.1) in real terms as:
(5.2)
This expression relates real money demand to the real GDP Y. This is logical; in the absence of money illusion, real variables ought to be related to real, but not to nominal variables. Now let us go a step further. Money, we noted, is created by the central bank.
What happens if the central bank doubles the stock of money overnight and distributes the new banknotes to everyone? People will be happy at first! They will want to buy more goods and services. The question is how producers and sellers will respond. In the
short run, they might well supply more goods and services to their customers, a process that we will examine in great detail in Chapters 11 and 12. But this is unlikely to last very long; otherwise, it would mean that growth could be increased simply by printing
more and more money! Most would think that this is too good to be true, and indeed it is. Every few years, another country tries the recipe and fails. The unfortunate fact is that, for reasons to be fully spelled out in Chapter 13, faster money growth will ultimately
only lead to higher inflation. More precisely, the neutrality principle asserts that money has no long-run effect on output and, more generally, it has no long-run effect on any real variable.
In the extreme, high inflation tends to lead to higher inflation, which sometimes even leads to hyperinflation. A hyperinflation is somewhat arbitrarily defined as an episode during which the rate of price increase exceeds 50% per month. Hyperinflations occur
rarely, but when they do they cause great disruption and social damage, for reasons discussed in more detail in Chapter 16. In the twentieth century, spectacular hyperinflations were observed in Germany (1922–3), Greece (1943–4), Hungary (1945–6), Taiwan
(1948–9), Bolivia (1983–4), Yugoslavia (1993–4), and Bosnia-Herzegovina (1994–5). Box 5.2 tells the story of Zimbabwe, which wins the prize for the highest inflation in the twenty-first century so far.
Let us try to make this reasoning more formal. If the nominal money stock M doubles, the left-hand side of (5.2) also doubles, all other things being equal. The monetary neutrality principle says that, in the long run, real GDP is unaffected. This implies that the
right-hand side remains unchanged. From a purely logical standpoint, this can only work in equation (5.2) if k or P changes. If k is constant, as we have assumed thus far, P must perform the balancing act. Indeed, if P also doubles, then the real money stock M/P
remains constant and nothing has changed. Thus the long-run neutrality principle implies that money and prices must remain proportional to each other in the long run. Formally, the neutrality principle implies that inflation π, the growth rate of the price level, is
equal to the growth rate of money:
(5.3)
We have reached this conclusion by logical means, but there is an important economic story lurking in the background. When we say that the central bank prints more money, we are really talking about an increase in the supply of money. When we say that
households and firms collectively want to hold cash in a proportion k of their purchases, we mean the demand for money. Then, we can interpret (5.1) or (5.2) as a money market equilibrium condition: demand (on the right-hand side) must equal supply (on the left-
hand side). Equilibrium in the money market requires that demand and supply be kept equal. How does money market equilibrium come about? With k constant, M exogenously set by the central bank, and Y unaffected by M as a consequence of the neutrality
principle, it is the price level P that brings supply back in line with demand.
Box 5.2 Zimbabwe: The Top of the Inflation Heap
One of the highest inflation rates in modern history and certainly the highest in the last half-century is due to Zimbabwe in southern Africa. By the time the calculation of price indices was halted in August 2008, inflation stood at 231 million per cent. If one
counts 22 working days in a month, this means that prices were then rising at a rate of 120% per day. And that was not even the peak! Figure 5.2 presents quarterly measurements of the inflation rate, expressed at an annualized rate of change. The scale
is cut off at 800% per year, otherwise we cannot observe anything. In late 2008, banknotes were denominated in the trillions of Zimbabwean dollars, and yet they were nearly worthless! Figure 5.3 shows two banknotes from Zimbabwe, issued in mid-
2008.
The root cause of the hyperinflation was the fact that the central bank created money at full throttle, as Figure 5.2 shows. It did so because the government was broke and could not pay for essential services, like the police or the army. The government
simply borrowed money from the central bank, the Reserve Bank of Zimbabwe, which gladly accommodated and printed the banknotes. Yet it blamed the underground economy, threatening to ‘take stern and unprecedented punitive measures against
the dark forces of parallel market trading and financial disintermediation’. According to the central bank, ‘[t]he blossoming cash barons, smugglers and other illicit dealers will threaten the stability of our national payment systems’. In fact, most citizens
had stopped using the local currency. As soon as they were paid in Zimbabwean dollars, they converted them into US dollars.
At such high rates of price increase, it becomes difficult even to measure inflation. At some point, President Mugabe forced firms and stores to cut prices by half. Rather than complying, firms sold their goods on the illegal underground market, where
prices exploded even further. With shops empty, the Central Statistical Office simply stopped producing the consumer price index. Its chief statistician, who refused to sample prices in the informal markets, candidly explained that ‘there are too many data
gaps’. Finally, in late 2008, with inflation hitting 79.4 billion per cent per annum, and bank notes of billions of Zimbabwean dollars (Figure 5.3), the central bank bowed to reality and allowed trade to be carried out in any currency. The Zimbabwean dollar
has effectively disappeared, mostly replaced by the US dollar and the South African rand. In a matter of weeks, inflation stabilized. In 2014, it stood at about 1.5% and became negative in 2015.
Fig. 5.2 Annual Inflation in Zimbabwe,2001–2015
The figure displays the rate of increase of the consumer price index and of the money stock over the previous four quarters. Surprisingly, the price level seems to be growing less rapidly than money since 2006. This is when the government started to impose price controls. Observers
report that inflation is much higher than recognized by the official numbers. The figure suggests that the observers are probably right. After year-end 2008, when the government allowed US dollars and South African rand to be used in trade, inflation had fallen to less than 5%.
Sources: Reserve Bank of Zimbabwe and IMF.
This reasoning is represented in Figure 5.4, which depicts the supply of money by the central bank (S) and the demand for money by the private sector (D). Equilibrium occurs at the intersection of the two schedules and determines the price level. An increase in
the nominal money supply is met by a proportional increase in the price level.
Fig. 5.3 Zimbabwean Banknotes, Mid-Year 2008

When inflation reached its peak in 2008, prices were rising so fast that many banknotes were worth less than the paper they were printed on and were burned for fuel! During the year 2008, the banknotes raced through denominations from 10 dollars to the bill portrayed above for
50,000,000,000 dollars and even higher. On one day in the autumn of 2008, the lower bill for 100 billion Zimbabwe dollars could buy three chicken eggs.
Fig. 5.4 Money Market Equilibrium
Nominal supply, the right-hand side of (5.1), is represented by the vertical schedules S1 for M1 and S2 for M2. According to the left-hand side of (5.1), demand is proportional to P (for given k and Y), which is represented by the line D. Money market equilibrium occurs at A1 when the
supply is M1 and at A2 when the money supply is M2.
5.2.3 Money, Prices, and Output

Let us now be a little more sophisticated. The left-hand side chart in Figure 5.1 provides only limited support for the neutrality principle as stated so far. The various observations tend to line up along the diagonal, but there are many deviations from the general
pattern. For example, so far we have reasoned as if output is constant but, in the long run, we expect real GDP to grow along the principles developed in Chapter 3. In line with the neutrality principle, money was never mentioned in that chapter. The view that
nominal variables do not affect real variables in the long run, also called the dichotomy principle, implies that long-run growth is independent of the evolution of money. On the other hand, growth proceeds on its own path and affects real money demand; this
could well explain some of the apparent departures from neutrality in Figure 5.1.
Box 5.3 Arithmetic Rate of Change
In several parts of this book, we use two arithmetic principles concerning rates of changes. If we look at a variable x, we may be interested in its rate of change Δx/x, where Δx represents the change over a given period (a day, a month, a year, etc.). Rates of
change are most often expressed in per cent per annum (per year). For example, if x increases from 10 to 12, Δx = 12 − 10 = 2, so Δx/x = 2/10 = 0.2, which is also 20%. The two principles concern the rates of changes of products and ratios. With two
variables, say x and y, we can ask what is the rate of change of their product xy. The intriguing answer is that it is the sum of the rates of change of the individual variables x and y:
Similarly, the rate of change of a ratio is roughly the difference of the rates of change of the individual variables:5
As an example, consider the real money stock M/P. Its rate of growth is the difference with the nominal growth rate and inflation, the rate of growth of the price level P:
If the real money stock M/P is constant, Δ(M/P) = 0 and it must be the case that ΔM/M = π.
To see why, note that as the economy grows, more transactions will occur, and more money is needed to conduct those transactions. Indeed, the logic of (5.1) is that nominal money demand increases proportionately to GDP as long as k remains constant.
Nominal money demand also increases proportionately to the price level. Using the arithmetic presented in Box 5.3, we can be even more precise and amend equation (5.3), which was derived by assuming zero output growth—Y was assumed to be constant—to
take into account GDP growth. Then we find that the growth rate of money demand is equal to the sum of GDP growth and inflation:
(5.4)
Now, we can reason as before. If the central bank determines the money supply and if GDP growth is driven by other, non-monetary factors, as explained in Chapter 3, this relationship in fact explains inflation. To see this formally, we can rearrange (5.4) to find:
(5.5)
This result can be interpreted as follows. Money demand is driven by GDP growth. Money supply is set by the central bank. If, for instance, the central bank lets the money supply grow faster than GDP, there is too much money around. People, i.e. households
and firms, have more money than they wish to hold. Of course, no one throws money away. People spend it. There is now too much money chasing too few goods and services (the GDP) and inflation plays the balancing act. Rising prices reduce the purchasing
power of money, i.e. the real value M/P of money, until (5.1)—or (5.2)—is satisfied. Conversely, if money grows more slowly than GDP, inflation is negative. Put differently, the only possibility of reconciling discrepancies between real money demand ( M/P) and
nominal supply (M) growth is for inflation to be an endogenous variable which can help ultimately to restore long-run money market equilibrium. Note that (5.3) is a special case of (5.5), corresponding to the situation where GDP is constant.
How does this analysis help explain the left-hand side chart in Figure 5.1? Consider two countries: one grows fast and the other grows more slowly, say not at all. In the second country, in the long run, inflation is equal to the money growth rate, as in (5.3), while
it is the excess of money growth over GDP growth that determines inflation in the first country, as in (5.5). Now imagine that money growth rates are equal in both countries. Obviously, inflation will be lower in the first country, by an amount equal to its GDP
growth rate. Graphically, it means that the point representing the first country in the left-hand side chart in Figure 5.1 will lie below the point representing the second country. If all countries were growing at the same rate, we would expect to see all the points
neatly set along a straight line, parallel to the first diagonal, but below it. If growth rates differ among countries, we expect deviations from this line. These deviations should seem small if the growth rates are not too different. This observation explains much, but
not all, of what we see in the chart. To explain more, we need to understand why k is not necessarily constant, which will be shown in Chapters 10 and 14. For the time being, we accept the rule of thumb given by (5.5) and the money neutrality principle.
Table 5.1 Inflation and Money Growth in the Long Run: A Rule of Thumb (assuming that real money demand grows at 3% per annum)
Nominal money supply (%) Inflation rate (%)

0 –3
3 0
8 5
50 47
103 100
This rule of thumb is illustrated in Table 5.1, where we assume that real GDP grows in the long run, on average, at a rate of 3%. This means that the real demand for money also grows by 3% per year. If the central bank allows the nominal money stock to grow at
the same 3% per year, demand and supply coincide when the average inflation rate is 0%. Were the nominal money stock to grow at an average annual rate of 8%, inflation would have to be 5%. Only then would the real money stock grow at the same 3% rate as
real money demand (8% nominal money growth less 5% inflation). This is what we mean when we say that, in the long run, inflation is a monetary phenomenon.
5.2.4 Nominal and Real Interest Rates

When you borrow from your bank, you are told the interest rate that will apply. If you borrow €1,000 at a rate of 4% per annum, you will have to pay €40 in interest, in euros, at the end of the year. We would say that 4% is the nominal interest rate. But if
inflation is positive, say, 1%, €40 in one year’s time will be worth less than €40 today. In ten years’ time, if inflation remains positive, it will be worth even less. Thus, a positive inflation rate has the effect of reducing the real cost of borrowing. To reflect this effect,
it is useful to think in terms of the real interest rate:
(5.6)
.
This relationship will play an important role in Part III, when we study inflation. Until then, we will effectively ignore it and assume that the inflation rate is zero—so the distinction between the nominal and real interest rates is moot. Yet, the neutrality principle
applies here as well.
In the presence of positive inflation, borrowers and lenders are well aware of the distinction between the nominal and real interest rates. For a given nominal interest rate i, an increase in inflation means a lower real interest rate r, as can be seen in (5.6). This is
good news for the borrower, but bad news for the lender. Conversely, a reduction of inflation will raise the real interest rate, to the benefit of the lender and at the expense of the borrower. Fairness—and market forces—imply that neither the lender nor the
borrower should benefit from something that is outside their control. This is why it is the case that, under normal conditions, the interest rate i rises one for one with the inflation rate, leaving the real interest rate r constant. This is another manifestation of the
neutrality principle, and it is subject to the same qualifications as those already mentioned, and more. We will study this relationship again in Chapter 14.

The neutrality principle also applies to the exchange rate, which is an important variable that we have not yet encountered. This section will give a few definitions and then show an implication of money neutrality.
5.3.1 Nominal Exchange Rates

Everyone who ever travelled to a foreign country has encountered nominal exchange rates. These rates are posted in foreign exchange rate booths at airports and border posts and they tell us how much it costs to swap one currency for another. These exchange
rates are an annoying aspect of tourism for most of us, yet as the relative price of money, they are an essential part of our lives and it is useful to understand how they work. Nominal exchange rates are set in foreign exchange markets. These markets are studied in
detail in Chapter 15, which also explains how markets determine nominal exchange rates.
Exchange rates can be quoted in either of two ways. The first, as the number of foreign currency units per domestic unit, is called British terms, e.g. $1.4 per £1 from the perspective of UK residents, or £0.8 per €1 from the point of view of the European
Monetary Union. A second way of quoting the foreign exchange rate is sometimes called European terms because, before the euro, most European countries used it to quote the US dollar or British pound sterling. It is the inverse of British terms and is the
number of domestic currency units needed to buy one unit of foreign currency. For example, we have CHF1.1 per €1 for Switzerland, or DKR7.5 per 1€ for Denmark.
We will adopt the convention of British terms since it is commonly used for quoting the euro and the pound. With this convention, an appreciation of a currency conveniently corresponds to an increase in its value in terms of foreign currencies, so the
exchange rate rises (e.g. from $1.1 to $1.2 for €1). Conversely, a loss of value, or depreciation, implies a decrease in the exchange rate.
5.3.2 Real Exchange Rates

Nominal exchange rates compare the relative prices of two different monies. They don’t say anything about how much one can buy with each currency. This is the role of prices. The higher the price level, the less one can buy with, say, one unit of currency.
However, because domestic prices are measured in the domestic currency and foreign prices in the foreign currency, price levels cannot be directly compared. Before we do any comparison, we must convert prices into the same currency, and this is where the
nominal exchange rate comes in.
Consider the following simple example. In Berlin, a haircut costs €20. In New York, you have to pay $36 for the same service. To decide which city has the cheaper service, you need to take the exchange rate into account. Suppose that it is 1.5 dollars per euro.
This tells us that in dollars, the Berlin haircut costs $30, less than in New York. What happens if we compute the price in euros instead? The New York haircut costs ($36)/(1.5$/€) = €24, still more expensive than in Berlin. Obviously, it does not matter which
currency we use. The relative price of haircuts in Berlin relative to New York is the ratio of these two prices expressed in the same currency. It is 0.833, the ratio of €20 to €24 or, equivalently, the ratio of $30 to $36.
Real exchange rates normally compare broad price level indices such as the CPI of the GDP deflator rather than individual goods.6 Let the price index in the euro area be denoted P, and P* in the US. The price index represents the cost of a basket of goods and
services; a European basket that is worth €100 in the euro area and another US basket worth $200 (the baskets need not be the same). Denote the nominal exchange as S. In our example, S = 1.5, so the dollar value of the European basket is SP = $150. The real
exchange rate is the ratio of these two baskets expressed in the same currency. In dollars this works out to be SP/P* = 1.5 × 100/200 = 0.75. For this comparison, we used the dollar but we could have looked at the euro values of these baskets. The euro value of the
dollar basket is P*/S = 200/1.5 = 133.33. The relative price is therefore €100 (P) divided by €133.33 × (P*/S) which is again 0.75.
Note that in the first example we compared a haircut in both places, and could decide where it was cheapest. Now the real exchange rate is the ratio of two index numbers, or average prices for different baskets of goods in two different places. Thus the value of
the real exchange rate does not mean much. The European basket collects goods and services in proportions that represent European tastes and habits. It is usually set equal to 100 in some reference year. It is likely that the US basket includes goods and services
in different quantities (more baseball equipment, fewer soccer balls) and it may be at 100 in a different year. In the next section, we see that what matters are movements in the real exchange rate.
In the end, the real exchange rate is an index defined as:
(5.7)
.
The real exchange rate can be thought of as the nominal exchange rate ‘doubly deflated’ by foreign and domestic goods prices. Like their nominal counterparts, real exchange rates appreciate when σ increases, and depreciate when σ declines.
5.3.3 Movements in Nominal and Real Exchange Rates

Nominal and real exchange rates move in tandem as long as inflation is the same at home and abroad, that is when P/P* remains unchanged. Now let us look at the case when the nominal exchange rate S is constant. If inflation at home is higher than it is abroad,
domestic prices are rising faster than foreign prices and the ratio P/P* must be increasing. The real exchange rate σ appreciates. Conversely, a lower inflation at home results in a real depreciation.
The real exchange rate can thus appreciate for two different reasons: (1) when the nominal exchange rate appreciates and inflation rates at home and abroad are equal; and (2) when the nominal exchange rate is stable but inflation is higher at home. More
generally, using the arithmetic presented in Box 5.3, we can compute the rate of change of the real exchange rate:
(5.8)
.
This decomposition shows that the real exchange rate σ is driven by two factors: (1) changes in the nominal exchange rate S, and (2) the inflation differential, the difference between domestic (π) and foreign (π*) inflation. The logic is simple. A real appreciation
is a loss of competitiveness, as our goods become expensive relative to foreign goods. This happens when the nominal exchange rate appreciates, or when domestic inflation exceeds foreign inflation. In the first case, the foreign price of domestic goods (SP) rises
even if the domestic price is unchanged, or the domestic price of foreign goods (P*/S) declines even if the foreign price is unchanged. In the second case, even if the nominal exchange rate is unchanged, domestic goods become relatively more expensive since π >
π* implies that P/P* increases. These various possibilities are illustrated in Figure 5.5, which presents effective exchange rates—effective rates are explained in the next section—for the Swedish krona.
5.3.4 Measuring the Real Exchange Rate in Practice

In practice, measuring real exchange rates poses two problems. The first concerns the definition of ‘foreign’. ‘Foreign’ stands for the rest of the world, the various countries with which our country has significant trade and financial relationships. Exchange rates
are often quoted in terms of the US dollar. This practice reflects both the size of the US economy and the role of the dollar as an international currency. Yet, the USA cannot really stand for the rest of the world, the large number of countries with which a given
country has bilateral exchange rates. How should one go about aggregating these rates? The same question can also be applied to the foreign price level P*.
Fig. 5.5 The Swedish Krona’s Nominal and Real Exchange Rates, 1970–2015
Three exchange rates are displayed: (1) the nominal bilateral rate between the Swedish krona and the US dollar; (2) the nominal effective exchange rate of the krona; and (3) its real effective exchange rate. All rates are expressed as indices that equal 100 in the year 2010. Because
Sweden is deeply integrated in the EU, the effective exchange rate does not follow closely the dollar value of the krona. From year to year, when inflation differentials are smaller than the exchange movements, the nominal and real effective exchange rates tend to move together. Over the
longer run, however, inflation differentials accumulate to sizeable magnitudes. While the effective real exchange rate is relatively stable, the effective nominal exchange rate displays wider fluctuations and a long-run depreciation.
The solution to this problem consists of computing an ‘average exchange rate’ and an ‘average foreign price level’ using weights that reflect the relative importance of each partner country (ideally, all of them). The weights assigned to each foreign country are
chosen to represent its importance to us, e.g. in trade or in financial flows. Box 5.4 explains how this is done. The corresponding values of S and σ are called nominal and real
effective exchange rates, respectively. These are indices—we can no longer express the nominal exchange rate in value terms, e.g. dollars per euro—which are normalized to take a simple value, such as 1 or 100, in some base year. Figure 5.5 presents an
example.
Box 5.4 Computing and Comparing Effective Exchange Rates7
Effective exchange rates are computed using a number of partner-countries. Each partner-country receives a weight typically representing its importance in trade for the country in question. For example, its share of our exports or our imports, or the
average of both. Geometric averaging is applied to price indices in these countries and to our bilateral exchange rates vis-à-vis their currencies.8 If n partner countries are selected and Si is our bilateral nominal exchange rate vis-à-vis country i with trade
weight wi, our effective nominal exchange rate is:
where the weights are fractions which sum up to 1 (Σwi

= 1). The effective foreign price level P is computed by applying the same weights to each partner-country i's price index Pi:
Then the effective real exchange rate is simply an

average of all our real exchange rates vis-à-vis each partner-country:
The effective real exchange rate is thus a weighted geometric average of the individual bilateral exchange rates which characterize a country’s international trade. This approach can be applied to the consumer price index, to the GDP deflator, to wages,
and other indices.
Another issue concerns which prices to use. In fact, a real exchange rate is simply the price ratio of two baskets of goods and services, one of which represents domestic goods and services and the other foreign goods and services. Many baskets, and many
definitions of ‘the’ real exchange rate, are possible. So far, we have looked at the ratio of consumer price indices, but there are a number of alternatives:
One frequently used real exchange rate is the ratio of domestically produced exports to foreign-produced import prices. It is called the external terms of trade because it indicates how many imports we get in return for our exports.
Another definition considers the ratio of non-traded to traded goods prices, sometimes called the internal terms of trade. This recognizes that many goods are not traded internationally and measures the price of those goods relative to goods that are.
Still another possibility is to measure the real exchange rate on the basis of the price of labour, using wages, total earnings, or total labour costs, which include employer contributions to social insurance.
In the end, there is no uniformly appropriate measure. Different questions are best answered with different indices.
5.4 The Exchange Rate in the Long Run: Purchasing Power Parity
We now present a second implication of money neutrality, which is displayed in the right-hand side chart in Figure 5.1. For each of the 155 countries, it plots the average annual rate of nominal exchange rate depreciation vis-à-vis the US dollar (i.e. − ΔS/S, or the
negative of the rate of appreciation) against the inflation differential vis-à-vis the US, the difference between the country’s inflation rate and the US inflation rate (π − π*). The chart suggests that, in the long run, countries which have a high rate of inflation also
have currencies which depreciate at the same rate. Conversely, currencies with a low rate of inflation appreciate in the long run. If we treat foreign inflation π* as a given, the lower the domestic rate of inflation, the higher the rate of appreciation:
(5.9)
.
This property is expected to hold in the long run. It carries a pretty striking implication, which can be seen by noting that together (5.8) and (5.9) imply:
(5.10)
.
In other words, the real exchange rate is constant over the long run. This conclusion should not surprise those who looked carefully at Figure 5.5. Over 40 years, the effective real exchange rate of the Dutch guilder has fluctuated, but in a narrow margin, and most
importantly, the absence of any trend is suggestive of long-term constancy. The conclusion that the real exchange rate is constant in the long run is sometimes presented as the principle of purchasing power parity (PPP).
The logic behind PPP is quite simple. Suppose that monetary policy is permanently more expansionary at home than abroad. We expect inflation eventually to be higher at home. If the nominal exchange rate remains unchanged, the real exchange rate
appreciates. This means that domestic goods and services become expensive relative to foreign goods and services. As a result, domestic producers gradually lose competitiveness. This obviously cannot go on forever, and indeed it will not. The nominal
exchange rate will depreciate, and this will work towards restoring competitiveness. If it depreciates by the full amount of the accumulated inflation differential, the loss in competitiveness is entirely erased. Thus PPP can be seen as simply asserting that, in the
long run, a country must retain its competitiveness.
There are two versions of PPP. The version presented here is called relative PPP. It is a natural implication of monetary neutrality. Because they are nominal variables, the nominal exchange rate and the price level are not expected to affect real variables in the
long run, including the real exchange rate. This is a manifestation of the dichotomy principle.
An alternative version of purchasing power parity, absolute PPP, is a much stronger proposition. It starts with the Law of One Price. Not really a law, this hypothesis holds that the same good should trade everywhere at the same price, when prices are
expressed in the same currency. If prices were to differ significantly, it is asserted, traders would buy the goods where they are cheapest, sell them where they fetch a higher price and make a profit. Demand would rise where prices are low and supply would rise
where prices are high. Arbitrage, as this process is called, should continue until profit opportunities have disappeared. This means that prices have been equalized throughout the world.
Absolute PPP is a much stronger hypothesis than relative PPP. Relative PPP requires that the real exchange rate σ be constant, while PPP equates it to unity. Indeed, according to the Law of One Price, P* = SP, which implies σ = SP/P* = 1. In practice, the Law of
One Price is known to be grossly violated, not just across countries but within the same country, using the same currency! In contrast, relative PPP is a pretty good rule of thumb and is supported by empirical evidence, as Figure 5.1 shows.
This discussion highlights the difference between the long run, which has been the subject of the last four chapters, and the short run, which lies before us. While it is a good guess that a continuously depreciating currency reflects persistently high inflation,
Figure 5.5 indicates that relative PPP is not entirely robust, certainly not from quarter to quarter. Rather, a closer look suggests that nominal and real effective exchange rates tend to move together. Especially when inflation is low, it does not move one-to-one with
the rate of depreciation. Over many years, however, inflation differentials will add up to sizeable magnitudes and the effective nominal and real exchange rates will move differently. This is what happened until the late 1990s. The nominal exchange rate appreciates
sharply while the real rate remains trendless. From equation (5.8), we can deduce that inflation in Sweden was lower than abroad. In fact, the negative inflation differential and the nominal exchange appreciation offset each other. Since the late 1990s, inflation
differentials have vanished and the nominal and effective rates move closely together.
Notice that the dollar bilateral rate is much more volatile than the effective rate. This is normal since the latter is a weighted average of many bilateral rates, one of which, not the most important, is the US dollar. Yet, dollar exchange rates receive considerable
attention, which may give an exaggerated impression of volatility.
The nexus between nominal and real exchange rates is one of the most important in macroeconomics, especially for small open economies. There are interesting cases to come, in which the nominal exchange rate is not solely driven by inflation differentials. In
the short run, sharp movements of the nominal exchange rate imply sharp movements of the real exchange rate, with implications for the real economy. This represents an important transmission mechanism from nominal to real variables and will be the subject of
extensive discussion in Chapter 11. Finally real exchange rates themselves can change, depending on the fundamental real economy. Chapter 15 will explain the conditions under which real exchange rates are constant in the long run.
Summary
1 Money is necessary because we need it to carry out day-to-day transactions. We demand money because we receive income and because we spend it.
2 The Cambridge equation asserts that the demand for money is a constant fraction of nominal GDP, a measure of total income.
3 The monetary neutrality principle asserts that money does not affect real variables. This is a manifestation of the dichotomy principle. The principle is understood to hold in the long run. It implies that, in the long run, prices, nominal exchange rates, and all nominal variables change at
the same rate as money is growing.
4 The nominal exchange rate defines the price of a currency in terms of another one.
5 Real exchange rates measure the price of domestic goods in terms of foreign goods.
6 Effective nominal and real exchange rates are weighted averages of a country’s exchange rates vis-à-vis the rest of the world—in practice, its main trading partners.
7 Relative purchasing power parity is obtained when the equilibrium real exchange rate is constant. In that case, the rate of change of the nominal exchange rate is equal to the inflation differential, the difference between foreign and domestic inflation. The inflation differential is a
fundamental determinant of the long-run evolution of the nominal exchange rate.
8 Absolute purchasing power parity states that the equilibrium real exchange rate is not just constant but equal to unity. The Law of One Price asserts that prices for the same good are the same in different countries. If that were true, absolute purchasing power parity would tend to be
verified. The evidence tends to reject absolute purchasing power parity for most countries.
9 While the evidence supports relative purchasing power parity in the long run, there are significant deviations from this proposition in the short run. This suggests that movements in the exchange rate are not immediately offset by exchange rate changes, and that nominal and real
variables interact with each other in the short run.
Key Concepts
principle of monetary neutrality

real economy
exchange rate
external competitiveness
nominal and real exchange rate
neutrality principle
demand for money
Cambridge equation
hyperinflation
dichotomy principle
nominal interest rate
real interest rate
British/European terms
appreciation/depreciation
inflation differential
effective exchange rates
external terms of trade
internal terms of trade
absolute and relative purchasing power parity (PPP)
Law of One Price
arbitrage
Exercises
1 Table 5.1 implies a relationship between money growth and the inflation rate when the GDP trend growth rate is 3% per annum. Represent this relationship graphically. Show how this relationship would change if the GDP trend growth rate were 8% instead. What conclusions can you
draw?
2 What happens to the real exchange rate between two countries if the price level at home doubles, all other things given? What if the price of foreign goods doubles? What if the nominal exchange rate doubles?
3 The following figure displays the evolution of the nominal and real effective exchange rates of Portugal since 1970. What can you conclude about the rate of inflation in that country?
4 Money is neutral in the long run, but is it neutral in the short run? Looking at the Cambridge equation (5.2), how could money increase without the price level increasing proportionately?
5 Consider an economy where the stock of money is growing by 5% per year while GDP expands by 2.5% annually. Assume that inflation abroad is 3.5%. What will be the rate of change of the exchange rate in the long run?
6 Use equations (5.5)—written for both the home and foreign countries—and (5.8) to find an expression that describes the long-run evolution of the nominal exchange rates as driven by the domestic and foreign money and GDP growth rates. Denote domestic and foreign money supply as
M and M*, respectively, and domestic and foreign GDP as Y and Y*.
Essay Questions
1 Most governments regard a depreciation as a sign of weakness. Does it make sense to resist a depreciation when inflation is high?
2 ‘Money neutrality simply says that there is no way to cheat with money.’ Comment.
3 A strong currency is a currency that tends to appreciate. In what sense is this a desirable objective?
4 The Law of One Price does not hold. Can you imagine reasons why?
5 Box 5.2 tells the story of Zimbabwe. It also notes that hyperinflations often occur after a war. In your view, why do governments allow hyperinflation to occur?
1 David Hume (1711–76) was arguably Scotland’s greatest moral philosopher and economist (although Adam Smith is pretty strong competition). Hume’s most noteworthy contribution to economics was his essay Of Money (1752), from which this quotation is taken.
2 An IOU ( ‘I owe you’) is simply a promise to pay—a bill or simply a commitment to deliver something in the future, which is accepted by others as something of value.
3 One could assume that money is held in proportion to total transactions or nominal spending, but such spending includes purchases by firms of intermediate inputs, which are not counted in GDP. For this we would need much more detailed information on what occurs in the economy on a periodic
basis than the information in the national income accounts.
4 This famous relationship is called the Cambridge equation because it was first elaborated formally by Cambridge (UK) economist Arthur Cecil Pigou (1877–1959). A variant is the quantity theory of money, which defines the velocity of money as V = 1/k to obtain MV = PY. It is called velocity
because it measures how often, on average, one unit of money—say one pound—is used during the year to support spending (indeed V = PY/M relates the total value of all final transactions to the money stock). The quantity theory of money was formulated by the American economist Irving Fisher
(1867–1947).
5 Strictly speaking, these are approximations which are acceptable if the rates of changes are small. Using calculus, this can be shown to be exactly true for infinitesimally small changes, that is:
6 Price indices are presented in Chapter 2.
7 Nominal and real effective exchange rates are computed and published by various sources. Among them, International Financial Statistics, a monthly publication of the International Monetary Fund, presents a variety of real exchange rates (using GDP deflators, export prices, CPIs, labour costs)
computed using a sample of advanced economies.
8 There are two ways commonly used to compute weighted averages. The most common approach is arithmetic averaging, which sums up values of the variables of interest x1, x2…, etc., multiplied by their respective weights w1, w2, etc., which sum to one: w1x1 + w2x2 +…. Geometric averaging
employs products instead, with same weights as exponents:

PART III

7 Asset Markets
8 Private Sector Demand: Consumption and Investment
In Part III, our attention turns to the short term, say, the next two or three years. We study the behaviour of households,
firms, governments, and financial institutions. These chapters will form our building blocks for Part IV, where we will bring
them together in an overarching framework, which will also link the short run and the long run.
We start by laying out the budget constraints of those households, firms, and governments. Next, we present the asset
markets, where these agents borrow and lend and where interest rates and exchange rates are determined. We are then
ready to examine what factors shape their spending decisions. The pivotal role of interest rates and the financial system
leads us to look at money: what is money exactly, how is it created, and what role does it play in a modern, open
economy? Finally, we will see how central banks formulate monetary policy and try to influence conditions under which
money is borrowed and lent.
Borrowing, Lending, and
Budget Constraints 6
6.1 Overview
6.2.1 The Future Has a Price
6.2.2 The Rational Expectations Hypothesis
6.2.3 The Parable of Robinson Crusoe
6.3.1 Consumption and Intertemporal Trade
6.3.2 The Real Interest Rate
6.3.3 Wealth and Present Discounted Values
6.4.1 Firms and the Investment Decision
6.4.2 The Production Function
6.4.3 The Cost of Investment
6.4.4 The Intertemporal Budget Constraint of the Consolidated Private Sector
6.5.1 The Public Budget Constraint
6.5.2 The Consolidated Public and Private Budget Constraint
6.5.3 The Ricardian Equivalence Proposition
6.5.4 When Ricardian Equivalence Fails
6.6.1 The Primary Current Account
6.6.2 Enforcement of International Credit Contracts and Sovereign Borrowing
Summary
Many people despise wealth, but few know how to give it away.
François de La Rochefoucauld
What’s the quickest way to become a millionaire? Borrow fivers off everyone you meet.
Richard Branson
6.1 Overview
Borrowing and lending is a fundamental act of economic life. Each of the main economic players identified in the circular flow diagram of Chapter 2 —households and firms of the private sector, the government, and the rest of the world—have their own reasons to
borrow and lend. In doing so, they shift income and spending between the present and the future. Borrowing brings future income forward to be spent today. Lending or, more generally, saving, defers the use of current income to some later date.
Expectations of the future motivate borrowing and lending decisions. Those who reasonably expect their incomes to grow will want to borrow and raise their standards of living now instead of waiting. In contrast, the lucky winner of a lottery is likely to save a
large fraction of the prize, because it is unlikely to occur again. Firms’ investment decisions, which means adding capacity for future production, are a gamble on future demand. Not the present, not the past, but expectations of the future exert the greatest
influence on firms’ investment decisions—to acquire capital to enable production in years to come.
The shifting of spending over time can be seen as an intertemporal trade. One could think of a lender as a seller of money today, and the borrower as a buyer. Like any trading activity, there must be both a price linking the present and the future and each actor
must face a budget constraint. Indeed, because people are impatient, time has a price. This price is determined by the interest rate.
Similarly, borrowing and lending must be limited by total available resources. In this case, however, the resources are not just current income and wealth, they also involve future income. By the same token, income from lending and commitments to pay back
borrowing will gradually emerge in the future. Present and future wealth can be thought of as assets of a household, while present and future commitments are like liabilities. An intertemporal budget constraint requires that liabilities be repaid someday, while
accumulated assets will eventually be spent.
The intertemporal budget constraint provides a powerful framework for understanding many fundamental aspects of macroeconomics. Because the future is unbounded, it can be rather overwhelming to think about it. For that reason, this chapter adopts two
simplifications. First, we reduce the course of time to just two periods, today and tomorrow, the present and the indefinite future. Second, we will continue to employ Robinson Crusoe, already introduced in Chapter 4 to think about the choice between work and
leisure, and as a parable for consumer, producer, and include the government, all at once. These steps will make abstract and complex considerations a bit easier to handle.
6.2.1 The Future Has a Price

Anything of value must have a price. This includes money and goods delivered at future dates. In fact, markets exist for the sole purpose of pricing future deliveries of commodities. These are the stock markets like Euronext or New York Mercantile Exchange
(NYMEX), exchange markets, or specialized commodity markets. Such markets exist in many countries, both developed and emerging. An important common feature of asset markets is to place a value on payoffs in the future. This includes company shares, loan
repayments, steel deliveries, or foreign currencies. In a sense, asset markets price the future.
Microeconomic principles can be readily used to understand how the future is priced. There is a parallel between intertemporal consumption choices (between present and future goods) and intratemporal consumption choices (among goods at a particular point
in time). When we choose between consuming now or in the future, we effectively decide whether to save or to borrow. As rational households plan spending over time, they take into account their future incomes and needs, and balance these against the interest
rate at which they can borrow or save. Similarly, firms need to forecast the profitability of plant and equipment in which they invest. They need a benchmark for that profitability—what could they or their owners have otherwise done with their money? They
compare the profit, or return, from those projects with that available from lending their money at some available interest rate, which represents either the cost of funds or, if funds are available, the best alternative use for them.
6.2.2 The Rational Expectations Hypothesis

Expectations about the future are crucial to all this. But how exactly do firms and households form their expectations? Do they get it right or wrong? In this book, we will generally take the modern view that economic agents’ forecasts are right on average. This is
the rational expectations hypothesis.
The rational expectations hypothesis does not mean that households and firms never make mistakes, or that they always forecast the future perfectly. It is simply a way to apply the rationality principle to the way economic agents think about the future—we
assume that they do not make systematic errors. Clearly, alternative assumptions about expectations are possible. In fact, laboratory experiments on human subjects show that we all get sidetracked, sometimes, often responding to emotions rather than to cold
calculation. These experiments also show that we have limited ability to process mentally all the information that we have, or should have. So why do we adopt the rational expectations hypothesis? For three reasons:
First, because there are so many ways of being irrational that there is no clear or obvious alternative.1
Second, standard economic theory is based on the hypothesis that agents ‘optimize’ in the sense that they behave rationally—that they take the best possible decisions in a logically consistent way. If agents are rational in planning their consumption, work,
and production, then they should be rational when thinking about the future.
Third, even if most people are not fully rational all the time, it is unlikely that they are repeatedly and systematically wrong. If they are, they would most certainly suffer continuous losses. Isn’t it natural to expect them to take steps to avoid such errors in
the future?
There is an even better and subtler justification for rational expectations. We are ultimately interested in how prices, interest rates, incomes, and spending interact on the marketplace. It is often enough that a few well-informed agents behave rationally to drive the
markets. If trade unions act on behalf of their members, it suffices that their expectations will be correct on average. In financial markets, all that is required is that a number of professional traders are well informed and have sufficient resources at their disposal. If
they perceive that prices are too low compared with their valuation, they will buy, forcing prices upwards. If prices are too high, they will sell. Less well-informed customers end up accepting the market prices because it is foolish to ignore them and possibly too
costly to try to do better.2
As a short cut, this book will frequently use a simplified version of rational expectations known as perfect foresight. Perfect foresight simply assumes that people anticipate the future correctly. The difference with rational expectations is that we ignore
uncertainty. Perfect foresight can be thought of as an exploration of what the world would be like if the future was in fact perfectly known. Of course, no one thinks that this is realistic but, as with rational expectations, if surprises are sometimes good and
sometimes bad, perfect foresight is a reasonable starting point.
6.2.3 The Parable of Robinson Crusoe

This chapter sets up the intertemporal budget constraint facing households, firms, the government, and the nation as a whole. We rturn to Robinson Crusoe as our representative household, already familiar to us from Chapter 4. As we deal with the infinite future,
it is convenient to collapse time into two periods, ‘today’ and ‘tomorrow’, where tomorrow is a metaphor for the future. And beyond tomorrow? Crusoe is rescued and will no longer need to concern himself with the economics of his island!
6.3.1 Consumption and Intertemporal Trade

Let’s start by imagining that Crusoe’s island does not even have coconut trees. Rather, the coconuts simply wash up on the beach. The number of coconuts that he (rationally) expects to have today and tomorrow is his exogenous endowment. This endowment
includes both present and future resources. Using subscripts to denote the relevant period in which they become available, we can represent his endowment of Y1 coconuts today and Y2 coconuts tomorrow by point A in Figure 6.1.
Until Robinson learns how to plant coconuts, he has no choice but to consume what nature gives him. Since coconuts cannot be carried over to the next period, Crusoe’s consumption is also given by point A. This is the autarky point. A household or a
country operates in autarky when it does not trade and consumes its endowment.
If we bend the story a little and allow for neighbouring islands inhabited by other economic agents, trade is possible. Because Crusoe’s coconuts are just as good as his neighbours’, we might not expect to observe any trade between them. Yet Crusoe may well
be interested in intertemporal trade, or exchanging resources across time. He might lend his neighbours some coconuts today, if he expects to find only a few tomorrow, or if he prefers to consume more tomorrow. On the contrary, if today’s ‘harvest’ is
abnormally low or if Crusoe is impatient, he could borrow coconuts now and repay later when times are better.
Fig. 6.1 Endowment, Wealth, and Consumption

Resources available today and tomorrow—the endowment—determine wealth and available consumption choices along the budget line BD. In the figure, the same level of wealth (OB) is attainable by a professional athlete (point P) or a university student (point M).
The rest of the chapter will explore borrowing and lending, and the role of financial markets will follow. As a way of moving away from autarky, borrowing and lending is just like selling or buying goods and services. In a way, it is like buying or selling the
command over goods in the future. Yet banking and financial markets suffer from reputation problems, and the recent financial and credit crisis hasn’t improved matters. Box 6.1 explores some of these reservations and possible reasons for them.
Box 6.1 Neither a Borrower nor a Lender Be: Economics and the Sociology of Credit
In a well-known scene from Shakespeare’s Hamlet, Polonius gives his son Laertes some parting advice: ‘Neither a borrower, nor a lender be: For loan oft loses both itself and friend, and borrowing dulls the edge of husbandry.’ In other famous plays,
such as The Merchant of Venice, Shakespeare gives lenders a pretty tough time. Is it morally wrong to borrow (or to lend)? And what is wrong with charging interest, if borrower and lender freely agree to it? Although the economic arguments against
autarky are convincing, many important religions of the world—including Islam, Christianity, and Judaism—have banned lending at a positive interest rate at one time or another in their histories. Why the ambivalence?
Perhaps it is because lenders have an unconditional claim on the resources of individual borrowers in an inherently risky world. If the fortunes of a borrower go south, those of the lender do not—that is, unless the borrower declares bankruptcy.
Perhaps it is because borrowers appear to be in a poor bargaining position, often seeking credit when all else has failed. Perhaps it is because individuals are frustrated when their bank won’t give them the loan they think they deserve, because the
credit officer deems them ‘too risky’, which they feel is insulting. And the market has ways of dealing with individual risks which are distasteful to many. One is charging higher interest rates, which appears opportunistic since ‘risky’ poor people pay more
than the ‘safer’ rich. Stories abound in the USA and the UK of indebted families with credit card debt paying interest rates of 30% per annum and more. Then there is loan-sharking, illegal and possibly violent enforcement mechanisms, and ‘payday
loans’, which amount to selling one’s own wage packet in advance at effective annual interest rates sometimes in excess of 600%.
And yet, borrowing and lending is as natural as breathing. Even when lending at interest is prohibited, as in many Islamic countries today, the market always manages to find ways around the ban, for example by declaring loans to be ‘equity stakes’
which participate in profits and losses of the enterprise. In the end, the fundamental truth is that the market for loans exists because there are gains from trade: different degrees of patience, different wants, different opportunities, and different information.
6.3.2 The Real Interest Rate

Crusoe and his neighbours must agree on the terms of repayment: how much should he pay (or receive) tomorrow for one coconut borrowed (lent) today? This is what we call the real interest rate and denote by r. The real interest rate is taken by Crusoe as
given or exogenous; it is ‘real’ because Crusoe and his neighbours do not use money. Borrowing 100 coconuts will require paying back 100(1 + r) tomorrow—the principal of 100 plus interest payments of 100 r. If the real interest rate is 3%, or 0.03, interest
payments will be 3 coconuts. Another, equivalent way of thinking of this transaction is that Crusoe agrees to sell 100(1 + r) coconuts tomorrow for the price of 100 today. Similarly, if Crusoe takes the other role and lends 100 coconuts today, he will receive 100(1 +
r) coconuts tomorrow; to buy 100(1 + r) coconuts tomorrow he must save 100 coconuts today. The relevant intertemporal trade involves swapping 1/(1 + r) coconuts today for each coconut tomorrow. We can say that a coconut tomorrow is worth 1/(1 + r)
coconuts today. 3
The price of tomorrow’s consumption in terms of today’s, 1/(1 + r), is called an intertemporal price. As the real interest rate r is positive, 1/(1 + r) is less than 1, which means that goods tomorrow are cheaper—or less valuable—than goods today. The real
interest rate measures the cost of waiting. Valuing future goods in terms of goods today (here, dividing by the interest factor, 1 plus the real interest rate) is called discounting. Discounting is a very important concept in both economics and business. It helps
explain the inverse relationship between bond prices and interest rates, as well as the fact that surprising changes in market interest rates often move stock prices in the opposite direction. It explains why borrowers—including governments—love to postpone
repayment of loans, and why lenders resist such attempts with vigour. Box 6.2 presents the important concept of discounting more generally.
Box 6.2 Discounting and Bond Prices
Discounting is used in economics and finance to value future incomes or expenditures in terms of income today. It is frequently used to put a value on financial assets and liabilities (debts). It asks: what is the amount of money required
today, given an interest rate, to generate some payment or payments in the future? By valuing a coconut tomorrow only as worth 1/(1 + r) coconuts today, Robinson Crusoe has successfully applied discounting to a practical problem.
Let us consider a simple bond that pays € 100 in one year’s time. (This type of bond is called a pure discount bond.) If the interest rate given by the market is 5%, what is the value of this bond today? It is the amount that, if invested now, yields € 100
next year. If that amount is B, then it must be true that B (1 + 0.05) = 100, so B = 100/(1.05) = €97.24. Similarly, the value of a two-year discount bond must be discounted twice, once for each year. From next year’s viewpoint, the second-year € 100 is worth
100/1.05. From this year’s viewpoint, this is worth [(100/1.05)] / 1.05 or 100/(1.05)2 = € 90.70. The further into the future the payout is, the more heavily any amount is discounted, and the lower is the price of a discount bond. A n -year € 100 discount bond
is worth 100/(1 + r) n today.
More generally, given discount rate r, the present value of a stream of payments at over n years, t = 1, …, n has present value given by:
Now consider the case of a bond that promises to pay a fixed amount a forever, called a consol. Is it possible to put a price on that income stream, even though the payments are infinite? As long as the interest rate is strictly positive, the answer is yes!
The price of a consol C which pays the constant amount a each period is simply the present discounted value of its payments a1=a2=a3=…=a:
where we have applied the well-known formula for a sum of a geometric series to the term in brackets. The price of a consol is inversely related to the interest rate. Other bonds have a finite maturity so the formula is more complicated, but the general
principle survives that higher real interest rates imply lower bond prices.
Intertemporal trade allows Crusoe to choose combinations of consumption today and tomorrow which are different from his autarky point. These combinations are represented by the line BD in Figure 6.1. This line must go through his endowment point A, since
he can always choose not to trade at all. At point B Crusoe would forego consumption tomorrow completely. In that case he would borrow fully against his future endowment Y 2, receive Y 2 / (1 + r) coconuts, and consume Y1 + Y2 / (1 + r) coconuts today. At
point D he would fast today and lend all his current endowment Y1 in order to consume Y1(1 + r) + Y2 coconuts tomorrow. The line BD represents all the possibilities open to Crusoe, in between the extremes just described. It is called the budget line. The budget
line is the simplest representation of the intertemporal budget constraint. Its slope4 is given by - (1 + r). If the rate of interest increases, the budget line becomes steeper. For a given amount of saving today, more will be available tomorrow.
6.3.3 Wealth and Present Discounted Valuess

If Crusoe’s income ‘from nature’ in the first period is Y1 and he consumes C1 in the same period, his saving is Y1 - C1. If Y1 - C1 is positive, Crusoe is lending; if Y1 - C1 is negative, he is borrowing. In the second period, consumption C2 will equal the sum of
income Y2 and (1 + r)(Y1 - C1), i.e. the interest and principal on his savings from period 1. If saving was negative in the first period, this means paying back principal plus interest. The budget line can be represented formally as:
(6.1)
This fully describes Crusoe’s intertemporal budget constraint. Dividing both sides by (1 + r) and rearranging:
(6.2)
The left-hand side is the present discounted value of consumption. It is the sum of today’s and tomorrow’s consumption valued in terms of goods today. The right-hand side is equal to the present discounted value of income (his endowment). It is the
maximum consumption that Crusoe could enjoy today, given his resources today and tomorrow, and is represented by point B in Figure 6.1. Put differently, OB is the present discounted value of Crusoe’s total endowment. In fact, in this situation it represents his
total wealth, which we denote by the symbol Ω.
Lending and borrowing enable individuals with the same total wealth but with very different income profiles to enjoy the same menu of possible consumption over both periods. It doesn’t matter whether Crusoe is a university student with low current and high
future income, as represented by point M in Figure 6.1, or a professional athlete with high current and low future income (point P). As long as these points are on the same budget constraint, the present discounted value of income is the same and intertemporal
trade allows income to be shifted across time by borrowing and lending.
Fig. 6.2 Inheriting Wealth or Indebtedness

When wealth B1 > 0 is inherited, the budget line shifts BD to B'D'. Debt B1 < 0 shifts the budget line to B"D". The lines are parallel because the real interest rate is assumed unchanged.
Now suppose Crusoe had initial tradable wealth B1 (an initial cache of coconuts). His wealth will increase by this amount and the budget constraint will be modified as follows:5
(6.3)
If B1 > 0, Crusoe can consume more in both periods. But B1 could be negative, if Crusoe began his existence with debt. In that case, he will have to consume less (in present value terms) in order to repay the debt and interest. In general, total wealth Ω is the sum
of inherited wealth or debt B1 and the present value of income: . This is shown in Figure 6.2, where the inherited wealth or debt is added to the present value of income. At a given real interest rate, it implies shifting the budget line
BD to B′D′ (if wealth increases) or B″D″ (if it decreases).
6.4.1 Firms and the Investment Decision

Crusoe’s income has been exogenous until now. In reality, income mostly comes from planned activities. As we saw in Chapter 3, production requires that resources are diverted from consumption and used to acquire productive capital. Crusoe could plant
coconuts today which would grow into trees bearing coconuts tomorrow. Naturally, once planted, a coconut cannot be consumed: it is useful only for its future production. The use of valuable resources to produce more goods later is called investment or fixed
capital formation. Many goods produced in modern economies are designed solely to make future production possible, and have no consumption value at all.
The notion of investment was already explored in Chapter 3 in the discussion of long-term economic growth. The investment decision also has a fundamentally intertemporal aspect. Firms decide to accumulate capital when it is sensible, i.e. profitable to do so,
and profitability depends on expected future outcomes. In order to finance their investments, firms can either obtain resources in the capital market (stock exchanges, bond markets, or banks) or use their own funds (retained earnings). In Chapter 3, capital was
accumulated as the result of available savings, which was assumed to be an exogenous fraction of national income. In this chapter and those which follow, we will begin to describe investment and savings as conscious choices of firms and households.
6.4.2 The Production Function

The investment decision depends upon the amount of output that can be produced with the available equipment (the number of coconuts to be obtained from a tree). The production function F(K) captures this relationship between capital input and output and
is depicted in Figure 6.3. It can be thought of as a special case of the production function of Chapter 3, in which labour input is fixed (Robinson’s available time). 6 The shape of the curve implies that, as more capital is accumulated, the resulting additional or
marginal output declines. That marginal output decreases when input increases is the same principle of diminishing marginal productivity that we encountered in Chapters 3 and 4.7
Fig. 6.3 The Production Function
As more input is added, output increases, but at a decreasing rate. This is the principle of declining marginal productivity.
6.4.3 The Cost of Investment

Starting with no capital stock (assuming no coconut trees on the island at the outset), today’s investment represents the total stock of capital available for production tomorrow. ( Box 6.3 considers the more realistic case when previously accumulated capital
already exists.) Crusoe understands that he can either invest K in productive equipment, or save K. In the first case, he will receive output F(K) tomorrow. In the second case, he will receive (1 + r)K tomorrow. The real interest rate measures the opportunity cost
of the resources used in investment. Because of the option of lending at rate r, the investment in this case must yield at least 1 + r to be worth undertaking.8
Box 6.3 Gross Investment, Depreciation, and the Capital Stock
When previously accumulated capital already exists in the amount K1, the situation in the next period is more complicated. The stock of capital may differ in the future from the previously accumulated stock in two ways. First, new capital I1 may be
invested. Second, depreciation—wear, tear, and obsolescence—may remove some of the value of the old capital stock. It is a proportion δ of the capital stock. The new capital stock is:
The realized change in the productive capital stock, ΔK = K2 - K1, is therefore equal to I1 - δK1, the difference between gross investment and depreciation of previously accumulated capital. For the capital stock to grow, new investment spending must
exceed depreciation.
Fig. 6.4 Productive Technology

The cost of borrowing to finance investment is given by OR. As long as output exceeds the cost of borrowing, the technology is productive and the producer makes a profit. Beyond A, she makes losses.
Figure 6.4 shows the opportunity cost of invested capital K as the ray OR from the origin, which is given by (1 + r)K. As long as the resulting output exceeds the cost, the technology is sufficiently productive and investment is worthwhile. At point A,
investment just covers its cost. There is no economic profit possible. To the right of A, investment uses up more resources than it produces. Positive economic profits occur only to the left of A.
The interest rate r is crucial for the valuation of investments. For example, if the rate of interest were to increase, the OR line would rotate counterclockwise, moving point A to the left and reducing the volume of investment that has any positive value at all. Put
differently, a given stock of capital must be more productive to make up for higher borrowing costs. The principle of declining marginal productivity implies that only less capital in use will generate high enough marginal, and therefore average, productivity.
Another approach, that will prove useful later, is to ask what is the net return V from investing K — what Crusoe gets from planting K coconuts as an irrecoverable expense. The value of the investment project, or the value of Crusoe’s enterprise, is simply the
difference between the present value of output tomorrow and investment today:9
(6.4)
.
Note again that to compare goods available tomorrow with goods available today, we must price the former in terms of the latter, i.e. we discount the former by applying the intertemporal price 1/(1 +r).
Fig. 6.5 Unproductive Technology
Given the interest rate, no firm will operate with the production function shown in the figure. A technological innovation which shifts the production function upwards can make an unproductive technology productive again.
An investment project is economically justifiable only if it has a positive present value. In terms of (6.4), that means V > 0 or if F(K) > K(1 + r). Figure 6.5 illustrates a case when the technology is not productive enough given the real interest rate. In that case it
does not pay to invest anything at all: it is more profitable simply to lend at the rate r. It would require either an improvement in technology (the production function schedule shifts upwards as in Figure 6.5) or a decline in the interest rate (the ray OR rotates
downwards) for some investment to be worthwhile.
6.4.4 The Intertemporal Budget Constraint of the Consolidated Private Sector

The budget constraint of Section 6.3 took endowments as given by nature as on Crusoe’s island. Once investment and production are taken into account, income tomorrow is no longer simply given by nature. The budget constraint now depends on the amount
that is invested and on its profitability. As long as the investment project has positive present value, investment increases wealth. Figure 6.6 shows how this happens. Starting from point A, Crusoe can save either by lending, or by investing an amount I1 up to a
maximum of his endowment Y1. In the latter case, Crusoe’s savings are equal to investment, which is equal to the capital stock for tomorrow’s output production (remember: the island is barren upon his arrival, so initial capital K1= 0). This is the difference between
today’s endowment Y1 and consumption C1:
Fig. 6.6 Investment Increases Wealth

Investing I1 in a productive technology (which becomes K2 in period 2) allows a household to increase its wealth over and above that corresponding to the initial endowment A. Here wealth increases by BB′ as FE additional goods become available in the second period.
(6.5)
The more he invests—the more we move to the left in Figure 6.6—the larger will be tomorrow’s production. This is why the production function AE is now the mirror image of the one shown in Figure 6.4: as consumption declines and saving rises, we move
leftwards from the endowment point A, investment increases and tomorrow’s output becomes larger. Tomorrow’s income—and consumption as it is the last period—is the sum of the endowment Y2 (the coconuts lying on the beach) and produced output F(K2):
(6.6)
The intertemporal budget constraint determines the present value of consumption C1 + C2/(1 + r) as equal to total wealth Ω. Recognizing that C1 = Y1 - I1 is given by (6.5) and C2 given by (6.6), the intertemporal budget constraint can be rewritten as:
(6.7)
Wealth now consists of two parts. 10 The first part is the present discounted value of the endowment as before in (6.2). The second part is the increase in wealth represented by V, the net value of the investment activity, as in (6.4). In Figure 6.6 the outcome of
investment I1 is shown as point E. Note that E lies above the initial budget line. This is because the production technology is productive at the rate of interest r. The distance OB still represents the present value of the endowment. But now, for a choice of
investment I1 that brings Crusoe to point E, new total wealth is the distance OB′. Since the value of future output is discounted at the same rate r, the new budget line is parallel to BD. The distance BB′ represents the net present value of the investment project.11
In the parable, Crusoe stands for the private sector as a whole, which consists of individuals and the firms they own. Firms ultimately belong to their shareholders, and the net return from investment raises their wealth. If shareholders anticipate that a firm will
become more profitable in the future—because of a technological advance, as represented by the shift in Figure 6.5—then net expected returns rise and they are richer. This wealth gain takes the form of an increase in the value of the firm. In the real world, this
would be reflected as an increase in the firm’s value in the stock market.12

6.5.1 The Public Budget Constraint
There was no government on Robinson Crusoe’s island. In the real world, there is a public sector, which collects taxes, purchases goods and services, and makes transfers to households. Yet the government is little different from other economic agents. It can
borrow, but is expected to repay its debt with interest; if it lends it will expect to be repaid by its debtors. Consider a government, which spends G1 today and G2 tomorrow, and raises net taxes T1 and T2.13 The government also has inherited debt D1 from the past.
D1 already incorporates interest from the past so the government can either pay it off in its entirety, or ‘roll it over’ into the next period. In latter case, debt incurred must be serviced (interest must be paid) at interest rate rG, and it is carried forward into the next
period. This means that the debt resulting in period 2 from period 1’s government activities is:
(6.8)
If the government is solvent, all debt at the end of period 2 must be paid off in its entirety, i.e. equal to zero, so the government in the second period must obey (1 + rG)(D2 + G2 - T2) = 0. Combining this with equation (6.8), we find that:
(6.9)
A solvent government is one in which initial debt (D1) is financed by present value of current and future primary budget surpluses, denoted by T1 - G1 and T2 - G2. The primary surplus is defined as the government budget balance (the difference between
receipts and expenditures) from which interest receipts or payments have been removed. A government which is borrowing has a deficit, or a negative surplus. Thus, the total borrowing requirement of a government (‘headline deficit’) is the sum of (1) the
primary deficit, the amount by which non-interest-related expenditures exceed revenues, and (2) interest payments and debt repayment. The distinction between primary and ‘headline’ deficits is used frequently when reporting the budget of highly indebted
governments. For example, the IMF reports that in 2015, Italy had an overall budget surplus of –€43 bn euros (a deficit) or about 2.6% of GDP, but at the same time showed a primary surplus of €22.7 bn, or about 1.4% of GDP.14
The lesson of the two-period government of Robinson Crusoe is a central one: in the second and last period, the government must repay its obligations in full. Equation (6.9) means that tomorrow’s primary surplus ( T2 - G2) must be sufficient to repay today’s
deficit (G1 - T1) plus initial debt burden D1, plus the interest service on the borrowing:
A government with obligations tomorrow in some combination of indebtedness today (D1 > 0) or deficits today (G2 > T2) will have to run surpluses tomorrow to repay them. This is the government’s budget constraint, which can be rearranged as:
(6.10)
The Government Budget Line
Fig. 6.7 The Government Budget Line

A deficit today must be matched by a budget surplus tomorrow, or vice versa, if the government is to obey its intertemporal budget constraint.
For the government to obey its intertemporal budget constraint, the sum of the present value of primary budget surpluses is equal to initial outstanding debt. This also means that the present value of government income must be sufficient to cover the present
value of purchases plus the initial debt. The government budget constraint is illustrated in Figure 6.7 for the case of no initial debt or assets (D1 = 0). The budget line has slope - (1 + rG) and passes through the origin.
The two-period parable contains a strong message: governments with debt and deficits today must run primary surpluses tomorrow; similarly, governments with a comfortable fiscal position today can relax (a little) in the future. Yet do governments really obey
their budget constraints? The European debt crisis—Box 6.4 discusses its origins—may give some reason to doubt it. In fact, governments throughout history have been faced with budget problems, sometimes resulting in spectacular defaults, or repudiation of
past debts. Yet an error commonly made by politicians and the general public is to look only at the current year’s deficit as a measure of the government’s solvency. Especially when the economy is growing robustly, tax revenues can be expected to grow over time
as well, thus easing some of the burden.15 For that reason, it is always a good idea to measure expenditures and tax revenues relative to GDP. Still, in order to avoid defaults, today’s primary deficits require primary surpluses later, and conversely. Given spending
plans, lower taxes today are followed by higher taxes tomorrow. Alternatively, for a given path of taxes, more spending today requires spending cuts tomorrow. How long does ‘today’ last before a government is hit ‘tomorrow’ by the budget constraint?
Figure 6.8 shows that governments do generally obey their budget constraints. It shows the evolution of primary budget balances for four countries over time, relative to the size of the economy measured by the GDP. Some countries (the UK) show a
succession of primary deficits and surpluses. In other cases (Ireland, Italy, the USA) deficits have been sustained over many years, yet eventually the primary budgets were corrected, sometimes moving into spectacular surpluses. The financial crisis of 2007–2008
has led to sudden relapses in Ireland, the UK, and the US. The case of Ireland is unusual (note that the scale is not the same for all four countries). In 2010, the government had to scramble to save its collapsing banks. With a deficit of some 30% of GDP, the Irish
government itself went into crisis and had to be bailed out by the other Eurozone countries with assistance from the International Monetary Fund. Italy, on the other hand, managed to keep its budget roughly balanced. Yet it went into near-crisis when markets
took a dim view of several decades of large deficit and smaller surpluses that left a legacy of a public debt of some 110% of GDP. These examples show how sudden unexpected spending needs or long strings of insufficient surpluses can stretch the intertemporal
budget constraint to its limits.
Box 6.4 The European Debt Crisis
Many people were surprised by the sudden panic in the market for Greek government debt in the spring of 2010, especially as it seemed to spread to other Eurozone countries. Several years later, a number of countries are still running deficits. Even
though these deficits are generally smaller, there are lingering doubts that some countries will be economically or politically able to meet their budget constraints. It is often considered that a sovereign default is inconceivable, yet history is littered with
government defaults, frequently accompanied by sharp political upheavals; examples include the turbulent years of the French Revolution, the October 1917 revolution in Russia, the end of the Weimar Republic in 1933, and Castro’s revolution in Cuba.16
In most cases, however, defaults are just the end of a long period of debt accumulation, reflecting a long string of budget deficits which don’t seem to demonstrate any budget constraint at all. This was true of much of Latin America in the 1980s, Russia
in 1998, Argentina in 2001, and Iceland in 2008.
The history of government finance shows that default is a subjective concept.17 Technically, it occurs when the borrower country has failed to service its debt, meaning that it has missed a payment due—even by a day! Yet financial markets, which
finance most of the deficits, can be enormously patient or excessively optimistic. It is always possible to arrange a short-term ‘bridge’ loan, even on short notice, for a few years’ time, if there is a high chance that the loan will be repaid. This is precisely
what happened with Greece in July 2015.
In the end, it is not really possible to know with certainty whether a government is meeting its budget constraint at any point in time. In reality, the intertemporal borrowing constraint features public spending and tax revenue streams that extend into the
infinite future. If there are lenders out there with strong nerves and enough patience, current deficits associated with a temporary decline in tax revenues in a recession can be overlooked. This is why the current situation—which will be examined in more
detail throughout the rest of the book—is so disturbing.
Fig. 6.8 Primary Consolidated Government Budget Surpluses, Four Countries, 1970–2015
Over time, primary budget balances must add up, in present-value terms, to initial public debt. Some governments, like the UK, have maintained primary budget balances on average over many years. Those that have allowed deficits to cumulate into large indebtedness will eventually have to
run surpluses, as has been the case in Ireland, Italy, and the USA.
Source: Economic Outlook, OECD.
6.5.2 The Consolidated Public and Private Budget Constraint

Both households and firms—which are owned by households—ultimately have to pay the taxes. They cannot ignore the public sector budget constraint. Much as they must include the budget constraints of the firms that they own, households must also see
through the public sector financing veil. In this section we follow this logic and integrate the private and public budget constraints to face intriguing and important consequences.
For simplicity, we ignore the existence of firms and set initial government debt to zero. The private and public intertemporal constraints are side-by-side as follows:
(6.11)
(6.12)
In the first budget constraint, the private citizens pay the taxes, which reduces disposable income as defined in Chapter 2, while in the second, the government receives them. Note that we do not assume that the government and the private sector face the same
interest rates when they borrow and lend. Traditionally, the private sector is considered as less safe than the public sector (but a crisis may be changing this presumption). The government sector borrows and lends at rate rG, while the private sector borrows and
lends at rate r > rG. Combining the private and public budget constraints yields the consolidated budget constraint.18
(6.13)
A comparison of the consolidated budget constraint (6.13) with the private constraint (6.11) shows that both link private consumption to private wealth—remember that to simplify, we ignored inherited debts or assets. Before consolidation, private wealth is the
present value of disposable incomes (net of taxes) over both periods. Equation (6.12) shows that after consolidation, private wealth includes two parts. First, the present value of incomes net of public spending, not taxes. This means that households can only
consume the output that the government has not taken for itself. As long as the government respects its budget constraint, its spending will be paid for by taxes, now or in the future, and it does not seem to matter when!
Yet it does matter. The second part of private wealth reflects the difference between the interest rates at which the government and the private sector can borrow. If, as is normally the case, the government can borrow more cheaply than the private sector, r > rG,
this part is positive: the more the government borrows, the better off the private sector is. In order to understand this surprising result, consider the case when the government reduces taxes today and raises them tomorrow to meet its budget constraint. This
means that it will have to borrow today at rate rG and pay back tomorrow. The private sector will pay less taxes today but more tomorrow; it makes sense for the private sector to save the corresponding amount, at interest rate r. The private sector benefits from this
operation because it earns r on its saving and will have to pay more taxes to cover the public borrowing at the lower interest rate rG. In effect, the government borrows on behalf of the private sector, allowing the private sector to save at a higher rate than it
borrows indirectly. Of course, it is also true that if the government borrows on worse terms than its citizens, that is when r rG, it can reduce net wealth of its citizens.
Fig. 6.9 Ricardian Equivalence
The government’s spending and taxing activities reduce private wealth. Given government purchases, the precise scheduling of taxes does not matter.
6.5.3 The Ricardian Equivalence Proposition

The story gets even more interesting. Suppose for the moment, that the interest rates of the private sector and the government are exactly equal, so r = rG. In that case, the consolidated budget constraint (6.13) collapses to:
(6.14)
This looks very much like the private sector budget constraint (6.11), except that now taxes do not appear, so that it does not matter at all when taxes are levied as long as the government abides by its budget constraint. Once the government has ‘helped itself’ to
G1 and G2 of output, the private sector will take the rest whenever it wishes in the form of C1 and C2, borrowing or lending as needed. In fact, the private sector has fully internalized the public sector’s budget constraint. The hypothesis that the private sector fully
internalizes the public sector’s budget constraint is known as the Ricardian equivalence proposition.19 In Figure 6.9, point A represents Crusoe’s endowment measured before taxes. Once public spending is taken into account as in (6.13), the private
endowment is represented by point A'. The government reduces Crusoe’s private wealth by an amount represented by the distance BB', which is either the present value of taxes or the present value of public spending—the two are equal because of the
government budget constraint. As long as the public and private sectors borrow and lend at the same rate (r = rG), these intertemporal shifts are equivalent and the public borrowing can be matched one for one by private saving along the same private budget line.
The Ricardian equivalence proposition can be stated in a number of different ways. The first is that total national spending—the sum of private and public spending on goods and services—cannot exceed the country’s wealth. The country can borrow or lend
abroad, but it must respect its (national) budget constraint.20 The second is that private sector wealth—which can be spent on private consumption—is the difference between the present value of production or income on the one hand and public purchases of
goods and services on the other. The implication is that the pattern of taxation over time has no effect on private wealth. What matters in the end is public spending, which represents resources taken away from the private sector. Finally, Ricardian equivalence
means that its citizens do not treat government debt as net wealth. Government’s indebtedness does not appear as part of private wealth on the right-hand side of (6.14). The private sector sees through the veil of government. It recognizes that the government’s
promises to pay the principal and interest on public debt are matched by taxes levied to service the debt, today or tomorrow. Public bonds are an asset to households, which is exactly offset by the present value of their future tax liabilities.
6.5.4 When Ricardian Equivalence Fails

For obvious reasons, the Ricardian equivalence result is highly controversial. It means that the path of taxes is irrelevant for the behaviour of the private sector. It implies that public borrowing and the resulting stock of government debt do not, on net, contribute
to the wealth position of households. Holding constant the path of government purchases of goods and services, budget deficits do not matter! This controversial Ricardian equivalence result requires a number of assumptions, and this section reviews them
critically. In the end, the result of this discussion is that budget deficits probably do matter, and that at least some fraction of public debt is regarded by the private sector as wealth.21
Different interest rates
A central assumption behind the Ricardian equivalence result is that the government and the private sector face the same interest rate. Is that realistic? It has long been taken for granted that, in any country, governments borrow at the lowest interest rate. The
reasoning is that the government is considered a less risky borrower than most private businesses or individuals because governments can always tax to pay back their debts while private agents may find themselves unable to reimburse some loans. In the
developed countries, the difference between the rates at which a government borrows and those that apply to firms in good standing has traditionally been some 1–2% more for businesses, and much more for households. Table 6.1 shows that this still holds for
the UK, US, and the Eurozone as a whole, but it is definitely not the case for individual countries.
Table 6.1 Interest Rates for Government and Corporate Bonds, 29 February 2016 (% per annum)
10-Year Government Bond 10-Year

Corporate
Bonds
A-rated BBB-rated
United Kingdom 1.32 3.46 4.25
United States 1.74 3.57 4.74
Euro Area 1.05 1.67 2.66
Source:: Macrobond.
This was a nearly universal observation before the crisis hit the Eurozone. Figure 6.10 shows that when the crisis started to unfold in 2009, the Italian government borrowing privilege gradually began to erode. By late 2010, Italian firms with solid credit ratings
were paying interest rates on debt that were considerably lower than those paid by the government. While government and private interest rates are taken as given for the purposes of the analysis, they are in fact endogenous and depend on a number of factors.
Box 6.4 already hinted that the reason for increases in government interest rates could be a perceived increased risk of default. Chapters 7 and 17 will make this case more clearly. In any case, when rG > r the government should be wary of running deficits, since
according to the logic of the budget constraint, they are hurting, not helping their citizens; in this case deficits reduce, rather than increase, their net wealth!
Mortal or new citizens
Another objection is that Ricardian equivalence must fail because citizens are not all alike when they face the taxman: some pay a lot more taxes than others. So the burden of public debt service is not equally borne by all citizens. Similarly, some hold government
debt, and some don’t. Yet, this does not imply that the aggregate household sector can escape the implications of equations (6.12) and (6.14). In the aggregate some pay more than average, others pay less and, as a first approximation, it does not matter.22
Fig. 6.10 Public and Private Borrowing Rates in Italy, 2003:1–2016:1
The interest rates refer to 5-year maturity loans to non-financial corporations and to the government in Italy. Normally, the private sector borrows at a higher rate than the sovereign. The reversal, which lasted from September 2010 to November 2013, was a potent signal that Italy was in
the midst of a public debt crisis.
Source: ECB.
On the other hand, citizens are certainly mortal. If they are not alive in period 2, they have little reason to care about the implication of the intertemporal budget constraint of the government. Of course, no one knows whether they will be alive next period but
there is always a possibility, unfortunately, that the answer will be negative. Collectively, the private sector currently alive may factor in only a fraction of all future tax liabilities. In that case, government debt represents private wealth as we realize that we will not
be alive next period and therefore we will not pay taxes to cover the public debt. To those who will not be alive, holding the current debt represents wealth. In a similar vein, new agents—immigrants, perhaps—who enter at some future date will pay taxes to
reimburse the public debt issued before they arrived. This too breaks the link between the budget constraint of the presently living generations and future government revenues.
Restrictions on borrowing
Many households cannot borrow as much as future expected income would justify; sometimes they cannot borrow at all. They may be unable to convince lenders—typically banks—of their creditworthiness. For their part, lenders only possess incomplete
information on the creditworthiness of borrowers when they apply for credit. In addition, future incomes are never really certain, so lending to households is risky. Borrowing rates exceed lending rates to compensate for this risk. 23 In the worst case, no lending is
extended and individuals are said to be credit rationed. The case of credit rationing is represented in Figure 6.11. With a net private endowment represented by point A, the agent can only move along her budget line on the segment AD. The segment AB is not
attainable through private borrowing. This means that the consolidated budget constraint—and Ricardian equivalence—is irrelevant for that agent. For instance, consider the case where the government runs a deficit today, so that T1 G1, borrowing at rate rG
which we assume equal to r for simplicity. The agent may now reach point A' and could consume Y1 - T1, which is larger than Y1 - G1. She benefits when the government does the borrowing that she cannot do.
Most often, individuals face higher and rising costs of borrowing. Lending institutions frequently demand higher interest rates from individuals to compensate for additional risk. The situation is similar to the case studied in the previous section and is also
illustrated in Figure 6.11. When lending, the constrained agent can move along AD, but for borrowing she moves along AB'. The budget line is now kinked at the endowment point. In this case, public debt contributes to citizens’ wealth, and the time profile of
taxes affects the private sector budget constraint. At point A' the constrained citizen is better off than at any point along AB'. Once again, when the government borrows on behalf of its citizens, it increases the wealth of those who cannot borrow on those terms.
Fig. 6.11 Borrowing Constraints

When the household cannot borrow at all, its budget line is restricted to the segment AD, because it cannot consume today more than what is left of the endowment after public spending (Y1 - G1). If the government reduces taxes and borrows instead (here, from abroad), the household’s
borrowing line extends to the segment A'D. When borrowing constraints take the form of a higher private bor-rowing rate r', the budget line is the kinked line B'AD. A budget deficit at A' at a lower interest rate r = rG relaxes the private household’s budget constraint.
Distortionary taxation and unemployed resources

There are many other reasons why the Ricardian equivalence can fail. An important one is that taxes are distortionary in the sense that they alter people’s behaviour. For example, taxation on labour income reduces take-home pay, which may lead some to work
less, as explained in Chapter 4. This will reduce output. This is not the case in the parable of Crusoe, because the supply of coconuts is exogenous but, in the real world, output is endogenous and affected by taxes. Another important argument against the
Ricardian equivalence is the presence of unemployment. Changes in taxes or public spending may affect the level of economic activity in ways that are explained in Chapter 8 and later chapters.
Evidence
Given the long list of qualifications, it would seem quite unlikely that Ricardian equivalence could ever hold in practice. Yet, it receives some empirical backing, especially when the public budget moves by large amounts that are clearly perceptible to the private
sector, possibly signalling important policy shifts. One piece of evidence that partly supports the Ricardian equivalence is presented in Figure 6.12, which considers the case of the United Kingdom. The figure shows that changes in the UK’s government primary
budget balance are partly mirrored by household savings: when the government borrows more, that is when it runs a bigger deficit or smaller surplus, households save more as if they were putting at least some money aside to face future tax liabilities. More formal
studies, using advanced statistical techniques, indeed suggest that households save about half of tax cuts. Ricardian equivalence is thus ‘50% true’—this is an oversimplification, of course, but worth keeping in mind.
Fig. 6.12 Ricardian Equivalence in the UK, 1970–2016
The figure plots the primary budget balance of the British government (in per cent of GDP) and the household saving rate, the proportion (in per cent) of disposable income that is saved. Remarkably, the two curves tend to move systematically in opposite directions, although not always
and not exactly. This suggests that at least some part of the spending of British households behaves in a Ricardian fashion.
6.6.1 The Primary Current Account

The consolidated budget constraint of the private and public sectors can be thought of as the intertemporal budget constraint of the nation as a whole. In Chapter 2 we saw that national net saving vis-à-vis the rest of the world occurs through the current account.
Like the public sector budget surplus, the current account can be decomposed into the primary current account and net investment income, which is the part of income account that represents the net investment income (interest service) on foreign assets and
liabilities:
(6.15)
where F stands for the country’s net asset position (sometimes called net investment position) vis-à-vis the rest of the world, and r, as before, is the real interest rate paid on F. Net investment income is positive when the country holds more assets than liabilities
(F > 0), or negative in the case of an indebted country (F 0).24 Translated into the two-period framework, the budget constraint of the nation requires that the present value of a country’s primary current account surpluses be no less than the value of international
assets in the first period:
Box 6.5 Global Imbalances
Figure 6.13 show estimates of net foreign asset positions—what we have called F1—for selected countries, expressed as the percentage of each country’s GDP. At one end of the spectrum, countries like Switzerland, Japan, or the Netherlands could in
theory run current account deficits for years and still not violate their budget constraints; in fact, these countries are still running surpluses and further accumulating net foreign assets. At the other end of the spectrum, countries like Greece, Portugal,
Ireland, and Spain must eventually start paying back their debts, and yet they are still running current account deficits.
Fig. 6.13 Net Asset Positions, 2014 (%of GDP)
The figure displays the difference between what residents (households, firms, and government) own abroad and what they owe to the rest of the world. This includes not only loans to foreign countries less borrowing from abroad, but also ownership of financial assets, foreign
exchange reserves, com-panies, production facilities, and buildings. The net position is measured in relation to each country’s GDP.
Source: World Economic Outlook, IMF.
Box 6.6 Pyramids: Is it Possible to Beat the Budget Constraint?
Failure to understand the budget constraint can be costly for ordinary citizens and governments alike. The view that debts must be repaid is frequently lost on investors in ‘pyramids’. Pyramids are dubious investments which promise depositors outsized
returns, but pay those returns to the earliest investors by using newly invested money from others. Word of mouth spreads the news of the wonderful opportunity and more and more gullible people line up to invest, especially when the first depositors are
repaid and sceptics are silenced by the ‘evidence’. So the scheme grows and grows, as it must if only to pay back maturing deposits. But it cannot grow indefinitely, simply because there is not an infinity of people in a country, or even in the world.
Pyramids eventually collapse and the people who set up such schemes know it. So they wait until they have enough investors, and then suddenly disappear with the money, and thousands discover that they have just lost their lifetime savings.
Pyramids are often called Ponzi schemes, after Charles Ponzi, an Italian immigrant to the USA who operated a large-scale scheme in the early twentieth century. Poor Ponzi did not run away fast enough; he went to jail, was later released, deported, and
died a pauper in a Rio de Janeiro hospital.25 Most countries ban pyramid schemes, but they flourished in the early years of transition in several former communist countries (with huge ones in Bulgaria, Romania, and Russia). Apparently, most ordinary
citizens didn’t grasp the mathematics of intertemporal trade and budget constraints, or were convinced by smooth-talking salesmen that they could pass the potato off to others in time. The collapse of an Albanian Ponzi scheme in late 1996 wiped out the
savings of tens of thousands of already poor people, many of whom had sold their cattle and houses in response to promises of 300% return and more. Massive street demonstrations and social unrest subsequently brought down the government,
which had failed to close down the pyramids after they had collected an estimated €1.1 billion in a country with a GDP at the time of €2.3 billion.
Modern pyramid schemes are not the exclusive preserve of Eastern European countries. For more than a decade, New York financier Bernie Madoff ran what was probably the biggest pyramid ever. He was able to attract the money of savvy professional
investors and rich grandmothers from around the world. He now sits in a US federal prison after defrauding his clients of an estimated $65 billion. The recent ‘sub-prime’ crisis in the USA, in which mortgages (long-term housing loans), were traded
between banks and investors, has many aspects of a Ponzi scheme. Fundamentally, loans of poor quality were made, then pooled together, repackaged, and sold off to other investors, who rationalized the investments with expectations of ever-
increasing house prices. As long as there is greed and overestimation of one’s own chances, there will be attempts to beat the intertemporal budget constraint. They are successful only for the few brilliant and greedy criminals who originated the
scheme in the first place. Assuming they get away with it.
(6.16)
If a country has net wealth at the beginning of period 1 (F1 > 0), it can draw on it to run future current account deficits. If there is external debt (F1 < 0), the present value of current accounts must be positive, by an amount sufficient to repay the external debt plus
interest. Box 6.5 presents the net external position of a few countries.
This pattern has been called global imbalances, a topic that is a recurrent theme of countless international summits and finance minister meetings. This situation cannot go on indefinitely, because budget constraints cannot be violated. The fear is that the
constraint will reassert itself through a crisis, pretty much as evolving violations of the government budget constraint eventually led to the Eurozone crisis.
The implication for the country as a whole is the same as for the private and public sectors. A primary current account deficit in the first period ( PCA1 < 0) must be repaid by primary current surpluses (in present value) in the second unless the previously
accumulated assets (F1 > 0) are sufficiently large. Symmetrically, surpluses in the first period enable a nation to spend more than it produces in the future. It would seem wasteful for a country not to do this; otherwise it is literally giving away resources in return
for claims on the rest of the world that it will never use. For that reason, it seems likely that countries with large surpluses today will eventually get wise and start using these surpluses to improve the standards of living of its citizens.
6.6.2 Enforcement of International Credit Contracts and Sovereign Borrowing

If a country persistently fails to satisfy its budget constraint, it will face a host of problems. Many of the crises of the 1990s can be traced back to growing fears that some countries were not going to meet their budget constraints. But ultimately international
borrowers have more or less honoured their obligations. Research shows that despite spectacular exceptions, international borrowers more often than not actually repay their debt in present value terms.
Governments, just like private households and firms, face an intertemporal budget constraint that limits their ability to borrow at any point in time to the present value of lifetime resources. ‘Lifetime’ has a clear definition for individuals; for firms and
governments less so, since the existence of firms and governments is neither guaranteed nor necessarily limited. Nevertheless, within a legal jurisdiction, private borrowers and lenders will generally be able to rely on special institutions to enforce the budget
constraint. Firms or individuals who simply walk away from debts face bankruptcy—exclusion from future borrowing—and possibly jail. Of course, there are always exceptions, but they generally involve fraud, either via outright default (‘take the money and run’)
or more complicated schemes such as financial pyramid schemes described in Box 6.6. These tricks are either illegal or declared to be soon after they are detected. In principle, these rules should also apply to governments, regardless of whether they borrow at
home or abroad. As soon as they appear to be violating their budget constraint, the source of credit should dry up rapidly. International institutions with the purpose of monitoring international borrowing exist—the International Monetary Fund, the Paris Club, or
the London Club are examples—but function imperfectly. Arguably, the European sovereign debt crisis starting in 2010 arose because such an institution was absent within the European Monetary Union.
All the same, it is important to distinguish between international borrowing by private entities and sovereign borrowing, or borrowing by national governments from foreigners. A country cannot be bankrupted or jailed. Unlike private lending within a
country, enforcement of sovereign loan contracts is legally difficult. What happens when a country’s government is unable to service its debt? The first reaction is that foreign lending immediately stops, and this often affects would-be private borrowers. The
country must at least balance its current account, since it cannot borrow, which forces painful adjustments in private and public budgets. Thereafter, negotiations start with the creditors to try to arrange a rescheduling of debt service. Rescheduling means that the
terms of repayment are changed from the original loan agreement, but without changing the present value of those repayments. Debt forgiveness, in contrast, involves a reduction in the present value of the loan burden to the borrowing country, and a loss to the
creditor.
Summary
1 Because households may borrow or lend, their budget constraint is fundamentally intertemporal. It incorporates all current and future spending on the one hand, and all current and future income on the other. Future spending and incomes are discounted using the interest rate at which
households can borrow or lend.
2 Wealth is the sum of the present value of current and future income and inherited assets less debts. The intertemporal budget constraint requires that the present value of spending be less than, or equal to, wealth. It applies to all economic agents, households, firms, the public sector, and
the nation as a whole.
3 When firms invest, they forego—on behalf of their shareholders—current consumption for future output. The profitability of investment depends both on the technology and on the rate of interest. The rate of interest is the opportunity cost of capital that investors apply to investment
projects because it is available on other assets.
4 Budget constraints can be added together, or consolidated. Consolidating the households’ and the firms’ budget constraints gives the budget constraint of the private sector. To a first approximation, firms are a veil: they provide their owners or shareholders with a means of increasing
their wealth.
5 The public sector intertemporal budget constraint implies that, for a given time profile of government purchases, tax reductions today imply tax increases later on, and conversely. Alternatively, given a tax profile, more government spending today implies less spending later on, and
conversely.
6 The Ricardian equivalence proposition asserts that the private sector internalizes the public sector budget constraint. Public debt does not represent private wealth, and the path of taxes over time does not affect the private sector budget constraint. If the private sector can freely borrow
at the same rate as the government, additional public dissaving is matched one for one by private saving.
7 Ricardian equivalence is unlikely to hold for several reasons. Interest rates are different and usually lower for governments than private agents. Individuals may rightly expect that some current public debt will be repaid after they die. Many households face borrowing constraints. Yet
there is some evidence that the private sector internalizes part of government debt.
8 The national budget constraint is the consolidation of the private and public sector budget constraints. It states that the present value of primary current account deficits cannot exceed the nation’s net external wealth. It also implies that, all things being equal, higher primary current
account deficits today will require primary current account surpluses in the future.
9 Although it must also obey an intertemporal budget constraint, sovereign borrowing by a nation may differ from private international borrowing by its individual residents. One difference is that governments and countries cannot be bankrupted, and the assets of defaulting governments
are hard to seize.
Key Concepts
intertemporal budget constraint

rational expectations hypothesis
endowment
autarky
intertemporal trade
real interest rate
intertemporal price
discounting
consol
budget line
present discounted value
investment
fixed capital formation
production function
opportunity cost
primary budget surpluses
primary deficit
Ricardian equivalence proposition
primary current account
global imbalances
sovereign borrowing
Exercises
1 Draw a budget line for Crusoe in a two-period world, assuming an interest rate r of 5% and an income of 100 in the first period (Y1) and 200 in the second (Y2). What is the value of total wealth Ω? Why are your answers different when instead Y1 = 200 and Y2 = 100?
2 Use the example given to consider the case of a higher interest rate r = 10%. In terms of wealth, which version of Crusoe has more to lose from an increase in the interest rate? Compare your answer to an individual with Y1 = 300 and Y2 = 0. Explain.
3 Suppose that Crusoe cannot trade intertemporally with his native neighbours, but coconuts no longer spoil completely, so he can store them for consumption tomorrow. Consider the case Y1 = 200 and Y2 = 100, and suppose that 10% of the stored coconuts are lost because of spoilage.
Represent this budget constraint graphically. Why does opening the market for loans always make him better off?
4 In the text, Robinson Crusoe does not want to leave any wealth beyond tomorrow because he will be rescued. The situation would be different if he wanted to leave his friend Friday a.pngt of a fixed amount B3 in the second period
(B3 might also be thought of as a bequest).
Write down Crusoe’s budget constraint and represent it graphically.
5 The real interest rate is 5%. What is the value of a new firm which invests €200,000 initially and expects to have profits (sales minus costs) of €100,000 next year, €70,000 the year after, €40,000 the third year, and then to close down with equipment valued at zero? How does your
answer change if the equipment bought initially is instead sold for €50,000? How does your answer change if the interest rate rises to 10%?
6 Sometimes when a firm announces a plan to take over another firm using its cash and not to distribute dividends to its shareholders, its share price decreases; and when the plans are cancelled, the share price rises again. At other times, the opposite pattern is observed. Can you explain
why? Under what conditions would you expect such an action to have no effect at all?
7 Suppose the production function has the Cobb–Douglas form: Y = AKαL1−α, and assume labour input is fixed at L = 1. Let α = 0.5 and A = 1. At an interest rate of 5% and no depreciation, what is the level of the capital stock K for which the project is just profitable? How does your
answer change when the interest rate is 10%? When the depreciation rate is 5% per annum? When A = 2?
8 Write down the value of the firm when the production function is Y2 = F(K2, L) with L = 1, and the capital stock in the second period is given by K2 = I1 + (1 − δ)K1. How does the initial and given stock of capital in the first period affect the firm’s value? The depreciation rate? The
rate of investment?
9 Consider a government in a two-period world starting with €1,000 in debt. G1 = G2 = 500 and T1 = 400. If the interest rate rG = 0.05, what do taxes in the second period need to be to guarantee the solvency of this government? How does your answer change when rG = 0.10?
10 Starting from your answer to the previous problem, show how a tax cut in period 1 can increase wealth if rG = 0.05 and r = 0.10? What happens if rG = 0.15? If rG = 0.10?
11 Compare the two measures of the Italian budget shown in Figure 6.8 with that of Table 17.3 (in Chapter 17). Explain the difference.
Essay Questions
1 Why do you think the interest rate is positive? What would be the consequence of a negative (real) interest rate?
2 For the government to honour its budget constraint, it is sufficient but not necessary that its budget be balanced every year. Why? Why might a balanced budget law not be a good idea? What conclusions do you draw?
3 Some contend that the ‘pay as you go’ system of social security in many European countries, in which the pension contributions of the currently employed are used to pay the pension benefits of older workers already in retirement, is a pyramid scheme. Do you agree or disagree?
Explain.
4 When a country defaults on its external debt, a frequent controversy concerns whether the country is unable or unwilling to honour its debt. Discuss this distinction and why it is difficult to resolve the controversy.
5 Argentina defaulted in 2002 in the midst of an economic and political crisis. Initially, the government refused to negotiate with its creditors. Eventually, in 2005 it reached an agreement with most creditors, which included a debt reduction of 70%, and started to pay back the remaining.
Yet, a minority of creditors insisted on being paid back in full. After years of legal battles in the US, these creditors won in 2014. Many of them, sometimes called ‘vulture funds’, had bought the debt from previous creditors at much discounted prices and stand to make huge profits.
This story raises many issues such the sanctity of debts, the responsibilities of lenders, equal treatment of creditors, or the role of vulture funds. What is your own view?
1 Some alternatives to rational expectations are presented formally in the WebAppendix to this chapter.
2 This point highlights an important tension between rationality of individual behaviour and aggregate outcomes. We will see in Chapter 7 that financial markets can be susceptible to problems when too many people blindly believe the signal thought to be contained in prices.
3 As a simplifying assumption, we have assumed that the interest rate is the same, whether one is borrowing or lending. The world is more complicated, but the logic is unchanged, when we consider different rates of interest for borrowers and lenders. Section 6.5 provides more details.
4 The slope of the budget constraint is negative and is given by –1 times the ratio OD/OB. From the text we know that
OD/OB = [Y1(1 + r) + Y2]/[Y1 + Y2/(1 + r)] = 1 + r.
5 This is obtained by noting that today’s available endowment is Y + B so that (6.2) is changed to C = Y + (B + Y − C )(1 + r).
1 1 2 2 1 1 1
6 If we set L = 1 and Y = F(K, L) as in Chapter 3, then Y = F(K, 1).
7 The reason behind this principle is that, given the existing amount of labour (Crusoe’s time) used to operate the equipment, adding additional equipment is less and less effective in raising output.
8 Alternatively, Crusoe could borrow coconuts for investment purposes. The interest rate then is literally the cost of this debt-funded investment. This is discussed in the WebAppendix.
9 We assume that the trees have no resale value; they simply die after the second period. If they didn’t, it would be necessary to add back the resale value of the depreciated trees in the second period, which would increase the value of firm. This modification is described in detail in Chapter 8.
10 To see this, write wealth as the present discounted value of net income and rearrange using , yielding
11A subtle, but important point: This valuation of the firm is independent of whether Crusoe finances the investment himself out of savings (or ‘retained earnings’ in the language of business) or whether he borrows funds to finance it (and discounts the project returns using the same interest rate). In
this benchmark case, it doesn’t matter. This result is known as the Modigliani-Miller Theorem, and is discussed in the WebAppendix to this chapter.
12 Because the production function still lies above the new budget line B'D', total wealth could be further increased by investing a little bit less than I1. Chapter 8 shows that, when Crusoe strives to do the best he can—behaves optimally—he will invest to push out his new budget line as far as possible,
i.e. he maximizes the value of his total wealth.
13G denotes government purchases of goods and services. It is not the same as total government spending or outlays, which include transfer payments. In our notation, transfer payments are deducted from taxes, resulting in net taxes T. Although interest payments are treated like transfers in the
national income and product accounts, they are such a central component of the intertemporal budget constraint that we will always distinguish them from other transfers throughout this book.
14Source: World Economic Outlook database. As a further example, consider Robinson Crusoe’s government in the first period. The stock of debt to begin with is D , the primary surplus is T – G while interest ‘income’ is –r (D + G – T ). The headline deficit in the first period is therefore the
1 1 1 G 1 1 1
surplus multiplied by –1, or G1 – T1 + rG(D1 + G1 – T1).
15The implications of economic growth for government budget constraints and stabilization policy will be explored in detail in Chapter 17.
16The public debt should be distinguished from the external debt, although in some instances the public debt is held by foreigners and represents the bulk of the external debt. This chapter assumes that the public debt is held by domestic residents.
17We will have much more to say about this in Chapter 17, which looks more carefully at government deficits, debt, and the role of the central bank.
18 To derive this result, multiply both sides of (6.12) by , and rewrite as or Substitution of this last expression into (6.11) yields (6.13).
19Named after English economist David Ricardo (1772–1823), who first formulated this idea, only to dismiss it as unlikely. The idea has been revived and championed by Harvard economist Robert Barro.
20 This can be readily shown by rewriting (6.13) as
21Other potential failures of the Ricardian equivalence proposition are related to the behaviour of agents under uncertainty, and go beyond the scope of this book.
22 It is true that poor people do not save and spend all of their income, while rich people only spend a fraction of what they earn. Taking from the rich to give to the poor does raise spending. Yet, the effect is typically very small, hence the ‘first approximation’ conclusion.
23 More details on the rates of return paid on risky assets are provided in Chapter 7.
24In Chapter 2, it was noted that when writing (2.8) as income less absorption (Y – A = CA) then the proper measure of income is GNDI. If we ignore the difference between rF and the international income account balance, equation (6.15) shows that if Y is GDP, we have Y – A = PCA.
25 For more details on Charles Ponzi’s life see <http://en.wikipedia.org/wiki/Charles_Ponzi>.
Asset Markets
7
7.1 Overview
7.2.1 Characteristics of Financial Markets
7.2.2 Implications: Volatility and Profitability
7.3.1 Intermediation
7.3.2 The Price of Time
7.3.3 Allocation of Risk
7.3.4 The Price of Risk
7.3.5 The Risk-Return Trade-off
7.4.1 Bonds
7.4.2 Stocks
7.4.3 More Sophisticated Assets
7.5.1 Arbitrage
7.5.2 The Bid–Ask Spread
7.5.3 Three Puzzling Implications of Market Efficiency
7.5.4 Market Efficiency or Speculative Manias?
Summary
Every great crisis reveals the excessive speculations of many houses which no one before suspected, and which commonly indeed had not begun or had not carried very far those
speculations, till they were tempted by the daily rise of price and the surrounding fever.
Walter Bagehot
History demonstrates that participants in financial markets are susceptible to waves of optimism. Excessive optimism sows the seeds of its own reversal in the form of imbalances
that tend to grow over time.
Alan Greenspan
7.1 Overview
The previous chapter stressed the role played by saving and borrowing as households, firms, and the government strive to match spending and income over time.
We also saw how firms raise income when they borrow to invest in productive equipment. Saving and borrowing are essential actions for the prosperity of people
and for the success of firms. Collecting savings from millions of people and allowing millions of people and firms to borrow is achieved through the financial
markets. These markets include banks, organized exchanges (for stocks, bonds, and other securities to be discussed later), hedge funds, and many other formal or
informal networks. Taken together they form a giant market for assets, or claims on future resources. Most of these claims also represent liabilities of other
economic agents. As noted in Chapter 2, assets are traded internationally, which often involves converting them from one currency to another. To that effect,
exchange markets operate as an integral part of the asset markets.
The asset markets do not just shuffle assets and liabilities around. They also set their prices by balancing demand and supply. Asset markets which are quite
different from goods markets, not only because they are impersonal—transactions are almost never face to face—but also because they can move large sums of
money at a moment’s notice. As a result, asset prices exhibit a high degree of volatility. They often appear driven by the whims of traders whose preoccupation is
with very short-term gains, without much apparent concern for the impact of their actions. We often read and hear about speculators who seem to gamble in these
markets, casino-style, make a lot of money, and occasionally run the economy into the ground. We will examine this possibility, although the truth is much more
complex.
An especially distinctive feature of assets is that they are durable. They are not consumed, but stored for later use, and later is at the discretion of the owner.
Their value is not in today’s use, but in tomorrow’s resale price. For this reason, they are driven entirely by the future. Almost by definition, the future is uncertain,
so assets almost always represent the assumption of risk—the possibility of large or total loss.
In this chapter, we offer a simple, but unified treatment of asset prices and markets as regards their role in the macroeconomy. We start by describing asset
markets and explaining some of their key features. They are big because they deal in stocks, not flows. They are fast-moving because profit opportunities can be
huge, but dissipate in seconds. Asset markets put a price tag on uncertainty—a special but unavoidable characteristic of every economy. They offer signals as to
where profit opportunities are arising, and thus facilitate the intertemporal shifting of resources already described in Chapter 6. We explore why markets sometimes
embark on apparently senseless behaviour, producing successive phases of exuberance and bust. We will conclude that this aspect, while sometimes unproductive
for the general economy, is a feature of market economies which can be regulated, but not eliminated.
7.2.1 Characteristics of Financial Markets

There are many asset markets—money markets, stock markets, bond markets, commodity markets, foreign exchange markets, derivative markets, and more—but they
share a number of unique characteristics.
First, unlike goods and services which are bought for consumption and often perishable (fruits and vegetables don’t last long, cars and computers become
outdated, and services are consumed at the same time they are produced), financial assets are durable. This distinguishing characteristic, plus the fact that they can
be held with negligible storage costs, are two reasons why they are a natural vehicle for saving for future consumption—just as Robinson Crusoe reasoned in
Chapter 6. They can be bought now and sold later, either when the time comes to use savings for consumption or when they are exchanged against other assets to
cash in profit or avoid a loss.
Second, in contrast with markets which trade in flows of goods or services, financial asset markets deal in stocks . The volume traded at any point in time can be
completely divorced from the creation of new assets, reflecting instead decisions by all asset holders regarding their total holdings. Consider Table 7.1, which
displays the total value of shares—partial ownership of companies—listed on five large stock exchanges (New York Stock Exchange, London Stock Exchange, the
Deutsche Börse, Zurich, and Euronext) and the total value of trades during one month.1 The last column reports the average volume of daily transactions during
that month as a percentage of the corresponding market capitalization. On any given day, only a tiny fraction of existing assets are actually traded; the rest is simply
held.
The paradox of asset markets is that, irrespective of currently trading volumes, the entire stock of tradeable assets could be brought to market at once if the
owners so desired. Indeed, while normal days are the rule, there can be hectic intervals when market participants become concerned, or even panic, and trading may
rise to a large multiple of usual levels. This explains the size and potential volatility of financial markets. In early 2016, the capitalization of the London Stock
Exchange was about 165% of UK GDP, so it is easy to understand why turmoil in financial markets is bound to affect the economy.
Table 7.1 Stock Market Capitalization and Trading, February 2016
Total value of outstanding shares in billions of Total market value, (% Monthly trading volume in millions of Average daily trading total
local currency of GDP) local currency value (%)
London 3168.1 164.8% 196.7 0.30%
New 17042.9 2.7% 1550.5 0.45%
York
Frankfurt 1408.2 45.7% 111.3 0.38%
Euronext 2916.7 83.5% 161.8 0.26%
Zurich 1365.3 212.8% 86.5 0.30%
Source: World Federation of Exchanges and OECD. The ‘GDP’ used to scale Euronext capitalization is the euro value GDP of the Netherlands, Belgium, France, and Portugal.
Third, financial markets are typically well-organized trading platforms for standardized assets that can handle large sudden volumes with ease. In earlier times,
asset trading occurred in trading halls or exchanges. The action was dominated by shouting, gesticulating traders wading through seas of hastily scribbled papers.
Then as now, asset markets were constructed as efficient conduits and processors of information. Today, most markets are computerized, and market participants
are linked through telecommunication lines from terminal screens virtually anywhere in the world. A great deal of trading is executed by computer algorithms, with
no human beings involved at all. The internet has created a single, global market in stocks and other financial instruments, with traders constantly in touch with
each other. As Figure 7.1 shows, at any moment in time, 24 hours a day, a financial market is open somewhere in the world, and any individual in the world with
access to a telephone or an internet connection may trade on it. This is perfect competition!
Fig. 7.1 Trading Hours of Stock Markets Around the World, Greenwich Mean Time
Stock markets are located around the world. As the world turns, some market somewhere is open, processing new information and pricing assets accordingly.
Sources: Deutsche Börse.
7.2.2 Implications: Volatility and Profitability

These characteristics force us to modify the usual demand and supply analysis in two ways. First, net flows can be misleading. The annual net savings by
households represents only the increment to their stock of wealth. Newly created assets—from newly printed money to new shares of a company—are
indistinguishable from existing ones; trading at any moment in time involves both new and old assets. In this way, asset markets are very different from markets for
goods and services, in which demand and supply of flows must be balanced. In asset markets, prices move to clear the demand for and supply of the whole stock.
Only a small part of the stock is traded because most owners keep what they have; their very inactivity is, in fact, a market decision.
Second, durability means that a key concern of market participants is the future value of each asset that they hold. After all, in contrast to non-durable goods that
are promptly consumed, assets are acquired for keeps, not for immediate enjoyment. Two implications follow. First, asset markets are necessarily forward-looking.
Market participant decisions are driven by their expectations of what can happen next—in five minutes or in five years. Then, since the future is unknown, financial
markets and their participants must deal with constant uncertainty. Unsurprisingly, they are jittery because they know that they do not have complete information
about the future. More often than not, they tend to pay close attention to what others are doing. While no one has all the information, it is very likely that someone
out there does possess useful information. One of the greatest challenges of trading on the stock markets is inferring what other traders know—or think they know
—and of that, what represents the truth. In a matter of minutes, changing expectations, or mere rumours, can radically alter demand and supply, swelling the volume
of trade, or drying up markets. Prices, which equate demand and supply on a second-by-second basis, can therefore fluctuate widely over time. This reasoning
applies not just to strictly defined assets (stocks, bonds) but to any durable object that can be (relatively) easily stored and sold—artwork, commodities, or
contracts for the future delivery of goods which may not yet exist (such as oil, electricity, wheat, or pork bellies). In fact, most durable goods are traded on markets
(oil in Rotterdam, wheat, cattle and orange juice in Chicago, art at Sotheby’s) which closely resemble those for financial assets.
Despite the widely documented volatility of asset prices, the nature of asset markets make it unlikely that profitable information will remain a secret for very long.
In fact, it is well-established that profit opportunities in asset markets are quickly, if not immediately, dissipated by the activities of participants. We call this
proposition the no-profit condition, and it is applied many times in this chapter. It states that on average, financial dealings of similar characteristics are likely to
yield the same interest rate or bear the same rate of return. The no-profit condition does not mean that excessive profits are impossible; but rather that they are the
exception rather than the rule—due to forces of competition. We will have much more to say about this important implication later.
The intense activity associated with asset markets, combined with the often phenomenal profits of some market participants, can easily create an impression that
financial markets are giant casinos with little economic purpose. This would be a dangerous oversimplification. Asset markets serve essential economic functions,
without which we would be significantly worse off for a number of reasons (some of which were already alluded to in Chapter 6). Markets for financial assets
perform three essential economic functions. The first is to bring borrowers and lenders together—more generally those who need resources now, versus those who
are willing to part with them for a time. The second is to put a price on the future. The third is to create a market for risk by putting a price on uncertainty and
enabling participants to control the risk they are exposed to. We now look at these functions in more detail.
7.3.1 Intermediation
Financial markets provide a meeting place for millions of households and firms who want to shift resources intertemporally—either saving or borrowing—or
intratemporally—from one form of asset or liability to another. But most individuals do not deal directly on asset markets. This is not only because the quantities
they desire to trade are small, but also because transacting in asset markets requires a great deal of expertise. To avoid these problems, they can act indirectly
through financial intermediaries. Financial intermediaries channel of new resources from savers to businesses who want to invest and households which want
to borrow. They also help households and businesses purchase and sell existing assets in financial markets. Professional traders accept and execute large orders on
the basis of mutual trust and charge each other relatively small fees. On foreign exchange markets, an average trade is about €1–2 million; converting this sum from
pounds sterling to euros and back again involves a transaction cost of around 0.05% or about €500–1,000. Similar fees are charged for large transactions involving
stocks or bonds. In return for a fee, intermediaries place orders for several smaller customers at once, or may even ‘make a market’ by maintaining a large inventory
which they can sell from or add to. In this way, intermediaries themselves may become asset-holders.
Banks are the most prominent form of important intermediary in financial markets. As is stressed in Chapter 9, banks perform many functions in the modern
economy. Most importantly they take deposits from and make loans to households, firms, and the government and in doing so provide the means of payments for
transactions. In Europe, the intermediary role of banks is relatively more important for the raising of funds than in the US, where asset markets are generally used
directly.
7.3.2 The Price of Time

The price of the future—or the reward for waiting—is the interest paid by borrowers, as explained in Chapter 6. By setting interest rates, financial markets price the
future. Bonds, which are tradable and standardized loan contracts, represent fixed payment streams in the future based on an agreed-upon interest rate. Shares are
titles of partial ownership in companies and promise dividend payments over the lifetime of those companies. We will see later how share prices are related to
interest rates. Market interest rates are the result of bringing together the demand for assets by lenders with the willingness of borrowers to trade repayment later for
cash now. In that sense, market interest rates are tightly linked to the overall impatience of borrowers and lenders.
There are many interest rates. A first distinction is the perceived riskiness of the borrower; we return to this aspect in Section 7.3.3. A second distinction is the
maturity of the loan. Typically, for a given riskiness, interest rates are higher the longer the maturity is. This is represented by the yield curve, as shown in the
left-hand chart in Figure 7.2. Two main reasons explain the shape of the curve. First, uncertainty grows as the time horizon lengthens, an issue that is examined in
Section 7.3.3. Another reason is flexibility or liquidity. For the lender, a longer maturity means that it will take more time to recover the money; it makes sense to ask
for a higher interest rate to compensate for being locked in. The borrower is willing to pay such a maturity premium because the funds will remain longer at its
disposal. All other things being equal, this premium increases with maturity.
While a yield curve is expected to like Figure 7.2(a), its exact shape is particularly interesting as it reveals a lot about the opinions of investors trading in the
market. This is clear from the two yield curves shown in Figure 7.2(b). In contrast to the simple intertemporal trade-off in Chapter 6, lending or borrowing over
several periods can be carried out in a number of different ways. For example, a two-year loan could be a single loan of two-year maturity, or two successive one-
year loans, or 24 one-month loans, etc. Because borrowers and lenders can choose any of these combinations, it stands to reason that they must be roughly
comparable: they must impose the same cost to the borrower and the same reward to the lender. If this were not the case—let us suppose that a two-year loan
carries a lower interest rate than two successive one-year loans—then no lender would be willing to lend for two years and no borrower would be willing to borrow
twice for two successive years. In fact, clever financiers would borrow for two years, lend twice consecutively for one year, and make a handsome profit. As many of
them rush to take advantage of this opportunity, brisk demand for two-year loans drives the corresponding interest rate up as much as the rising supply of
consecutive one-year loans drives the corresponding interest rates down. In the end, the two strategies will indeed imply the same lending cost and the market will
be in equilibrium.2
Fig. 7.2 Yield Curves

The yield curve relates interest rates, sometimes called yields, on assets of varying maturity, where the maturities are ranked on the horizontal axis. Chart (a) shows a ‘normal’ yield
curve. Chart (b) displays two yield curves as observed on 11 May 2011 and 29 February 2016 for high-quality government bonds in euros for maturities from 3 months to 15 years.
The first one accords well with the standard shape. The second one shows that the market expects interest rates to be low for another two to three years and then rising but not
above levels previously anticipated.
Sources: European Central Bank.
Box 7.1 explains why the yield curve is a consequence of the no-profit condition defined in Section 7.2. The reasoning applies to all maturities, and carries an
implication: long-term interest rates can be seen as averages of the current and expected future short-term rates. This means that longer-term interest rates must
reflect the expected evolution of shorter-term interest rates. The yield curves reveal what the market expects to see over the time horizon relevant to each maturity. In
Figure 7.2(b), the distorted curve of end-February 2016 indicates that the market expects the short-term interest rates to remain low, possibly declining over the
ensuing two to three years. The market also believes that the interest rates are not expected to reach previously seen levels, even 15 years hence. Chapters 8 and 9
will shed light on why this is so.
7.3.3 Allocation of Risk

Assets are inherently risky because the future on which they are based can hold many surprises. Borrowers who raise money using bonds can default—totally or
partially. The dividends paid on a firm’s shares depend on profitability, and firms can even go bankrupt. If you think about the future, the spectrum of possible
outcomes is very wide indeed. Some investors are willing to accept some risk in exchange for higher expected returns, but to some degree only, and they wish to
protect themselves from catastrophic events, such as a collapse of prices. Not all asset-holders are equally risk averse, and their horizons also differ. This leads us
to the second major function of financial markets: to price risk and allocate it to those who are most willing to bear it.
Just as it is possible to purchase insurance to protect against losses involving car accidents, fire, unemployment, and death, it is possible to use asset markets to
insure against financial uncertainty. This is done through diversification, the holding of a mix of several different assets. Doing so dilutes the impact of any
particular bad outcome by mixing a number of risky assets. Box 7.2 explains how diversification works in more detail. In fact, many financial firms offer diversification
to their customers by proposing ready-made investment funds, a mix of well-chosen risky assets. They can tailor the riskiness of these portfolios to accommodate
the different tastes of their customers. ‘Funds of funds’ have also been proposed to achieve further diversification.
Box 7.1 The Term Structure of Interest Rates3
Consider a long-term interest rate with a maturity of L years. Interest rates are generally quoted in annualized form, i.e. the interest rate paid each year. If the
annual interest rate on the long-term loan is RL, the total return over the entire period is (1+RL)L, because the interest is compounded over L years. For the
sake of the argument, suppose that investors knew for sure the one-year interest rate to prevail in each year t into the future, where the superscript e
denotes the value expected at that date. Then a succession of L one-year loans which reinvest the proceeds at the end of each year is given by
The no-profit condition implies that the two alternative investments should offer comparable returns. If they are set
equal, we have:
As a first approximation, this equality implies that the long rate at any time t, is an average of expected future short rates in the L periods that follow:4
In the more general and realistic case when future yields are not known, the two strategies will not have equal yields. Investors may demand compensation
for investing in the risker one. If denotes the compensation paid to investors for holding the bond with maturity L, the relation is modified to:
A more detailed explanation of this last expression is offered in Section 7.3.4.
7.3.4 The Price of Risk

Risk can be reduced through diversification, but it cannot be fully eliminated. Someone must be willing to bear some risk. The way to induce people to bear risk is to
compensate them for it. This compensation is called a risk premium. It means that the rate of return increases with the riskiness of assets. Box 7.3 explains why the
willingness to pay for various assets determines the risk premium and why.
Both borrowers and lenders benefit from the existence of financial markets. Taken in isolation, each investment is risky and would have to pay a high yield to
attract savers. Financial intermediaries, banks, and other specialized institutions pool these individual risks by offering funds to savers. The borrowers pay lower
risk premiums, the lenders bear less risk, and the intermediaries collect a fee. Because diversification tends to be more complete when more assets are involved, the
best services tend to be offered by the major financial centres dealing in stocks of companies from all over the world. This is precisely why asset trading tends to
concentrate in few places and also why worldwide electronic trading is growing so fast. High stock market trading volumes and concentration are the consequence
of the endless search for diversification to the benefit of all, whether they are lenders or borrowers, and whether they are large or small.
We have thus discovered another reason why the yield curve in Figure 7.2 tends to be upward-sloping: longer-term loans are generally associated with greater
risk. We saw in Section 7.3.2 that the maturity premium rises with maturity to reflect the impatience of both borrowers and lenders. Since lending is almost always
risky, the interest rate also includes a risk premium. To the extent that, at least in normal times, uncertainty increases as the investment horizon extends deeper into
the future, a lender is also likely to ask for a higher premium when they part with their money for a longer time. It follows that the risk premium should also rise with
maturity.
Box 7.2 Risk Diversification
Don’t put all your eggs in one basket’ is folk wisdom. The principle is remarkably simple: if you pool many risky outcomes, the result can be much less risky
than any of the outcomes taken individually. Under some conditions, averaging can reduce variability, sometimes significantly. To see how this is done,
consider the following, simple example with five different assets.
Assets A and B are based on independent flips of a coin. In either case, ‘heads’ means receiving €100, ‘tails’ means receiving nothing. On average, if
repeated many times, both A and B yield €50. We say that each has an expected value of €50.
Asset C mixes assets A and B, and is simply one-half of A plus one-half of B. The expected value of C is also €50. Asset D also combines A and B in equal
proportions, but now the returns to those underlying assets are decided by a single coin flip: ‘heads’ means A pays €100 and B nothing, while ‘tails’ means A
pays nothing and B delivers €100. By construction, D always pays €50.
Asset E is also based on an equal mix of A and B. Like D it is based on a single flip of a coin but now both yield €100 if the coin turns up ‘heads’ and both
yield nothing if it is ‘tails’. Its expected value is €50, since it yields €100 half of the time and €0 in the other half.
All five assets described have an expected return of €50, but they are very different as far as their variability is concerned. A and B pay €100 with a probability
of 50% and nothing with the same probability. The returns from asset C are less variable. To see this, list the four possible outcomes and payoffs from asset
C:
A = heads & B = heads: Payoff = €100
A = heads & B = tails: Payoff = €50
A = tails & B = heads: Payoff = €50
A = tails & B = tails: Payoff = €0.
Like assets A and B, asset C offers extreme returns of €100 or nothing, but each with only one chance in four; C further pays out €50 with a probability of 50%
with A and B. Asset C is less risky because it diversifies the risks attached to A and B by combining them. Since the returns from A and B are decided by
different flips of a coin, they are independent events. In the language of statistics, we say that they are uncorrelated.
Asset D offers even more diversification than C because in this case, A and B always move in opposite directions; their payoff is determined by the same
coin toss; they are no longer uncorrelated. In fact, they are perfectly negatively correlated, since it is always the case that either A or B pays €100 and the other
pays nothing. In this case, diversification eliminates risk completely. This is a special case of the general proposition that pooling assets is a way to reduce
riskiness. This is what diversification is all about. Asset E, on the other hand, is equivalent to owning either asset A or asset B, which in terms of their payoffs
are indistinguishable. A single flip of a coin determines that both A and B either yield €100 or nothing. This is a case of perfectly positive correlation. The
composite asset E does not reduce risk at all. Perfectly positive correlation prevents risk diversification.
Diversification can reduce risk in investors’ portfolios, but it cannot eliminate it (unless the risks are perfectly negatively correlated, which occurs extremely
rarely). If much of the risk is macroeconomic (business cycles, policy actions), risk cannot be diversified away. Foreign assets, however, are likely to have
different risk characteristics. This creates an incentive to pool markets across national borders, the key reason for the existence of globalized financial
markets.
7.3.5 The Risk-Return Trade-off
This textbook is about macroeconomics. There are many excellent finance textbooks for guiding interested students through the maze of theories of asset pricing
and portfolio choice; in short, how to use diversification to tailor investments to investors’ preferences. Macroeconomic risk, on the other hand, is a central aspect
of our lives, because it is not diversifiable. This risk may have to do with technological innovations which suddenly arrive or become suddenly obsolete.
Macroeconomic risk may also have to do with politics, migration, or international trade, or catastrophic events such as earthquakes or global warming. Someone has
to bear this fundamental risk. And after an investor has taken all possible measures to eliminate individual risks, the average risk which remains must be held by
someone, and that risk will be only borne for a price.
Box 7.3 The Price of Risk
Return to the first example of Box 7.2. Investments A and B have the same expected value, €50. How much would you be willing to pay to acquire either
investment? Most people are risk averse, and they would rather get €50 for sure than buy a risky investment with the same expected value. (Those who don’t
care and would pay €50 are said to be risk neutral; risk lovers would pay even more than €50 for the thrill.) If you are willing to pay, say, €48 for A, the risk
premium is €2 or 4% of the average or expected value of the gamble (€50). If total demand and supply of that asset are equated at €48, then the risk
premium represents the market price of risk. Since B has exactly the same characteristics, it should also be priced at €48 (by the no-profit condition).
What about investment C? It has the same expected value as A and B, but it is less risky. An investor would be willing to pay more for investment C than for
either A or B. If the market price is, say, €49, the risk premium on the new asset is only €1 or 2%. This is how diversification is an efficient way of spreading
the risks of individual assets in the market.
Now consider investment D of Box 7.2, when a single flip of a coin produces two investments A and B. Since this composite investment is riskless, it
should sell for €50, with no risk premium at all. This asset diversifies away all risk, which should benefit both sellers and the buyers of investments A and B.
The last investment E makes A and B perfectly positively correlated, there is no diversification at all. Composite asset E should sell for €48, with exactly the
same risk premium required for A and B.
Finally, for a more realistic example, suppose that you pay now, but the coin is flipped only in a year’s time. Above and beyond the risky gamble, any
investor will expect to receive interest for the time the money was tied up in the investment. Compensation for risk in this case can be expressed as a
premium added to the risk-free interest rate available on, say, a government bond. Suppose that, as before, the ‘investment’ has an expected value of €50
and the market currently prices it at €46 today. Suppose, further, that the risk-free rate is 3%. Then the risk premium can be computed as 5.7% (the total
expected rate of return ((€50 − 46)/€46 = 0.08696 = 8.7%) minus the risk-free rate (3%)).
A nice way of summarizing the role and price of risk is the risk-return curve displayed in Figure 7.3. It says that risker assets must offer a higher rate of return.
For the riskiest assets, the return might be so high that the asset cannot exist, because no one is willing to purchase and hold it.
But what is ‘risk’? We face innumerable risks without even knowing it. Riskiness changes over time—its variability as well as its comovement with other risky
events. Even though gambling markets do take many future events into consideration, some are so unimaginable that they are hardly discussed. Sometimes only
after a discussion has taken place does a market and the research associated with it emerge. The idea of diversification is that each asset must be evaluated in
comparison to all others. Two highly risky assets may offset each other perfectly so that, taken together, they are riskless. We do not pursue this involved analysis
but it is important to realize that the notion of risk is quite subtle.
Fig. 7.3 The Risk-Return Curve

The market will accept macroeconomic risk only for a price. The marginal investor determines the price of risk in the market. The curve depicts the willingness of the market to bear
that risk. The higher the overall undiversifiable risk in the market, the higher the rate of return that will be expected by the market.
So far, we have looked at assets as paying interest. Being constantly bought and sold, assets have prices attached to each of them. This section explains the
important link between prices and yields.
7.4.1 Bonds
Much lending is carried out by banks for their customers. These loans, which are assets to the banks, are usually not traded.5 It is possible for large players—
usually governments and large corporations—to borrow money directly from financial markets and bypass banks completely. They do this by selling bonds, which
are standardized forms of debt contracts specifying a schedule of payments concerning both interest and the principal. Once issued, these bonds can be resold in
markets to others, at a price that ultimately determines their yield, or rate of return.
Buying a bond gives the holder the right to receive all the interest payments initially promised as well as repayment of the principal—the amount initially
borrowed—when the bond reaches maturity. The no-profit condition implies that, at any moment in time, a bond’s current price must represent the value of all the
payments to which the bond’s owner is entitled. Since these payments will be spread over time, impatience—and risk when there is some—must be taken into
account. This gives rise to discounting, the way future payments are priced already discussed in Chapter 6.
A few examples are presented in Table 7.2. We first look at how the price P depends on the yield i. Bond A is the simplest case of a so-called discount bond of
one-year maturity with face value of €10,000, i.e. the bondholder will receive €10,000 in one year. An investor can either buy the bond at price P and receive €10,000
in one year or lend the same amount directly and receive in one year’s time the principal P and interest iP (the interest rate i is applied to the total amount lent P),
hence a total of (1 + i)P. Thus, in both cases, the investor initially spends P. In the case of purchasing the bond they will receive €10,000 in one year, and in the case
of the loan they will receive (1 + i)P in one year. The no-profit condition means that both investments should have the same valuation, i.e. the price P must be such
that (1 + i)P = €10,000. The price must therefore be P = €10,000/(1 + i). More generally, if V is the face or redemption value, the price of a safe one-year bond must be P
= V/(1 + i). This is a general expression of the principle of discounting first introduced in Chapter 6: the present value of a payment stream is simply the equivalent
value which, when invested at rate i, produces that payment stream. In the case of bonds A, B, and C, the payment is discounted by the number of years until
payment.
The last example D is the ‘ultimate bond’ or consol: it exists forever—its maturity is infinite. Such a bond is also called a perpetuity. The owner of the consol is
entitled to a coupon of €1 forever and the principal will never be repaid (because infinity is never reached!). Someone who pays P for such a consol has the option of
giving P to a borrower and asking indefinite payments that correspond to the prevailing interest rate i, applied to the bond price P, i.e. iP each year. The no-profit
condition implies that the two loans are identical so we have 1 = iP. Thus P = 1/i must be the price of the consol.6
Table 7.2 Bond Prices and Yields: Some Examples
Description of the payment stream Price in euros given yield i Yield given price P
A. One-year pure discount bond paying 1 euro 1/(1 + i) 1/P−1
B. Two-year pure discount paying 1 euro in 2nd year 1/(1 + i)2 1/P1/2 − 1
C. Ten-year discount bond paying 1 euro in 10th year 1/(1 + i)10 1/P1/10 − 1
D. Consol paying 1 euro per annum, forever 1/i 1/P
7.4.2 Stocks
Shares in firms, or stocks as they are also called, are held by households or their intermediaries. For firms, it is an alternative way of raising money. Instead of
borrowing, they ‘sell a piece of themselves’. Lending to firms is already risky because firms can sometimes go bankrupt. Stocks are even riskier because, in contrast
to bonds that regularly serve a predetermined interest, shareholders are paid according to profits and only after firms have covered their costs, including interest on
bonds and loans.
How are stocks valued? Once again, we make use of the no-profit condition. In contrast to what precedes, we will now reason in real terms—and invoke the real
interest rate r—because we wish to focus on the real value of shares. A share in a company gives rise to the payment of dividends, a portion of profits (in real
terms). Let us denote by dt the dividends paid at the end of period t. The alternative to buying a share is to hold a riskless bond with a real yield r, which we assume
constant. The shareholder does not just receive dividends, however. The value of the share that they hold can increase or decrease, which means capital gains
when the share price rises, or losses when the share price declines. If qt is the real share price at the beginning of period t, the rate of return on the company share is
the dividend yield, dt /qt , plus anticipated capital gain, (qt + 1 − qt )/qt (a gain if qt + 1 > qt , a loss if qt + 1 < qt ), valued as a proportion of the share value to be
comparable to a rate of return. For simplicity we will start by treating the future price qt + 1 as known with certainty and abstract from risk. The no-profit condition
implies that both assets have the same yield over period t:
(7.1)
This condition can be transformed into:

(7.2)
This no-profit condition says that today’s stock price qt is equal to the present discounted value of the dividend to be paid at the end of the period plus next
period’s price qt +1. An important aspect of this result is that today’s share price depends entirely on the future: dividends paid out at year’s end and the future
share price. As already noted in Section 7.2, expectations—qt +1 is not known today—of the future price drive today’s stock price.
But this is not the end of the story. If qt depends on qt +1, then qt +1 must depend on qt +2 in exactly the same way. Of course, the same will apply to qt +2, qt +3,
etc., endlessly. In the end, substituting (7.2) for itself in an endless process of telescopic recursion, we find:
(7.3)
This formula says that the current price (in real terms) of a share is the present discounted value of future dividends, forever. It shows that the market values a
company on the basis of what it is expected to earn, now and in the indefinite future. It also explains why stock prices can rise suddenly when market expectations of
future profits rise, e.g. when new technologies are developed. It can also abruptly decline if the markets take a dimmer view of the firm’s profitability in the future.
We have noted that shares are very risky. In the previous reasoning, we did not take that aspect into account. Since a share is riskier than a safe asset, its rate of
return will have to carry a risk premium. Assuming that this premium is constant, we need to modify (7.1) to state that the expected return on the share is equal to the
safe interest rate r plus the risk premium, denoted by ψ:
(7.4)
Following the same reasoning as before, we find:

(7.5)
We see that the risk premium, like the safe interest rate, reduces the share price. Quite logically, the riskier the firm is perceived to be in the future, the lower its share
price today.7
7.4.3 More Sophisticated Assets

Bonds, shares, and bank loans are the basic products of financial markets. As you might expect, modern finance has developed a broad spectrum of financial
products. As time goes by, these products become increasingly complex and, indeed, require the talents of sophisticated mathematicians, often referred to as
‘quants’ (for quantitative types) or even ‘rocket scientists’.
These products strive to deal more efficiently with risk, through various means. One obvious approach is diversification, putting together many basic assets just
as shown in Box 7.2. Indeed, most investors do not buy individual assets, but shares in investment funds, i.e. a collection of assets, assembled by specialized
financial institutions. Some funds also offer guarantees such as minimum returns, protection of invested capital, etc. More generally financial institutions are in the
business of securitization, that is creating new securities out of pools of existing ‘underlying’ assets, which can be anything that pays to its owners under certain
conditions. Often these underlying assets are hard to sell or value individually because they are risky. In the spirit of the asset C in Box 7.2, securitization creates
new, and more marketable mixtures of underlying distinct assets, with the beneficial effect of averaging out or reducing risk to their owners.8
Derivatives are financial instruments that pay out only under certain circumstances. They can be based on the value of a particular asset (or a fund). They can
allow the holder to sell the asset before maturity, or to buy it, at some guaranteed price at some future date. The owner of a derivative does not have to exercise the
transaction if it is not profitable for her, in which case the derivatives imply elapses. Derivatives allow investors to take positions to benefit from movements of the
price of the underlying asset without ever having to hold it; they only need to buy it at the last minute if they contracted to sell it, or they sell it at the same time as
they receive it if they contracted to buy it.
A popular and controversial form of derivatives is called credit default swap, or CDS. A CDS is issued by a bank, an insurance company, or a wealthy investor
over a specified period of time. If, during this period, the asset specified in the CDS misses a payment or loses its value—that is, if there is a ‘credit event’—the
issuer of the CDS is contractually obligated to pay the owner of the CDS a specified amount of money. In some contracts, the missed payment is paid; in others, the
entire debt is assumed. In return, the purchaser of the CDS makes stipulated payments every period for the ‘protection’ that is provided.
The CDS market has grown by leaps and bounds in the past decade, with outstanding volumes now well into the trillions of dollars. Although it may sound to
some to be little different from a punt in a betting shop, the CDS market serves two important functions. First, it provides a tool for transferring risk. Important
financial institutions such as banks, pension funds, or insurance companies can purchase CDSs to reduce, or hedge, risk on assets that they hold on their books
(own outright). For example, banks holding government debt which looks increasingly risky can purchase a CDS which pays if the country should default. Naturally
the risk does not disappear from the economy; it is assumed by the issuer of the CDS. The second function of the CDS market is to put an explicit price on risk, quite
simply the price of the CDS contract. As long as they are traded in competitive and transparent markets, the price should reflect the collective market assessment of
the probability that the credit event will occur. Later in Chapter 17 we will see how CDSs have been used to reveal how much markets mistrust government
borrowers in the Eurozone.
Because these financial products can become quite complex, many investors do not fully understand their properties. Sometimes, even the financiers that issue or
underwrite them fail to understand fully the risks they are taking. The global financial crisis of 2008–2009, which was characterized by the issuance of many new
types of poorly understood securities, is a case in point. Some of these aspects of the crisis are discussed in Box 7.4.
Box 7.4 Mortgage Securitization and the US Financial Crisis of 2008–2009
Traditionally, banks lend lots of money to households so they can buy the flat or house they live in. These housing loans, called mortgages, are held as
assets by banks on their balance sheets. While still risky, these loans are ‘secured’ or guaranteed by the value of the house. If the borrower fails to honour
their commitment, the bank takes possession of the house—a process called foreclosure—and sells it to recover its money.
Starting in the late 1990s, financial innovation in the USA began to allow banks to ‘originate’ loans for houses without ultimately holding them on their own
books. Instead, originators could resell mortgages in bundles to larger financial institutions, and pocket a profit while unloading the risk. The practice of
originating and reselling mortgages to large financial institutions is an example of securitization. It represented a drastic departure from traditional banking,
which considered housing loans a special commitment to the local community and ‘for the long haul’. From an economic point of view, ownership of the
mortgage aligned the bank’s own interests—in making a profit—with those of the economy as a whole—making good loans and avoiding bad ones.
In the brave new world of securitization, lenders who originate mortgage loans are in business only because larger financial institutions are willing to take
these assets off their hands. This was hardly an incentive to act prudently. Many loans were of inferior quality, meaning that the borrower was unlikely to repay
the mortgage.9 Financial intermediaries sold the bundles of these inferior mortgages to other investors all around the world. Supposedly, securitization was
justified by diversification: lending to many persons or many regions should be less risky than lending to a single person, or to a single region. If those
mortgage-backed securities were in fact less risky than the simple sum of the individual underlying mortgages, they could fetch a higher value, as explained
in Box 7.3. This explains why intermediaries could buy the mortgages at a good price from the initial lending banks and yet make a profit by immediately
reselling them repackaged as securities.
With hindsight, most analysts now agree that these fancy securities were overvalued—simply because the value of underlying houses that stood behind
them was driven by the US real estate boom and thus much more correlated than individual default risks would predict. By the end of 2006, after many years
of rapid growth, house prices started to decline in many parts of the USA. In addition, many sub-prime borrowers faced higher interest rates and realized that
the value of their debt—their mortgages—now exceeded the value of their homes. They defaulted on their loans. Foreclosures led to more forced sales of
homes in a market in which prices were already falling. Further declines in housing prices triggered more defaults, more foreclosures, and more sales. The
mortgage-backed securities were guaranteed by houses whose prices were plummeting, so the value of the securities declined. Many of the world’s largest
and most prestigious banks lost hundreds of billions of dollars in securities with high ratings from the industry’s most sophisticated and reputed rating
agencies. Due to hubris, ignorance, or simply the hot potato syndrome, bankers were blinded to systematic, macroeconomic risk of the simplest sort—
movements in house prices—so mortgaged-backed securities were in fact hardly diversified at all.
Asset prices reflect the collective judgement of market participants—borrowers and lenders alike. This judgement, in turn, is based on all the information collected
by market participants. The information concerns each single asset, its underlying value and future performance—how much it will pay under which circumstances,
as in the examples of Box 7.2. It is a hallmark of properly functioning financial markets that participants are engaged in a never-ending search for profit opportunities,
either for their customers or for their own accounts. Both the amount of money at stake and the speed at which information moves make it unlikely that opportunities
not involving additional risk will be left unexploited for any significant period of time. This is the no-profit condition presented in Section 7.3.2. Markets which
satisfy this condition—that publicly available information cannot earn consistently above-average returns—are said to be efficient. The efficient market
hypothesis maintains that asset prices fully reflect all available information.10 This observation carries a number of implications that we now examine.
7.5.1 Arbitrage
Arbitrage is an example of the no-profit condition in action. It involves financial trades that do not imply taking on additional risk.11 It is customary to distinguish
three types of arbitrage: (1) yield arbitrage, (2) spatial arbitrage, and (3) triangular arbitrage.
Yield arbitrage concerns two assets which happen to offer different returns. Strictly speaking, it applies to riskless assets, like Treasury bills, but it is
sometimes applied to risky assets which bear similar risk (like assets A and B in Box 7.2). Consider the example of two riskless government bonds which offer
different rates of interest. Holders of the less attractive bond will sell it and buy instead the one with the superior yield. In fact, the entire stock of the less attractive
bond could be put up for sale at once. As the higher-yielding bond becomes more expensive and the lower-yielding bond becomes cheaper, the implied yields
converge (the link between yields and prices is explained in Section 7.4). Arbitrage prevents such pricing misalignment from occurring: yield arbitrage imposes
identical returns for identical assets.
Spatial arbitrage concerns the same assets traded in different locations. For example, large commercial banks borrow from each other at the interbank interest
rate. If the interbank rate in two cities were to diverge, enterprising banks present in both cities would immediately borrow in the cheaper market and lend in the
dearer one. With high capital mobility and negligible risk differences, the yields should be virtually identical. This is indeed the case.
Triangular arbitrage applies mostly to foreign exchange markets and is possible when the relative prices of three—or more—currencies are not consistent
with each other. If the euro costs one US dollar and one euro costs 8 Danish kroner (DKR), then the DKR/$ rate must be (DKR8/€)/($1/€) = 8 DKR/$. Otherwise
limitless profit would be possible by buying the euro where it is cheap and selling it where it is more expensive. Figure 7.4 displays this example.
7.5.2 The Bid–Ask Spread

The risk premium takes another interesting form. To carry out the various operations just described, investors ‘go to the market’. There they need to find a
counterpart, someone interested in exactly the same trade, for the same amount, but in reverse. This is extremely unlikely, so the market would be useless were it not
for the presence of market makers. Market makers are usually large financial institutions that have an interest in keeping the market liquid at all times, meaning
that trades can be executed quickly and without large swings in price. They stand ready to satisfy any demand, buying or selling any amount—within generous
limits—that appears on the market. Each market maker specializes in certain trades, e.g. the euro–sterling exchange rate or bunds (the debt of the German
government). Of course, as they do so, they provide a service. They also take a risk and holding an inventory can be expensive. Quite naturally, the market maker
needs to be compensated for both the service and the risk. This takes the form of a bid–ask spread.
The bid–ask spread is familiar to anyone who has travelled abroad and has bought foreign currency at an exchange booth. There, as on all foreign exchange
markets, exchange rates are quoted in pairs: a lower ‘bid’ price for those who want to sell the foreign currency, and a higher ‘ask’ price for buyers. The difference is
the market maker’s profit. The bid–ask spread in wholesale foreign exchange trading is currently quite small—roughly 0.2% on a five-million euro transaction
between euros and sterling. Yet it can be much higher for currencies which are inherently risky, or are thinly traded. Box 7.5 gives a nice example of how a bid–ask
spread evolved over time for the ostmark, the money of the German Democratic Republic, the communist German country which disappeared from the European map
after German unification in 1990.
Fig. 7.4 Triangular Arbitrage

When two exchange rates among three currencies are known, the exchange rate between the remaining pair of currencies should be determined by triangular arbitrage. In the
absence of transactions costs, any discrepancy between purchasing a currency directly and acquiring it using a third currency is eliminated.
7.5.3 Three Puzzling Implications of Market Efficiency

A surprising implication of the efficient market hypothesis is that at any point in time, the price of bonds, stocks, foreign exchange, and other financial instruments
traded in asset markets should represent a consensus based on information available to traders in the market. What will happen tomorrow—or the next minute—
depends entirely on new information that does not exist today; it is perfectly unpredictable. Three conclusions follow:
First, we shouldn’t have to work very hard to obtain information about those financial assets that are traded—because the markets do this for us already!
Second, it is impossible for anyone to systematically outperform the market. Reports of investors systematically beating the markets are more likely a sign of
good luck than much else.
For every winner we hear about, there are as many losers, some of whom have disappeared from the market either because they ran out of money or because they
were sacked by their bosses.
Third, as argued by the late Nobel laureate Milton Friedman, speculation cannot be destabilizing. Those traders who are responsible for pushing asset prices
away from their ‘fundamental’ prices, e.g. such as shown by equation (7.3), are those who buy high and sell low. Prices ultimately return to their fundamental
values, so these destabilizing traders should consistently lose money and ultimately exit the market. Successful speculators buy low and sell high, in effect
bringing prices back into line, stabilizing the market.
Box 7.5 The Short-Lived Market for Ostmarks
Even before the Berlin Wall fell on 9 November 1989, trade in the ostmark (OM), the East German currency, was significant, and quotes for OM in West
German deutschmarks were published daily in major West German newspapers. After this historic date, volume increased by an order of magnitude as East
Germans tried to convert their savings into harder currency. It was unclear until March 1990 whether monetary unification would occur. This implied automatic
conversion of OM currency and bank deposits into DM. The conversion rate of one DM for one OM applying to a part of East Germany’s holdings and one for
two for the rest—resulting in an average estimated by the Bundesbank at 1.8—was first officially suggested in March. It was then formalized as part of the
state treaty of monetary and economic union between the two German states in May 1990.
Considerable uncertainty characterized this period. Furthermore, before the Berlin Wall fell, the markets were relatively thin and trade was exclusively a
Western business. This is reflected in the bid–ask spread which stood at more than 30% in early 1989, as seen in Figure 7.5. As the situation became
clearer, trade moved to the streets of East and West Berlin and most banks entered the game. With the decision to establish a monetary union between
West and East Germany by July 1990—in effect, replacing OMs with DMs—uncertainty declined, and so did the spread.
Fig. 7.5 Ostmark–DM Rate and Bid–Ask Spreads, August 1989–June 1990
As monetary union approached, the risk involved in holding ostmarks, the currency of the vanishing German Democratic Republic, declined. This is reflected in the bid–ask
spread, which fell significantly.
Sources: Burda and Gerlach (1993).
7.5.4 Market Efficiency or Speculative Manias?

The next question is whether the financial markets really are efficient. The huge financial crisis of 2008, and the ensuing market gyrations, strongly suggest that the
markets behave irrationally. Indeed, some studies do turn up statistically measurable deviations from market efficiency. For example, if stock prices decline today,
there is a tendency for them to revert over time to their previous values. This could imply that markets overreact to news, and that the markets may not be fully
efficient. But then, we must understand why no one buys these ‘oversold’ assets when it is profitable to do so. The no-profit condition implies that prices should be
restored to their fundamental values immediately.
How and why, then, might asset prices deviate from their fundamental values? It is always tempting to write off asset markets as irrational and prone to fads.
There are several reasons why markets appear to behave irrationally as a whole while, in fact, individual traders act rationally. They all involve recognition that not
all traders are alike. They may differ in the information that they have, they may be more or less adverse to risk-taking, they may operate under different constraints
and objectives.
Take risk aversion, for a start. It could be that too few traders are willing to take risky positions long enough for the expected reversion to correct prices to occur.
In this case, the deviation from the fundamental value may be consistent with the absence of profit opportunities, given that the risk involved in correcting it has its
own price as measured by the risk-return curve: the expected ‘profits’ are insufficient to compensate for the riskiness of betting against the irrational price. Betting
against the market involves considerable risk, and deviations from fundamental values can get worse before they get better.
Noise traders
The role of access to information can be illustrated through an example. It seems that only a subset of traders are informed and have access to information about the
true underlying value of assets. The remainder are noise traders who act on limited and ‘noisy’ information. The result is that they systematically lose money to
the informed traders. Noise traders arrive continuously on the scene, with new ones replacing those who systematically lose and quit in disgust, so that there are
always some of them around. Despite perfectly efficient and rational behaviour on the part of the professionals, stock prices may diverge from their fundamental
value for long periods of time.
Herds
Information is never exhaustive and traders know it. They are keenly aware that others may know things that they do not. When they observe that the price of an
asset declines and yet do not understand why, they may assume that others do. In this situation, it is perfectly rational for them to sell the asset, which further
brings its price down and worries other investors, who sell the asset, and so on. It may well be that the initial move was caused by misleading information, and yet
the market will amplify this move. This will last until the correct information surfaces. Meanwhile, it is safer for each trader to stay with the pack. Figure 7.6 provides
an example of a sharp movement, that was surprising to observers and traders and was eventually reversed after six weeks of intense anxiety.
Bubbles
Herd behaviour may help understand the phenomenon of speculative bubbles, persistent and growing deviations of asset prices from their fundamental values.
To see how bubbles may arise, consider the share valuation example from Section 7.4.2. Let us assume, for simplicity, that the real dividend of the stock is fixed
forever at d. Ignoring the risk premium, some computation shows that the share value should be:
(7.6)
This is called the fundamental value of the share. The puzzling observation is that an infinity of other price paths also satisfy the pricing relation (7.2). Consider the
case where the stock price is , which is higher than its fundamental value This violates (7.6). The dividend is too low to justify such a price. But imagine now
that market participants observe and also believe that the share price will increase to . Then, using (7.1), we can find:
(7.7)
satisfied for the ‘right’ expected future price . Thus, it is possible for the share price to exceed its fundamental value—and violate condition (7.6)—and yet to
satisfy the pricing formula (7.2) from which the fundamental value is derived. Oddly, although the fundamental value (7.6) assumes that the price remains constant
forever (the dividend is assumed constant forever) the share price may be expected to rise tomorrow. In fact, if it rises to that higher
price today becomes justified as the no-profit condition is not violated—at least for that period. The expected capital gain offsets the ‘too low’ dividend d.
Then, in period t + 1, the story will repeat itself. With we need to rise again above , and so on.
Fig. 7.6 The SP500 Index in Early 2016

The year 2016 started with a surprising fall of stock markets worldwide. The chart shows the evolution of the SP500, an index that measures the average price of stocks traded on
Wall Street. The index declined by more that 10% until mid-February. Market participants were unable to explain this surprising development.
Sources: Standard and Poor Dow Jones.
Thus an overvalued share (when its price exceeds its fundamental value) can be justified by subsequent further price increases, on and on. Any price today can
thus be validated by its subsequent evolution. This means that, if the price is once above its fundamental value, it could be justified by further price increases.
Figure 7.7 plots possible evolutions of the share price over time, for given r and d. Only one initial price does not ‘explode’, and that price is precisely the
fundamental value.
The non-fundamental paths, which are exploding without any apparent fundamental justification, are self-fulfilling, and there can be an infinity of them that all
satisfy (7.7). Without violating any market efficiency condition, prices rise because they are expected to. The apparently inexorable growth of the share price is
called a speculative bubble. It is a bubble because it keeps growing until it bursts, and speculative because its growth is due to the expectation of future capital
gains. A bubble is rational in the sense that it will continue to grow as long as traders believe that the bubble will continue to grow, validate market expectations,
and offer the ‘normal’ return.
There is a catch, however. History shows that bubbles do eventually burst. Why? Because if an asset price were to grow indefinitely, it would eventually exceed
the world’s wealth, and become too expensive for anyone. And if no one can afford it, its price must decline. In the late 1980s, Japanese real estate had become so
expensive that the value of the Emperor’s gardens was greater than the entire state of California! Sure enough, not long afterwards the bubble burst, and Tokyo,
while expensive, no longer has streets paved with gold. Similarly, in the heat of the real estate bubble in the USA in 2006, house prices had risen to unprecedented
heights, more than four times historic averages in some regions.
Bubbles are puzzling because they appear and seem expected to grow for an extended period, but never fail to fall back to their fundamental value—at a
completely unexpected time. In fact, the correction is always violent, and its timing cannot be known. Logically, markets cannot know that a bubble will end!
Suppose there was a date, call it period T, when the some asset price bubble would end with certainty, after which the price simply stabilizes at its fundamental
value. Now imagine the period just before, T − 1. For the bubble to exist at T − 1, the price must be expected to rise. Since the price in T will be at its fundamental
value, in T − 1 it must be below the fundamental value. This is a contradiction, however, with the earlier observation that the price of a bubble always lies above its
fundamental. So the period just before stabilization of the price cannot be below the fundamental, it can only be at the fundamental. Working backwards, it is easy to
see that the same reasoning applies all the way back to the present: there can be no bubble that is anticipated to stabilize! So a bubble can grow for a while, but
rational traders know that it must end with a bang—literally ‘burst’—and they must stand ever ready to jump. Just as in a game of musical chairs, someone can’t
move fast enough, and is left standing.12
Bubbles have all the features of economic rationality and market inefficiency, save for the end. Box 7.6 reviews a famous historical bubble-like episode that
occurred in the seventeenth century in Holland that involved tulip bulbs. History offers a plethora of suspicious episodes: the run-up of the world’s stock markets
before the crashes of 1929, 1987, and 1989, the explosion of property prices in the UK, Japan, and Scandinavia in the late 1980s.
Pretty much the same thing occurred in the US housing market bubble in 2007, as well as in Spain, the UK, and Ireland. Figure 7.9 displays an index of housing
prices in Ireland. Prices rose by 360% between 1978 and 1995, and then picked up speed rising by another 450% until the peak in the second quarter of 2007. Since
then, houses have on average lost a third of their peak values. The fast recovery since the fall may become yet another worrisome signal.
In each of these instances, market participants are convinced that the boom will continue. In each case, the bursting of the bubble is followed by serious
economic dislocation. Are market participants naïve? Take the information technology (IT) revolution, for instance. It is transforming the way we collect information
for both work and leisure and is seen by many as a technological ‘golden age’. When it started to take hold in the late 1990s, stock prices of IT companies started to
rise. Many of these corporations were not turning out any profit at all but some would grow enormously (this is when Google was created, and years before
Facebook came to be). It was not entirely irrational to be enthusiastic, but just how much? When scepticism overtook enthusiasm, stock markets promptly fell in
2000–2001, bubble style, as shown in Figure 7.10. The sharp rise of stock prices since 2010, once again, could be seen as a signal that a new IT bubble is on its way.
Fig. 7.7 Possible Stock Price Paths
Path (1) is the fundamental value of the asset. The price of the stock is equal to the present value of the dividend d, which is assumed constant, so it satisfies the arbitrage
condition. Paths (2), (3), and (4) also satisfy the arbitrage condition, but are explosive bubbles.
Box 7.6 Tulipmania
History has given us several instances of price behaviour that look like speculative bubbles. In Holland during the seventeenth century, the price of rare tulip
bulbs rose by extraordinary rates within the space of a month, only to collapse thereafter.13
The bubble involved tulip bulbs with non-negligible fundamental value because they were exotic varieties. Yet they became exorbitantly expensive. Figure
7.8 displays the price of tulip bulbs in the first two months of 1637, when they increased by over 3,000%, and then collapsed sharply. For example, the price
of the Switser variety is reported to have fallen to one-twentieth of its 2 January 1637 price. In one of the most authoritative accounts of the Tulipmania
episode, Charles MacKay (1841) wrote:
The demand for tulips of a rare species increased so much in the year 1636, that regular markets for them were established on the Stock Exchange of Amsterdam, in
Rotterdam, Harlem, Leyden, Alkmar, Hoorn, and other towns. Symptoms of gambling now became, for the first time, apparent. The stock-jobbers, ever on the alert for a new
speculation, dealt largely in tulips, making use of all the means they so well knew how to employ, to cause fluctuations in prices. At first, as in all these gambling mania,
confidence was at its height, and everybody gained. The tulip-jobbers speculated in the rise and fall of the tulip stocks, and made large profits by buying when prices fell,
and selling out when they rose. Many individuals grew suddenly rich . . . nobles, citizens, farmers, mechanics, seamen, footmen, maidservants, even chimney-sweeps and
old-clothes women, dabbled in tulips. People of all grades converted their property into cash, and invested it in flowers. Houses and lands were offered for sale at ruinously
low prices, or assigned in payment of bargains made at the tulip mart. Foreigners became smitten with the same frenzy, and money poured into Holland from all directions.
The prices of the necessaries of life rose again by degrees; houses and lands, horses and carriages, and luxuries of every sort, rose in value with them, and for some
months Holland seemed the very antechamber of Pluto.14
Fig. 7.8 Tulipmania, 1637

History has given us several episodes of price behaviour that resemble speculative bubbles. In Holland during the seventeenth century, the price of rare tulip bulbs rose by
extraordinary rates within the space of a month, only to collapse thereafter.
Source: Garber (1990)
Fig. 7.9 Housing Prices in Ireland, 1978:1–2015:3
The index is set to be equal to 100 in 1995. From 1970:1 to 2007:2, the index rose almost 60-fold. The moderate pace quickened in the mid-1990s and reached a peak in 2007:2,
when it dropped.
Sources: Private Property Price Statistics, Bank for International Settlements.

It should be clear by now that asset markets are central to a well-functioning economy. They channel savings into loans that finance expenditures. They put a price
tag on time and on risk. They allow for risk diversification, encouraging risk-taking by entrepreneurs. The next chapters will explain the crucial role they play in
transmitting the impulses of monetary policy. It is no wonder that they have grown to become an essential market.
At the same time, uncertainty lies at the heart of finance. The asset markets deal with the risks inherently faced by lenders and borrowers. Asset markets make the
rewards achievable but they cannot fully eliminate occasional catastrophic outcomes. In order to manage risk, the markets have become increasingly sophisticated
and risk-prone. This is why they are tightly regulated, but the best regulation cannot eliminate financial crises. While, in normal times, the asset markets greatly
contribute to a smooth functioning of the macroeconomy, occasionally they become the source of major upheavals.
Fig. 7.10 The Rise and Fall of NASDAQ Stocks, 1992–2016
NASDAQ stands for the National Association of Securities Dealers Automated Quotation (system). It is the US stock exchange that specializes in high-technology companies,
especially personal computers, telecommunications, and the internet. After more than quadrupling in value over the period 1997–2000, the NASDAQ index collapsed just as
spectacularly by more than 60%—before once again recovering to its previous peak 15 years later.
Sources: www.bigcharts.com.
Summary
1 Asset markets are special in several ways: (1) they concern stocks and not flows, (2) they deal with uncertainty, and (3) they are purely forward-looking. These characteristics
imply that financial markets are inherently volatile and sometimes even unstable.
2 By putting a price tag on the future and on risk, financial markets allow households and corporations to decide on saving and borrowing without having to gather the whole array
of uncertain information that affects their own future. They allow those savers who are most willing to bear risk to do so, at minimum cost.
3 Financial assets are traded with ease by professional intermediaries in large, well-organized markets. While financial intermediaries can be thought of as intervening in financial
markets on behalf of their customers, in fact most of the transactions correspond to trade among intermediaries.
4 The no-profit condition implies that returns among similar assets—similar in terms of risk and maturity—must be equalized.
5 Bond prices are inversely related to interest rates. The same tends to apply to stock prices, which are also driven by expected dividends. Beyond bonds and shares, there exist an
ever-increasing variety of derivatives.
6 The no-profit condition is a characteristic of efficient financial markets. In the absence of risk-taking, it takes the form of arbitrage.
7 In the presence of uncertainty and undiversifiable risk, the no-profit condition implies that expected returns are equalized up to a risk premium which rewards risk-averse agents
for bearing risk.
8 Markets are efficient when they gather and use all available information so that prices reflect fully what is known and the risks attached to any single asset. The evidence on
market efficiency—at least in its weak form, meaning that it is impossible to profit from publicly available information—is mostly favourable, in spite of phenomena such as
speculative bubbles or noise trading.
9 The forward-looking nature of financial markets can lead to phenomena like rational bubbles which occur when asset prices continuously rise because they are expected by
enough of the market to rise further. Bubbles eventually collapse, often imposing large costs on the overall economy.
Key Concepts
assets
bonds, shares
no-profit condition
financial intermediaries
maturity
yield curve
maturity premium
risk averse
diversification
risk premium
risk-return curve
yield
discounting
consol
perpetuity
investment funds
securitization
derivatives
credit default swap
efficient market hypothesis
arbitrage
yield arbitrage
spatial arbitrage
triangular arbitrage
market makers
market liquidity
bid–ask spread
noise traders
speculative bubbles
Exercises
1 Asset markets are considerably more volatile than goods markets. Asset markets balance stocks while goods markets balance flows. What is the link between these two
observations?
2 Why is the bid–ask spread the common way of organizing pricing of financial assets? Give two interpretations of the bid–ask spread.
3 Consider three assets A, B, and C with uncertain outcomes (payoffs) depending on three possible future states of the world.
Payoffs in state 1: A: €100; B: €50; C: €150.
Assume that each state has one chance out of three to happen. What mix or in which proportions could you purchase assets A, B, and C to diversify fully
(eliminate) risk?
4 The yield curve is said to be inverted when it is downward-sloping. How can you explain this occasional phenomenon?
5 It is often criticized that the price of stocks, which are often owned by wealthy individuals, rise in recessions. Can you give an explanation for this fact?
6 Consider the example of a speculative bubble in the text (Section 7.5.2). Now imagine that while, as before, there are two assets, investors no longer have perfect foresight. The
private asset can be purchased at variable real price qt and pays a fixed real dividend d. Now, however, there is a probability s that in the following period qt + 1 = 0 (i.e. the
bubble will burst), and a probability (1 − s) that it can be sold at qt + 1 > 0. Investors are risk neutral and equate the rate of return on the government ‘safe’ asset r with the
expected rate of return on the private asset.
(a) Write down the arbitrage condition.
(b) Solve for the ‘non-exploding’ value of current qt.
Essay Questions
1 How do asset markets extend and refine the model of intertemporal of Robinson Crusoe introduced in Chapter 6?
2 What are the conditions needed for a financial market to be efficient? Are they likely to be met in practice?
3 People have bad opinions of financial markets, seen as a source of illegitimate enrichment for already wealthy people. Discuss this view critically.
4 Financial derivatives have attracted a lot of attention recently, especially since the fiscal and financial crisis. Many advocate the abolition of such instruments. State arguments
for and against such a position.
5 Banks are the most important form of intermediary discussed in Section 7.3.1. In Europe 80% of intermediation occurs through banks (as opposed to 20% in the United States).
Why do you suppose that Europe has so many banks as opposed to the United States?
1 Euronext represents exchanges in Amsterdam, Brussels, Paris, and Lisbon. In 2007, the New York Stock Exchange and Euronext merged; in 2014, the merger was reversed.
Euronext remains the largest trading platform of continental Europe.
2 Naturally, the one-year loan rate at which the investor borrows after one year is unknown from the perspective of today, so the two options are not exactly equivalent. But it is
reasonable to expect the two options to be comparable in expectation in any case. Box 7.1 provides more details when they are known with certainty.
3 This box is for mathematically minded readers.
4 This equation is an approximation obtained by taking natural logarithms of the previous equation and using the fact that ln(1 + x) ≈ x.
5 Box 7.4 shows however, that banks have found innovative ways to sell such loans to investors.
6 We can reach this result by discounting each annual payment by the number of years until it occurs and adding these payments. The price of the bond is 1/(1 + i) + 1/(1 + i)2 + 1/(1 +
i)3 + … + 1/(1 + i)n + … which is equal to 1/i according to a mathematical formula used in Box 6.2.
7 The risk premium raises the discounting factor from 1/(1 + r) to 1/(1 + r + ψ) or the discount rate from r to r + ψ. This is not a coincidence. Risk, captured by its premium ψ, acts as a
permanent drag on the value of the share just like time, captured by the interest rate r.
8 Box 7.4 discusses the securitization of housing loans, which is particularly widespread, but the practice has been applied to credit card, automotive, and commercial debt.
9 These loans were called ‘sub-prime’ because loans were of worse than normal ‘prime’ quality. Just before the crisis, many loans were made to borrowers whose risk was worse than sub-
prime—in banking circles these loans were known as ‘NINJA’ loans: no income, no job, or assets.
10 The idea of market efficiency is related to the rational expectations hypothesis introduced in Chapter 6. Rational expectations assume that agents do not make systematic forecasting
errors. Market efficiency applies this concept to markets and price-setting.
11 In financial market jargon, this distinction is not always so clear. For example, traders who search for information on corporate takeovers in order to take a position in the stock ‘in
play’ are sometimes called risk arbitrageurs. Technically, this is a contradiction in terms: if a takeover is called off, the ‘risk arbitrageur’ may be left holding a great deal of stock and may
suffer a large loss.
12 Usually the informed professionals get out first, which highlights an important link between noise trading and bubbles. As the great New York financier Bernard Baruch put it: ‘When
beggars and shoeshine boys, barbers and beauticians can tell you how to get rich it is time to remind yourself that there is no more dangerous illusion than the belief that one can get
something for nothing.’
13 Part of this description is taken from Peter Garber’s (1990) survey of the Tulipmania boom.
14 Pluto was the ancient Greek god of wealth.
Private Sector Demand:
Consumption and
Investment
8
8.1 Overview
8.2 Consumption
8.2.1 Optimal Consumption
8.2.2 Implications
8.2.3 Wealth or Income?
8.2.4 The Consumption Function
8.3 Investment
8.3.1 The Optimal Capital Stock
8.3.2 Investment and the Real Interest Rate
8.3.3 The Accelerator Principle
8.3.4 Investment and Tobin’s
8.3.5 The Microeconomic Foundations of Tobin’s
8.3.6 The Investment Function
Summary
There is often misconception in reasoning about spending and investing. For example, Henry Ford’s remark has been widely reported: ‘No successful boy ever saved any money. They spent it as fast as they got it for things to improve themselves.’ In this remark Mr Ford drew no hard and
fast line between spending for personal enjoyment and investment for improvement. And there is no hard and fast line … Spending merely means expending money primarily for more or less immediate enjoyment. Saving or investing is spending money for more or less deferred enjoyment.
Irving Fisher (1930)
8.1 Overview
We learned in Chapter 2 that GDP is income paid to production factors, and it also represents value added as well as the sum of all final goods expenditures. More precisely, it is the sum of consumption, investment, government purchases, plus exports minus imports in a year’s time. These
components of total spending represent demand for goods and services by various sectors—households, firms, the government, and foreigners. The sum of these components is often called aggregate spending, or aggregate demand. In this chapter, we focus on explaining the behaviour of the
most important private elements of aggregate spending: consumption and investment.
As the epigram to this chapter suggests, consumption and investment are driven by different motives. Private consumption spending, about two-thirds of GDP, is a much more stable component over time than investment, which is volatile and is often thought to be a major reason for
business cycle fluctuations. We will see that while consumption represents the ultimate source of worldly existence and satisfaction, investment is only a means to an end, enabling the economy to produce more goods and services in the future. Among other things, this chapter will
contribute towards explaining the strikingly different behaviour of consumption and investment. Moving from our long-run model of economic growth to the short-run view of business cycle fluctuations is all about understanding these differences.
As in previous chapters, we will continue to study the decisions of a representative consumer and a representative firm. Both are taken to be rational: they strive to do the best they can, given their available resources and opportunities. We often say that they take decisions to optimize, or
achieve the best possible outcome. Although optimizing behaviour is sometimes understood as implying extraordinary intelligence or the ability to perform elaborate calculations, in fact it simply means that agents are rational and, possibly through trial and error, behave in a logically
consistent fashion.1 The final product of the chapter will be a consumption function and an investment function, two key building blocks of macroeconomic analysis.
8.2 Consumption
Households receive income from their labour or their asset holdings and have to decide what to do with it. The decision to consume is a decision not to save, and saving is a decision to postpone consumption. It is fundamentally intertemporal: now or later, which is better? Microeconomics
focuses on how households decide what to consume, e.g. apples or oranges. For macroeconomics, the emphasis is on when to consume. For this reason, we make the simplifying assumption that there is only one good to consume (Robinson Crusoe’s coconuts) and the focus is the choice
between now and later.
8.2.1 Optimal Consumption

Contemplating what do to with the coconuts that he finds on the beach, Crusoe realizes that he need not consume them on the spot. With access to a capital market, he can borrow or lend: this is called intertemporal trade. In Chapter 6, we saw that the intertemporal budget constraint allows
him to choose from many different combinations of consumption today and consumption tomorrow (Fig. 6.1), but were silent about the choice he ultimately made. His optimal choice will depend on his tastes or preferences, which are summarized in Figure 8.1 using indifference curves.2
Each curve corresponds to a given level of utility, or well-being. A particular indifference curve represents combinations of consumption today and consumption tomorrow that leave Crusoe equally happy, or indifferent. Higher indifference curves correspond to higher levels of utility.
Two central aspects of indifference curves are their slope and their curvature. The slope of an indifference curve shows Crusoe’s willingness to swap consumption tomorrow for consumption today, holding utility constant. Where the curve is steep, for example, he is willing to give up a
lot of future consumption to increase today’s consumption. A relatively flat segment indicates reluctance to give up consumption tomorrow for consumption today.
The second aspect of indifference curves, curvature, shows how the willingness to substitute depends on the relative abundance of consumption in the two periods. The more abundant consumption tomorrow is relative to consumption today, so the greater the willingness to swap
tomorrow’s for today’s. Moving along an indifference curve upwards and to the left, Crusoe is less and less willing to give up coconuts today as the consumption of coconuts tomorrow relative to today grows larger and larger. Box 8.1 provides more details on the phenomenon of
intertemporal substitution.
Naturally, Crusoe wants to consume as much as possible in both periods, but he is limited by his intertemporal budget, which was derived in Chapter 6. It is represented as the straight line in Figure 8.2. The best that he can do is point R, where the highest possible indifference curve just
touches the budget line. A more desirable indifference curve, such as IC3, is beyond his means as it lies above his budget line. He can afford the utility level corresponding to IC1 because this curve cuts the budget line, but can do better. The highest utility level Crusoe can reach is associated
with IC2, which is tangent to his budget line. Box 8.1 explains why other choices available to Crusoe are inferior.
When Crusoe is on his budget line, he spends his total wealth (OB) in the course of the two periods:
(8.1)
.
If he can borrow or lend as much as he wants at the going interest rate, his consumption pattern over time depends on the present value of his income—his budget constraint—and not on the particular timing of his income. In Figure 8.2, a ‘student Crusoe’ (with endowment M) borrows
because his current income Y1 is low relative to his future income Y2, while a ‘professional athlete Crusoe’ (endowment point A) with high current and low future income will save. Since both individuals lie on the same budget line, they have the same wealth OB. If both have identical tastes,
as described by their indifference curves, saving and borrowing allows them to have identical consumption patterns. Different borrowing and lending patterns reflect the differences in endowments. Access to financial markets makes this possible.
Box 8.1 Indifference Curves, Intertemporal Substitution, and Optimal Consumption
Each indifference curve represents combinations of consumption today and tomorrow for which utility or satisfaction is held constant. Moving to the right along a given indifference curve, today’s consumption increases while tomorrow’s declines. At any particular point, the slope of the
indifference curve shows how many units of goods tomorrow we are willing to give up for an additional unit of goods today. The technical term for this willingness to trade goods tomorrow for goods today is the marginal rate of intertemporal substitution.
In most cases, the willingness to substitute consumption across time changes when moving along a representative indifference curve. As we move left to right, it becomes flatter and more horizontal, because we are willing to give up less and less consumption tomorrow for
consumption today. The opposite occurs as we move from right to left. The curvature of the indifference curve captures the sensitivity of the marginal rate of intertemporal substitution to relative consumption. In Figure 8.1, Panel (a) describes the normal situation, but it is worth thinking
about two opposite extremes. At one end of the spectrum, there can be no substitutability at all. The consumer is better off only if consumption is increased in fixed proportion in both periods. The indifference curves would look like a series of right angles in Panel (b) of Figure 8.1;
happiness can only be increased by raising consumption in both periods. At the other extreme, the marginal rate of substitution is constant. The consumer is always willing to substitute the same amount of consumption today for consumption tomorrow, regardless of how much is
consumed. Then indifference curves are straight lines, as in Panel (c).
Consider Figure 8.2, which characterizes optimal consumption. The key feature of point R in Figure 8.2 is that the marginal rate of substitution is just equal to the slope of the budget line, or 1 + r. To see why, suppose that Crusoe obeys his budget constraint but has a marginal rate of
substitution of 1. He is willing to exchange one coconut today for one tomorrow. By lending 1 today instead, he gets (1 +r) tomorrow—a good deal for him—and can make himself better off, i.e. move towards R. If he goes too far, the marginal rate of substitution will exceed (1 +r). He
would then prefer to shift consumption back to the present, increasing his utility by doing so. Only when the marginal rate of substitution is equal to the intertemporal price of consumption has Crusoe exhausted all gains from intertemporal trade.
Fig. 8.1 Indifference Curves
Along any particular indifference curve, utility is constant. In Panel (a), consumption tomorrow can be substituted smoothly for consumption today, but at a rate which decreases as today’s consumption increases. In (b), the consumer can be made better off only by increasing consumption today and
tomorrow in fixed proportions. In (c), consumption today and consumption tomorrow are always substituted at the same rate. In all cases, indifference curves lying above and to the right correspond to higher utility levels.
8.2.2 Implications
The principles developed here have wide-ranging and important implications both at the individual level and for a country as a whole. We examine some of these implications.
Permanent versus temporary changes in income
A first implication is that we can understand how the consumption of households reacts to changing economic circumstances. We will take three examples.
Case 1: a temporary increase in income
Imagine that, today, Crusoe’s harvest is unusually plentiful, rising to Y′1 in Figure 8.3, while next period’s harvest Y2 is expected to remain unchanged. For simplicity, the figure represents the case where, initially, consumption was exactly matching income in both periods so that there was
no need to borrow or lend (points A and R overlap). The endowment point now shifts from A to A′, on the new budget line B′D′, which is parallel to the initial line BD since the real interest rate remains unchanged. It is natural to expect that Crusoe will consume more. However, the key
insight is that his consumption (point R′) will rise in both periods. Consumption today increases less than the windfall, since part of it is saved and spread over time. A temporary increase in income is accompanied by a permanent, but smaller, increase in consumption.
Case 2: a permanent increase in income now and in the future
What if, instead, the increase in income is permanent, in the sense that both Y1 and Y2 rise by equal amounts? (Think of a lasting improvement in the harvest outlook.) The new endowment point is A″ and the corresponding budget line is B″D″. Optimal consumption moves to point R″. As
a first approximation, points A″ and R″ coincide and consumption rises in both periods.3 Being equally better off in both periods, Crusoe sees no reason to save or borrow. A permanent increase in income is absorbed in a permanent increase in consumption of similar size.
Fig. 8.2 Optimal Consumption

The budget line shows how much can be consumed today and tomorrow for given endowment (represented by point M) and real interest rate (the slope). Optimal consumption is achieved at point R. In this case the consumer borrows C1 − Y1 today and repays Y2 − C2 tomorrow. Consumption at R is
also possible for an individual with endowment A, who lends today and dissaves tomorrow
Case 3: an expected future increase in income

Finally, consider the case when income is unchanged today, but is correctly expected to increase tomorrow. If Robinson knows that his future crop will be more plentiful, he will borrow today against his future income to afford a better standard of living right away. Far from reckless
behaviour, borrowing makes him better off. In a setting with many periods, forward-looking consumers would try to assess the permanent component of current shocks to better assess the future, and change current behaviour accordingly.
Fig. 8.3 Temporary and Permanent Income Changes
The shift from A to A′ describes a temporary increase in income. Consumption rises both today and tomorrow (the household moves from R to R′). Part of today’s income windfall is saved to sustain higher spending tomorrow. The shift to point A″ represents a permanent increase in income. It does
not require consumption smoothing through saving or borrowing. The best course of action is to increase consumption permanently (to point R″).
An implication of this reasoning is that only new information about the future should alter existing consumption behaviour. If future incomes are correctly anticipated, they are incorporated into current wealth and current consumption fully reflects this information. The only reason why
consumption should change is if unexpected disturbances affect current or future income significantly enough that wealth changes. Since all that is known of the future is already taken into account in the evaluation of wealth, only true surprises can alter wealth and therefore consumption. Put
differently, changes in consumption must be unpredictable. This is known as the random walk theory of consumption, because changes in consumption should be random.4 This logic, previously encountered in Chapter 7, is a consequence of the rationality of expectations.
Consumption smoothing
The common theme of the three cases is that people dislike variable consumption patterns. When faced with a temporary change in income, rational consumers save or borrow to spread the effects on consumption over time. In bad times, this may take the form of dissaving (spending from
accumulated savings) or borrowing (from the bank, from relatives, or using a credit card). In good times, consumers accumulate assets or repay their debts. This phenomenon is known as consumption smoothing. It explains why consumption is less variable than GDP, and is in general
the most stable component of aggregate demand. Box 8.2 describes an example of consumption smoothing when the shock is perceived as temporary and of consumption change when the shock is perceived as permanent. This being said, it is very difficult to detect permanent from
temporary shocks; consumers (and economists) may be misled by their perceptions.
Permanent income and the life cycle
Most households do not expect a constant flow of income over their lifetimes. Typically, young people earn less than older people, as represented in Figure 8.5. The principle of optimal consumption implies that they should borrow when young and repay debts or even save when older in
order to smooth the pattern of consumption over time. It is rational for young people to borrow many times their annual income to buy an apartment or a house, which they plan to inhabit for a long time and to pay off the loan over the course of their lives. Naturally, they must assume
what they will continue to earn in the future, and no one can be sure of anything. Nevertheless, mortgage lending is a common fact of everyday life in all modern economies, usually combined with a life or other type of insurance policy.
Box 8.2 Germany and Greece During the Crisis
The financial crisis and the ensuing Eurozone crisis played out very differently in Germany and Greece. Figure 8.4(a) shows that German GDP growth declined sharply in 2009, but resumed its trend afterwards. Figure 8.4(b) shows that Greece went into a massive recession as its
real GDP declined by more than 25%. These two charts indicate that consumption fell sharply in Greece while hardly changing in Germany. Figure 8.4(c) presents the ratio of consumption to GDP in both countries. Before the crisis, this ratio was on a declining trend, widely believed to
be driven by population ageing, as Germans realized that they would have to support growing cohorts of pensioners. Faced with a temporary income shortfall in 2009, the Germans reduced their savings or, equivalently, consumed a larger share of their incomes. As a result,
consumption barely departed from its trend.
In contrast, Greek households faced a string of bad shocks, deepening a recession into a veritable depression. At each turn, Greeks cut consumption but less than GDP; the ratio of consumption to GDP, long around 75%, actually rose to 80%, clear evidence of consumption
smoothing. The interpretation is supported by evidence that the Greeks did not believe that they would go into an ever deeper depression. Each year was seen as a temporary nasty surprise justifying lower savings, and actual dissaving—running down previously saved wealth—for a
part of the population.
Fig. 8.4 Consumption Smoothing in Greece and Germany, 1995–2016
Panels (a) and (b) display GDP (as a measure of income) and consumption in Germany and Greece surrounding the sharp slump in output associated with the Great Recession beginning in 2009. Consumption in both countries more or less tracked output, with stable consumption/GDP ratios until the mid-2000s. A trend break in
Germany is observable after 2001. The initial impact of the crisis is similar in the two countries, but Germany (Panel (a)) recovers quickly and consumption is stable over the episode. In Greece, the drop in GDP becomes successively worse (Panel (b)). The perception that the boom years have come to an end seems to have
pushed consumption into a severe and lasting decline. Nevertheless, Greek households continue to shield their consumption from the ravages of the economic crisis.
Source: AMECO on line, European Commission.
Fig. 8.5 Life-Cycle Consumption

When income is expected to increase over a lifetime, consumption smoothing implies borrowing when young and paying back when older.
The principle of life-cycle consumption is illustrated in Figure 8.5. To maintain a constant flow of consumption, individuals should spend each year an amount corresponding to their permanent income.5 Permanent income is that income which, if constant, would deliver the same
present value of income as the actual expected income path. It is a good measure of sustainable consumption over one’s lifetime. In the two-period thinking of this chapter, permanent income Y P would therefore be defined as:
(8.2)
.
The life-cycle principle can also be applied to a country as a whole. If income is expected to grow, it is optimal for a country to borrow abroad, and thus run current account deficits for a while, and pay back later through current account surpluses. This may explain why the new EU
members, who can reasonably expect to catch up with the old members, often run current account deficits. One puzzle which merits much attention is China, as Box 8.3 points out.
Consumption and the real interest rate
When the real interest rate rises, so do the rewards of saving. Why? Because the price of consumption tomorrow relative to consumption today declines. Does saving always increase and consumption always decline? The question is harder to answer than it first appears. Figure 8.6 shows
that one determinant of consumption today is whether Crusoe is a net borrower (e.g. a student) or a net lender (e.g. a professional athlete). In both panels of the figure, endowments today and tomorrow are held constant, so the budget line rotates around point A. Optimal consumption shifts
from point R to point R′. Net lenders (Panel (b)) gain from higher interest rates, moving to a higher indifference curve and consuming more in both periods. The borrower (Panel (a)) faces higher interest costs and is in the opposite situation, as he must devote more resources to servicing the
debt, and his current consumption declines. Increases in interest rates have important redistributive effects between borrowers and lenders: an increase hurts the former and benefits the latter.
Box 8.3 Does China Defy the Permanent Income Hypothesis?
In the late 1980s, China was a poor country with a GDP per capita about 4% of that of the Netherlands—and a population 70 times larger. By 2013, China had grown to 26% of the Netherlands’ size. At the current rate of growth, China is likely to reach Portuguese levels of GDP per
capita in 15–20 years, and few doubt that China will continue to close the gap. The permanent income hypothesis suggests that China should run a substantial current account deficit during this catch-up period. In fact, its current account has been steadily increasing, reaching 10% of
GDP.
While there are many explanations of the high Chinese saving rate, none of them alone is really convincing. One view is that Chinese citizens cannot borrow against future income because the banking system is rather underdeveloped. Another is that citizens save so much because
there is no system of social security as it is known in Europe, so they need to save for health expenditures and retirement income. Education is not free, so parents often save for the education of children, which in a country with strong families might also be important for retirement.
State-owned firms, which are highly profitable because they are sheltered from competition, are also big savers. Yet another reason is fear that borrowing on international capital markets could bring even greater problems, including loss of government control over the economy that
comes with international investment. Most frequently cited is the view that China intentionally keeps its currency undervalued to artificially boost competitiveness, and by doing so makes imports of cheap consumptions goods unattractive. (The precise meaning of undervaluation has
been hinted at in Chapter 5 and will be explained more fully in Chapter 15.) At any rate, each of these reasons implies that Chinese households are not on their optimal intertemporal consumption paths and could easily increase their well-being with a few well-placed reforms.
Fig. 8.6 The Effect of an Increase in the Interest Rate

As the interest rate increases, the budget line becomes steeper and rotates about the endowment point A. The response of the consumer depends on whether she is a borrower or a lender. The borrower (a) will tend to consume less today because the interest rate at which resources are brought
forward has increased. The lender (b) consumes more today, since the same amount of lending can increase the amount of consumption possible tomorrow without reducing today’s.
In general, the effect of the interest rate on consumption is ambiguous because it works through two channels. First, an increase in the interest rate makes the budget constraint steeper, since it determines the slope of the intertemporal budget constraint. It increases the cost of goods today
relative to those tomorrow, making it more attractive to save (consume tomorrow). Second, it reduces the value of wealth Ω, which is the present discounted value of all income. This effect will depend on how much of our wealth stems from future, as opposed to current, income. The more
wealth we have today, the more we profit from an increase in interest rates and the more likely this wealth effect will predominate.
8.2.3 Wealth or Income?

An old tradition in macroeconomics, which can be traced back to John Maynard Keynes, links consumption spending by households to current income—which we have called disposable income Y d and defined as GDP less net taxes. Keynes argued that most people simply set aside a
fraction of disposable income for saving, and consume the rest.6 The evidence seems to support this hypothesis. Figure 8.7 shows the evidence for France, plotting consumption expenditures in each year in the period 1980–2006 against disposable income in the bottom chart and, in the top
chart, against a measure of household wealth, which includes liquid financial assets of families as well as the value of fixed assets and real estate. The link between consumption and disposable income is strong, in fact stronger than that between consumption and wealth.
Fig. 8.7 Consumption, Disposable Income, and Wealth in France, 1978–2014
The link between real consumption and net wealth of households is manifest (Panel (a)), but appears less tight than the link between consumption and disposable income (Panel (b)).
This evidence challenges a key implication of the theory developed in the previous sections: that consumption is driven by wealth, not current income, and that households strive to smooth their consumption relative to income. One possible reconciliation is that income and wealth grow in
tandem, so that the observed consumption–income relationship may reflect a common dependence on wealth. Yet wealth appears more volatile than disposable income, partly because of fluctuations in share prices on stock markets already discussed extensively in Chapter 7. It is likely that
households also regard short-term stock market gains and losses as temporary, paying attention only to long-term increases in wealth. In addition, private wealth is difficult to measure precisely, in part because people are reluctant to provide information about their assets, in part because
expected future income—an important component of total wealth—is not measurable and therefore left out.
A second explanation is related to a household’s ability to borrow and lend. In our parable, Crusoe can borrow freely at a given interest rate. This might be the case if present and future incomes of individual households—against which borrowing is pledged—were known with certainty to
lenders. In real life, banks and other lending intermediaries cannot know the repayment prospects of all individual borrowers with certainty. A common banking practice is to demand collateral—the borrower pledges tangible wealth, such as a house, in case of non-payment. This option is not
available to all households. Banks charge higher interest rates to customers who appear riskier and sometimes refuse to lend at any rate, as explained in Chapter 7, or place ceilings on the amount that can be borrowed. Consumers who cannot obtain credit in spite of future earnings potential
are said to be credit rationed.
Fig. 8.8 Credit Constraints
If Crusoe cannot borrow, his budget constraint shrinks from CD to CA. He would like to be at point R, however, borrowing today and paying back tomorrow. The best outcome for him under the circumstances is to consume at point A, with consumption equal to income both today and tomorrow.
In the presence of credit rationing, spending is governed by current disposable income, not wealth.7 This is shown in Figure 8.8, which uses Figure 6.11 as its point of departure. Because Robinson Crusoe is prevented from borrowing, his consumption possibilities are limited to the kinked
line CAB. In particular, he cannot reach the segment AD of his intertemporal budget constraint, and his preferred consumption plan R is not possible. In that case, the best option for him is point A, where he consumes exactly his income in both periods. If a significant proportion of
households is rationed in credit markets, disposable income will be the dominating influence on consumption, whereas it will not matter at all for non-rationed households. Box 8.4 illustrates the importance of national borrowing constraints in the early phase of the process of economic
transformation in Eastern Europe. But even in advanced countries, credit rationing affects a substantial proportion of households. It is therefore not surprising to observe the tight link between consumption and income in Figure 8.7.
Box 8.4 Current Income and Spending in East Germany and Poland
The rapid conversion of East Germany (the former German Democratic Republic) and Poland to market-based economies in 1990 provides a unique example of an anticipated increase in permanent income. In both countries, the adoption of market-based institutions implied that income
levels would eventually reach those of Western Europe. However, the transition to a market economy was bound to be painful. Some dislocation was inevitable. Income initially fell as inefficient production was shut down and workers had to change occupations and industries. Yet,
remember that wealth is defined in (8.1) as the present value of current and future incomes. This is why, despite the fall in initial observable income, wealth increased because future incomes were bound to be much higher than before. The permanent income hypothesis implies that
consumption had to rise ahead of income. This could only happen if people are able to collectively borrow abroad, or receive transfers. As part of German unification, the citizens of the Eastern Länder had access to a well-developed domestic financial market. For the former East
Germany borrowing ‘abroad’ meant getting loans from West German banks, or receiving credits or grants from the government. On the other hand, Poland started out with a large external debt, which made it difficult for the government to borrow more, and its citizens and firms certainly
did not have access to credit, despite high growth. Figure 8.9 shows the dramatic difference. While GDP fell in East Germany after 1990, private consumption there rose to equal total East German GDP. Public spending and private investment also rose. As a result, the current account
deficit—the difference between total spending and income—reached nearly 100% of GDP. In credit-constrained Poland, spending tracked income. In fact, because of its large external debt, Poland had to run a primary surplus. In contrast, East Germany’s large external debt was
assumed by West Germany.
Fig. 8.9 GDP, Domestic Demand, and the Current Account: Poland and East Germany
Spending in East Germany rose after unification as firms, citizens, and authorities were able to borrow against higher expected future income. There is no link between spending and income, because ‘foreign’ borrowing is almost as large as income. In Poland, which has similar long-run growth
prospects, spending follows income because of the impossibility of borrowing large amounts abroad.
Sources: DIW Wochenbericht; World Bank; CSO; DGII.
8.2.4 The Consumption Function

How can we summarize all we have learned thus far? Consumption is driven primarily by wealth, and wealth is based on current and future incomes of households. By this argument, current income should matter less than future expected income. At the same time, current income may be a
good predictor of what is to come. In addition, many people are credit-rationed and cannot borrow even when their expected future income is higher. For them, income is the effective determinant of consumption.
In the end, we can write down the consumption function as a compact way of linking aggregate consumption in the economy to its two main determinants:
(8.3)
.
The plus signs underneath the arguments of the function remind us that consumption increases with both wealth Ω and personal disposable income Y d, that is that part of GDP that reaches households, subtracting corporate savings and taxes net of transfers. The consumption function is a
fundamental tool that will be used throughout the rest of this textbook.
What about the interest rate? Section 8.2.2 noted that the real interest rate affects consumption, but that its direct role is ambiguous; this is why we do not include it in (8.3). On the other hand, a higher interest rate is bound to reduce consumption indirectly through wealth. So the real
interest rate affects consumption after all, but through its effect on wealth.
8.3 Investment
The second component of aggregate demand to be explained is investment, sometimes called gross domestic capital formation in the national income and product accounts. Investment goods include machine tools, computers, office furniture, land-moving equipment, buses and lorries, and
construction of new factory buildings, as well as increases in inventories of goods to be sold at a future date. Software programs used in computers that are purchased by businesses are also treated as investment spending. All these goods have the common trait that they are not intended for
consumption; instead they make the production of goods and services possible in the future. The decision to invest is therefore an intertemporal decision.
8.3.1 The Optimal Capital Stock

As in Chapter 6, we will continue to assume that labour input is constant. Accordingly, we set L = 1 so it is possible to write the production function using the shorthand F(K) for F(K,1). This represents the amount of output that can be produced by a representative firm, giving output Y
available tomorrow when capital K is in place. The production function is depicted in panel (a) of Figure 8.10. A related concept, already introduced in Chapter 3, is the marginal productivity of capital (MPK), and is plotted in Panel (b). This is the extra output produced when an
additional unit of capital is installed (ΔY/ΔK). It is the same thing as the slope of the production function.8 Because of the principle of declining marginal productivity, the MPK declines as more and more capital is put in place.
The MPK is the return from an additional unit of capital. What about the cost? When he saves, Crusoe can choose between planting coconuts himself, or lending them in the capital market. In the latter case, he can expect to receive the coconut plus interest tomorrow. Alternatively, he
could borrow a coconut that he doesn’t own, plant it himself, grow the tree, and pay back principal and interest tomorrow. The same is true for any firm. If the investment is funded with resources that could instead be invested in financial assets, the opportunity cost of the investment is
(1 + r). If the investment is financed by borrowing, the marginal cost of capital is (1 + r).9 In both cases, the cost is the same: it is shown in Panel (a) of Figure 8.10 as the ray OR. The ray represents the total cost, (1 + r) K, of capital installed today and productive tomorrow, the sum of
the principal and the interest charged. (The cost of equipment here is unity because it takes one coconut to start a tree.) The marginal cost of capital, or the cost of one incremental unit of productive capacity, is simply (1 + r). It is represented in Panel (b) by a horizontal line.
The firm’s profit in the second period is the difference between what it produces and the cost of production:
(8.4)
.
In Panel (a) of Figure 8.10, this is measured as the vertical distance between the curve depicting the production function and the ray OR. To maximize profit, the firm chooses the optimal capital stock such that the distance between the two schedules is as large as possible. This occurs
where the slope of the production schedule (given by its tangent) is equal to the slope of the cost-of-capital schedule OR. Then the marginal productivity of capital (MPK) is equal to the marginal cost, here the opportunity cost 1 + r:
Fig. 8.10 The Optimal Capital Stock

The optimal stock of capital is reached when the firm’s production function is farthest vertically from the line OR, which represents the cost of capital. There the marginal productivity of capital is equal to its marginal cost (MPK = 1 + r). Investment is the difference between the desired capital stock
and the previously accumulated capital stock K.
(8.5)
.
In Figure 8.10(b), the optimal capital stock corresponds to the intersection of the MPK and marginal cost curves. The key threshold for investment is that the marginal productivity exceeds the opportunity costs, 1 + r.
In the real world, capital investments typically last more than two periods—sometimes many decades. Capital goods can be resold to third parties. It is important to capture this aspect of investment and capital, and is relatively easy to do so within the story of Robinson Crusoe. Box 8.5
shows that while the logic is unchanged, the threshold for assessing the profitability of investment is considerably lower when capital does not depreciate completely and can be resold after the second period.
If firms behave optimally on average, the same principles can be applied to the economy as a whole. Two conclusions follow. First, the optimal capital stock depends positively on the expected effectiveness of the available technology, captured by the marginal productivity of capital. An
improvement in technology means that more output can be produced with the same capital stock. In Figure 8.11, the production function in panel (a) and the MPK schedule in panel (b) both shift upward. The optimal stock of capital increases from to ′.
Second, the optimal capital stock depends negatively on the real interest rate. If the interest rate increases, the cost schedule OR rotates counter-clockwise in Figure 8.11(a) and the marginal cost schedule shifts upwards in Panel (b). The intuition behind this important result is that, for a
given state of technology, higher opportunity costs of capital reduce the amount of capital that can be optimally employed and still be more profitable than simply ‘lending’ the resources in the financial markets.
Box 8.5 Optimal Capital Investment: Looking Beyond Two Periods
Suppose that Crusoe, instead of abandoning his capital stock, sells it to Friday, who chooses to stay on the island and continue his economic pursuits there. A ‘resale market’ changes the investment decision for Crusoe in an important way. As before, the optimal decision means that
MPK equals marginal cost, but now the MPK includes the resale value of capital. For simplicity, suppose first that the price of trees is equal to one (coconut). Then the optimal condition is:
(8.6)
.
The cost of capital is 1, the same as the value of output by assumption (1 coconut). This condition can be simplified to
(8.7) .
When Crusoe can recoup the principal of his investment, the marginal product of capital is set equal to the real interest rate r, rather than 1 + r. In practice, firms can resell at least some of their equipment at some price, so an investment can be justified by a lower MPK than would be the
case were it to be abandoned entirely. Reality will tend to be somewhere between equations (8.7) and (8.5), depending on depreciation, or loss of value of the installed capital over time.
This loss of productive equipment can be quantified by a rate of depreciation δ, which we have already seen in our discussion of economic growth in Chapter 3 and of intertemporal resource constraints in Chapter 6. With depreciation, tomorrow’s value of a unit of today’s capital
is no longer 1, but 1 − δ. Taking this into account, (8.7) becomes:
(8.8)
.
which simplifies to:
(8.9)
.
The optimal capital stock is reached when the marginal product of capital is equal to the sum of the interest rate r and the depreciation rate δ. Because depreciation can be thought of as an additional cost of capital, the right-hand side is often called the user cost of capital. The original
rule for the optimal capital stock (8.5) can be regarded as a special case of complete depreciation (δ = 1).
8.3.2 Investment and the Real Interest Rate

Investment occurs for two reasons: (1) to bring the capital stock to its desired level, and (2) to make up for capital lost through physical or economic depreciation.10 In Figure 8.10, we can find the optimal stock of capital and the stock of capital inherited from past investment K —
perhaps there were already some coconut trees around when Crusoe came to his island. Ignoring depreciation, optimal investment is simply the difference - K. Thus, given the present stock of capital and the rate of depreciation, the determinants of optimal investment are the same as those
of the optimal stock of capital. An increase in the real interest rate, which lowers the optimal stock of capital, also lowers optimal investment since the capital stock brought forward from the last period and the rate of depreciation remain unchanged. Accordingly, the aggregate investment
function could already be formulated as:
(8.10)
.
where the minus sign ‘−’ indicates that higher real interest rates depress investment.
Fig. 8.11 Technological Progress

Technological progress makes more output possible with the same stock of capital. In Panel (a) the production schedule shifts upward. In Panel (b) the MPK schedule moves up to the right to MPK′. The optimal stock of capital is now ′, which is larger than the initial value
8.3.3 The Accelerator Principle

In Chapter 3, we stressed the relative stability of the capital–output ratio as one of Kaldor’s stylized facts of economic growth. It is reasonable to expect that, for the capital stock to reach its optimal level derived in Section 8.3.1, investment would have to move in roughly the same
proportion. This idea gives rise to a simple way of thinking about investment. Suppose the optimal capital stock in the second period is proportional to the expected output level: K = νY, where v is a constant.11 If firms invest to keep the capital stock at its optimal level, an increase of
GDP from Y1 to Y2 requires a change from K1 = νY1 to K2 = νY2. Ignoring depreciation, this means an investment of:
(8.11)
.
This relationship captures the accelerator principle. It is called the accelerator because in order for investment to remain constant, output must continually increase. Increases in investment are therefore associated with an acceleration of output. In practice, the capital–output ratio is
between 2 and 3 in most economies. Put differently, annual GDP represents between one-third and one-half of the installed capital stock. GDP movements therefore are associated with much larger swings in investment.12 A second, central insight of the accelerator principle is the
dependence of investment today on expected growth in output tomorrow. This provides a convincing reason why investment is more volatile than GDP. It is based on expectations of the future. In the following sections we will develop this idea in greater detail.
8.3.4 Investment and Tobin’s q

Looking at stock markets, Chapter 7 noted that they continuously look at future returns. Stock prices, it was shown, are the present discounted value of future profits of firms. As titles of ownership, they represent claims on these profits. Profits are the difference between sales and costs.
In the aggregate economy, these costs are primarily wages and non-wage labour costs. When the markets are efficient, share prices can be thought of as the market’s best estimate of the value of those present and future profits.
The stock valuation of a firm is the price at which it can be bought. It is the price to be paid for acquiring the firm, its buildings and equipment, i.e. its capital stock. Arbitrage suggests that the stock market valuation of a firm and its installed capital should be equal. But does it also imply
that the value of installed capital is equal to the cost of purchasing the underlying capital goods and putting them together reproducing that firm from scratch? It turns out that, for a number of reasons, the market valuation of a firm need not equal the replacement cost of the capital stock.
One such reason is the existence of intangible assets which include such factors as the firm’s know-how and its trained labour force, its network of distributors and retailers, its reputation among customers, etc. These factors require time to establish. Another reason is the fact that ‘Rome
wasn’t built in a day’. Establishing a new firm from scratch requires time and resources. These costs are greater, the more rapidly an investment project is undertaken. Sometimes, the time required to invest in projects cannot be reduced no matter how many additional resources are brought to
bear on the undertaking.
To summarize these tensions between market’s assessment or valuation and the replacement cost of capital, we use a ratio, called Tobin’s q, which is defined simply as follows:13
(8.12)
.
The numerator of Tobin’s q is the firm’s value as priced by the stock market, the total value of all existing shares. The denominator is the amount that would have to be spent to replace the capital goods incorporated in existing firms.
The q -theory of investment relates the behaviour of aggregate investment to Tobin’s q. When Tobin’s q is greater than one, installed capital in existing firms is more valuable than what it would cost to purchase the necessary investment goods to start a new firm from scratch. For this
reason, entrepreneurs (people who start and run businesses) take this as a signal to purchase new plant and equipment—either in new market entrants or in established enterprises. Investment is then positive. For example, a Tobin’s q of 1.2 would imply that a firm that spends 100 on new
investment in plant and equipment could increase its market value by 120 at the margin. Installation and use in production adds a value of 20 to uninstalled equipment. Given the principle of declining marginal productivity, investment reduces the return on the capital over time and therefore
reduces Tobin’s q.
The positive relationship between investment and Tobin’s q is depicted in Figure 8.12. As long as Tobin’s q is greater than one, firms will continue to invest, increasing the capital stock and reducing the marginal product of capital, until q has returned to unity. Alternatively, when q is
lower than 1, net investment is negative—firms will invest less than depreciation, so the capital stock will shrink. If possible, selling off existing, undepreciated equipment at replacement cost is profitable and therefore desirable from the point of view of the firm’s shareholders. In the
aggregate economy, this can occur only if gross investment is exceeded by depreciation or capital is dismantled, sold, or scrapped.
Fig. 8.12 The q-Theory of Investment
When Tobin’s q is larger than 1, it pays for firms to invest. When Tobin’s q is less than 1, there is no incentive to invest, but rather an incentive to disinvest, or to dismantle or abandon productive capacity.
How is the q-theory related to the previous section, which showed that the interest rate is a key determinant of investment? The stock market values firms by discounting future earnings using the real interest rate. As explained in Chapter 7, any increase in the interest rate leads to heavier
discounting and therefore to a decline in stock prices, ceteris paribus. Thus, the negative effect of the real interest rate on investment is already incorporated into Tobin’s q.
But Tobin’s q does more than just take the interest rate into account. It also incorporates two other factors in the investment decision. First, gains in productivity of capital raise future income, and thereby increase share prices and q. An excellent example of how this can occur was the
massive rise (and later, decline) in stock prices in the wake of the internet and new communications technologies discussed in Chapter 7. Second, q incorporates the role of expectations. Investment is almost always a bet on the future. Firms buy and install equipment now to produce output
for several years under uncertain conditions. How they will be able to take advantage of the equipment is not known when the investment occurs. Uncertainty ranges from the general economic situation, to competition in domestic and foreign markets, to the evolution of technology and even
political developments. All these aspects are continuously evaluated by the stock markets, which explains why share prices are so volatile: forward-looking markets factor into share prices volatile expectations of many volatile underlying factors. In the end, this explains why investment is
the most variable component of GDP. It was Keynes who linked the behaviour of investment to the animal spirits of entrepreneurs, i.e. their fickle and volatile expectations of the future profitability of investment. To see how well Tobin’s q predicts investment expenditures, let us
consider Figure 8.13, which presents data for Germany for the period 1980–2010 as well from a much earlier period, the late 1920s and early 1930s, which included the Great Depression. The link is evident, with stock prices leading investment with a variable but relatively short lag. Not all
companies are traded on stock markets. Many are too small to issue shares, and many larger ones are reluctant to ‘go public’. It is often cheaper for firms to draw on their own savings (retained earnings) because profits are usually taxed less when reinvested than when distributed. In
addition, some firms prefer to finance investment by borrowing rather than by issuing shares, especially in continental Europe, where bank lending plays an important or even dominant role. Despite these limitations, Tobin’s q still does a pretty good job measuring the incentive to invest for
these firms. It reflects both expected profitability—the numerator—and the real cost of borrowing through the discount factor and the cost of capital—the denominator.
Tobin’s q explains the link between the stock market and the state of the economy. As stressed in Chapter 7, the economic function of stock exchanges is to evaluate the future profitability of the large number of listed firms and to place a value today on shares, which represent ownership
rights to a stream of future earnings. Present and especially future economic conditions affect stock prices. Conversely, we should expect stock markets to affect economic conditions since stock prices influence investment through Tobin’s q.
8.3.5 The Microeconomic Foundations of Tobin’s q

This section is a more advanced presentation of the q-theory of investment based on installation costs. It is similar to the reasoning used to establish the optimal stock of capital and can be skipped without any loss of continuity. The WebAppendix to this chapter presents a formal analysis.
Installation costs
We now have two ways of thinking about the firm’s investment decision. The first is that the firm invests to reach its optimal capital stock. The second approach sees investment as taking advantage of the difference between the market value of capital already in place (installed) and its
replacement cost—the cost of new capital goods. As we note below, when Tobin’s q is unity, the two approaches are equivalent. From either perspective, one might expect firms to seize such opportunities quickly. In practice, however, they do not adjust the capital stock instantaneously
to its optimal level. For this reason, q can and will differ systematically from its long-run value—unity—for some time. One reason for this is that firms face installation costs in addition to the direct costs of capital considered so far. It turns out that taking these installation costs into
account allows us to integrate the previous theories.
Fig. 8.13 Investment and Tobin’s q, Weimar and Modern Germany
Both in historical and more modern contexts, investment has consistently followed movements of Tobin’s q, as proxied by a stock price index divided by the consumer price index.
Sources: IMF; Ritschl (2004).
The idea behind installation costs is straightforward. With adequate resources, it could have been possible to dig the Eurotunnel—or build the Berlin airport—in just six months. Doing so, however, would have been enormously ‘costly’ in many ways, so these investment projects took
several years instead. Intuitively, the bigger the investment per unit of time, the more costly it is to install it. Each addition of new equipment in a factory requires workers and resources to install it, disrupts existing production, and requires the training of workers to operate it.
Installation costs can explain why Tobin’s q is not always equal to unity. Recall that in the absence of installation costs, firms would always set capital input to its optimal level, equating marginal productivity and marginal cost as in equation (8.5), rewritten as:
(8.13)
.
MPK/(1 + r) is the present value of the return on investment. It is next period’s return on the latest addition of capital discounted back to today.14 It must be equal to the marginal cost of equipment, which is unity. When installation is costly, however, the cost of investing is not just the
price of equipment. It now includes an additional cost, the marginal cost of installing new equipment, These installation costs are denoted by φ. It is assumed that φ is an increasing function of the volume of investment undertaken; the faster you expand the capital stock, the more expensive
it is.15 The optimal investment decision is to invest until the present value of the MPK of new equipment is equal to the augmented marginal cost of equipment:
(8.14)
.
Comparing (8.13) and (8.14), we see that installation costs raise the MPK required to justify the investment in each period. This means that less capital is invested initially. Marginal productivity will remain higher after the first period, since MPK is decreasing in capital; as a result, the firm
takes smaller steps to reach the desired long-run capital stock . Since installation costs increase with the amount of investment each period, the next round of installation is cheaper, φ is lower, so firms will engage in more investment, and so on until φ is driven down to zero and (8.13)
holds. That way firms break the path towards into small steps that entail lower adjustment costs.
Installation costs and Tobin’s q

In Robinson Crusoe’s world, the stock market should value the return on an additional unit of investment by the present value of its marginal return, MPK/(1 + r). The replacement cost of capital is simply the cost of an additional coconut (uninstalled equipment) which is 1. Tobin’s q is
therefore:16
(8.15)
.
Equation (8.15) establishes the link between the two investment principles. The optimal capital stock of capital is reached when (8.13) is satisfied, that is when Tobin’s q is equal to 1. When q is above 1, the MPK exceeds 1 + r, and investment is warranted. As new capital is put in place,
the MPK declines, as does q. Investment becomes smaller and installation costs φ decline until they become negligible. As q approaches 1, MPK approaches 1 + r and the stock of capital approaches its optimal level. Installation costs cause firms to move towards the optimal capital stock
incrementally. Along the way the return on investment in present value terms exceeds the replacement, or user, cost of capital.17 Box 8.6 provides more details on both the user cost of capital and its relationship to Tobin’s q.
Box 8.6 User Cost of Capital, Tobin’s q, and the Price of Investment Goods
In Robinson Crusoe’s world, coconuts were used for both consumption and investment. In the real world, investment and consumption goods have different prices. This complicates slightly, but does not invalidate, the main line of reasoning. Let us suppose that a coconut in the ground
(investment) is identical to a coconut in the mouth (consumption), but in the second period, the undepreciated part of the tree can be sold for pK in terms of the consumption good. Now, the profit of the firm in units of the consumption good is:
(8.16)
.
The optimal stock of capital will obey a modified version of (8.9) in Box 8.5:
(8.17)
.
where πK = pK − 1 is the rate of price increase for investment goods.18 The user cost concept of Box 8.5 allows for a changing relative price of investment goods over time. An increase in the real interest rate and depreciation lowers optimal investment, all things being equal; they
raise the user cost of capital. An increase in the rate of change of investment goods prices increases investment. Why? By selling his investment, Crusoe reduces the effective cost of using the capital goods in production. It is easy to see that any government policy that increases the
resale value of the investment, including tax breaks, will similarly reduce the user cost of capital.
Similar to equation (8.12), a reasonable approximation of Tobin’s q is:
(8.18)
.
The only difference is that the extended version of Tobin’s q now compares the present discounted value of the marginal product of capital with the value of new capital goods, the price of investment goods, which is equal to 1 in the first period.
Finally, if installation costs φ must be borne for a marginal unit of investment, optimal investment must obey:
(8.19)
.
At any point in time, the marginal gain from an additional unit of capital equals the marginal costs of that new capital, which includes the installation costs of investment.
The geometry of installation costs

Installation costs have two particular properties. First, they increase with the size of the investment. Big steps are proportionally more expensive than small ones. Second, they are transitory. Once the equipment is in place, the only relevant cost is the interest rate and depreciation—the
opportunity cost of resources employed in production. Figure 8.14 modifies Panel (b) of Figure 8.10 in two ways. First, investment is measured on the horizontal axis as the investment rate I/K, which gives a better indication of the intensity of disruptions which give rise to installation
costs.19 Second, on the vertical axis, marginal costs and returns are expressed in today’s present discounted values. The marginal cost of capital in present value is equal to 1, because one unit of capital implies giving up one unit of consumption good today.
The horizontal schedule represents the marginal cost of investment in the absence of installation costs. With installation costs, the cost of investment will exceed the cost of capital. The more equipment is put in place per unit time, the higher is the marginal cost. Hence the upward-sloping
marginal cost of investment curve. The marginal return on investment is MPK/(1 + r). It is downward-sloping because of the principle of declining marginal productivity.
Fig. 8.14 Tobin’s q

In Panel (a) profits are maximized at point A, where the marginal return of investment in present-value terms is equal to its marginal cost. The marginal cost of capital is 1 (unit of forgone consumption). The optimum rate of investment is (I/K)′. Tobin’s q corresponds to point A: it is the ratio of the
marginal return on new investment to the cost of new capital. In the absence of installation costs, the optimum rate of investment (I/K)′ brings the capital stock immediately to its optimum level (point C). In Panel (b) investment starts at point A as before. With a higher stock of capital, the MPK then
declines as represented by the shift from MPK1 to MPK2. With Tobin’s q still above unity, investment continues but at the lower rate (I/K)2. The process continues until q is equal to 1, and no further investment is warranted.
Firms invest until the marginal cost equals the marginal return at point] A in Panel (a) of Figure 8.14, where the two curves intersect. The value of an additional unit of capital installed—Tobin’s q —exceeds the replacement cost of capital. Without installation costs, the firm would choose
point C instead and invest more. It may be surprising that the marginal return is higher with installation costs than without them. Rather than reducing the long-term profitability of investment, installation costs simply induce firms to invest at a slower rate. In the long run, firms achieve the
same desired capital stock as in the absence of adjustment costs.
Panel (b) of Figure 8.14 shows how the investment rate moves over time. With q above 1, investment first occurs at rate (I/K)1 corresponding to point A. Each MPK schedule is drawn for a given stock of already installed capital. Moving along the schedule, we find the profitability of
further additions to the existing capital stock. Once the capital stock has increased as a result of investment, however, these additions become less productive—a direct result of the principle of declining marginal productivity. So, as further investment accumulates, the MPK schedule shifts
downwards in the figure, from MPK1 to MPK2.20 Investment will continue as long as Tobin’s q is greater than unity, but will do so at a declining rate, here at rate (I/K)2 corresponding to point B. The process will continue until q is driven back down to 1 and the capital stock has reached its
long-run, optimal level.
8.3.6 The Investment Function

The investment function summarizes all the macroeconomic relationships that have been developed until now in a convenient and compact way. First, investment is inversely related to the interest rate because it measures the opportunity cost of the resources invested. Higher interest rates
imply lower investment spending. Second, the accelerator mechanism captures the stable long-run relationship between the capital stock and output. Since the rate of proportionality is greater than 1, increases in output lead to magnified increases in investment expenditures. Finally, Tobin’s
q reflects the fact that some firms finance investment expenditures by issuing shares on the stock market. High stock prices mean that the market places a high value on existing, installed capital, so firms can raise more resources per share issued, and this encourages investment. They provide
a central indicator of the market’s assessment of the profitability of new investment. Because a higher interest rate discounts future profits more heavily and reduces q, Tobin’s q also incorporates the effect of real interest rates on investment (for firms that raise money by issuing bonds or
borrowing from banks, the interest rate determines the cost of investment). These results can be summarized by the following investment function, which will be used in various forms throughout the book:
(8.20)
.
This function states that aggregate investment depends negatively on the real interest rate r, positively on the change in GDP, and positively on Tobin’s q.21 Holding interest rates constant, an increase in Tobin’s q increases investment.
Summary
1 Rational consumers attempt to smooth consumption over time, borrowing in bad years, saving in good ones. Consumption is driven primarily by wealth, the present discounted value of current and future incomes, and initial net asset holdings. Over a life cycle, income typically increases. To
smooth out consumption, agents typically borrow when young and pay back later.
2 Individual consumption smoothing means that aggregate current account imbalances (national saving or dissaving) are the best response to temporary disturbances. In contrast, permanent disturbances lead to immediate consumption adjustment rather than to borrowing or lending.
3 The world is uncertain and financial markets are imperfect. Uncertainty about future incomes and the inability of banks to assess individual future prospects without error will make it difficult if not impossible for some households to borrow against future expected income. For these households,
current disposable income also affects aggregate consumption.
4 The effect of changes in the real interest rate on current consumption is ambiguous. Lenders tend to increase, while borrowers decrease, consumption in response to increases in the interest rate. In the aggregate, however, higher interest rates are likely to reduce consumption by reducing wealth.
5 The consumption function relates aggregate consumption to wealth (positively) and disposable income (positively).
6 The optimal capital stock equates the marginal productivity of capital to the marginal cost of capital. The optimal capital stock increases when the real interest rate declines and when technological gains raise the marginal productivity of capital.
7 Investment over and above capital depreciation increases the capital stock. As is the case with the optimal capital stock, investment is driven by the real interest rate.
8 The accelerator mechanism links investment to changes in output. This is both a mechanical relationship (in the long run the capital-output ratio is constant) as well as a symptom of credit rationing for some firms, which then react sensitively to their own current cash flow.
9 The ratio of the market value of installed capital to the replacement cost of installed capital is called Tobin’s q. It is an approximation of the ratio of the present discounted value of the marginal return of investment to the marginal cost of capital. This ratio is equal to unity when the capital
stock has reached its optimal level. When Tobin’s q is larger than unity, the capital stock is below its optimal level and firms benefit from further investment.
10 The market value of installed capital, the numerator in Tobin’s q, is determined in the stock market. In determining this price, stock markets look ahead. The forward-looking nature of Tobin’s q mirrors how firms take into account expected future earnings when they make investment decisions.
11 Because of various installation costs, firms do not acquire their optimal capital stock immediately. Rather, they spread investment over time, gradually bringing capital to its optimal level.
12 The present discounted return to investment exceeds the marginal cost of capital to compensate for installation costs. Investment proceeds until the present value of its return, at the margin, equals the marginal cost of investment, the sum of borrowing and installation costs.
13 The investment function states that aggregate investment depends upon: (1) the real interest rate (negatively); (2) GDP growth (positively); (3) Tobin’s q (positively).
Key Concepts
consumption function
investment function
indifference curves
utility
marginal rate of intertemporal substitution
random walk
consumption smoothing
life-cycle consumption
permanent income
credit rationed
marginal productivity of capital (MPK)
opportunity cost
marginal cost of capital
optimal capital stock
rate of depreciation
user cost of capital
accelerator principle
Tobin’s q
q-theory of investment
animal spirits
installation costs
Exercises
1 A household has income of £10,000 today and £50,000 tomorrow.

(a) If the real interest rate is 5%, what is its wealth (i) in terms of today’s consumption, (ii) in terms of tomorrow’s consumption? Compute the household’s permanent income (see Equation (8.2)).
(b) If today’s income unexpectedly increases by £1,000, what is the change in permanent income?
(c) If income goes up by £1,000 permanently (in both periods), what is the effect on permanent income?
(d) Now answer the same questions with a 10% real interest rate.
(e) Suppose the utility of the household is given by ln(C1) + β ln(C2), where β is a positive constant less than 1. Derive optimal consumption today and tomorrow for (a), (b), (c), and (d).
2 Show, graphically, the effects on current and future consumption of an income windfall gain expected in the future. Now apply this reasoning to the national level. How should a consumer react to a tax cut that is announced to be temporary (in period 1 only)? How would consumers react to an
expected future tax cut?
3 Define the permanent income hypothesis. How does it relate to the more general theory of consumption under intertemporal utility maximization?
4 Suppose you expect to live for six decades. Your income is €15,000 in the first, €100,000 in the second, €150,000 in the third, €250,000 in the fourth, €200,000 in the fifth, and €100,000 in the last. What is your permanent income if the interest rate is 3%? 5%? If you consume your permanent
income each year, what does your savings pattern look like? How does your answer change if you decide to bequeath €50,000 to your favourite charity?
5 There is evidence that spending on durable goods drops faster than spending on non-durables during a recession. Can you give some economic reasons why? (Hint: think of durable goods expenditure as a form of saving.)
6 A company is considering expanding its computer equipment today by €50,000. Suppose the interest rate used in discounting future profit is 10%, and suppose that the equipment is scrapped after one period.
(a) If the new investment generates €60,000 of profit, should the company undertake the investment? How does your answer change if the firm only expects to earn €54,000?
(b) Now suppose the equipment can be sold on the resale market for €15,000. How does your answer change? Discuss how the emergence of a more efficient resale market (eBay, etc.) could increase investment.
7 Governments sometimes try to stimulate the economy by offering firms a temporary tax credit. Firms that purchase capital goods before a deadline receive a tax reduction on income made in the next year. Explain why this measure should increase expenditures on investment. Would you expect a
permanent or a temporary measure to have more effect?
8 Suppose a firm produces Y2 output in period 2 using the production function Y2 = Aln(K1), where K1 is the capital in place in period 1. Suppose the firm has no outlays besides K1, the price of output is given at p2, and that (1 − δ) of the capital can be resold for pK 2 per unit. Future outlays and
revenues are discounted at rate r.
(a) Write down a mathematical expression for the profits of the firm. How do profits depend on p2? On pK 2? δ? r? A? K1? Use informal arguments, a spreadsheet, or calculus to explain your answer.
(b) Use calculus to find the optimal stock of capital K1 as function of p2, pK 2 δ, r, and A.
(c) Now suppose the same firm can instead rent capital at rate U for one period, after which the capital must be returned to the rental agency. Write down a mathematical expression for the profits of the firm. At what point should the firm be indifferent between renting and ‘owning’ the
capital it uses?
9 Should a temporary increase in taxes to finance a temporary increase in public spending reduce the current account deficit? What about a permanent increase in both taxes and public spending? What is the effect on the current account in an open economy?
10 It is sometimes argued that the Eurozone crisis is related to the mismanaged boom in spending and investment that followed the adoption of the euro. How can you explain this boom and what do you expect the effect to be on the (primary) current account? (Hint: think of the impact on the
interest rate, on private wealth, and on stock prices.)
11 Over the next 10 or 20 years, Greece, Portugal, and many other European countries will have to raise taxes to reduce their public debts. What does this mean for consumption, now and in the future?
12 In Ireland, Spain, and the UK, housing prices increased considerably over the 2000s, until they fell abruptly in 2008. Is there any reason why this price decline should affect consumption?
Essay Questions
1 A great deal of debate has arisen in Germany on the financing of the expenditure necessary to improve the much-neglected infrastructure in its new eastern states. One side favours increased taxes, which would fall largely on households. The other side favours an increased budget deficit. Which side
is right? How important is it to know whether the spending increase is permanent or temporary?
2 Should wars—temporarily but abnormally high expenditures of the government—depress or raise consumption of households? In formulating your answer, think about all the budget constraints discussed in Chapter 6.
3 Over the past 20 years the price of computers and software—an essential investment good for modern companies—has dropped steadily. At the same time, the rapid rate of technical change has increased the rate of obsolescence of equipment and programs. In addition, the development of new
service companies has made it easier to install and employ these new innovations. Explain, carefully, using the concepts you have learned, how these developments should affect the user cost of capital, Tobin’s q, and investment.
4 The public debt crisis forces some countries to precipitously cut their budget deficits because governments cannot borrow any more. Most raise taxes on households and firms, which are taxpayers. Some economists argue that this is a self-defeating strategy. What is your view?
5 Booming economies frequently show negative current account balances. International organizations like the International Monetary Fund must often judge whether a current deficit is ‘good’ or ‘bad’. Explain what this might mean or imply. Use your answer to interpret the case of Poland: in the
period 2003–2007 Poland grew 5.1% per annum, as compared to 2.4% in the EU. During this period the current account deficit was 2.6% of GDP, declining from an average deficit of 4.6% of GDP in the period 1998–2002.
1 Introspection often makes us sceptical about such assumptions. Who hasn’t given in to the tempta-tion of buying a pastry when not really hungry or a new sound system or some new clothes despite being short of cash? Such departures from rationality are infrequent enough to be
outweighed by a majority of reasoned decisions. This is why rationality in economics is the right way to think about reality—at least to a first approximation!
2 Indifference curves were introduced in Chapter 5 to study the labour supply decision of households in an intratemporal setting.
3 This is only an approximation. Impatient consumers may bring forward some of tomorrow’s windfall, whereas more patient agents would save some of today’s windfall for tomorrow’s consumption. The exact answer will depend on the shape and position of their indifference curves.
4 The random walk theory of consumption was first formulated by Robert E. Hall of Stanford University. It is surprisingly difficult to reject this theory empirically.
5 The permanent income hypothesis was first proposed by Chicago economist Milton Friedman (1912–2006) and was cited as his main contribution when he was awarded the Nobel Prize in 1976. The life-cycle theory of consumption was also recognized by the Nobel Prize committee as an important contribution of MIT economist Franco
Modigliani (1918–2003).
6 This assumption regarding saving and consumption is also central to the Solow growth model studied in Chapter 3.
7 Chapter 6 provides more details on interest rates and credit rationing.
8 To see why, take one point on the curve. An increase ∆K in the stock of capital is represented as a horizontal move from the initial point. How much output ∆Y is available? It is measured as the vertical distance which returns us to the production function. The ratio ∆Y/∆K is simply the slope of the line connecting the initial and terminal
points along the production function. For smaller and smaller steps ∆K, this slope ultimately becomes the slope of the curve itself (formally, it is the line tangent to the curve at the initial point). In the language of calculus, it is the first derivative of the production function with respect to capital: F′(K).
9 The careful reader may appreciate the following detail: presumably, capital can be resold. Here we assume that Crusoe abandons his coconut grove upon rescue. If he
could sell it, the expected resale value comes as a deduction from investment cost. Boxes 8.5 and 8.6 elaborate on this point.
10 In Chapters 3 and 6 we saw that the change in the capital stock ΔK is equal to I − δK, where δ is the rate of depreciation, and therefore δK is the amount of capital that is used up in each period. This relation can be rewritten as
I = ΔK + δK, which shows that I—gross investment—must cover both ΔK—the intended net investment—as well as replacing depreciated capital.
11 This proposition is both empirically and theoretically tenable. The long-run stability of the capital–output ratio was one of Kaldor’s stylized facts introduced in Chapter 3. Theoretically, consider a Cobb–Douglas production function with constant employment equal to 1: Y = AKα. Then the MPK is given by ∂Y/∂K = αAKα − 1 = αY/K.
The optimal capital stock is K* such that MPK = 1 + r, so αY/K* = 1 + r, and
K* = αY/(1 + r). Setting ν = α/(1 + r) implies (8.11).
12 To account for depreciation, (8.11) is simply changed to
I1 = νΔY2 + δK1. The same conclusions apply.
13 Tobin’s q is named after US economist and Nobel laureate James Tobin (1918–2002), who in 1969 pointed out that the ratio of a firm’s market valuation to the replacement value of its capital stock should predict investment activity. In the meantime, more sophisticated analyses have shown the conditions under which this ‘average q’
concept—which is readily measured using stock market values and investment good prices—is equivalent to the marginal concept presented in the text. For our purposes, we will simply assume that these conditions are met.
14 Remember, there are just two periods, so ‘tomorrow’ is shorthand for the indefinite future. Otherwise we would have to discount and sum up all future MPKs.
15 The volume of investment is usually measured relative to the total capital stock. For example, the investment costs associated with a 10% increase in the capital stock might be four times the costs implied by a 5% increase.
16 We cheat a bit here. The definition of Tobin’s q is based on the market value of firms, while what matters in theory is the marginal return on investment. The market value of firms depends on the average return on all capital, not just the latest addition. We overlook the difference between these two definitions of average and marginal q.
17 A related approach, pioneered by Professors Finn Kydland of University of California, Santa Barbara and Edward Prescott of University of Arizona, both Nobel laureates, stresses that it takes time to design, acquire, and put in place new equipment. The implications are similar to those of installation costs.
18 The expression (1 + r) − (1 − δ)(1 + πK) = r + δ − πK + δπK is well approximated by r + δ − πK for small δ and πK.
19 In the absence of depreciation, I/K = ΔK/K. Focusing on the investment rate is justified by the idea that a given amount of investment is more disruptive in a small firm (or economy) than in a large one.
20 The observant student will note that for linear marginal product and marginal cost schedules and in the absence of depreciation, the intercept of each successive MPK curve with the vertical axis is determined by the value of q in the preceding period.
21 In theory, Tobin’s q should contain all information necessary about the profitability of investment. Since many firms finance investment with retained earnings or by borrowing, the interest rate also matters directly. In addition, as with households, many firms are rationed on the credit market, so they have to rely upon current income to
finance spending on productive equipment. This would rationalize a third term related to the discussion of the accelerator in Section 8.3.3.
Money and Interest
Rates 9
9.1 Overview
9.2.1 Technical Definitions of Money
9.2.2 The Money Makers: Central and Commercial Banks
9.2.3 Details on the Money Creation Process
9.2.4 Central Bank Control of the Money Supply
9.3.1 The Money Market and its Players
9.3.2 The Interest Rate is the Price of Money
9.3.3 The Demand for Money
9.3.4 Money Market Equilibrium
9.4.1 A Private Convenience
9.4.2 A Public Good Provided by the State
9.4.3 A By-product of Banking and Credit Creation—or not?
Summary
The invention of a circulating medium, which supersedes the narrow, cumbrous process of barter, by facilitating transactions of every variety of importance among all sorts of people, is a grand type of advance in civilization.
Chambers Encyclopaedia, 1870
Money frees you from doing things you dislike. Since I dislike doing nearly everything, money is handy.
Groucho Marx (1890–1977)
9.1 Overview
Until Chapter 5, we discussed the macroeconomy in real terms, that is, things that really matter—goods, labour, capital, technology. But goods are rarely exchanged in barter for goods or for labour. Workers’ wages, interest rates, and profits are not measured in
goods for that matter, either. In fact, introducing money was an essential step towards a market economy: it establishes a means of payment which does not in itself use up resources. In addition, money is a measuring rod for the value of goods, labour, and assets
in a common, value-neutral way. Instead of stating wages and prices in haircuts, cigarettes, or cups of coffee, or weeks of rent, we quote them in pounds, dollars, euros, or yuan. In Chapter 5, we saw that the monetary neutrality principle implies that the sheer
volume of the money stock should not affect our standard of living directly. Rather, as the great Scottish philosopher and economist David Hume put it, money is ‘not, properly speaking, one of the subjects of commerce; but only the instrument which men have
agreed upon to facilitate the exchange of one commodity for another. It is none of the wheels of trade: It is the oil which renders the motion of the wheels more smooth and easy.’1
Even if it is merely the oil which lubricates the wheels of our economic system, the quality of that oil matters. At the very minimum, money has this convenience function. It also represents wealth, it is widely accepted, it can be readily exchanged for goods
and services and it is a yardstick for the value of goods and services. It also creates a veil which can mask the real value of those things. In addition, money plays a crucial role in determining the interest rate, which drives financial markets (Chapter 7). So far, we
never really explained what money is ,or who creates it, how it does so, and for what purpose. Answering these questions is the task of this and the next chapter. It seems strange to have to explain something that we use every day, yet it is surprising how little
most people understand what it is.
We will start by defining money. We will then look at the supply of money, where in fact it comes from. This will lead us to introduce the central bank, an important institution with great influence over the supply of money and monetary conditions in general.
The next step is to study the demand for money. When this is done, we will be ready to see how demand and supply are equilibrated in the money market, as well as how that market works. Along the way, we will discover how money is created, in a complex dance
between the central bank, the banking system, and the real economy. Finally we will ask hard questions about the future of money in our economic lives.

Virtually every civilization has used one form of money or another. We have defined money as a form of wealth—a means of transferring consumption across time—which is generally and readily accepted by others in market transactions. Despite its very low rate
of return—banknotes yield no interest at all—people and companies use it. This is because money uniquely facilitates commercial transactions, big and small alike. As a result, it lies at the heart of any economy.
The nineteenth-century British economist William Jevons defined money in his time as having four attributes or functions: (1) means of payment, (2) unit of account, (3) store of value, and (4) standard of deferred payment. These features described well the
coins and sterling banknotes as well as the growing use of paper cheques, promissory notes, and other negotiable instruments of his day. Yet do they describe adequately what we use today and what we will use as money in the future?
9.2.1 Technical Definitions of Money

Despite our definition of money, there is room for disagreement on the details. Most of us would agree that banknotes and coins are money. But what about bank account balances? Savings accounts? Other financial instruments? Foreign exchange? Bitcoin?
Once upon a time, only gold, silver, and other commodities were used as money in the form of coins or bulky ingots. Over time, paper money (banknotes) edged out these commodity monies. Paper money ‘backed’ by commodity money became the dominant
form of money by the end of the nineteenth century. At the beginning of the twentieth century, central banks began issuing paper money and coins that were not really backed by any precious metal. Devoid of any intrinsic value, fiat money—from the Latin
word ‘fiat’ meaning ‘let it be done’—is legal tender simply because the government decrees it to be. Money changed again with the widespread use of bank accounts for making payment, technically known as sight deposits or demand deposits, which can be
converted at banks for cash on demand or accessed at points of sale using ‘debit cards’—electronically secure pieces of plastic that grant their owners rapid and safe access to means of payment. In a growing number of countries, including those with only poor
access to banking services, smartphones have assumed this function.
To understand the nature of money is to understand that it is accepted by others as a means of payment. Paper money arose as a convenient way of safekeeping precious metals; a certificate of gold ownership issued by a trustworthy trustee of gold and silver
was as good as the real thing. Other IOUs (IOU = ‘I owe you’) also emerged over the ages as a means of payment. Bills incurred by individuals, banks, or companies of excellent reputation have also been accepted as payment for hundreds of years, but not by
everyone. In the nineteenth century, there were doubts whether paper money could ever be as trustworthy as coins made from precious metals. Today, we use paper cash or metal coins—together known as currency—to pay for whatever we want to buy without
caring about lack of intrinsic value of the banknotes. In many countries, currency itself is slowly becoming something of a relic. Figure 9.1 displays the stock of currency in circulation as a proportion of nominal GDP.
The figure shows two different trends in the four different monetary areas. In the UK and the US, the demand for currency is stable, even slightly rising since 2000 (to be sure, more than half of US cash circulates outside the country!) In contrast, the volume of
currency used in Sweden and Norway has fallen sharply. What do Swedes and Norwegians use instead? To pay for the goods or services or houses or assets they buy, they increasingly move funds electronically using the internet and computer terminals or
smartphones, which means that they transfer ownership of parts of their bank accounts to others. What are bank accounts? They are liabilities of banks. As Table 9.1 shows, most of what macroeconomists consider to be money is in fact bookkeeping entries in
bank ledgers—that is, bank accounts. Most people see little difference between coins, banknotes, and bank deposits, and treat them as almost identical in transactions. This is our first definition of money: currency in circulation plus immediately accessible, or
‘sight’ deposits. This monetary aggregate is known as M1:
Fig. 9.1 Currency in Circulation (in per cent of nominal GDP)
The stock of currency (banknotes and coins) is displayed in relation to GDP in four countries. In the two Scandinavian countries, it is clearly trending downwards, reflecting the increasing use of sight deposits to pay for goods and services, accessed instead by bank transfer, credit, and
debit cards. In contrast, the use of cash in both the UK and the US exhibits a slightly increasing trend, due most likely to increasing circulation of banknotes abroad as well as a rising tendency to underdeclare market income.
Sources: IMF.
Sight deposits at banks—commonly known as ‘giro’, cheque, or current accounts—have three main characteristics: (1) they may be converted into cash on demand at the issuing bank, (2) cheques can be written or bank transfers can be made against them, and
these claims are accepted by other banks, (3) the interest paid on those deposits is either nil or very low. They are convenient for making purchases, but not particularly attractive as a way of holding wealth. This is why banks often offer other types of accounts
that bear higher interest rates, but whose funds cannot be used as easily as normal money—i.e. by writing cheques or stopping at a cash machine. Yet such funds can be conveniently transferred into regular sight deposits—usually a few keystrokes on a
smartphone and an internet connection are enough, possibly with a little waiting time on top. Ease of transfer renders these assets very similar to sight deposits in the eyes of their holders. This is why they are included in a broader definition of money, M2:
An even broader measure includes instruments such as large certificates of deposit, or time deposits with a longer term and possibly restricted access, foreign currency deposits, and deposits with non-bank institutions. The precise meaning of ‘larger’ and ‘longer
maturity’ depends on national rules and regulations. The distinction is one of degree: these instruments are less liquid, meaning that they are more costly or difficult to convert into cash or demand deposits. Thus the definition of M3:
Because M1 is highly liquid, it can be used for commercial transactions, unlike the other components of M2 and M3, which must generally be converted into M1 for that purpose.
Every country has its own preferences for means of payment and it is not surprising that definitions of monetary aggregates differ across the world. For instance, the Bank of England focuses on M4, a concept slightly wider than M3. While not fully
comparable, the examples in Table 9.1 show that, in the UK, people use less cash than on the continent and in the USA, but that the other monetary aggregates are proportionately larger—a sign of more sophisticated retail banking. Indeed, in Poland, the
difference between M2 and M3 remains small, suggesting that customers do not yet have access to higher-yield bank deposits in M3 or that these are not very attractive relative to the more liquid alternative M2.
The pace of technological development in banking has revolutionized the definition of money over the past century, and more change is certainly in store. Many believe we are not far from the limit of this process, a cashless society that the Swedish economist
Knut Wicksell once imagined and which is described in Box 9.1; indeed, Figure 9.1 and Table 9.1 suggest that the Swedes themselves are leading the charge. We will generally refer to money without being specific about its definition, calling it simply M and
thinking of it as:
9.2.2 The Money Makers: Central and Commercial Banks
Table 9.1 shows that currency (coins and banknotes) is only a small fraction of what we call money, regardless of the definition we use. Bank deposits—liabilities of the banks—make up the rest. While currency is produced by the so-called monetary authorities—
the central bank and/or the Treasury—the bigger part, bank deposits, is created by commercial banks. Think of bank deposits as liabilities of banks—effectively what they owe to their customers.
Table 9.1 Money in Five Places, 2015
Currency M0 M1 M2 M3
United Kingdom (£ bn) 67.4 373.3 1464.3 2130.6 2401.3
% of GDP 3.6 20.0 78.5 114.3 128.8
Euro Area (€ bn) 1110.8 1930.8 6630.8 10234.5 10836.8
% of GDP 10.7 18.6 63.7 98.3 104.1
United States ($ bn) 1380.8 3611.4 3147.0 12401.5
% of GDP 7.7 20.1 17.5 69.1
Poland (Zl bn) 163.0 215.0 692.1 1145.7 1155.4
% of GDP 9.1 12.0 38.7 64.0 64.6
Sweden (SEK bn) 73.5 144.4 2289.4 2728.9 2788.7
% of GDP 1.8 3.5 55.1 65.7 67.1
Note: US stopped publishing M3 statistics after 2006.
It may surprise readers that money is actually created by the private sector! Although it is seen as a symbol of national identity, most modern money is not directly produced by the government or a government agency. How does that happen? Can banks do
this without limits? Ultimately the answer is no. First, deposit-talking banks create money only if it is profitable for them to do so. Second, and perhaps more importantly, they create money under the control and supervision of the central bank. This is what we now
examine.
Box 9.1 Wicksells Cashless Society
Once upon a time, money was gold or silver, or seashells, or large stones on South Pacific islands. Such commodity money has an intrinsic value, since it can be used for other purposes. These goods are ‘wasted’ when used as money, and this is one
reason why paper and cheap metal have replaced silver or gold. A century ago, the Swedish economist Knut Wicksell (1851–1926) went further. He asked: why have money at all? He envisioned a central record keeper who would keep a tally of all credits
and debits. Whenever an individual worked, his balance would be credited; whenever he spent, the balance would be debited. In principle, it would be possible to run a negative balance, i.e. to borrow from the system, as long as the debt was repaid.
Instead of producing currency, the central bank would simply operate and guarantee this record-keeping system and determine the value of the unit of account.
At the time, Wicksell’s moneyless society was dismissed as impractical science fiction. A century later, the technical problems of a ‘moneyless society’ have been largely solved. Computers and the internet allow accurate, up-to-date record-keeping
and investigate the creditworthiness of households and businesses. Credit and debit cards already represent the preferred means of payment. Because banks regularly exchange information about the creditworthiness of their customers, the use of
credit cards has made it easy to borrow money in most countries (up to agreed limits). While credit cards themselves do not represent money per se, they allow individuals to access credit—which takes place at the same time as money is created and
paid to the vendor of goods or services purchased with the credit card. Throughout Europe, payments systems have been introduced, some more successful than others. All can be activated by a few keystrokes on the computer or a hand-held telephone.
Some of the greatest strides have been made recently in Africa, where the low cost of cellular phones and internet access has made it easy to leapfrog older, traditional forms of payment.
Central banks
The central bank is a public or quasi-public agency with an explicit, exclusive legal mandate to control money and credit conditions.2 Think of it as the ‘bankers’ bank’. Just as households and firms keep money at the bank to execute their daily transactions, banks
themselves maintain deposit account balances at the central bank.3 They can use these funds to settle payments against other commercial banks, for example, when customers transfer funds from one bank to another. Central banks generally do not take deposits
from households and firms. However, they often serve as the bank for the national and local governments in which they reside.
Central banks produce two sorts of money. Both represent liabilities vis-à-vis those who own them. One is currency, that is, banknotes held by the non-bank public (households, firms, government), including coins. These liabilities of the central bank can be
verified simply by inspecting a banknote, sometimes in fine print.4 Second, they create bank reserves, those funds held by commercial banks at the central bank. Bank reserves do not circulate outside commercial banks, but they can be used immediately and
thus are close substitutes for currency. The sum of currency in circulation and commercial bank reserves is known as the monetary base, and is known as M0. Other terms sometimes used are ‘high-powered money’, ‘base money’, or ‘central bank money’.5
Table 9.1 shows that M0 is not much larger than the amount of currency in circulation, implying that the reserves of commercial banks are relatively small. Yet, the supply of bank reserves is the tool through which central banks can control money creation by
commercial banks.
Commercial banks
Most people associate banks with the taking of funds from depositing customers and lending to others who want to borrow. As discussed in Chapter 7, this is why they are also called financial intermediaries. But in fact, banks play many different roles in our
economic lives. They advise customers on how to manage their wealth, provide insurance, and may even trade financial assets on their own account. Most important, they are custodians of the payments system: they create most of the means of payment and
enable households, firms, and governments to transact with each other. These functions are all intimately related. During the financial crisis which started in 2007, the payments system was under threat because other functions had run into difficulties. It started
first in the United States and Ireland, spread to the UK and the European continent, and continues to leave its trace on the macroeconomies of the world today. Box 9.2 gives an impression of the role and importance of banks in that crisis.
Box 9.2 Money, the Payments System, and the Role of the Banks
We saw that money has many functions in our economy. The most important is the means of payment. While banknotes are legal tender—that is, decreed by law to be acceptable in transactions, they represent only a small fraction of the money supply
that we call M1 or M2. The bulk of the money supply is made up of claims on the banking system (look for those liabilities in Figure 9.2). In most countries, banks are privately organized entities; so why are claims on banks owned by firms and
households generally accepted as a means of payment? Several things are necessary: (1) credibility of the deposit-issuing bank, (2) trust of the owners of the deposits, (3) trust of those who accept the deposits, and (4) trust of other banks involved in the
transactions. Trust and credibility among banks are the most important ingredients of the monetary system. Banks lend to each other and hold demand deposits with each other. While they can always ‘clear’ debits and credits with the central bank, much
clearing is done without the central bank’s direct involvement. In the next chapter, we will see that the regulation of banks also increases trust in their activities.
All around the world, the financial crisis 2007–2008 severely damaged the credibility of banks and destroyed trust that they had in each other. Assets owned by some banks lost a great deal of value in a short period of time. Because information is
incomplete, many depositors and creditors simply assumed that their own bank was also involved, either that they were hiding their investments in offshore operations (‘vehicles’) or were themselves a depositor at a troubled bank. Banks began to wind
down (sell off, cash in) their asset holdings and call in their money lent to other banks. While reducing these loans does not affect the money supply directly—remember that only deposits held by non-banks count as money—many financial institutions
were thereby forced to call in loans to non-banks to generate liquidity. This activity had real consequences, leading to a ‘liquidity squeeze’. In Chapter 10 we discuss in detail how the collapse of trust and credibility has attenuated the ability of monetary
policy to affect credit conditions in the economy.
Paying insufficient attention to banks can have dire consequences, as the experience of the euro—one of the greatest macroeconomic experiments in modern times—has shown. While the introduction of the common currency in 1999 (bank reserves)
and 2002 (banknotes) is generally considered a stroke of technical genius, the role of banks in creating the money supply was largely ignored. Insufficient integration of national banking systems, lack of coordination of national deposit insurance, and the
failure to introduce uniform euro-area banking regulation remain unsolved issues of the euro area.
When you deposit money in your bank, you may have assumed that it is there, waiting for you to withdraw or spend it whenever you would like. In fact, the money you deposited was most likely lent to one of the bank’s many customers. For the bank, your
deposit is a liability of immediate or short maturity, most frequently ‘due on demand’, sometimes after some period of notice. For this reason they are often called ‘demand deposits’ or ‘sight deposits’. Banks make loans to people who buy cars or houses, or to
firms that buy inventories, raw materials, or wages, and these loans are generally paid back over months or even years. These loans are the bank’s assets, and they tend to be of much longer term or maturity than its liabilities. 6 The practice of taking on short-term
liabilities and using them to acquire long-term assets is called maturity transformation.
Maturity transformation is the primary business model of banks—borrow short, lend long. In doing so, they provide an important liquidity service to lenders: the ability to convert more profitable loans into cash in short time. Although it is standard practice,
maturity transformation is a fundamentally risky activity, and this is why banks are regulated. To see why, it is useful to present a snapshot of the state of a bank’s financial operations, its balance sheet. The balance sheet is a summary statement of what the
bank owns—its assets—and what it owes—its liabilities—at a particular point in time, such as the end of the year. Figure 9.2 presents a simplified balance sheet of a typical bank. The balance sheet shows that the bank’s primary liability is the deposits of its
customers. Additionally, it may have borrowed funds from the central bank. On the asset side, a commercial bank keeps some cash in its vaults or as a deposit (bank reserves) with the central bank. It also holds various securities that can be traded on financial
markets. Its main activity, however, is to make loans. These loans are assets, since its customers owe them to the bank. The excess of assets over liabilities is the bank’s net worth. Net worth is treated like a liability on the balance sheet since the bank ‘owes’ the
net worth to its owners.
To see why maturity transformation can be risky, imagine that all customers of a bank withdrew all the money that they had deposited. This would be their right, because those deposits are means of payment, perfect substitutes for cash; banks are legally
obligated to convert these deposits on demand into currency. Since most of the money has been lent out, it is no longer in the bank. A mass conversion of deposits to currency at short notice would simply be impossible, despite the fact that the bank is a going
concern with positive net worth. Such sudden, overwhelming surges of withdrawals are called bank runs or bank failures.
In a sense, every bank is gambling that such an event will not happen, just as insurance companies do not expect everyone’s house to burn down on the same day. Bank runs are unlikely events associated with widespread panic; they are usually caused by
worries about the financial health or solvency of one particular bank which is presumed to suffer large losses from disappointing deals. When a run occurs, the afflicted bank is unable to meet its obligations and it must close, leaving angry customers pounding at
the door, unless it can be rescued or ‘bailed out’, either by another bank or the government in some form.7 A bailout simply means that depositors, creditors, and owners are shielded from losses that they would incur, had the bank gone out of business.
Fig. 9.2 Balance Sheet of a Typical Commercial Bank, Abstract and Concrete
The left-hand side of Figure 9.2 is an abstract representation of a commercial bank’s balance sheet. The right-hand side displays the balance sheet of a prominent euro area financial institution which is active in many countries.
Sources: UniCredit S.p.A., 2015 Consolidated Reports and Accounts, authors’ calculations.
*Refinancing operations collateralized by securities and loans.
So why, you may ask, is the economic system willing to trust bank deposits—backed by loans to the economy made by commercial banks—as Hume’s oil to grease the wheels of commerce in the first place? There are two good reasons. First, it is in the
business interest of any commercial bank to be seen as acting responsibly. Any fear can trigger a bank run, which normally leads to bankruptcy. Second, banks are tightly regulated and supervised, precisely to reassure their customers that their deposits are safe.8
9.2.3 Details on the Money

Creation Process
Yes, banks actually create money when they grant loans—when they extend credit to customers. This can be seen in the representative balance sheet in Figure 9.2—the snapshot of the banks shows that only a fraction of their deposits is actually ‘backed’ by
cash. But how did this come to be? In this section and the next, we will answer this question by telling two stories.
Consider Ms A, who receives €1,000 in cash from abroad and deposits it in her bank, Bank No. 1. This represents an injection of cash that had not been in the banking system before, perhaps locked away in a safe.9 The new deposit, to the extent that the foreign
cash had not been counted, is now part of the money supply. Now Bank No. 1 has extra funds which can be lent or used for other purposes. Looking for a profitable use for these funds, Bank No. 1 lends this money to Mr B, another trustworthy customer. Mr B
needs the money to buy a sofa. Soon enough, the €1,000, initially deposited by Ms A, will be handed over to the store that sold the sofa to Mr B. The shopkeeper now owns €1,000. Yet Ms A also owns €1,000. As far as she is concerned, her money is in the bank.
In fact, she owns a deposit in her bank which is backed by a loan in the amount of €1,000, yet the €1,000 in currency has long since left the bank. Through the creation of a credit, the initial €1,000 has given rise to €2,000 in balances at banks. And the story does
not end here. If the sofa storeowner deposits his money in his own account at Bank No. 2, which could lend it to Ms C, who will spend it to buy a laptop for her business, the computer store will rightfully think that it possesses an additional €1,000—because it
does! Each loan amounts to the creation of new credit and money, and the process can seemingly go on and on. Will it ever end?
To see why the process of credit and money creation does in fact stop, imagine that Ms A—the original character in this play—wants to withdraw her money from the bank, say to take a vacation abroad. Will Bank No. 1 tell her that the money is gone and will
only be back in two years’ time? Of course not! A natural part of the banking business is to anticipate this possibility; Bank No. 1 will have planned for it. One possibility is to have some cash on hand, that is, not to lend it out. Another possibility is that someone
down the chain is likely also to be a Bank No. 1 customer, so that at least some of the initial €1,000 will return. On average, Ms A is not expected to withdraw all of the money that she initially deposited.10
For this reason, modern banking is often called a ‘fractional reserve banking system’. If the bank determines that, say, only 10% of the sums deposited will be withdrawn at any point in time, it needs to keep only 10% of its deposits in reserves, and can lend out
the rest. In the case of Ms A’s deposit, the bank will keep €100 in cash and lend the remaining €900. In that case, the initial amount of money created is only €900. After it is deposited in Bank No. 2, a new loan of €810 (this is €900 less 10%), will be arranged, and
the process will go on as shown in Figure 9.3. At each step, the amount of newly created money declines by 10%. After a while, it will have become tiny. This is why the money creation process is not infinite. A little bit of algebra tells us that the initial €1,000 will
lead to a succession of loans that add up to €9,000:11
Fig. 9.3The Money Multiplier

An initial deposit triggers a succession of loans, paid in the form of deposits. When the bank keeps 10% of any deposit as reserves (cash or deposit at the central bank), each new loan is 90% of the previous one. Thus the chain of loans and deposits eventually dies out.
There is another side to this story. Bank No. 1 received €1,000 and put aside €100, Bank No. 2 received €900 and put aside €90, Bank No. 3 received €810 and put aside €81, etc. Jointly, all banks put aside €1,000. 12 In the end, the initial €1,000 has led to a series of
loans for a total of €9,000 and to the setting aside of €1,000. It all worked as if the €1,000 were fully set aside (more cash in the vaults) by the banking system taken together to support the creation of new money for a total of €9,000 (shown in the asset side in Figure
9.2). On the liability side, deposits increased by €10,000 as well: the initial €1,000 deposit and the €9,000 loans which have been credited to customers’ accounts.
This chain of money and credit creation is known as the money multiplier process. It says that any time new money is injected in the economy—in the example we assumed that the initial €1,000 came from abroad—the result is an increase in money stock
which is a multiple of that initial increase. This process explains two fundamental characteristics of modern fiat money systems. First, as we just saw, money is created by private commercial banks as they make loans to customers who want to borrow. Second, as
we explain in the next section, central banks can still control the size of the money stock if they want to.
The new money is created ‘at the stroke of a pen’—this old-fashioned expression is better described as a series of keystrokes in the bank’s computerized accounting systems. The money is ‘backed’ by trust in the banks—trust in the underlying value of the
loans and other securities on the asset side of the banks’ balance sheets. Money will be trusted ultimately as long as those assets are known to have adequate value.
9.2.4 Central Bank Control

of the Money Supply
In the previous example, commercial banks set aside 10% of any money deposited on their customers’ accounts. Presumably, they kept it in the form of cash—which often requires an expensive and secure (robber-proof) vault. The more convenient alternative is to
deposit these amounts with the central bank. Just as households and firms own sight-deposit accounts at commercial banks, commercial banks also maintain sight-deposit accounts at their central banks. They can draw on these accounts to obtain cash if needed.
In effect, cash in the vault and deposits at the central bank are two forms of bank reserves:
Table 9.2 Reserve Ratio Requirements in Selected Countries, 2015
Country Deposits subject to reserve requirements Mandatory reserve ratio requirement

Australia None
Canada None
Denmark None
China Deposits at large financial institutions 18.5%
Czech Republic Deposits with maturity up to 2 years 2.0%
Country Deposits subject to reserve requirements Mandatory reserve ratio requirement
Euro area Deposits with maturity up to 2 years 1.0%
Hungary All deposits 2.0%

New Zealand None
Poland All deposits except funds from repurchase agreements 3.5%
Sweden None
Switzerland Deposits with maturity up to 3 months 2.5%
United Kingdom None
United States Transaction accounts in excess of $15.2m but less than $110.2m 3.0%
Transaction accounts in excess of $110.2m 10.0%
Source: National central banks.
Together with the amount of currency in circulation, bank reserves make up the monetary base M0:
The proportion of deposits set aside in the form of bank reserves is called the reserve ratio.13 In the previous discussion, we presented the reserve ratio as a prudential measure taken by banks to be able to pay out requests for withdrawals by customers. In
some countries, banks are free to choose their reserve ratios; in others, the reserve ratio is imposed by the central bank. Table 9.2 provides some examples of mandatory reserve ratios.
In the previous example, the initial bank deposit by Ms A came from abroad. Most of the time, however, new money is created domestically. In our second example of the money supply process we illustrate how the act of borrowing money from the bank by
firms or households can give rise to money creation.
Our second story starts when Mr D requests a loan of, say, €1,000 from his bank. After carefully checking Mr D’s ability to repay—possibly even asking Mr D to put up collateral, meaning to pledge his own assets that would be forfeited should he fail to
repay his debt—the bank then makes the loan and does so by crediting Mr D’s bank account with €1,000. As before, money creation by banks occurs hand-in-hand with lending activities. The only hitch is that the new loan, hence the new deposit, immediately
raises the amount of reserves that the bank wants or needs to hold. If the reserve ratio is 10%, the bank must find an additional amount of reserves of €100. It may have these excess reserves at hand, but if it doesn’t, it will have to find them somewhere.
Because banks are in the business to make profits, they have an incentive to lend as much as they can while holding as few reserves as they must. This is where the reserve ratio kicks in. Whether it is imposed by law or simply chosen for prudential purposes,
the reserve ratio rr implies that reserves must be at least a fraction of deposits, which can be formally stated as follows:
Fig. 9.4 The Reserves–Money Stock Link

When reserves are a constant proportion (rr) of deposits (R= rrD), deposits cannot grow without an increase in reserves. Conversely, a change in reserves ΔR allows banks to increase their deposits—by granting loans—in much larger amounts. The reserve multiplier is the inverse of the
reserve ratio.
(9.1)
If banks have no incentive to hold more reserves than this, the inequality becomes an equality:
(9.1)
This relationship is represented in Figure 9.4 as a pyramid linking reserves to existing deposits. But the figure may also be viewed in reverse: the volume of deposits that banks issue cannot exceed a multiple of existing reserves. Rearranging (9.1) results in:
(9.2)
The factor (1/rr) is sometimes called the reserve multiplier.14 Equation (9.2) means that together, commercial banks cannot expand their deposits, i.e. create money beyond a multiple of existing reserves. If, as in the previous example, rr = 10%, then 1/rr = 10,
which explains why reserves are rather small, as we noted earlier when looking at Table 9.1. If (9.1) holds as an equality, then (9.2) does as well, and we now have:
(9.2)
which implies that, by choosing the volume of reserves, the central bank can control total bank deposits. A small catch: technically the central bank controls M0, defined as reserves of commercial banks plus currency in circulation. Since M = currency in
circulation + bank deposits, the relationship is somewhat more complicated than simply (1/rr) times M0. Only if no currency is held by firms and households will this be the case. The larger the fraction of M held by the non-bank public in the form of cash, the lower
the money multiplier. It is easy to show this mathematically under a few simple assumptions and the details are spelled out in Box 9.3.
Ultimately, the central bank can set M0 rather precisely and control the money supply for any particular behaviour of the banking system. How central banks do this in practice is the subject of the next chapter. In this chapter, we will assume that the central
bank can control the money supply with the precision necessary to conduct monetary policy.15
9.3.1 The Money Market and its Players

We have learned that banks create most of the money that we use, and we have learned how they do so in the process of taking and lending money to the economy. Banks are usually busy places, making loans to customers and taking deposits on a daily basis.
Because they are custodians of the payments system, banks are handling funds of their customers and others on a constant basis; loans are repaid, funds are withdrawn, new customers arrive, and old ones may leave. For a large bank, deposits and withdrawals
are likely to more or less cancel out at the end of the day. If the day has been good, more loans were made than paid off, and the loan portfolio grew. The discussion of the previous section implies that the bank might need additional reserves. On another less
active day, total loans outstanding might contract, possibly leaving the bank holding excess reserves.
Box 9.3 The Money Multiplier
Both bank reserve and currency-holding behaviour exert a strong influence on the money multipler. Suppose that the public (firms and households) hold currency in proportion cc to M, defined as the sum of currency and bank deposits, in the form of
currency. Further assume that banks hold reserves R at the central bank as a fraction rr of their customers’ deposits D. Then:
This is a system of three equations in four unknowns, M, R, M0, and D. Thus, if the central bank fixes M0, then the money multipler can be determined. Solve the first equation in the form D = M(1-cc). Use the second equation to eliminate R from the last,
obtaining M0 = ccM + rrD. Combining these results and eliminating D yields M0 = ccM + rr M(1-cc), which can be rearranged to yield:
(9.3)
According to equation (9.3), an increase of M0 in the UK banking system of 1 pound leads to an increase in the money supply of pounds. Since both cc and rr take values from 0 to 1, the money multiplier can range from one to
infinity. It is easy to show that the larger cc or rr, the smaller the money multiplier. Note that if cc = 0, we obtain the reserve multiplier, the money multipler in a world without currency. If there are no banks and only currency, the multiplier equals one. M1/M0
in the European Monetary Union has been roughly 4 over the past 10 years; in the UK, where there are no reserve requirements, it is well over 15.
Computing the money multiplier is interesting and informative, but rests on the assumption that cc and rr are constant. In the next chapter, we will see that the financial crisis changed reserve-holding behaviour of financial institutions (less so the
currency-holding behaviour of households), making the multiplier quite volatile.
The normal reaction of an individual bank short on reserves will be to try to borrow from some other bank holding excess reserves that it is willing to part with—for a price. It will go to the money market, often called the interbank market or the open
market. This is not a physical marketplace, rather it is a network of banks that facilitates buying (borrowing) and selling (lending) of reserves, i.e. deposits at the central bank (high-powered money). If the total reserves in the banking system are just sufficient to
cover the reserve requirements of the outstanding volume of deposits in the banking system, then some banks having excess reserves implies that others are short. Overall, they can just deal among themselves to resolve the shortfall.
It becomes more complicated when all banks collectively face a growing demand for loans and try to satisfy that demand, and collectively need more reserves. The central bank will then have to decide whether or not to create these additional reserves. However,
normally the central bank does not respond face-to-face to a given bank. Instead, it generally deals with the money market as a whole.
When the central bank creates reserves, it makes a loan to commercial banks through the money market, pretty much like commercial banks provide loans to their customers. It also charges interest for these loans. Similarly, when banks lend reserves to each
other, they also charge interest. Money markets vary from country to country, but they typically bring together large banks and financial institutions, such as insurance companies or mortgage lenders, which handle large amounts of cash. In normal times, they all
are highly reputable institutions that know each other well, they lend and borrow large amounts, usually without asking for collateral or guarantees.
9.3.2 The Interest Rate is the Price of Money

The interest rate at which they do so is the rate at which large financial institutions can fund themselves. In effect, for them, the interbank interest rate is the ultimate cost of financing their needs. This rate is the basis for all interest rates. Indeed, a bank
charges its customers the money market rate plus a premium that represents their riskiness. The interbank rate is called EONIA (European Overnight Interest Average) in the euro area, the Fed Funds in the US, the Sterling interbank rate in the UK, etc. Figure 9.5
shows that the various rates charged by commercial banks follow the same evolution as in the interbank rate in the euro area.
9.3.3 The Demand for Money

Households and corporations need money to carry out their daily transactions. This is the basis for demand for money M. The private sector is either holding the money already and paying an opportunity cost of forgone interest, or it can borrow at a bank, at the
cost of interest paid. In order to satisfy this demand, commercial banks create much of what we call money when they grant loans to their customers. As they do so, they need to acquire sufficient reserves to meet their reserve ratio. Thus the demand for money M
by the private sector translates into the derived demand for central bank money M0.
Fig. 9.5 Interest Rates in the Euro Area, 2007–2015

EONIA is the interbank interest rate in the euro area. The figure also displays the interest charged to corporations for large (more than €1 million) and smaller (less than €1 million) loans and the interest charged for consumer credits, in all cases for a one-year loan. Lenders charge a higher
interest rate to borrowers that they perceive as riskier, but all rates tend to follow the EONIA.
Sources: ECB.
First, we need to understand the determinants of the public’s demand for money, M. The first simple answer is that it is driven by the volume of transactions measured as value in the local currency, since this is what makes money useful in the first place. We
discussed this issue in Chapter 5, where money demand was represented as a constant proportion k of nominal GDP PY. That proportion is the inverse of the velocity V of money (k = 1/V)—the number of times money is ‘spent’ on GDP each year on average (i.e. V
= PY/M). To assume that k is constant, however, is simplistic and is contradicted by the evidence. In fact, we will see that the velocity of money moves positively with the interest rate, or that the factor k moves in the opposite direction of interest rates.
The logic behind the interest rate as the cost of money is simple. Suppose that you want to hold more money (cash and ‘money in the bank’). How do you get it? One way is to borrow from your bank and pay the corresponding interest rate. Another way is to
sell assets and bank the proceeds. In that case, you give up the interest that you can earn on your assets to receive cash, which does not yield interest, or for deposit at the bank, which yields a very low interest. In both cases, the interest rate emerges as the
opportunity cost of holding money. This is why the demand for money declines when the interest rate increases, as shown in Figure 9.6. Thus, while money demand is a proportion k of nominal GDP, this proportion is not really constant. It becomes smaller when
the interest rate increases.
Fig. 9.6 The Composition of Money Demand and Interest Rates, Euro Area, 2000–2016
The left-hand scale is the fraction of M3 held in the form of M1, cash, and demand deposits (which pay little or no interest); the scale on the right is the 3-month interest rate (annualized). This figure shows that interest rates and holdings of low interest-bearing money move in opposite
directions.
Sources: ECB.
As we saw, different borrowers face different interest rates, depending on how risky they appear to their banks. On the other hand, Figure 9.5 shows that all these rates move together. For simplicity, we think of a single interest rate, denoted by i. The demand for
money is represented as:
(9.4)
where k(i) declines when i rises (formally, k(i) is a decreasing function of i).
Now the derived demand for M0 by commercial banks is a fraction of the public demand for M, since any new bank lending, which means more bank deposits, must be accompanied by an increase in reserves as indicated by (9.2′). It follows that the derived
demand for M0 also declines with the interest rate. It is represented by the downward-sloping schedule D in Figure 9.7.
9.3.4 Money Market Equilibrium

Suppose now that, starting from point A, nominal GDP increases. This means more transactions will occur, so an increase in money demand Md is to be expected. Households and firms will try to borrow the extra money that they need from banks. As banks
respond by granting loans, their own need for reserves increases. This is captured in Figure 9.7 by the rightward shift of the derived demand schedule from D to D′.
How can the central bank respond to the new situation? It has many options, since it is the sole producer of M0. One option is to keep the initial interest rate unchanged. This requires the central bank to provide the additional monetary base M0 needed, moving
to point B. Another option is to keep M0 unchanged, which means that we move to point C and the interest rate rises accordingly. In that case, the interest rate must rise enough to offset, via a decline in k(i), the impact on money demand Md due to the increase in
nominal GDP. In fact, the central bank can pick any point on the new derived demand schedule D′. Where it decides to go is the central question of monetary policy. We return to this choice in Chapter 10. A stunning development of recent years, told in Box 9.4, is
that central banks can even choose negative interest rates.
Fig. 9.7 The Money Market

The public’s demand for money is negatively related to the interest rate i, which represents the cost of borrowing from commercial banks. This demand translates into a derived demand for the monetary base (M0) by commercial banks, denoted in the figure by the negatively sloped
schedule D. If the public wants to hold more money, the derived demand schedule shifts to the right from D to D′. The central bank may decide to keep the interest unchanged (point B), or not to respond, in which case the interest rate rises (point C), or it can pick any combination of M0 and
the interest rate as long as it lies on the demand schedule.

In the next chapter we will explore monetary policy in detail: how the central bank of a nation or economic zone controls interest rates in the money market or affects the supply of money in an economy in general. This is an important practical dimension of the
monetary sphere, because the neutrality principle predicts that the inflation rate is determined by the growth rate of money less the rate of economic growth. Yet we learned in this chapter that money is a social convention, meaning that our theory of money
should be flexible enough to help us imagine a world in which money looks very different from that which we use today. Indeed the world is changing rapidly and the digital age holds many surprises in store for the years to come.
Box 9.4 How Can Interest Rates be Negative?
A close look at Figure 9.5 shows that in 2015, the EONIA turned negative. Interest rates are negative when a creditor lends, say €1,000 and receives less than €1,000 in a year’s time! In this case—the EONIA applies to loans between banks—it means
that it pays to borrow! How could that be possible? The following can help clarify why interest rates are different for different borrowers and lenders of different types and highlights the limits of monetary policy examined in Chapter 10.
Can a household really be paid to borrow money? Not really, unless banks want to give away money to their customers. Suppose the interest rate was –3%, meaning that a borrower of €1,000 would have to pay back €970 in a year’s time. Why not then
borrow €1,000, put €970 under the mattress to be paid in a year’s time, and get €30 for free? If you want €3,000, just borrow €100,000! This doesn’t sound sustainable, because finding lenders under such circumstances is likely to be difficult. Even if
there was deflation—if the price of goods was falling and was expected to fall in the future, repayment in more valuable money would still be possible by the ‘mattress operation’. Suppose prices were falling by 5% per year, meaning that repayment of
debts occurred in more valuable money (in terms of goods and services). Then a –3% loan is not a free lunch for the borrower, since the €1,000 must be repaid a year later in 5% more hard-earned goods and services. Putting the cash under the
mattress would be a better investment for the lender, with a 5% return compared with only 2% for the loan (–3 – (–5) = 2). The difficulty of reducing interest rates in an environment of low interest rates is called the zero lower bound problem. As long as
there is currency that can be hoarded—and as long as central banks do not tax money directly—it is difficult to push interest rates in the economy below zero.16
Negative interest rates observed in Figure 9.5 are a special case which apply to financial institutions and their dealings with the central bank. Recently, banks in the euro area, Denmark, Sweden, and elsewhere have paid to deposit their reserves with
the central bank. This is like a tax on banks for not lending their reserves. If instead financial institutions wish to borrow, interest rates are still positive or close to zero. Commercial banks could only charge negative interest rates and make a profit if the
central bank itself were giving away money—which they generally don’t.
9.4.1 A Private Convenience

As Groucho Marx states at the beginning of the chapter, money is handy—it is convenient to have in hand. Who has not been caught waiting for that taxi in the rain, only to be out of cash, or after an expensive meal at a restaurant which doesn’t accept credit
cards? While being short on cash has become less and less of a problem over time, cash can still be extremely useful in emergencies, and there are still businesses that do not accept credit or debit cards.
More significant is the desire to use money to remain anonymous. Those who engage in illegal transactions prefer not to be associated with those deals. Organized crime generally uses cash when doing business or investing, and the laundering of profits
requires the use of cash. Restaurants and other businesses sometimes try to ‘manage’ provable income and reduce tax liabilities, especially in hard times. Some vendors give discounts on transactions paid in cash. The technology of paper money means that cash
transactions are untraceable, and some people like not only the convenience, but also the privacy of doing things not immediately observed by banks, other financial agents, or the government. As Figure 9.1 suggests, it may be difficult to eliminate currency from
the economy, especially if the incentives to use it are strong enough.
9.4.2 A Public Good Provided by the State

Money’s fascinating history points us to two interesting episodes—and there are many others. In Scotland in the eighteenth century—in the age of David Hume and Adam Smith—the rise of banks gave them an unprecedented level of prominence. Scottish kings
were dependent on the banks for finance, probably because the Bank of Scotland, established one year after the Bank of England in 1695, was not allowed to make loans to the King without parliamentary approval. At any rate, the Bank of Scotland became an
issuer of high denomination banknotes for the wealthy, leaving a gap in the market for the means of payment. After the Bank of Scotland’s monopoly lapsed around 1716, the Royal Bank of Scotland 17 and other banks emerged as competitors, also issuing
banknotes. The interesting aspect of this competition was potential takeover as a disciplining device for excessive banknote issue—the competitor banks could simply purchase the less valuable notes and then use them to buy the over-issuing bank! This period
of free banking led Scotland to important innovations in finance, such as the overdraft credit and the taking of deposits. However, by the 1760s, a proliferation of banks had led to an expansion of the money supply of often dubious quality. Widespread bank
failures in the Commercial Crisis of 1772 led to the first regulatory efforts to prevent excessive banknote issue. Later, because Scottish banks were generally more successful than their English counterparts, the English Parliament tried to restrict the circulation of
Scottish banknotes in the south. In 1845, the Peel Act regulated the issue of banknotes by Scottish banks and to this day, three Scottish banks—Bank of Scotland, Royal Bank of Scotland (RBS), and the Clydesdale Bank—may issue banknotes, all 100%-backed
by deposits at the Bank of England, the central bank of the UK. In practice, only the banknotes of the RBS are in circulation and are not legal tender in England and Wales—but are still accepted in large cities and by cash processing machines.
Similarly, after gold was discovered in large quantities in California in 1848 and the territory joined the US in 1850, an economic boom ensued which led the new US state to be one of the richest in the country. As always, strong economic growth led to a
proliferation of banks. At the time, the United States had no central bank to manage monetary conditions; as in Scotland, private banks in California stepped into the breach, issuing banknotes against gold deposits. But the quality of the money’s reputation is
ultimately only as good as the issuing bank. The history of California free banking was not one of great confidence; wildcat banks often took gold and moved far away to make withdrawals difficult. Banknotes traded at discount to each other and to gold dollars.
Bank failures were rampant in the years surrounding economic and financial crises. Yet during the early twentieth century California banks introduced the branch banking concept, which remains to this day an important aspect of the business.
One important lesson from these two episodes is that while confidence may allow a successful private money to emerge and compete, the establishment of a central bank and legal tender was a crucial improvement in terms of the management of trustworthiness
and credibility. At the same time, innovation in money and banking often comes during times when the banking sector is expanding rapidly and regulation is light. This is a dilemma for regulators and we will return to this subject at many points in the rest of the
book.
9.4.3 A By-product of Banking and Credit Creation—or not?

Fast forward to the modern era. Since the late 1800s, the money supply has become increasingly elastic and responsive to the private and public sectors’ needs for means of payment. The increasingly professional practice of central banking has contributed to this
development. In the next chapter, we will see that fluctuations in interest rates may be quite extreme in the absence of a central bank. This follows from inspecting Figure 9.7. Yet banks are subject to the same forces present in financial markets: boom and bust,
exuberance and fear. More often than not, banks participate and fuel the frenzies associated with speculative bubbles. This was the case in the South Sea Bubble and the Darien Scheme (the Bank of England and Scottish banks), the US stock market boom which
preceded Great Depression, and the US real estate boom and bust preceding the Great Recession 2007–2009 (US mortgage financing companies and US banks).
Since the financial crisis of 2007–2008 and the recession which ensued, there have been many discussions surrounding the fundamental soundness and stability of the banking system. Some have argued that the banking system is so important to the
economy’s health that it should be owned by the government. This is an extreme view which downplays the importance of private initiative and profit in guiding investment decisions. The opposite view, that the central bank should be abolished, has also found
proponents. Yet others have challenged the fractional reserve system as the cause of recurrent crises. The extreme vulnerability of the financial system to bank runs—originating either with customers or other banks—is certainly a cause of concern, leading to
economic disturbances of increasing severity. A much-discussed proposal is positive money or Vollgeld (full money), which would separate the means-of-payment function from the credit function, effectively turning money creation into a utility like water,
sewage treatment, or electricity. Chapter 10 will deal with these issues in more detail.18
Because the payments system—the money supply—is so fundamentally important to the economy, it seems reasonable to shield it from crises. At present money is a mixture, primarily a privately supplied inside liability (bank deposits) and a publicly provided
one (currency). The central bank is, when supervision is tight, in a position to control monetary conditions in order to keep the economy on an even keel. Most importantly it is tasked with preserving the functioning and integrity of the payments system.
Nations are made and destroyed by the quality of their institutions; one of the most important is the means of payment. It is for this very reason that a currency is normally an attribute of the State, but there are a few exceptions such as the European Monetary
Union. Central banks must keep an eye out for innovations in the future which may render their control and influence obsolete or superfluous. One is the arrival, with the innovations of the internet, large databases, and social media, of payments systems that
might suddenly displace those created or sponsored by central banks. PayPal has become an enormous player in payments systems, as have online retailers (Amazon), search engines (Google), and social media (Facebook); these well-endowed companies could
eventually render banks obsolete in their payment function. Even more interesting is the rapid evolution of ‘crypto-money’ such as Bitcoin, the product of developments in encryption and so-called distributed ledger technology which keeps records in a
decentralized and unfalsifiable fashion. Never before has the emergence of payments systems like those envisioned by Wicksell (Box 9.1) been within our reach—possibly without any need for government involvement at all.
Summary
1 Money is a form of wealth which is generally accepted as a means of payment in an economy. Holding and using it has a convenience function, but comes at a price: forgone interest.
2 There are different definitions of money, but broadly speaking it is the sum of currency in circulation and deposits in banks and, possibly, other financial institutions.
3 Most of what we call money and employ in everyday transactions are demand (sight) deposits created by commercial banks at the same time they make loans to customers. These deposits—the banks’ liabilities—are considered a close substitute for currency and are therefore accepted
as a means of payment. The rest of the money supply—banknotes, and coins—is produced by the monetary authorities.
4 The banking system is therefore a crucial participant in the monetary economy. The acceptance of money depends on the trustworthiness of the banks as well as their solvency.
5 Banking involves maturity transformation, a risky operation which produces a mismatch between liabilities (which are short-term and convertible into cash) and assets (which may be long-term investments subject to valuation changes). Good banking practice is a juggling act, requiring
holding sufficient reserves to cover withdrawals while making sound, profitable investments.
6 For both prudential or regulatory reasons, commercial banks therefore hold cash—currency or deposits at the central bank—in proportion to their customers’ deposits. These reserves are liabilities of the central bank and are the main source of influence of central banks on the money
creation process.
7 The reserve ratio forces commercial banks to acquire more reserves whenever they create more money. The banks’ demand for reserves is derived from the public demand for money, which in turn is equivalent to the volume of bank loans to non-banks.
8 The central bank directly controls the amount of currency in circulation. It controls indirectly the other component of the money stock, bank deposits, by deciding whether to provide commercial banks with bank reserves.
9 Facing the derived demand for the monetary base, central banks can decide on which quantity to supply, or to supply whatever quantity is demanded to achieve a particular interest rate. An additional instrument available to the central bank is the required reserve ratio, when it exists.
10 The history of money shows that central banks serve a key function in the economy. Cash will have a demand as long as there is a demand for anonymous transactions.
11 The acceptability of money is a public good, and central banks can help provide this public good. Experience with privately issued money without government sanction has proved to be innovative, yet subject to recurrent crises.
12 New innovations in payments systems will continue to pose a challenge to central banks in the present and in the future.
Key Concepts
convenience function
commodity money
fiat money
currency
monetary aggregates (M0, M1, M2, M3)
bank deposits
bank reserves
monetary base
financial intermediaries
payments system
legal tender
maturity transformation
liquidity service
balance sheet
bank run, bank failure
bailout
money multiplier
bank reserves
reserve ratio
collateral
reserve multiplier
money market, interbank market, open market
interbank interest rate
derived demand
zero lower bound
free banking
positive money, Vollgeld
Exercises
1 Table 9.1 shows that the size (relative to GDP) of the monetary aggregates can vary significantly from country to country. What might account for these differences?
2 We saw that households in the UK hold M1 in relation to nominal GDP of about 0.20 in 2015. Suppose that the function k(i) is well-described by the form where i is the nominal interest rate in per cent, and that the interest rate for short-term non-money bank
deposits (the opportunity cost of holding M1) is 1%. Suppose that in the next few years the nominal interest rate rises to 4%. What is the resulting demand for money (as a fraction of GDP). What is the elasticity of the demand for money, measured in terms of the observed change in
interest rates?
3 In Exercise 2, we saw that the real demand for money declined in response to an increase in interest rates. What happens to the supply of money? Draw a graph of before and after, assuming that the central bank is fixing the interest rate 1% and raises it to 4%. Suppose instead that the
nominal supply of money is held constant while the interest rate is raised. How could equilibrium be restored? How would your answer change if the price level is fixed instead?
4 Suppose that the demand for money is (M/P)D= 0.1Y/i, where i is the nominal interest rate in per cent. If real GDP is given by €17 trillion and the interest rate is 1%, what is the demand for money? What is the consequence of a rise of the interest rate to 2%? If the nominal money
supply is constant, how can the real money supply achieve this? If the price level instead is constant?
5 Given the demand for money in the previous problem: suppose now that the interest rate is 1% and is maintained constant by the central bank. What is the consequence of an increase in real GDP of 2% for the real demand for money? How must this real demand for money be satisfied
if the nominal money supply is constant? If the price level instead is constant?
6 What is a balance sheet? Why do you think the balance sheet is important in the economic system? Name some ways in which balance sheets are important for macroeconomics.
7 Consider the following commercial bank (based on real data): cash and bank reserves: €25 bn; investments in securities €250 bn; loans: €550 bn; other assets: €25 bn; liabilities to central bank €50 bn; total deposits: €500 bn; total (non-central bank) debt: €200 bn; other liabilities: €50 bn.
(a) Construct the balance sheet in the manner of Figure 9.2. How ‘long’ is the bank’s balance sheet (what are total assets)? What is the net worth of the bank?
(b) What is the reserve ratio for this bank?
(c) Compute a crude capital ratio of the bank—defined as the ratio of net worth to total assets.
(d) Suppose there was a sudden decline in the value of investments and securities of €20 bn. If the bank reduces (‘marks down’) the value of those securities it shows in its balance sheet, how do your answers to questions (a), (b), and (c) above change?
8 In macroeconomics, balance sheets are usually reported by sectors in consolidated form, that is, netting out (cancelling) all claims (assets and liabilities) on any other entity in that same sector. For example, the ECB reports data for all commercial banks, savings banks, and other financial
institutions taken together, in which loans or bonds issued by banks and held by banks are cancelled out. Data for households, firms, and the government are sometimes reported in consolidated form, sometimes they are not.
(a) Consider the following example of a banking sector comprised of only four banks with assets and liabilities as follows, in billions:
Bank A Bank B Bank C Bank D

Reserves at central bank 20 40 30 30
Securities held that are issued by other banks 60 50 60 30
Securities held that are issued by non-banks 100 140 200 300
Loans to banks 80 40 60 20
Loans to non-banks 370 400 300 500
Other assets 20 30 10 20
Debt owed to central bank 40 30 25 60
Deposits 450 520 460 650
Securities issues to other banks 80 80 20 20
Debt owed (securities issued) to other banks 70 60 70 0
Debt owed to non-banks 20 30 30 80
Construct the consolidated balance sheet of the banking sector.
(b) What is the money supply in this economy? Why is the sum of all deposits at all banks much greater than the money supply as defined by central banks? When reporting money supply statistics, why do central banks report consolidated deposits of the
banking sector?
(c) Why might consolidation be a good idea when constructing balance sheets? When might it be misleading?
9 Suppose the non-bank sector in a particular country holds 10% of its money holdings as currency, and that the consolidated banking system holds reserves in the same relation to its deposits as the hypothetical bank in Question 7.
(a) Using the tools of Box 9.3, compute the money multiplier for this country.
(b) How would your answer change if, as in Denmark and Sweden, the non-bank public ‘goes off’ cash, choosing to hold only 2% of their total M1 in the form of currency?
(c) Suppose that banks choose to hold more reserves at the central bank than the previously derived rate; in fact, the consolidated banking system chooses to hold reserves in the amount representing 10% of their deposit liabilities. What is the money
multiplier now?
10 Explain why trust in the banking system is the same thing as trust in the payments system. How might a run on the bank damage the viability of money? How can a banking system deal with this threat?
Essay Questions
1 State Jevons’ definition of money. Compare it with central banks’ definitions as well as that provided in this book. Which definition do you think is the most practical? Which is most likely to endure the new technical innovations in payments systems?
2 Gresham’s Law states that ‘bad money drives out the good’. This means that, when two monies circulate at the same time, firms and households tend to spend the ‘inferior’ form of money and keep or hoard the ‘better’ one. Is this behaviour consistent with our definition of money or
those features stressed by Jevons? Explain.
3 Name some ways in which the central bank can establish trust in the currency. How can trust be damaged or destroyed? How can the government, to the extent it is a separate entity from the central bank, damage or reinforce trust? Do you think that political influence on bank regulation
is a positive or a negative aspect, and why or why not?
4 Hong Kong, while a part of the People’s Republic of China, does not use the Chinese yuan (also called renminbi), but rather the Hong Kong dollar. Even more remarkable, Hong Kong does not have its own central bank, but uses banknotes and reserves issued by the three major private
banks. This system functions extremely well without a central bank. How can you explain this? How might the Hong Kong banking system differ from that in California in the 1850s?
5 Cybermoney is, simply put, a decentralized system based on the internet—of which Bitcoin is the best-known example—in which ownership of each unit of the means of payment is validated by all members of the cybermoney-using community in question. In this way it is very
difficult to counterfeit money. Explain why Bitcoin money fulfils the definition of money. Some have argued that the central bank should not regulate or prohibit the use of bitcoin, but should rather encourage it and increase the confidence that the public has in its use. What do you
think? What are the advantages and disadvantages of such regulation?
1 Hume (1752) Essays, Moral, Political, and Literary: Part II, Essay III, 1.
2 The Bank of England was a private institution from its founding in 1694 until it was nationalized in 1946. Similarly, the Banque de France began its existence in 1800 as a private entity and nationalized in 1945. The Deutsche Bundesbank was established in 1949 as a post-war successor to the
Deutsche Reichsbank, founded in 1876. The oldest central bank is the Swedish Riksbank, founded in 1668. Other central banks are: Bank of Japan, 1882; Banca d’Italia, 1893; Austrian National Bank, 1816; Swiss National Bank, 1905. The Federal Reserve of the USA started out in 1913 and is owned
by member banks, although profits above a statutory maximum are remitted, as in most countries, to the government (Goodhart 1988).
3 Details vary from country to country, and sometimes smaller banks can satisfy their clearing needs by using a much larger partner, ‘correspondent’ bank. The central bank performs this function for the largest of banks.
4 For example, pound sterling banknotes are marked ‘Bank of England’ and by the Queen’s promise ‘to pay the bearer on demand the sum of … pounds’, reminiscent of times past when they could be redeemed for gold or silver. The currency of the euro area is marked simply with ‘ECB’ and the
equivalent abbreviation of ‘European Central Bank’ in other languages of the European monetary union. US banknotes simply feature the words ‘Federal Reserve Note’.
5 Bank reserves and currency are highly substitutable for each other. To guarantee this, the central bank prints and warehouses (under great security) large quantities of banknotes ready to be converted at a moment’s notice for bank reserves. A bank need only place a telephone call and the
banknotes will be delivered to the commercial bank, usually in an armoured truck. Until the moment of delivery, however, those banknotes are simply paper.
6 The term or maturity of an asset was discussed in Chapter 7.
7 The spectacular failure of Northern Rock in the autumn of 2007 is a relatively recent example of a bank run. Bank runs are relatively rare events in most countries—before that, the last one in the UK took place more than 140 years ago. Bank runs will be examined in more detail in Chapter 10.
8 8Naturally there are always some bankers who bend or break the rules, or pursue dangerous lending policies. More details on bank supervision and regulation can be found in Chapter 10.
9 The money could also come from under her mattress, or from a long-lost treasure chest. The important fact is that it represents an injection of cash (currency) from outside the banking system.
10 If Bank No. 1 were the only bank in the country and if people never held currency, the money would always come back. In practice, because there are many other banks and because people keep some money in cash, Bank No. 1 knows that it cannot rely on returning monies to pay back Ms A. But
the bank also has many other customers and it knows that they tend to keep a big part of the sums that they deposit on their accounts for a long time.
11 If a proportion p is lent at each step, the sum of all loans made is 1,000p + 1,000p2 + 1,000p3 + … = 1,000p(1 + p + p2 + p3 + …) = 1,000p/(1 − p). If p = 0.9, we find indeed 9,000.
12 Continuing to denote the proportion that is lent at each step by p, the amounts put aside are 1,000(1 − p) + 1,000p(1 − p) + 1,000p2(1 − p) + … = 1,000(1 − p) (1 + p + p2 + …) = 1,000(1 − p)/(1 − p) = 1,000.
13 In some countries, it is called the liquidity requirement.
14 Since reserves are a (small) fraction of deposits, rr is less than one. Then 1/rr is (much) larger than one.
15 Technically, we are assuming at this stage a constant money market multiplier. For more than half a century since the end of the Great Depression this was a valid assumption, and economists were not much concerned with excess bank reserves, This was because the opportunity cost of holding
them (interest rates) was high and the risk of lending them out was low. In Chapter 10 we explore why this relationship has broken down again and may be a long time in mending.
16 The real interest rate is the one that matters. If there is inflation, it eats away at the value of the money under the mattress, so lenders are unlikely to lend unless they are compensated for this loss (higher nominal interest rate). Borrowers will understand this as well. Only if the marginal
productivity of capital is negative will lenders and borrowers be happy with negative real rates. Also, a risk premium for ‘unsafe’ assets (as discussed in Chapter 7) could explain negative interest for safe ones. More on this in Chapter 10.
17 The Royal Bank of Scotland still exists today. It emerged from a bailout of the Darien Company, which was a failed overseas trading firm that had tried unsuccessfully to establish a colony in Panama. It was also in the business of issuing banknotes.
18 The proposal is noteworthy because it achieved the status of a referendum question in Switzerland and has been advocated by some economists. It has features similar to the famed Glass–Steagall Act of 1934, a US law which forced deposit-taking banks to divest themselves of investment banking
and securities trading operations.
Monetary Policy, Banks, and Financial Stability
10
10.1 Overview
10.2 The Instruments of Monetary Policy: Nuts and Bolts
10.2.1 Review: The Influence of the Central Bank on Banking and Money Creation
10.2.2 Open Market Operations
10.2.3 Direct Lending to Commercial Banks
10.2.4 Legal Minimum Reserve Requirements
10.3.1 Objectives
10.3.2 Instruments and Targets
10.3.3 The Taylor Rule
10.4.1 The Interest Rate Channel
10.4.2 The Asset Price Channel
10.4.3 The Credit Channel
10.4.4 The Zero Lower Bound
10.5 Financial Stability as a Prerequisite for Monetary Policy
10.5.1 The Inherent Instability of Fractional Reserve Banking
10.5.2 Confidence-building Measures
10.5.3 The Central Bank as Lender of Last Resort
10.5.4 Technological Innovation and Financial Stability
Summary
First of all let me state the simple fact that when you deposit money in a bank, the bank does not put the money into a safe deposit vault. It invests your money in many different forms of credit—bonds, commercial paper, mortgages and
many other kinds of loans. In other words, the bank puts your money to work to keep the wheels of industry and of agriculture turning around. A comparatively small part of the money you put into the bank is kept in currency—an amount
which in normal times is wholly sufficient to cover the cash needs of the average citizen. In other words the total amount of all the currency in the country is only a comparatively small proportion of the total deposits in all of the banks.
Franklin D. Roosevelt1
Out of the increasingly unwieldy gyrations, the Frankenstein mechanics of an uncontrolled supply of money, the need for a strong central bank has been found to be essential to our economic stability. The banks, the people, and the
government realized that panics and crises, caused by periodic irregularities in the flow of our money supply, must no longer be permitted to rock our country back and forth every few years. The Central Bank was designed to minimize these
convulsions, create more stable values, and thus make possible the smooth functioning of monetary machinery so necessary to promote the growth of the country and to improve standards of living. This was the purpose and this is the ideal.
William McChesney Martin2
10.1 Overview
Seventy-five years ago, in the midst of a national banking panic, US President Roosevelt used the radio to explain to nervous citizens how a bank works and why they shouldn’t withdraw their money all at once from
their accounts. While he was talking about banks, in fact he was talking about money. Because the banking system creates most of the means of payment in an economy, it merits our special attention. Then, as now, the
system of money and banking was rather complicated, and in a state of constant change. It is a central component of the macroeconomy.
We saw in Chapter 9 that the natural reaction of banks to an expanding economy is to increase the supply of money and credit. Banks will generally do this to the extent that it is prudent and profitable to do so.
Prudent means that they have enough reserves to cover expected withdrawals, perhaps with some excess reserves on top. Profitable means that the bank can earn money on its operations. Yet from time to time, banks
may fail or be subject to runs, often causing other financial institutions to be pulled into panic and despair. Because only central banks can create the reserves that banks need for their day-to-day operations, they
naturally have a special role as guarantor of the financial system.
Most importantly, the central bank sets monetary policy; it controls interest rates, exchange rates, or financial conditions in general. By doing so, it indirectly but decisively regulates the supply of money and credit.
Because the banking system provides 90–95% of the money supply in most European economies, its health and stability determine monetary conditions and the overall state of the real economy. In the last chapter, we
learned that when one bank is in trouble, other banks are not far behind. Banks are frequently active in capital markets, and fluctuations in asset prices can have dramatic implications for them as well.
Monetary policy affects our economic life in many ways. We will see that the practice of monetary policy has evolved over the years, marked by both great progress and severe setbacks. In developed countries,
conventional wisdom focused on the control of inflation. Recent economic and financial turbulence has led economists to adjust their views somewhat. After having understood the basics of modern monetary policy, we
conclude the chapter by reminding ourselves why banking is risky, why the health of the financial system affects the transmission of monetary policy, what can be done to avert financial crises, and how central banks
should proceed when crises occur.
10.2 The Instruments of Monetary Policy:

Nuts and Bolts
10.2.1 Review: The Influence of the Central Bank on Banking and Money Creation
The convenience of modern money allows economic agents to execute transactions without ever having to touch a banknote. On top of this, banks keep detailed records of these transactions and are custodians of the
payments system. Banks also make loans and take deposits, provide financial advice and other services. They may also invest in securities and other financial products on behalf of their clients, and may even do so
on behalf of the bank’s owners themselves. All these activities generate profits that represent income to the banks’ owners.
Most of us think of banks as lenders. When you go to your bank and ask for a loan, it may be the case that fresh loanable money in exactly the right amount had just been deposited by some other customer, but this is
unlikely. Yet, even if the funds are not available, the bank is also unlikely to tell you that you have to wait until adequate reserves are around before a loan is possible—banks never do that. In fact, if the bank is
convinced of your creditworthiness, it is good business and standard practice to lend you the money—even if it doesn’t have it! It will simply credit a bank account that you will open at that bank by the amount that you
asked for—all assuming, of course, that you are willing and able to service the loan that it has just made, i.e. pay back principal and interest over some agreed-upon period. When the loan is granted, the bank creates an
asset for itself (the loan) and a liability that it owes to you (the deposit). All this is perfectly legal, and you are free to spend the money you have borrowed from the bank!
This process is not really much different from money creation described in Chapter 9. To recall, €1,000 in cash from abroad was deposited in one bank and the banking system as a whole went on to create €9,000. All in
all, the money supply increased by €10,000: €9,000 of bookkeeping (‘checking account’) money and €1,000 cash ‘injected’ exogenously from outside. With a reserves ratio of 10%, new bank reserves of €1,000
correspond exactly to that initial injection of €1,000 in cash (remember, currency is part of M0). 3 In the present example, however, the bank simply gives you a loan of €10,000 and credits your account balance with
€10,000. (In the jargon of bankers, it ‘originates’ the loan). The money stock increases directly by €10,000.
The key difference between this example and the one of the last chapter is that the loan-issuing bank still needs, by assumption, €1,000 in reserves and must obtain them from somewhere. One possibility is that the
bank is not fully ‘loaned up’ to begin with, and has excess reserves already on hand. But if excess reserves are not available, it can refinance its lending, i.e. borrow reserves either from another bank, or directly from the
central bank. If the central bank does the lending, it credits (increases) your bank’s reserve account directly. In this case, your bank created money by crediting your bank account with the amount of your loan, and the
central bank created reserves by depositing a proportion rr of your loan into your bank’s account. Easy!
Fig. 10.1 Balance Sheet of Typical Central Banks, Abstract and Concrete
The left-hand side presents an abstract balance sheet of a central bank. The right-hand side displays two simplified central banks’ balance sheets, those of the European Central Bank (ECB) and the Bank of England, in 2015.
Sources: European Central Bank, Bank of England.
We can further reinforce our understanding of the central bank’s action by looking at its balance sheet, shown in both abstract form on the left, and for two real cases on the right in Figure 10.1. Currency in circulation,
money created by the central bank, is technically its liability, even though central banks do not ‘reimburse’ money that they create.4 Its other liabilities are the deposits of its ‘customers’: households, firms, but also other
banks, and possibly the government. Its assets include foreign assets—called foreign exchange reserves—loans to banks and securities, and often government bonds or loans to the government. When a commercial
bank needs reserves, it either asks the central bank for a loan or it sells it some of its own securities. In either case, the central bank’s assets on its balance sheet increase—either as a loan to the commercial bank or the
asset it purchased from it. In return, the commercial bank receives cash or an increased account balance with the central bank, both of which appear on the liability side of the central bank balance sheet.
10.2.2 Open Market Operations

Figure 10.2 makes clear that, as monopolist producers of the monetary base, central banks exercise a decisive influence on the money market. Institutional details vary from country to country, but one feature is
common: they bring new high-powered money into circulation by buying assets and holding them on their balance sheets. Similarly, they can reduce the stock of high-powered money by selling their assets,
simultaneously taking their own liabilities out of circulation. For this reason, the ‘length of the balance sheet’, or the total volume of assets owned by the central bank, is an important indicator of the high-powered money
supply M0. When the central bank purchases assets or lends to banks, it injects reserves into the banking system. Money creation therefore ‘lengthens’ the balance sheet of the central bank, as shown in the right-hand
side of Figure 10.2. Most of these dealings involve the money market discussed in Chapter 9, and involve only safe or low-risk assets. In practice, this means debt of their own government and foreign exchange.5
Fig. 10.2 Balance Sheet of a Central Bank Before and After an Open Market Purchase
The left panel shows the balance sheet of the ECB before a hypothetical open market operation in the form of 100 billion euros in repurchase agreements. The right side shows the balance sheet after the operation, entered as an increase in
‘loans to banks’. A repurchase agreement is similar to a collateralized loan to a financial institution. As a result, bank reserves increase by 100 billion euros.
The features of the open market are standard around the world. Banks operating in the money market typically lend to each other over very short periods of time, from overnight to two weeks. The central bank
monitors the market closely and ensures that either the interest rate or the quantity of money is in line with its intentions. For reasons explained in detail later, most central banks fix the interest rate. This is effectively
represented in Figure 10.3 by the horizontal money supply schedule S. This schedule drives home an important point: as a producer of its own liability (bank reserves) a central bank can set any interest rate it chooses,
provided it stands ready to provide or withdraw liquidity in any amount needed to meet the derived demand6 corresponding to this rate.
Fig. 10.3 Setting the Interest Rate

In principle, the central bank can achieve any interest rate it chooses by supplying whatever amount of money the private sector wishes to hold at that rate.
Central banks typically announce the rate that they intend to set; the position of the supply schedule S in Figure 10.3 is public knowledge, and can be discovered by a visit to a central bank’s website. This rate is
known the ‘refi’ rate in the euro zone, as the Federal Funds or target rate in the USA, the Bank Rate in the UK, or the ‘repo’ rate in Sweden. 7 In order to enforce this rate, each central bank provides its interbank
market with whatever volume of reserves are demanded. To that effect, it carries out open market operations, which means it purchases and sells assets using reserves, which are its own liabilities. These open
market operations can take many different forms. For instance, the European Central Bank (ECB) conducts auctions of reserves every week as repurchase agreements, complemented by monthly auctions, as do many
other central banks (more details on the ECB are presented in Box 10.1). In the USA, the Federal Reserve Bank of New York simply operates directly on the money market whenever the federal funds rate deviates from the
targeted rate by outright purchases or sales of US government securities. Some banks intervene by purchasing and selling foreign exchange. All these mechanisms have the same intended effect: to set the interest rate
by providing the money market with reserves it needs at that interest rate.
Box 10.1 How the ECB Does It8
In order to keep the interbank market rate as close as possible to its target rate, the European Central Bank has set up a complex system for financing commercial banks. For the most part, it uses three kinds of
open market operations:
Main refinancing operations, which are weekly auctions for loans with a maturity of one week, are the main channel through which commercial banks obtain reserves. Banks submit competitive bids on rates and quantities and the ECB
chooses how much to allocate at which interest rate.
Longer-term refinancing operations, which take place once a month and serve as a modest complement to the main refinancing operations. The loans have a maturity of three months.
Fine-tuning and structural operations, which are conducted from time to time to deal with special circumstances, at the ECB’s discretion.
The interest rates at which the ECB deals with the open market are shown in Figure 10.4. The marginal lending facility rate is for short-term loans outside the auction at a higher penalty rate; the deposit rate is
the interest paid for the reserves that banks park at the ECB. These rates form a tunnel in which the market interbank rate EONIA moves, close to the bi-monthly auction rates.
In normal times, the economy grows and demand for bank credit increases. Through their liquidity-creating operations—auctions and open market operations—central banks allow bank reserves, and therefore the
monetary base, to grow along with that demand. This allows commercial banks to increase lending to their customers, which increases M1 and the wider monetary aggregates. But what happens if the central bank wants
to slow down the growth of reserves, because it wants to raise the interbank interest rate? If it raises the interbank rate, it must make reserves scarce by withdrawing them from the market. Since banks are
continuously repaying short-term loans to the central bank, the easiest way to do this is not to renew maturing loans or to renew them only in part.9
Lending is always a risky business for any bank. In particular, loans may not be repaid in the future, becoming non-performing loans. Banks with significant non-performing loans will suffer a loss of profitability
and may not be able to repay their own debts, including those to the central bank. To avoid taking on this risk, central banks take extensive precautions. First, they ask borrowing commercial banks to provide
collateral—i.e. to put up an equivalent value of high-quality assets, such as government debt or bills issued by large corporations of highest quality, which would become property of the central bank if the bank
defaults. Collateral is an excellent way to discipline banks to borrow only if they have a sound project to finance. It makes it very costly for banks to default, i.e. to walk away from their debts. Second, central banks lend
only for a short time period, usually a week or 10 days. When the loan comes due, the commercial bank repays the central bank and collateral is returned, but there is no guarantee of being able to borrow again. In
contrast to commercial banks, central banks are keen to avoid risks. This is why the interest at which central banks lend is the lowest of all, as shown in Figure 10.4—low risk means low return.
Fig. 10.4 ECB Interest Rates (January 1999–June 2016)

The ECB conducts weekly auctions at the pre-announced main refinancing interest rate. The marginal lending facility and deposit facility rates determine, respectively, a ceiling and a floor for the interbank rate EONIA (Euro OverNight Index
Average), that tends to follow the refinancing rate closely. Since 2012, however, in the midst of the Eurozone crisis, the ECB has used other means (e.g. long-term repurchase operations) to push EONIA down to its deposit rate, which is
negative. This means commercial banks must pay the ECB for the privilege of depositing reserves with them.
Sources: ECB.
10.2.3 Direct Lending to Commercial Banks

In less sophisticated financial systems, it is common for central banks to refinance commercial banks directly. Even in continental Europe before the 1990s, it was common for central banks to do so as well. There are
several operational difficulties with direct lending by central banks to private banks. The first is the risk of default when the loans are not properly collateralized. Because central banks generally prefer to avoid
discussions concerning quality of the assets behind the loans—which also may have political dimensions—they would prefer to only accept government debt, which is standardized. Furthermore, how does the central
bank decide which lucky bank should obtain the refinancing? Increasingly, central banks prefer to deal with financial institutions ‘at arm’s length’ and in a neutral fashion.
10.2.4 Legal Minimum Reserve Requirements

Legal minimum reserve requirements can also be used to affect the money supply, assuming they are binding—banks would like to lend out more but are constrained by the legal reserve ratio. If this is the case, raising
the required reserve ratio from 5% to 6%, with unchanged supply of reserves, will cause deposits to contract by roughly 20% (the percentage increase from 5 to 6 is (6 – 5)/5 = 20%). This is a dramatic move, which not
only can prevent commercial banks from lending, but might cause them to demand immediate repayment of some existing loans or refuse to roll over others. This would cause deposits of the banking system as a whole to
decline.
Because this can be very costly and disruptive to banks and their customers, reserve ratios are normally changed only in small increments, and only in emergency situations. It is more frequently used in countries in
which the internal money market is less developed. One example, discussed later in Chapter 12, is China. As Table 9.2 showed, in some countries the holding of a reserves ratio is purely voluntary. In addition, even if
there is a legal reserve ratio, banks may hold more reserves than legally required.
Now that we understand how a central bank can decisively influence monetary conditions, it is reasonable to ask: why would it want to do so? In this section, we first examine the objectives that a central bank might
pursue. Then we look at how it realizes these objectives, both in an intermediate sense and using the day-to-day tools discussed in the previous section. It is useful to take a step back and summarize our discussion of
the technical aspects of monetary policy in Section 10.2. Consider the following four fundamental facts that will guide our thinking:
1. By making bank reserves more or less abundant, the central bank affects interest rates—in the first instance, interest rates in the money market. It lowers interest rates by making money more abundant and raises them by making money
scarce.
2. By setting the cost of bank refinancing, central banks can affect the attractiveness to banks of increasing lending activities—all other things equal. It also increases the ability of banks to purchase securities or other existing assets.
3. Because it influences the growth of the money supply, monetary policy determines inflation in the long run. It takes something like two to three years for policy to begin to affect inflation and sometimes up to a decade for the effects to be
fully felt.
4. By influencing the level of interest rates, monetary policy can affect the level of economic activity—output and employment—in the short to medium run. It takes two to three quarters for these effects to become noticeable.10
10.3.1 Objectives
The money neutrality principle, presented in Chapter 5, asserts that the inflation rate is determined by the rate of growth of the money stock in the long run. The money supply is created by the central bank and by
commercial banks under its watch. In chapters to come, we will show that inflation does not bring lasting benefit to an economy. In fact, it can create much trouble. For that reason, price stability—which is taken to mean
a low rate of inflation—is socially desirable. It stands to reason that monetary policy should, first and foremost, be dedicated to achieving price stability, and central banks bear ultimate responsibility for fulfilling that
objective.
On the other hand, money is not neutral in the short run. Chapters 11 and 12 will show that, over a horizon of one to about two to three years, monetary policy actions affect output and employment. This implies that
central banks cannot avoid also being concerned with the level of economic activity. This seriously complicates their job. It is always easier to have only one objective on the radar screen. However, there are times when
central banks confront conflicting demands between keeping inflation low—which takes time to achieve—and keeping the economy and employment growing in the meantime. Such conflicts of policy objectives greatly
complicate the conduct of monetary policy.
They also raise political issues. Central banks are run by non-elected officials with a great deal of power. For this reason, central banks in some countries were even placed under the direct control of the finance
ministry. This placed an important responsibility in the hands of democratically elected officials, but politicians tend to focus more on the short run, i.e. upcoming elections, while inflation is a long-run consequence of
monetary mismanagement. Time and again, subservient central banks were pressured into paying more attention to the short than to the long run. Often, the result was high inflation, in some cases even runaway
inflation (see Box 10.2). To deal with this conflict, many countries have made their central banks formally independent of ruling governments and given precise tasks. The general thrust of these tasks is strikingly similar:
The long-run objective of low inflation comes first, while the shorter-run objective of stabilizing output and employment comes next.11
10.3.2 Instruments and Targets

Low and stable inflation, GDP growth, and high employment are the most cited objectives of monetary policy, and central banks around the world attach different weights to these goals. Yet none of these objectives is
directly under the control of the central bank. The challenge is to use available instruments to affect these variables, possibly by identifying and focusing on intermediate targets. This chain of causation between
instruments, targets, and objectives is the subject of the rest of this book. Logically, we will start with the instruments.
We have already discussed three instruments under the direct control of central banks: (1) the interbank or money market interest rate, (2) the supply of bank reserves, and (3) the required reserve ratio. Figure 9.7
showed that a central bank cannot control both the money stock and the interest rate. It can only choose a position on the schedule representing the derived demand for reserves. If it chooses the interest rate, it must
accommodate the demand for reserves corresponding to this interest rate. In Figure 10.5, this is shown as the horizontal Si schedule, which says that the central bank stands ready to let the money supply be at whatever
level is demanded at the chosen interest rate. The central bank could instead choose to fix the volume of reserves available in the market, but then it would have to accept whatever interest rate is necessary to equate
demand to that supply, shown as the vertical SM schedule.
Money growth targeting

That money growth drives inflation is, as we noted, the reason why central banks around the world are tasked with the price stability objective. It is also the reason why money growth was identified as a potential
intermediate target once that task was established. The logic is clear: achieving a suitably chosen rate of money growth is the way to achieve low inflation. That was not always the case, however. During much of the
1950s and 1960s, conventional wisdom was that central banks ought simply to maintain low and stable interest rates in order to stimulate investment and growth, with little or no regard for inflation.14 As it turned out,
with the interest rate set too low, money grew quite rapidly. Just as the theory predicts, this policy ultimately led to high inflation rates in the 1970s, as can be seen in Figure 10.6.
Fig. 10.5 Choice of the Monetary Policy Strategy

When the central bank fixes the interest rate in the interbank market, it effectively commits to supply whatever reserves are demanded at that interest rate. If it chooses instead to fix the supply of reserves, variation in demand is translated
into fluctuations in interest rates.
Box 10.2 Hyperinflation and Central Banks: Historical and Current Episodes
Time and again, a country—and in particular, the central bank of that country—will fall into the trap of letting the money supply grow at excessive rates for long periods of time. The consequence is always runaway
inflation. Hyperinflation was defined as a situation when monthly inflation rates exceed 50%, which means an annual rate of price increase of about 8,600%. On average, prices are rising by some 2% every day.
Almost always, excessive growth of the money supply is related to government budgets which have gone out of control. Rather than raising taxes, the government pays for its expenditures by borrowing directly
from the central bank in a process of monetary finance that eventually gets out of control.12
Many hyperinflations follow wars. This was the case in the 1920s in Germany, Greece, and Hungary. In Germany, prices rose by 29,524% during the month of October 1923. This meant an average daily
inflation rate of 19%. The situation was so bad that factories stopped production at lunch to pay their workers and give them time to go shopping, before prices had risen again and further eroded the purchasing
power of their pay. More recently, Serbia broke the world record when, in January 1994, daily inflation reached 60%. These episodes destroy the economy and put the basic fabric of society under severe strain,
sometimes even leading to war and civil unrest. The most recent episode of hyperinflation in Zimbabwe is recounted in Box 5.2.13
Fig. 10.6 Inflation in OECD Countries, 1970–2015

Inflation in OECD countries soared in the 1970s. It was reduced over the following two decades and appears to have stabilized at a low level since the late 1990s, with an uptick in 2008 followed by a sharp drop thereafter.
As an immediate reaction to the damaging inflation of the 1970s, many central banks began to target the rate of money growth. The strategy of monetary targeting rests on two links. The first one is the link from bank
reserves to bank deposits or, more generally, the money supply to M0 ( equations (9.2) and (9.3) in Box 9.3, respectively). If the money multiplier is stable, a choice of monetary base implies a choice of money supply,
and the central bank could control the rate of money growth, for whatever definition (M1, M2, etc.) is considered appropriate. The second link is the money neutrality principle developed in Chapter 5. In the long run,
inflation π is equal to the difference between the rate of money growth µ and the real GDP growth rate g:
(10.1)
Money growth targeting implies forecasting GDP growth g and allowing money to grow at a rate µ that delivers low inflation. The instruments are either bank reserves or M0, and the target is some
monetary aggregate, like M1 or M2.
Money growth targeting is sometimes credited for the successful disinflation of the 1980s. Yet it was also generally abandoned in the 1990s for a number of good reasons. One of them is that it was never clear which
monetary stock should be targeted: M0, M1, M2, M3, or some other aggregate? This control became difficult over time as the financial sector introduced innovation after innovation to banking, blurring distinctions
between monetary aggregates. As a result of this and other factors, the money multiplier became highly variable, so the money base M0 no longer provided a precise handle on the wider monetary aggregates. Figure
10.7. shows this in no uncertain terms. Perhaps the worst blow to monetary targeting was the instability of the public demand for money, which made the link captured by equation (10.1) between money growth, GDP,
and inflation unusable as a guide to policy.
Fig. 10.7 Money Multipliers (M1/M0) Before and After the Crisis
The three panels show the M1 money multiplier for the Eurozone, the UK, and the US. Its wide range reflects the ease of using deposits (UK) and preferences for banknotes (US, Eurozone). Its instability, especially around the financial crisis
in 2007–2010, renders monetary targeting difficult if not futile.
Inflation targeting
The practical difficulties associated with monetary targeting paved the way for alternative strategies for monetary policy. One of these, called expected inflation targeting, or simply inflation targeting, puts best
practice forecasts of future inflation at the centre of monetary policy. The central bank publicly states an objective (an acceptable rate of inflation), then makes forecasts for actual inflation over the policy horizon—two
to three years down the road, or roughly the delay between monetary policy actions and the resulting inflation rate. It then uses the monetary instruments described earlier, usually the interbank interest rate, to bring the
realized inflation rate as close as possible to the target. If the forecast (predicted) inflation rate is too high, the central bank will raise the interest rate, which will reduce the inflation rate, for reasons that will be presented
in Chapter 14. If future inflation is below target, the central bank will cut interest rates.
Why is the strategy described as expected inflation targeting? As time unfolds, new exogenous events can change our economic reality, sometimes dramatically. In the 1970s, oil prices jumped sharply and affected the
price level and inflation significantly—but in a one-off fashion. The arrival of China and India on the world economic stage meant falling prices for many imports for an extended period of time. Many countries have
experienced wrenching changes in nominal exchange rates in both directions. In some sad cases, war, civil unrest, or revolution may impair the economy’s productive capacity. These events will lead measured or realized
inflation to deviate for a time from forecasts that use all available information, but could not have anticipated the events described.
Expected inflation targeting has been adopted by central banks because it is easier to target interest rates than the volume of reserves. A quick look at Figure 10.5 makes this clear. The central bank can decide where to
move the economy along the money demand schedule, say point A. Under monetary targeting, the central bank would then determine the money supply that corresponds to that point represented by the vertical money
supply schedule SM. If it proves too difficult to control the quantity of money, the central bank can simply choose its price, the interest rate instead, represented by the horizontal money supply schedule Si.
Inflation targeting was first implemented by the Bank of New Zealand in the late 1980s. Since then, it has been adopted by a number of important central banks. While the three largest central banks, the ECB, the Fed,
and the Bank of Japan, do not target inflation actively, they refer frequently and publicly to inflation rates as objectives in major policy statements.15
The natural appeal of inflation targeting is that it is closest to the central bank’s responsibility to maintain price stability. With a simple, verifiable target, it is easier for the central bank to explain its decisions and to
reassure the public that its actions are driven by a clear and transparent commitment to price stability. But how do central banks actually know how to move the interest rate instrument to achieve the target? Inflation
targeting is similar to sailing: you move the rudder in small increments and observe where the ship heads to, which also depends on underwater currents and the wind direction. Because it looks two or three years ahead,
an inflation targeting central bank will do better by moving the interest rate in small increments, observing the outcome, and adjusting.
Invariably central banks will make forecast errors and miss their targets. Figure 10.8 shows the evolution of inflation in three countries whose central banks have explicitly adopted this strategy: Sweden, Great Britain,
and Poland. Each panel presents a ‘fan chart’ which displays each central bank’s forecast, plus an indication of how confident it is about its own forecasts (for example, the range of values between which the realized
value will be with 50% probability). In all three countries, actual inflation was significantly lower than the forecast value. As evidence mounted, central banks adjusted their rates. The Riksbank initially raised its interest
rate and then lowered it, even bringing it in negative territory.
Fig. 10.8 Inflation Forecasts of Central Banks and Outcomes
The panels display forecasts of national inflation rates published by the central banks of Sweden, the UK, and Poland in late 2011 for the following 3–4 years. These ‘fan charts’ are designed to emphasize the uncertain nature of inflation
forecasts. The innermost part of the fan indicates the range in which the central banks are most confident that inflation will be. The widest zone corresponds to a 90% probability. The farther out in time the forecast, the less certain it is. The
black curve in each figure represents the actual inflation rate and shows that all three central banks overestimated inflation in this period.
Sources: Monetary Policy Update, Riksbank (December 2011), Inflation Report, Bank of England (November 2011), Inflation Report, Narodowy Bank Polski (November 2011); OECD.
Exchange rate targeting

Central banks are sometimes tasked with fixing the country’s nominal exchange rate. We will see in Chapter 12 that in a world of international capital mobility, the central bank can only achieve this goal by giving up the
option of setting interest rates. This means that an autonomous monetary policy is impossible. The exchange rate effectively becomes the anchor of monetary policy. Why would a country fix its exchange rate? Quite
simply, if domestic control of monetary policy is politically difficult or impossible and a country is willing to give up the control of inflation to some foreign country—the country against which the exchange rate is being
pegged. The significance of a fixed exchange rate regime for the independence of monetary policy is the subject of Chapter 12, and the interaction of monetary policy, inflation, and the exchange rate regime is discussed
in detail in Chapter 15.
Box 10.3 The Taylor Rule
Monetary policy is a complicated process, usually taking place behinds closed doors and sometimes behind walls of secrecy. Some central banks have made exceptions but most do not publish minutes of
monetary policy committee meetings until weeks or months after they take place. Central banks very carefully craft their public statements because even small changes in perceived monetary policies can lead to
market turbulence.
Nevertheless, Prof. John Taylor of Stanford concluded in the 1990s, after observing central banks in action, that equation (10.2) was a good representation of what central banks do. The coefficients a and b
relate the relative importance attached to deviations of inflation and output from their targets; for example, in its original formulation, values a = 1.5 and b = 0.5 were chosen. This means that, holding the central
bank’s target inflation constant, a 1% increase in inflation leads to a 1.5 percentage point increase in the target rate. A boom leading to a 1% increase in output above trend would lead to an increase in the
interest rate of half a percentage point, or 50 basis points. Thus a typical recession, associated with a drop in inflation of 2 percentage points and 2 percentage point drop in output relative to trend, would thus
induce the central bank to decrease the interest rate by 400 basis points. Figure 10.9 confirms that rule tracks average central bank behaviour pretty well! Yet the Taylor rule can only be seen as a description of,
and not a prescription for, interest rate policy pursued by central banks. Large values of a correspond to central banks with strong anti-inflation views, while high values of b would match central banks that care
about stabilizing output.
Why would central bank raise interest rates when the economy is doing well? US Fed Governor William McChesney Martin, quoted at the beginning of the chapter, summarized the problem as follows: ‘The
Federal Reserve … after the recent increase in the discount rate, is in the position of the chaperone who has ordered the punch bowl removed just when the party was really warming up.’ The central bank sees
its role in stabilizing inflation and output, and that is what the Taylor rule is meant to capture.
10.3.3 The Taylor Rule

A simple way of summarizing the way central banks deal with the two objectives of inflation and output stabilization is the Taylor rule. The Taylor rule, which will be used extensively throughout the rest of the book,
states that the central bank raises the interbank market interest rate, denoted by i, when the inflation rate π exceeds its target inflation rate and when real GDP Y exceeds its current equilibrium or sustainable trend level
:16
(10.2)
with a > 0 and b > 0. Symmetrically, the central bank cuts interest rates when The target inflation rate is meant to capture the central bank’s price stability objective. The trend level
corresponds to the level of GDP observed if the economy is operating at sustainable level and unemployment is at its equilibrium level, as derived in Chapter 3; it is often called potential output. The term
is called the output gap and represents the difference between GDP and its trend.17 It is positive when the economy is operating at levels higher than potential or trend capacity, and is negative when
the economy is operating below its potential. The parameters a and b reflect the importance that the central bank gives to each objective. We can think of the Taylor rule as determining where the (money supply)
schedule S is placed in Figure 10.7. The Taylor rule anchors interest rate policy around a level that is called the neutral or natural interest rate. It is the interest rate that the central bank would want to set if both
inflation and GDP were at their desired levels. Box 10.3 explains the Taylor rule in detail.
For each country, Figure 10.9 compares the actual interest rate and the one predicted by the rule. Given the extreme simplicity of the rule, we would not expect a tight link. Remarkably, the Taylor rule gives a reasonable
picture of what central banks actually do. We will use it as a tool to represent monetary policy—the way that the central bank’s interest rate reacts to inflation and output. The main differences appeared during the crisis
in countries in which the output gap was negative and unusually large. This would have implied a negative nominal interest rate, which is implausible and unlikely under normal conditions.18 This situation is known as
the zero lower bound and has led some central banks to drive their interest rates to zero, or close to zero.
Fig. 10.9 Examples of Taylor Rules
The figures compare actual central bank policy interest rates with those predicted by the Taylor rule (10.2) with a = 1.5 and b = 0.5, and where the neutral interest rate is chosen on the basis of past experience or central bank statements.
Sources: OECD, Economic Outlook; IMF, International Financial Statistics.

We have not been very explicit, so far, about the how central banks eventually affect inflation. A full understanding of this issue is provided in Chapter 13. At this stage, we only note that monetary policy is meant to
affect the spending behaviour of households and firms according to principles developed in Chapter 8. We have seen that the interest rate affects the choice between consuming and saving as well as the borrowing
costs of firms and Tobin’s q. These are indeed the transmission channels of monetary policy. Throughout the rest of the book we will focus on four main transmission channels: (1) the interest rate; (2) asset
prices; (3) bank credit; and (4) the exchange rate. We now review the first three, leaving the fourth one for the next chapters.
10.4.1 The Interest Rate Channel

Chapter 9 has shown how central banks can control very short-term interest rates. In Chapter 8, however, we saw that the longer-term future matters for households and firms when they buy cars, remodel their kitchens,
refinance their debt, or invest in new equipment. It stands to reason that longer-term interest rates will shape their economic choices. In the terminology of Chapter 7, the central bank controls the short end of the yield
curve, while it is the rest of the curve that matters for spending decisions of the private sector.
How can the central bank affect the longer end of the yield curve? In Chapter 7 we saw that longer-term interest rates reflect market expectations of future shorter-term rates; this is the expectations theory of the term
structure. To exploit the interest rate channel, it is not enough for central banks to lower or raise the very short-term interest rate; it must shape expectations of future interest rates. For instance, in Figure 10.10, we see
that the entire yield curve shifted downward between 2009 and 2015, but in varying degrees at different points in time. Like central banks all around the world, the ECB responded promptly to the financial crisis by
lowering its policy (refi) interest rate. The longer end of the yield curve moved down as well, reflecting expectations that market rates would stay low for a while. Eventually, the ECB brought its deposit rate to zero by
mid-2014 and undertook further measures to flatten the yield curve, as explained in Section 10.4.4. This, combined with a global growth slowdown that included China, led market participants in 2015 to expect significantly
lower interest rates into the distant future, meaning a significant reduction of borrowing costs at long maturities.
These examples illustrate two important aspects of the interest rate channel of monetary policy. First, the central bank must affect long-term interest rates—those that matter for monetary policy transmission to private
spending decisions. Second, to that effect, the central bank must be able to convince the markets of its future actions. Put differently, communication is an integral part of monetary policy.
Fig. 10.10 Euro Area Yield Curves

Each curve displays the interest rate on government bonds from the euro area at various maturities ranging from the short term (here one month) to the long term (30 years). The yield curves stem from the years 2009 to 2015 and show that
monetary policy—which influences the extremely low short-term interest rates directly—is not always able to achieve the desired interest rates at longer maturities, especially during deep economic downturns.
Sources: ECB.
As inflation targeting has gained popularity for implementing monetary policy, central banks have honed their communication skills. They have done their best to win the trust of financial markets and the public at
large, partly by establishing a track record of competence, partly by becoming increasingly transparent. They publicize forecasts of inflation—to give an idea of what they might want to do—and have even begun to
reveal the interest rates they anticipate setting. This new approach is followed by the central banks of New Zealand, the Bank of Norway, the Swedish Riksbank, and the Central Bank of Iceland. The Federal Reserve
Board of the United States publishes the interest rates foreseen by all individual members of its policy committee.
10.4.2 The Asset Price Channel

Chapter 7 also showed how interest rates affect the market value of assets. For example, when the central bank raises the interest rate, stock and bond prices tend to decline, as do housing prices. This reduces private
wealth. Households react by consuming less, increasing their savings, and at least partially restoring their wealth positions. Similarly, declining stock prices lead firms to postpone issue of shares, depriving them of the
means to carry out investment projects. The asset price channel describes the effect of monetary policy on asset prices and economic outcomes.
10.4.3 The Credit Channel

Another way for monetary policy to affect the economy is via the availability of credit independently of its cost. The credit channel reflects the fact that if the lending channel of the banks is impaired, borrowers may
have difficulty getting a loan even if lendable funds are plentiful. For a number of reasons, commercial banks may be reluctant to issue loans, even if refinancing costs are low. For regulatory and prudential reasons, they
need to hold liquid assets in proportion to the credit that they grant. Banks may also become worried about the general economic, financial, or political situation, and may be wary of lending money. They may try to
increase their net worth, or to rebuild it after sustaining losses. This mechanism was important in Europe during the financial crisis, as the sovereign debt crisis spread and banks began to mistrust each other regarding
their exposure to countries with debt problems.
10.4.4 The Zero Lower Bound

Figure 10.4 shows that the ECB’s main policy interest rate was lowered to zero in 2014. Many other central banks, including those of the USA, the UK, and Japan, have also hit what is called the zero lower bound. Indeed,
until then, it was believed that interest rates can only be positive. Why else would anyone lend money?19 When interest rates reach the zero lower bound, central banks cannot reduce them any further, and the use of
the interest rate instrument is impaired. In the 2010s, many central banks explored alternative instruments, known as non-standard or unconventional policies, designed to restore effectiveness to various monetary
transmission channels.
Negative interest rates
In order to keep the interest rate channel alive, a few central banks (in the Eurozone as well as in Denmark, Sweden, Switzerland, and Japan) have moved their policy rates below the zero lower bound. This is only possible
because they are monopolist suppliers of bank reserves and because commercial banks need to hold those reserves. These experiments are too recent to lead to solid conclusions, but a couple of observations can be
made. First, the lowest rate achieved so far by any central bank is –0.75% (Switzerland and Denmark). Negative deposit rates can be seen as a penalty or tax on banks for holding reserves. One potential limitation of this
policy is they may lead banks simply to hold reserves in cash in their own vaults, thus saving the negative interest (but with potentially significant insurance and security costs). Second, banks themselves face a zero
lower bound issue, and it is very unlikely that they will charge negative lending rates (see Box 9.4).
Quantitative easing
After the crisis hit, the Fed and the Bank of England adopted a new strategy: they began to conduct large-scale open market purchases directly in financial assets markets.20 This policy, known as quantitative easing,
was later adopted by Japan in 2014 and the ECB in 2015. While not new—central banks regularly engage in open market operations in short-term government debt—quantitative easing represents a firm commitment by
the central bank to purchase government debt, private bonds, or even equities over a relatively long horizon of two years or more. Because these assets remain on the central bank’s balance sheet, quantitative easing
creates money, exactly as in other open market operations. What is special is the size and the long, pre-announced duration of the undertaking.
Quantitative easing operates through all monetary policy channels. Banks find themselves awash with money; yet this money is only M0, and the success of quantitative easing depends on the credit channel, the
willingness of banks to lend out further funds in the aftermath. Figure 10.11 shows how banks had sharply reduced refinancing at the central bank between 2012 and 2015, repaying significant volumes of loans (‘LTROs’)
from the ECB. Quantitative easing (‘Securities held for monetary purposes’) is an attempt to restore the ECB balance sheet to its pre-crisis trend, which as of mid-2016 had not yet occurred.
Quantitative easing also acts through the asset price channel by raising bond and share prices. Raising the price of longer-term bonds reduces long-term interest rates, flattening the yield curve as in Figure 10.10. The
evidence in the Eurozone is that, indeed, yield curves have been compressed along the zero lower bound (or below) and that asset prices have been raised. Some analysts have raised concerns that quantitative easing
could inadvertently trigger asset price bubbles, as described in Chapter 7.
Fig. 10.11 Quantitative Easing in the Euro Area

The size of the ECB balance sheet grew steadily until the global financial crisis in 2008–2009. In 2011–2012 the ECB sharply increased long-term refinancing to banks (LTRO) to stimulate lending. Banks however began to repay these loans
after 2013, despite low refinancing costs. Quantitative easing, visible after 2015, has yet to restore the ECB balance sheet to its trend before the financial crisis.
Sources: ECB.
10.5 Financial Stability as a Prerequisite for

Monetary Policy
10.5.1 The Inherent Instability of Fractional Reserve Banking

We have seen that modern money need not be made of precious metal to have value. The essential ingredient is trust. As long as others are willing to accept it as a means of payment, be it for something bought in a
store or for one’s own hard work, it will have value and be used, even though most of it is not even created by the government. The fact that its value depends on widely shared trust implies that money is a public
good—meaning that while individuals enjoy the benefits of a trustworthy money, they will tend to underinvest in maintaining its trustworthiness. Ultimately, this is why governments or central banks regulate and
supervise banks. Yet despite precautions, crises of confidence occur from time to time.
Crises of confidence arise when bank customers doubt the safety of their deposits and attempt to convert them into cash. Because commercial banks hold only a small proportion of deposits in currency or reserves at
the central bank, commercial banks have then no other choice but to suspend withdrawals or even close. Not only does this infuriate depositors who feel defrauded, it undermines the whole economy. Often bank runs
start with one failing bank and quickly spread to the whole banking system. It is not much different from the behaviour of people in a crowded theatre with a limited number of exits. Even if the objective likelihood of a fire
is low, people who catch the slightest whiff of smoke will want to be the first out the door to avoid being trampled by the others. To prevent a financial and economic meltdown, central banks must intervene decisively to
prevent the crisis from spreading and, at worst rescue the banking system. Box 10.4 shows that the problem is not merely a matter of panic among small depositors. The recent crisis showed that in the UK, US, and
Ireland, bankers themselves can behave like theatre-goers who rush to the door the instant their trust in each other evaporates. In what follows, we will see how central banks can serve as ultimate fireman in such cases,
and use some quite large firehoses to put out the fire.
Box 10.4 Bankers’ Mistrust, Eurozone Edition
In August 2007, banks’ trust in each other literally collapsed. Too many non-performing loans had been made to families to buy houses and the underlying collateral was based on a bubble; in other words, the
loans were not backed by good assets. Those bad mortgages had been part of a massive securitization wave which had created pools of loan repayments of thousands of homeowners. Some of those
securities lost a lot of value. Not being able to know which banks owned those securities or not knowing the risk involved, banks immediately suspected that other large banks were also facing similar serious
risk losses. On 9 August, suspicion began to spread that some banks could not repay their short-term loans. Within minutes, interbank markets seized up in the USA, in the Eurozone, in the UK, and around the
world.
This problem spread to Europe, even though European banks had not engaged directly in financing US home purchases. Figure 10.12 displays two different interest rates on 3-month interbank loans.
Eurepo® is the rate at which banks lend to each other using high quality assets as collateral; these loans are considered very safe. Euribor® is the rate that applies to loans without collateral, based only on trust
in the borrower. Since trust can always be misguided, the Euribor® is higher than the Eurepo®, but until August 2007 the two rates moved together in lockstep, with only a small spread between them.21 After
August 2007, they parted ways abruptly.
The spread narrowed in the USA after 2010, but not in the euro area. In fact, it even widened during 2011, when investors called into question the underlying value of the government debt of some European
governments. Until then, government debts were considered to be perfectly safe, but starting with Greek debt and spreading to many other euro area countries, this assumption was challenged by the poor
budgetary situation of these countries. Since banks hold large amount of ‘safe’ public bonds, partly to use them in the interbank market, banks began to worry about each other. Interbank markets froze up again,
just as they had in the financial crisis. One particularly egregious example was Cyprus, whose two main banks were large holders of Greek government debt. In no time, depositors, especially offshore
depositors from Russia, began to withdraw their funds, leading to the collapse of the Cypriot banking system in 2012.
Like many other central banks, the ECB worked hard during the financial crisis to keep banks afloat. In particular, it expanded massively its normally modest long-term re-financing operations. Although central
banks normally keep the supply of reserves on a tight leash by offering only very short-term loans, the ECB offered liquidity at up to 3-year repayment horizons, and in whatever quantity banks asked for. The
result: massive borrowing of central bank money by troubled commercial banks trying to avoid being shut out of the interbank market.
10.5.2 Confidence-building Measures

Given the severity of the consequences of a bank failure or a fully-fledged financial crisis, it makes sense for the government—either through the central bank or regulatory agencies—to implement measures well in
advance of a crisis. These confidence-building measures are the first line of defence against financial instability.
Fig. 10.12 Mistrust in the Interbank Market (3-month interest rates)

Euribor® is the interest rate that applies to uncollateralized loans in euros among large banks. Eurepo® concerns loans that are guaranteed by ‘best quality’ collateral. The difference is a measure of the risk that banks perceive in dealing with
each other without the protection of collateral.
Sources: European Banking Federation, http://www.euribor-ebf.eu/
Deposit insurance
One such measure is deposit insurance. In such schemes banks insure themselves against bank failures. They pay a premium for this insurance—which is generally passed on to the customers as fees or lower
interest rates. Up to a certain amount, the value of deposits of current or savings accounts is guaranteed. If the insurance programme is credible, smaller depositors will not rush for the exit and withdraw their funds at the
slightest hint of crisis. One problem with deposit insurance is that it may engender a lack of diligence, or even risk-seeking, on both sides of the table. Customers will tend to go to banks which pay the highest interest
rates and will tend not to ask questions about the quality of a bank’s management or its lending activities. Banks may engage in riskier lending to attract those customers.
To limit such risks, deposit insurance rarely applies to large depositors (usually more than €100,000) or to large creditors of banks. If the bank fails, large deposits tend to be lost, even though the deposits of the smaller
savers are protected. Sometimes larger depositors and lenders to banks are asked to forfeit some of their loans in order to save the bank from collapse by converting them into ownership equity claims on the bank
(shares). This is called a bail-in. The credible risk of a bail-in incentivizes larger investors in banks to exercise diligence and critical judgement when evaluating the banks that they invest in. This will be explained in more
detail shortly.
Bank regulation and supervision

Money is created when commercial banks grant loans to their customers, with the expectation that principal and interest will be repaid. It is a fundamental and unalterable fact that banks have less information about their
customers’ creditworthiness than the customers themselves do. A borrower will tend to misrepresent circumstances that would otherwise lead to a refusal of credit. This fundamental information asymmetry has
wide-ranging implications that affect virtually all financial dealings.22
One direct implication of information asymmetry is that banks cannot avoid making and owning non-performing loans. A bank that makes too many bad loans can go bankrupt, and in this case the depositors—who are
not investing, but simply banking—may lose their money. Depositors can thus never be perfectly certain that their money is safe and, because bank deposits are only as safe as the bank itself, the value of money itself is
constantly at risk. Just as confidence in banknotes and coins rests on the quality of the central bank that issues them, so the confidence in money created by the commercial banks is based on the expectation that those
funds are freely and immediately convertible into currency at any time. History is full of bank failures which turned into bank panics, as worried depositors attempted to withdraw cash from their bank accounts, not only
from the failing bank but from all financial institutions—just to be safe.
One bank failure is painful for its depositors, but not lethal for the economy. The greater risk is that the whole banking system could fail as a chain reaction, called contagion, takes hold which would bring healthy
banks down along with the bad ones. Systemic risk, as the phenomenon is called, arises because banks (and, more generally, financial institutions) hold each other’s liabilities. For example, correspondent balances are
the plumbing of the modern economy. They allow banks to deal with each other on the interbank market. Should one bank go bankrupt, its liabilities to other banks lose value, and this frequently leads to more
bankruptcies. When many banks engage in risky behaviour, excessive risk-taking is bound to result in even more frequent crises. In such cases, government and the central bank are hostages to the state of the financial
system and have no choice but to bail out the banks, for fear of far worse outcomes.
One response is for central banks or supervisory agencies to get ahead of the curve by closely supervising commercial bank practices, imposing regulations on inappropriate bank behaviour and, when needed,
requiring banks to correct the problem. If the instruction is not heeded, the misbehaving bank may lose its licence to operate. A second response is to impose macroprudential measures, such as increasing required
bank capital (issuing shares) as a function of risk taken. Some macroprudential policies even raise capital requirements in boom times and relax them in recessions. This is explained in more detail later and in Box 10.5. A
third response is for central banks to maintain a large degree of uncertainty as to what they would do in case of bank failure. This explains why central banks deny in public that they are the lenders of last resort. They
hope that depositors will be encouraged to keep an eye on their banks in good times, and withhold their patronage if there is cause for concern.
Financially healthy banks
Another way to motivate banks to pursue prudent lending behaviour and financial stability is to ensure that owners and large creditors of the bank pay for their mistakes. If the threat of bankruptcy and bail-in is credible,
banks are more likely to undertake measures to make runs less likely. Among other things, this means making fewer risky loans, holding fewer high-risk securities, and maintaining a sufficient amount of capital (net worth)
in their balance sheets. Banks may also be required to do so by national or international regulations. Like all ongoing businesses, banks need positive net worth to survive. The liquidity and value of assets can fluctuate
significantly, so net worth represents a cushion necessary to protect the value of deposits from swings in security prices and loan valuations.
The consequences of a significant valuation loss for a commercial bank are shown in Figure 10.13. In this example, the bank suffers a significant loss of asset value; this loss will reduce the value of the bank, because
the value of assets in the balance sheet must always equal liabilities plus equity, and any loss of asset value is ‘paid for’ by the owners of the bank; i.e. absorbed by net worth. To maintain confidence in its operations, it
is especially important for banks to have sufficient net worth (also called ‘equity’ or ‘bank capital’ in the jargon of bankers and their regulators).
Sometimes after an event such as that shown in Figure 10.13, the bank is forced by the regulator to increase its net worth, in light of the reduced value of its assets. This process is called recapitalization. Figure
10.14 shows how a recapitalization would look, starting from the second balance sheet in Figure 10.13.
Net worth is increased by bringing in resources from outside the bank. Most commonly, the regulator can force bank owners to invest fresh funds (new capital) or issue new shares to the public. In some extreme cases,
only the government or the central bank has the resources to recapitalize a bank. The bank must then be partially or completely nationalized. The troubled bank can also improve its net worth as a fraction of its assets by
simply selling securities and paying off debt or refusing to roll over lending to its customers. This option is painful because it reduces credit to the real economy and usually comes at the worst possible time.
Box 10.5 Basel Regulation and Capital Adequacy
As any honest banker will admit, all assets are not created equal. In Chapter 7 we saw that prices for financial assets traded on markets can fluctuate a lot. Other assets have stable values—the most stable
being reserve deposits at the central bank—but do not yield much income. Still other assets such as loans to businesses or households have a nominal or ‘book’ value—the amount of the loan—but their value
will decline if the loans are non-performing. Bank loans are illiquid, meaning that they are hard to convert into cash or reserves in times of crisis. As evident from Figure 10.12, changing asset values have
implications for the net worth of a bank.
Bank regulators distinguish between riskier and less risky assets in determining capital requirements. Banks which have made risky loans and hold risky securities are more likely to be subject to a bank run
and should show more capital in their balance sheets. In 1988, the Basel Committee on Banking Supervision published a set of recommendations known as Basel I. These voluntary recommendations were
first applied by the G10 countries, followed by many others.23 A second, more complex set of rules followed (Basel II). The common feature of Basel I and II is a minimum capital requirement, but with different
assets requiring different amounts of capital. Assets were divided into five risk groups and safer assets were given a lower weight when computing the total value of assets. Capital is costly for banks, because it
must be matched by assets which are not financed by loans, or leveraged. The rules are meant to deter excessive lending and risk-taking. A key difference between Basel I and Basel II is that the latter did not
rely on an arbitrary risk classification but left it to the banks themselves to assess the riskiness of their assets. In particular, the rules overestimated the extent of risk diversification (as explained in Chapter 7).
Overall, a bank is required to have net worth or bank capital in the amount of at least 8% of its risk-weighted assets.
The financial crisis of 2007–2010 revealed a number of shortcomings of Basel II. First, banks’ own risk assessments were far too optimistic concerning the overall riskiness of balance sheets. Second, the
capital requirement was too low to deal with large shocks, forcing governments to rescue banks with taxpayers’ money. Third, the crisis came not just from the asset side (bank loans) but also from the liability
side as banks had borrowed short-term directly from the markets, thus magnifying the maturity mismatch problem. Finally, a number of banks had grown so large that they were not just too big to fail but also too
complex to manage and supervise. This led to Basel III, adopted between 2011 and 2013, and additional proposals by the newly created Financial Stability Board, also based in Basel. These changes are to be
implemented by 2019. Capital requirements are higher than Basel II and even higher for larger ‘systemically’ important financial institutions. Short-term borrowing, as opposed to traditional bank deposits, is
also restricted—this is the so-called leverage ratio, to prevent banks from expanding their assets excessively when financed by deposits or short-term borrowing. The Basel III rules have made loans more
expensive than before, and are more likely to apply than the reserve ratio.
Because of the information asymmetry problem, bank regulation is common throughout the world. Outsiders (depositors and other lenders) have inferior information and owners should not be allowed to operate banks
on a shoestring. For decades, bank regulators have cooperated internationally to prevent financial institutions from playing one country’s regulation against another’s. These so-called ‘Basel rules’ impose minimal
capital adequacy (see Box 10.5), by adjusting the capital (net worth) requirements to reflect the underlying riskiness of the assets.
10.5.3 The Central Bank as Lender of Last Resort

Contagion-driven, systemic bank failures represent one of the most dangerous threats to modern market economies. Within days or weeks, a large fraction of the banking system may cease to function. This cripples most
daily transactions and precipitates countless bankruptcies and plant closures. Unemployment rises sharply and standards of living fall precipitously. GDP declined by 10% or more in the wake of systemic bank collapses
in Korea in 1998 and Argentina in 2002, especially hurting the poor who lost access to their bank accounts and sometimes lost everything. Europe and the US in 2009 were spared bank runs and the resulting economic
dislocation because central banks and government acted promptly. However, bank runs occurred later in Greece and Cyprus because the ECB refused to absorb losses that would then have to be shared by the other
Eurozone member countries.
Fig. 10.13 The Effect of Loan Losses on a Bank’s Balance Sheet

In this hypothetical situation, a bank’s loan losses cause the entry ‘Loans’ to decline by €40 bn, shown in Panel (b), which causes net worth (the value of the bank on the basis of the balance sheet) to decline by €40 bn, or 80%.
Fig. 10.14 The Effect of a Government Recapitalization on a Bank’s Balance Sheet

In this hypothetical example, the bank receives €50 bn from the government, paid for in cash, which is deposited at the ECB, while the government’s ownership of the bank (and its net worth) has increased by the same amount.
The point is that central banks are not just in charge of price stability; they cannot stand simply by when the banking system collapses. They are also entrusted with the task of preserving financial stability. In normal
times, they limit money growth and monitor the quality of bank loans. In bad times, when bank deposits threaten to vanish, they alone can produce unlimited quantities of money. When they do so, they act as lenders
of last resort. In principle at least, the central bank is expected to provide failing banks with a sufficient monetary base to avoid immediate bankruptcy and reassure depositors that they can continue to withdraw money
from their deposits normally. Sometimes they may even recapitalize banks, or help governments to do so. But the lender of last resort function must be handled with great care. Box 10.6 summarizes Bagehot’s Rules, a
remarkable set of instructions formulated more than a century ago to help solve the policy dilemma posed by a general bank crisis.
Box 10.6 Bagehot Rules!
Central banks frequently find themselves in a difficult position. On the one hand, commercial banking systems can be unstable in the face of large, sudden withdrawals, which may be based as much on
irrational fears or misinformation as real threats. Supplying liquidity in crises can spare considerable pain and suffering. On the other hand, if financial institutions believe that the central bank will always bail
them out when in trouble, they may well take on excessive risks. This is the moral hazard problem. The danger is that crisis management in the last resort can itself become a source of future crises.
Walter Bagehot (1826–1877), a renowned British financial economist of the nineteenth century and the editor of the Economist, understood these issues all too well. Bagehot (pronounced ‘badget’) proposed
in 1873 the following principles, to be made public as policy of the central bank in times of crisis:
Lend only against marketable collateral. This forces banks to hold an adequate amount of safe assets (with low returns) alongside risky loans (with higher returns).
Lend in large amounts at a higher rate than the market interest rate. Large amounts are meant to stop the bank run, higher rates are meant to make the operation punitively costly.
Sell or liquidate insolvent banks, with losses to be borne by their owners. The managers and shareholders should bear the cost of imprudent lending as a way of reducing the moral hazard problem.
The ‘Bagehot rules’ are still used to this day! The difficulty lies in practising what you preach. In recent bank bailouts in Europe, bank owners and bond holders have consistently received special treatment or
bailouts from national governments or central banks, usually justified by arguments that national banks and their reputations must be protected. These arguments by the owners and creditors are usually self-
serving, and should be considered as such.
We have noted that banking is a risky business. One reason why banks limit the risk that they take, and the returns that come along with risk-taking, is that they want to avoid bankruptcy. But if central bank lending in
the last resort always saves them from bankruptcy, they will have less reason to behave prudently, which undermines the banking system and the stability of the financial system. This is the moral hazard problem;
information asymmetry comes from the fact that central banks do not know how much risk commercial banks take and that, if they expect a bailout, banks have an interest in taking more risk than they really should.
10.5.4 Technological Innovation and Financial Stability

Banks are constantly devising new ways of satisfying the financial needs of their customers. Most innovations yield some benefits to both the consumers and bankers, but they are also often driven by the banks’ desire
to evade regulation and increase profits. In general, a central bank’s objective of reining in money and credit growth runs counter to individual banks’ interest in increasing profitability. Similarly, banking regulation may
aim to protect customers by limiting the range of banking activities, including risk-taking, but these measures will also reduce bank profitability. Pressed by competition and prompted by continuous technological
innovations—in financial instruments, computer power, and communications systems—banks constantly innovate by exploiting loopholes in existing legislation. As a result, monetary control is weakened and banks
may become more vulnerable, or fragile, because they have borrowed too much or have accumulated too little net worth.
Many banks borrow more than they should, reducing equity to asset ratios below levels set out by the Basel rules. One trick associated with the financial crisis was to hide risky investments by creating legal
subsidiaries—called Special Purpose Vehicles or SPVs—that kept the riskiest assets out of sight of bank supervisors, often offshore. These SPVs are part of what is called shadow banking. (Hedge funds, discussed later,
are another part of shadow banking.) When the crisis hit, many risky SPVs went bankrupt and the banks themselves had to refinance them or absorb their losses. All of a sudden, the riskiest assets were visible for all to
see.
The temptation for banks to trade is great, and for this reason financial regulators must be ever-vigilant. Advising President Obama, Former Federal Reserve Chairman Paul Volcker wrote that ‘adding further layers of
risk to the inherent risks of essential commercial bank functions doesn’t make sense, not when those risks arise from more speculative activities far better suited for other areas of the financial markets’ ( New York Times ,
30 January 2010). His proposal, called the Volcker Rule, was incorporated in US bank regulation and was due to take effect in 2012, but as of the summer of 2016, had not yet been fully implemented. It explicitly
prohibits proprietary trading and other risky activities by commercial banks. Needless to say, banks continue to resist and successfully lobby against such legislation.24
Summary
1 The central bank is at the centre of monetary policy. Facing the derived demand for the monetary base, central banks can decide on which quantity of money to supply, or to supply whatever quantity of money is needed to achieve a
particular interest rate.
2 Monetary policy is carried out through open market operations, lending to banks, or adjusting a legally binding reserve ratio, when it exists. Most central banks conduct open market operations or lend to banks against high-quality
collateral in auctions.
3 The neutrality principle implies that inflation is ultimately driven by money growth. This has led most countries to delegate the responsibility of achieving and maintaining price stability to their central banks. Central banks are usually
also expected to care about output growth and employment, which are affected by monetary policy in the shorter run.
4 Combining these two objectives is not always easy, so central banks have developed monetary policy strategies, which have evolved over time. The money growth strategy proved its mettle in bringing inflation down following the faulty
low-interest-rate strategies of the 1960s and 1970s. With low inflation and rapid changes in banking and finances, the money growth strategy gave way to inflation targeting.
5 The Taylor rule describes monetary policy as setting the interest rate in response to deviations of inflation from its designated target and to the output gap. Despite its simplicity, the Taylor rule offers a reasonable, reliable interpretation
of what central banks do.
6 A healthy financial system is essential for a robust economy and for the effectiveness of monetary policy. Because confidence in money is equivalent to trust in the banking system, it is essential that governments and central banks take
measures to prevent bank failures and financial crises. Public deposit insurance, effective bank regulation and supervision, and well-capitalized financial institutions are means towards this end.
7 Because commercial banks supply money and credit to an economy, their failure cannot be taken lightly. An isolated bank run can increase the risk of generalized contagion. This is why central banks must be ready to act as lender of last
resort, providing liquidity while sanctioning imprudent practices that are responsible for the crisis of confidence.
8 Financial institutions are constantly innovating by introducing new products and services. This is good for consumers and can increase bank profits. Yet some innovation increases the fragility of the banking system and can impede the
execution of monetary policy.
9 Preventing crises such as that which preceded the Great Recession from recurring will require preventing banks from engaging in risky investments with excessive amounts of borrowed money. It may require prohibiting trading in risky
assets altogether.
Key Concepts
monetary policy
payments system
reserves ratio
bank reserves
refinance
money market
derived demand
interbank market, open market
open market operations
non-performing loans
collateral
objectives
instruments
targets
money multiplier
money growth targeting
expected inflation targeting, inflation targeting
Taylor rule
target inflation rate
potential output
output gap
neutral interest rate
basis points
zero lower bound
transmission channels of monetary policy
asset price channel
credit channel
non-standard or non-conventional policies
quantitative easing
public good
deposit insurance
bail-in
information asymmetry
contagion
systemic risk
macroprudential
recapitalization
capital adequacy
lenders of last resort
Volcker Rule
Exercises
1 Suppose that the long-run growth rate of GDP is 3% and that the reserves ratio is 10%. What should be the money growth target if the central bank wants to achieve an inflation of 2% in the long run? How quickly, then, should it allow
banks reserves to grow?
2 Define and explain the distinction between objectives, targets, and instruments of monetary policy. Give some examples of conflicts between different instruments, and between different targets. Do you think it is necessarily better to
have many or fewer instruments and targets?
3 Table 9.1 shows that the size of the monetary aggregates relative to GDP varies from country to country. Explain why and how these ratios may reflect the degree of financial development of the country.
4 In some countries the reserves ratio is not set by the central bank, rather it is left to the discretion of commercial banks. Does it matter for the control of the money stock? Does the same problem exist where the central bank controls
interest rates instead?
5 Figure 10.5 highlights the difference between money growth and inflation-targeting strategies. Consider an increase in the demand
for money. How will the central bank react under the money growth strategy by keeping SM unchanged? Under the inflation-targeting strategy if Si is kept unchanged?
6 Sterilization is defined as a procedure in which the central bank conducts an open market operation and simultaneously conducts another to prevent the first from affecting the money supply. Consider the example of a foreign exchange
market intervention in which the central bank buys or sells its domestic currency against foreign assets. Using Figure 10.1, show the effect of a purchase of €500 million foreign exchange paid for with bank reserves, on the following central
bank balance sheet items: (1) foreign assets, (2) deposits by commercial banks, (3) net worth. What should happen to the money supply? Explain what type of open market operation could be carried out to sterilize the foreign exchange
market intervention.
7 During the financial crisis, some central banks bought large amounts of ‘toxic assets’ (packages of poor quality loans) from banks in financial trouble. Use Figure 10.2 or Figure 10.13 to describe these actions and their effect on the bank in
distress. Now imagine that these assets lose value: how will that be treated in the central bank’s balance sheet?
8 Why is deposit insurance a way of reducing the chances of a bank run? What are the limits of this scheme?
9 Consider the Taylor rule (10.2). Assume as in the text that a = 1.5 and b = 0.5 and that the neutral interest rate is 4%. What interest rate is implied by the rule when inflation is 2% above target while the output gap is zero. What if the
output gap is 3%? And −2%?
10 The target inflation rate in Sweden is 2%. Looking at Figure 10.8, what do you expect the central bank (the Riksbank) to do over the years 2012–2014?
11 Box 10.1 describes the open market operations of the ECB. Explain carefully why the marginal lending facility establishes a ceiling on the market rate and why the marginal deposit facility establishes a floor.
Essay Questions
1 The central bank mandate of the Eurosystem sets price stability as its overriding priority. In the USA, the Federal Reserve System is legally committed to pursue both ‘stability of the currency’ and a ‘high level of employment’. Compare
and discuss these mandates.
2 In prisoner-of-war camps it was common to use cigarettes as money. Explain why this makes sense. Do you see any problems with ‘cigarette cash?’ Why might the administrator of such a camp prefer this to, say, using regular money or
printed coupons?
3 Why do central banks need to have targets? Discuss and evaluate the various possible targets.
4 The Governor of the Bank of England is obliged to write a letter to the Chancellor of the Exchequer (the Finance Minister of the UK) to explain when it misses the inflation target. Read the Governor’s letter from August 2016:
http://www.bankofengland.co.uk/monetarypolicy/Documents/pdf/cpiletter040816.pdf and the response of Chancellor Hammond: http://www.bankofengland.co.uk/monetarypolicy/Documents/pdf/chancellorletter040816.pdf What are the
Governor’s excuses for missing the target and what do you think of the Chancellor’s response?
5 In order to maintain financial stability, one task of the central bank is to intervene as lender of last resort. One of the trickiest parts is to determine if a bank is insolvent (it is really bankrupt and cannot recover) or just illiquid (it has no
fundamental problem but is running out of cash). Explain why it matters and how this distinction might be made.
1 In a radio address on 12 March 1933, US President Franklin D. Roosevelt spoke to the American people on the national banking crisis. To listen, visit http://millercenter.org/scripps/archive/speeches/detail/3298.
2 In a speech to the Investment Bankers Association of America on 19 October 1955. William McChesney Martin was Chairman of the Board of Governors of the US Federal Reserve System from 1951 to 1970.
3 In Box 9.3 we showed that if households retain a positive fraction of their overall money holdings as cash, the money multiplier will be strictly less than 1/rr.
4 In the days before fiat money, banknotes promised to pay gold or silver when redeemed; now they offer little more than another banknote to replace it!
5 In principle, central banks could acquire gold bullion or other precious metals, stocks, real estate, or loans to governments, but generally do not. They do, however, continue to amass significant foreign exchange reserves generally held in interest
bearing form, as short-term debt of highly reputable governments.
6 Chapter 9 explained that the public demand for money translates into a derived demand for high-powered money by banks.
7 Markets use ‘refi’ as shorthand for Main Refinancing Rate and ‘repo’ for Repurchase Operations.
8 Full details are provided by the ECB on its website at https://www.ecb.europa.eu. Each central bank also presents in great detail its own procedures. For a list of all central banks, see http://www.centralbanking.co.uk.
9 Central bank lending is usually of short duration, with commercial banks continuously borrowing and repaying and re-borrowing in a process called ‘rolling over’. Central banks can also withdraw liquidity by borrowing from banks—by offering a
sufficiently high interest rate.
10 The transmission mechanisms of monetary policy are described in more detail in Chapter 11. At this point it is enough to recall from Chapter 8 that an increase in interest rates reduces investment spending by increasing the opportunity cost of
invested resources and lowering the value of capital already installed. It also reduces consumption indirectly by reducing the present value of wealth.
11 For instance, monetary policy at the Norwegian central bank ‘shall be oriented towards low and stable inflation’, but ‘shall also contribute to stabilizing output and employment’.
12 Since the central bank creates money, it is an ideal place for governments to look for cheap finance. Commentators often speak of ‘turning on the printing presses’. We will explain why and how this can happen more precisely in Chapters 13 and
17.
13 Speaking in February 2006 on government-owned television ZBC, President Mugabe not only gave advance warning for the country’s hyperinflation episode, but also gave us the reason why they happen: ‘Those who say printing money will cause
inflation are suggesting that you just fold your hands and say “aah, let the situation continue and let the people starve”. The Good Lord up there has given you a brain and the brain must function, not in a stereotyped manner but in a flexible manner
… so I will print money today so that people can survive.’
14 The disregarding of inflation followed the prevailing macroeconomic theory of the time, which focused on output and largely ignored the price level. This view, associated with Keynes, was vigorously attacked by the monetarist school. Today’s
macroeconomic theory, as presented in this text, is a synthesis of these views. These points will become clear in Chapter 13. The debate is presented in Chapter 20.
15 The ECB does not currently announce an inflation target but has indicated that it considers that price stability is achieved when the inflation rate is ‘close to, but below 2%’. The Fed has indicated that it wants inflation to be around 2%, between
1% and 3%. The Bank of Japan is officially committed to raising inflation above zero, its level for more than two decades.
16 The trend level of output corresponds to the long-run sustainable evolution of GDP, as explained in Chapter 3.
17 Both inflation and output are expressed in percentage deviations, respectively, of the rate of increase of the price level from target, or of output from its trend level.
18 While the central bank cannot give money away, lenders in a crisis situation may be willing to pay more than €100 for a guarantee of €100 tomorrow—especially when it comes from an especially reliable borrower. This situation occurs when
investors would rather move their funds into government debt because it is safer than deposits or loans to banks. For more on negative interest rates, see Box 9.4 in the previous chapter.
19 See Box 9.4 for a more detailed discussion of negative interest rates.
20 In fact, the Bank of Japan was the first to develop this strategy in the 1990s.
21 Eurepo® stands for euro repo (a contraction of repurchase agreement). Euribor® means euro inter bank offer rate.
22 Three economists, George Akerlof from Berkeley University, A. Michael Spence of New York University, and George Stiglitz, now at Columbia University, received the Nobel Prize in 2001 for their research on the economics of information
asymmetry.
23 The G10 includes 13(!) countries: Belgium, Canada, France, Germany, Italy, Japan, Luxembourg, Netherlands, Spain, Sweden, Switzerland, United Kingdom, and the USA.
24 In the United States, the Glass–Steagall Act of 1933 was a reaction to pre-Depression abuses, in which banks invested customers’ deposits in the stock market, often combined with borrowed funds. This law lasted until 1999, when—after intense
pressure by Wall Street financial institutions—the US Congress rescinded it. The abolition of the Glass–Steagall Act added much fuel to the housing boom and ultimately the financial crisis in the United States. Some things never seem to change!
PART IV

The most important step in macroeconomics is the integration of the different decisions by households, firms, and
governments. This type of economic reasoning, which is called general equilibrium, analyses the simultaneous
determination of output, employment, the interest rate, and finally the price level and the rate of inflation. The concept of
macroeconomic equilibrium is first developed in Chapter 11 for the case of a country that only trades goods and services
with the rest of the world, with capital flows playing a passive role. In Chapter 12, we examine how trade in financial
assets—financial openness—changes the equilibrium of an economy. It does so in profound ways.
Next, we incorporate inflation, the rate of growth in the general level of prices, in our study of macroeconomic
equilibrium. After the origins of inflation and aggregate supply are explored in Chapter 13, Chapter 14 presents the
aggregate demand–aggregate supply (AD–AS) framework as a powerful tool for explaining the behaviour of both output
and inflation over short-run horizons. Finally, in Chapter 15, we study the nominal exchange rate through the lens of
equilibrium, both in the short and the long run. The exchange rate turns out to be a special and important example of an
asset price, one that ultimately connects the short with the long run.
Macroeconomic Equilibrium in the
Short Run 11
11.1 Overview
11.2 Aggregate Demand and the Goods Market
11.2.1 The Market Equilibrium Assumption
11.2.2 Determinants of Demand
11.2.3 Goods Market Equilibrium
11.2.4 The Keynesian Multiplier
11.2.5 Endogenous and Exogenous Variables
11.3.1 The IS Curve
11.3.2 Off the IS Curve
11.3.3 A Key Distinction: Movements Along or Shifts of the IS Curve
11.4.1 The Taylor Rule and the TR Curve
11.4.2 Slope of the TR Schedule
11.4.3 Monetary Policy: Moving Along or Shifting the TR Curve
11.5.1 Macroeconomic Equilibrium in the IS–TR Model
11.5.2 Real Disturbances: Shifts of the IS Curve
11.5.3 Monetary Policy Disturbances: Shifts of the TR Curve
11.5.4 A General Approach
11.5.5 Monetary and Fiscal Policy in the IS–TR Model
Summary
The long run is a misleading guide to current affairs.

In the long run we are all dead.
J. M. Keynes
11.1 Overview
John Maynard Keynes said a lot in his short life, but this description of the long run is perhaps his best-known pronouncement. On the one hand, it is a normative
and controversial judgement that policies should care more about the present than the distant future. But it is also a statement of fact: the economy needs time to
reach that long run, and we need to understand what happens in the meantime. This chapter is concerned with the central question of macroeconomics posed in
Chapter 1: why do countries experience business cycles? These recurrent, irregular periods of ups and downs are presented in a stylized way in Figure 11.1. Recall
that we saw in Chapter 3 that GDP appears to increase without limit. This long-run growth is shown by the smoother curve displaying an upward trend. In the
short run, however, economic activity fluctuates around that trend, with periods of relatively fast growth followed by periods of slowdown, or even negative
growth, called recessions. This chapter shows that business cycles can be explained by disturbances that originate in goods and financial markets.
To study the short-run behaviour of the economy, we adopt a crucial Keynesian assumption that prices are constant. In most countries where inflation is
low, it is a reasonable assumption over, say, a couple of years. This is why this chapter is about the short run. The Keynesian assumption is a convenient
analytical short-cut. Not having to track prices will simplify the analysis a lot. It turns out that it is also a rather accurate description of the short run.
Fig. 11.1 Cyclical Fluctuations
The Keynesian assumption helps explain short-run fluctuations of real GDP around its long-run growth trend.
Macroeconomic price flexibility is a reasonable working assumption for the long run. Under price flexibility, the monetary neutrality principle applies, and money
and prices behave independently of output. This long-run independence, or dichotomy, of the real and monetary dimensions of the economy serves as an
important anchor for analysis. But by adopting the Keynesian assumption, we will upset this dichotomy. Under the condition that prices are constant, rigid, or just
‘sticky’ (i.e. slow-moving), it implies that the nominal and real sides of the economy interact with each other. Macroeconomics therefore incorporates two
fundamentally different perspectives. In the long run, prices are fully flexible and neatly separate out nominal and real sides of the economy. In the short run,
prices do not move, or move rather slowly, the dichotomy fails, and everything becomes more complicated. The working assumption of constant prices will help us
explain more easily the complex aspects of the business cycle.
The Keynesian model focuses on two markets: (1) the market for goods and services—which we will henceforth call the goods market, and (2) the money
market, which we studied in Chapters 9 and 10, along with the asset markets, that transmits the money market impulses to the goods market—and occasionally
becomes a source of disturbances. In the next chapter we will add the foreign exchange market, providing an open economy version of the Keynesian model. Our
reasoning follows the general equilibrium approach. General equilibrium is the simple requirement that all markets achieve equilibrium at the same time, and
that these equilibrium conditions are mutually consistent with each other. This is a very powerful method of analysis, which allows us to capture the many
interactions that take place among the different markets in one step. Figure 11.2 offers a preview of the full open economy analysis. Through the interest rate and
asset prices, the money market affects the goods market, and thus the level of output. The goods market, in turn, affects the demand for money, the policy interest
rate chosen by the central bank, and thereby market interest rates. Later on, we will also allow the real exchange rate to influence the demand for domestic goods.
We will see how interest and exchange rates affect each other, and therefore the goods market.
Fig. 11.2 General Macroeconomic Equilibrium

Conditions in domestic money and goods markets affect each other. Interest rates and exchange rates influence the level of aggregate demand, while income affects the interest
rate set by the central bank via the Taylor rule. General equilibrium occurs when equilibrium conditions in the three markets are consistent with each other.
11.2 Aggregate Demand and the Goods Market
11.2.1 The Market Equilibrium Assumption

We start with the decomposition of GDP presented in (2.5) in Chapter 2:
(11.1)
where all variables are measured in real terms, i.e. at constant prices. It describes the various categories of spending on domestic goods that make up the GDP. As
a definition of GDP, this equation is true by construction and we could leave it at that, but here we view it in a different light. We now think of the left-hand side of
(11.1) as the aggregate supply of goods and services by domestic producers, while the right-hand side brings together various components of aggregate
demand for goods and services produced domestically. To recall, these components are given by the demand by resident households for consumption goods
(C), investment spending by firms on capital goods (I), the public sector’s own demand for goods ( G), and net exports (X − Z), the difference between the
world’s demand for our goods, exports X, and the domestic demand for foreign goods, imports Z.
Viewed in this way, (11.1) is no longer a definition. Rather, it is a market equilibrium condition. We require it to hold not only because national accountants get
numbers right, but because we assume that the goods market is in equilibrium so that demand for goods equals the supply of goods. But how are the two
brought into balance?
When we look at markets for individual goods (apples, clothing, computers, machines tools, steel, oil, etc.), we usually think of that good’s price as the
mechanism which brings demand and supply in line. We apply this procedure to the macroeconomy, using the average price of goods and services as the price
level. In the short run, it is convenient and realistic to adopt the Keynesian assumption that prices on average are sticky, i.e. that they move slowly and in a
predetermined way. In the language of macroeconomics, the price level is assumed to be exogenous. Later on, in Chapters 13 and 14, we will return to that
assumption and allow the general price level to respond endogenously to evolving aggregate demand and supply forces.
To see what a fundamental difference this assumption can make, let us suppose that demand (the right-hand side of (11.1)) is greater than supply (the left-hand
side). If prices were flexible, they would rise, as prices tend to do when there is excess demand. Higher prices would lower demand, but would also raise supply. If
they are sticky, how can the market return to equilibrium? The Keynesian answer implies that demand determines supply and output. We can now interpret (11.1)
as an explanation of GDP’s movements over time: demand responds to exogenous forces, while supply responds passively to shifts in demand. For the short run,
this is the central and far-reaching implication of the Keynesian assumption.
11.2.2 Determinants of Demand

What, then, are those exogenous forces that shape demand? In order to answer this question, we need to think hard about the behaviour of consumers, firms, and
foreign customers. In doing so, we will return to many of the insights from Chapters 6 and 8: the behaviour and resource constraints of households, enterprises,
and the government. As a short-cut, we consider public spending G and tax receipts T as exogenous; in words, the behaviour of the public sector is independent
of economic conditions. To emphasize this assumption, we often write exogenous variables with an upper bar: _G and T .
In Chapter 8 we studied in detail the factors which drive private consumption and investment. Now, all we need to do is to synthesize these results in a simple
and compact way. We start with consumption, which originates in household behaviour. Measured in real terms, consumption is driven by wealth and disposable
income. Wealth, represented by the symbol Ω, is assumed to be exogenous. 1 Household disposable income Yd, income after tax, is the difference between GDP Y
and private sector tax payments T , measured net of transfers to households and firms. Our description is summarized by the following consumption function:
(11.2)
As before, signs underneath the two determinants of consumption indicate the effect of each variable on demand. More wealth and a higher disposable income
each raise consumption demand. Note that, while taxes are assumed to be exogenous at this stage, GDP is endogenous, and represents total aggregate income,
which is equal to total demand, which we are in the process of explaining.
In Chapter 8 we also studied the demand by firms for investment goods. Although it is a smaller fraction of total demand, it tends to fluctuate the most and is
considered by most macroeconomists to be highly relevant for the ups and downs of the business cycle. This demand is represented by the following form of the
investment function:
(11.3)
The investment function states that investment expenditures increase when Tobin’s q increases, and decline when the interest rate rises. Remember from Chapter
8 that, all other things being equal, investment is driven by business expectations or entrepreneurial ‘animal spirits’. This is captured by Tobin’s q. The interest
rate matters for investment for two reasons: (1) because a higher interest rate reduces Tobin’s q; (2) because firms borrow from banks to finance investment.2
When borrowing costs rise, firms invest less.
The interest rate is an important variable in the analysis of this chapter, so it will be useful to make clear which interest rate we are talking about. In Chapters 6
and 8 we studied the real interest rate r, which measures the real cost of borrowing or the return to lending. Later, in Chapters 9 and 10, we examined the nominal
interest rate, denoted by i, which is quoted by banks or reported in the financial press. We learned that the nominal interest rate is the cost of holding money, and
that includes a component which reflects inflation. Later, when we study inflation in Chapter 13 and afterwards, we will distinguish between the nominal and real
interest rates. The Keynesian assumption that prices are sticky implies that inflation is zero or negligible, so we can treat these two rates as the same for the time
being. For future reference, however, we note that it is the real interest rate r that matters for spending decisions and the nominal interest rate i that is relevant
when we look at monetary questions.
Then there is the last term in (11.1), net exports, the balance on goods and services, as defined in Chapter 2:
(11.4)
First consider imports. Recall that total domestic spending, also called absorption (A), includes spending on both domestically produced and imported goods and
services.3 Imports are thus a component of absorption: the more we spend, the more we import. This proportion need not be constant, however. In particular, it
will depend on the country’s competitiveness. In Chapter 5, we saw that competitiveness can be measured by the real exchange rate σ, the relative price of
domestic to foreign goods. A real depreciation—a decrease in σ—makes foreign goods relatively more expensive, and therefore discourages imports. Conversely,
a real appreciation—an increase in σ—boosts imports, which are now cheaper. These observations are summarized by the import function, which says that
imports rise with absorption and with the real exchange rate:
(11.5)
Turning to exports, we need only recognize that they are the imports of the rest of the world. Accordingly, they behave as our imports, from the foreign
perspective. Thus our exports depend on foreign absorption A* (and its determinants, foreign wealth Ω*, foreign disposable income Yd*, foreign Tobin’s q*,
etc.). If our real exchange depreciates—if σ decreases—our goods become cheaper in the foreigners’ eyes and stimulate our exports. Conversely, a real
appreciation—an increase in σ—depresses exports, which have become more costly. We bring these observations together in the form of the export function:
(11.6)
We combine these results in the form of the net export function, which is simply the difference between the export and import functions in (11.5) and (11.6):
(11.7)
The net export function tells us that any factor that boosts one of the components of domestic absorption A = C + I + G (increases in wealth, disposable income,
Tobin’s q, real growth, public spending or tax cuts, or a decline in the interest rate) will increase imports and reduce the primary current account (increase the net
export deficit). In contrast, anything that boosts foreign spending (increases in foreign wealth, disposable income, Tobin’s q, real growth, public spending and tax
cuts, or a decline in the foreign interest rate) will increase foreign absorption A* and therefore our exports, which improves our primary current account. Finally, a
real exchange rate appreciation (an increase in σ) leads to a deterioration of net exports as imports rise and exports fall. Conversely, a real depreciation improves
net exports.
Macroeconomists like to take short-cuts. Because the factors that affect absorption and aggregate demand also affect output—that is the Keynesian
assumption—then factors that affect absorption will ultimately affect output. The obvious short-cut is to replace the net export function in (11.7) with the
following version:4
(11.8)
In the end, we will work with this net export function, which states that net exports are negatively related to domestic real income, positively to foreign income, and
negatively to the real exchange rate. This completes the picture of aggregate demand in an open economy.
11.2.3 Goods Market Equilibrium
The groundwork having been laid, we can proceed to study equilibrium in the goods market. We know how much consumers want to buy, how much firms want to
invest in productive equipment, and the government’s spending intentions. In addition, we have characterized how much of all that should fall on domestic or
foreign goods and the intentions of foreign customers. Adding up the various demand functions according to the national income identity, we obtain the planned
or desired demand function:
(11.9)
Why is this called ‘desired’ and not actual demand? Because it describes what households, foreign customers, firms, and the government would like to spend,
given the variables that shape their preferences. However, there is no reason that desired demand DD adds up to actual output Y.
To begin with, let’s treat the interest rate and the exchange rate as exogenous, focusing our attention on real income Y. We proceed in three steps. First, we
assert that, given the exogenous variables, there is a value of Y such that the goods market is in equilibrium in the sense that desired demand is equal to supply:
(11.10)
This equilibrium condition—supply of goods equals demand—is essential for what follows. Next, we ask what happens out of equilibrium, when DD is not equal
to Y. Finally, in Section 11.2.4, we examine the effects of changes in the exogenous variables (W, T, r, q, G, Y*, and Σ).
The level of Y that satisfies condition (11.10) is called equilibrium GDP.5 The next logical step is to ask how Y affects desired demand. We note that (11.9)
shows that Y matters in two ways: first, it affects consumption positively; C rises. But, second, it also exerts a negative effect on net exports, since higher income
increases imports; NX falls. Which effect dominates? Theory and evidence say that it is the consumption effect; imports represent a fraction of domestic
spending. When an increase in GDP causes consumption to rise, imports too rise but by a fraction of consumption. As a result, the increase in C dominates the
decline in NX, so DD indeed increases. From Chapter 8, we also know that an increase in income will be partly saved. Just how much depends on whether the
increase is deemed permanent or temporary. At any rate, C will increase by less than Y. Aggregate demand DD ( = A + NX) thus increases by less than C (since
actually NX declines). It follows logically that, when Y increases by one pound or euro, desired demand rises but by less one pound or euro.
Keeping that important result in mind, we now ask what has to happen for the goods market to be in equilibrium, or when desired demand is equal to supply (DD
= Y). Now we use the Keynesian assumption that GDP adjusts in response to any disequilibrium. Suppose actual GDP is Y′, and exceeds its equilibrium level Y.
From previous reasoning, we know that desired demand DD′ that corresponds to Y′ also exceeds its equilibrium level. We also know that the increase from DD to
DD′ is smaller than the increase from Y to Y′. Since DD = Y, it must be that DD′ is less than Y′ (DD′ < Y′). We face a situation of excess supply. Since no one is
forced to spend more than they freely choose, firms can only sell the quantity DD′ but actually produced Y′. In other words, firms produce more than their
customers want to buy. The excess supply is stored aside as inventories of unsold goods or services. This is a situation that cannot last very long because firms
will not produce goods that they cannot sell. Sooner or later, they will reduce their production and the output level will decline. Less production means less
income and desired spending declines. The process will continue until GDP has declined enough to reach its equilibrium level Y, which is the level where desired
demand is equal to supply. Conversely, starting below equilibrium GDP, we observe a situation of excess demand. Firms satisfy this demand at first by drawing
down their previously accumulated inventories but, at some point, they will raise production, and output will increase until the equilibrium level Y is reached.
Let’s summarize what we have learned about the goods market in the short run. First, GDP returns to its equilibrium level because firms adjust the production
level to meet the market’s demand. Second, any disequilibrium in the goods market is relatively short-lived. Firms use inventories and unfilled orders as buffers.
They accumulate inventories and produce orders placed earlier when demand is unexpectedly weak, and they run down inventories and build up their order book
when demand exceeds production. This is why firms carefully monitor their inventory levels, and macroeconomic analysts study new orders and backlogs.6
11.2.4 The Keynesian Multiplier

When the exogenous variables that appear in (11.9) change, desired demand will change and so will equilibrium GDP. Understanding these effects provides the
first explanation of output fluctuations, the purpose of this and subsequent chapters. For example, what would happen if public spending G were to increase by
ΔG? Assuming that the economy starts from equilibrium with real GDP level Y, desired demand has now increased by ΔG and exceeds output Y. In the new
situation, for a while, producers will make up the difference by drawing down their inventories but, within a few weeks, they will raise their output to match the new
demand. They will continue to do so until equilibrium output has sufficiently risen to restore equilibrium. The really interesting question is by how much GDP will
rise. The short answer is that GDP increases by more than ΔG.
To see that, let us track down what happens after the increase in public spending, which is assumed to fall on domestic goods. In the first instance, excess
demand is eliminated when firms increase their production by ΔG. This is not the end of the story, however! As firms raise output, more income is generated,
which raises consumption and, therefore, desired demand. Firms will again raise output to meet this additional demand. And once again, more output means more
income and therefore a higher level of desired demand. The process will continue, but will it ever come to an end? The answer is affirmative, at least for all practical
purposes.
The first increment to aggregate demand was of size ΔG. The second increment in demand results from an increase of consumption in response to an increase in
Y by the amount ΔG. But consumers do not respond to an increase in income by spending all of it. As they save some part, the second step will be smaller than
ΔG. It will be further reduced because some of the extra spending falls on imports, and imports create income abroad, not at home. Thus while more demand creates
an equal increase in supply and therefore income, the extra spending generated by the additional income is smaller. Savings and imports operate as leakages
draining some of the newly created income away from additional spending. This is why each additional increment is smaller than the previous one. Eventually, the
steps become so minuscule that we can ignore them. In addition, the sum of these increments is finite and greater than the initial increase in desired demand
caused by the increase in government spending, ΔG. Box 11.1 gives a formal description of this process.
Box 11.1 The Multiplier’s Algebra
When government spending increases by ΔG, output and therefore incomes first increase by ΔG as well.7This so-called first-round effect is the size of the
exogenous change in desired demand. Assume that consumption responds by increasing by a fraction cof the corresponding income increase but that a
fraction z is spend on imports. (Those fractions c and z and called the marginal propensities to consume and to import.) This means that the second-round
effect is of size c(1 – z)ΔG. Because they are fractions, they obey 0 < c < 1, 0 < z < 1, and therefore 0 < 1 – z < 1. The second-round effect is smaller than the
first-round effect.
Applying the same logic again, the third-round effect is of size c(1 – z)[c(1 – z)ΔG] or c2(1 – z)2ΔG. Since c and 1 − z are smaller than 1, c2 is smaller than c
and (1 − z)2 is smaller than (1 − z). Continuing in the same way, in the end, the cumulative increase in income is:
an infinitely long series of increasingly smaller steps. A standard result from algebra is that, if a is less than one, then 1 + a + a2 + …. + an + … = 1/(1 – a).
Thus the total effect of ΔG is:
This shows that the multiplier is , which is larger than 1. Note that, if there is no leakage to savings (c = 1) and to imports (z = 0), the multiplier
would be infinite. Needless to say, this case is not of practical relevance.
The remarkable result is that the total increase of output ∆Y is a multiple of the initial exogenous increase in demand ∆G. This is why the effect is called the
Keynesian demand multiplier. It is very general: no matter which exogenous change triggers the process, and whether the disturbance is positive or
negative, equilibrium output always responds to demand, sometimes by a larger amount.8
The multiplier effect corresponds to the fundamental insight provided by the circular flow diagram in Chapter 2. Each individual’s spending is someone else’s
income. By raising incomes, an exogenous increase in demand generates additional desired demand, which means more spending and income, a never-ending
process, although at each stage, the effect becomes smaller, and eventually dies out. The circular flow diagram showed where these leakages occur: taxes, savings,
and imports each capture a portion of any additional income.9 These three leakages represent domestic income that is not automatically re-spent on domestic
goods and services. It should be clear, by now, that big leakages reduce the multiplier. This means that the multiplier will be reduced if a large proportion of any
additional income is saved or if a large proportion of any additional spending is imported.10
The word ‘automatically’ is important. For example, some of the additional taxes collected along the way could be used to support additional public spending
but it would be an exogenous decision by the government to raise G even more than the initial boost ∆G. Similarly, higher savings are available to finance new
productive equipment. However, firms must be convinced to invest more. Improved expectations of future profitability captured by Tobin’s q, which we treat here
as exogenous, would do this. Finally, a rise in imports will generate higher incomes abroad, triggering there another multiplier effect which might well lead to higher
foreign GDP Y* and more exports. All of these effects are plausible but, since we consider G, T, Ω, q, and Y* as exogenous, at this stage we cannot logically treat
them as responding automatically in the circular flow of income.
In practice, there is much debate about the size of the multiplier. Part of the debate, discussed in Chapter 20, is associated with doubts about, or even rejection
of, the Keynesian assumption. Another part of the debate is related to more technical issues, for example what is assumed to happen to some of the variables that
we have ignored in the previous reasoning (tax revenues, Tobin’s q, etc.). Finally, the speed at which GDP responds to fiscal policy actions varies, depending on
countries and circumstances.
11.2.5 Endogenous and Exogenous Variables

When using this framework, we return to the distinction made in Chapter 1 between endogenous economic variables we are trying to explain, versus those
exogenous variables we take as given to the analysis. Whether a variable is treated as endogenous or exogenous is an analytical assumption made for
convenience. Indeed, by declaring a variable to be exogenous, we free ourselves from the obligation of explaining its behaviour. Naturally, we would like to explain
everything at once, but being careful has its rewards. Treating most variables as exogenous is an effective way to get a handle on complex phenomena of interest.
Later, it is possibly to endogenize variables considered exogenous by introducing new theoretical mechanisms, i.e. the behaviour of government spending and
taxes to changes in output, unemployment, etc.
A review will be helpful at this stage. The Keynesian assumption implies that the price level P is exogenous. Instruments of government spending and taxation
policy, such as government purchases, are assumed to be under direct control of the government and thus treated as exogenous—henceforth they will be written
with the overbar symbol as G and T, respectively. 11 As in Chapter 8, total household wealth Ω is also treated as exogenous, as are foreign variables such as
foreign GDP (Y*) and the price level (P*), since they are also not influenced by domestic changes as long as the economy under consideration is small relative to
the ‘rest of the world’. Nominal and real exchange rates are also treated as exogenous, although we will change that in the next chapter. Finally, the interest rate
and Tobin’s q will be endogenized in the next section.
In this section, we add the interest rate to the list of endogenous variables. In the spirit of general equilibrium analysis, we seek to explain output and the interest
rate jointly. This objective will be achieved by the end of this chapter. To do so, we use graphical tools to assist us in our reasoning. The diagram that we will use
will depict two endogenous variables: output Y on the horizontal axis, and the interest rate i on the vertical axis.
11.3.1 The IS Curve

We can now ask: what happens to equilibrium output when the interest rate changes, holding everything else (the exchange rate, public spending and taxes,
private wealth, foreign demand) constant? This is the time to remember that Tobin’s q is inversely related to the real interest rate r. When r rises, future profits are
more heavily discounted and Tobin’s q declines. While expected future profits—animal spirits, as we called them—will be considered exogenous, Tobin’s q is
now endogenous to the interest rate.
In Figure 11.3, we start from point A, where equilibrium is achieved. Now suppose that the interest rate declines from i to i′. We know that Tobin’s q will
increase. Looking at the investment function (11.3), we see that investment will rise, since both of its determinants, the interest rate and Tobin’s q, act in the same
direction. Desired demand is now higher, which will trigger the multiplier mechanism. The new equilibrium is now achieved at point B, which means that equilibrium
output increases from Y to Y′.12 This reasoning can be repeated for other interest rate levels, to obtain more points similar to A and B. They will trace out a
negative relationship between the interest rate and equilibrium output depicted by the downward-sloping schedule known as the IS curve.13 For given values of
exogenous variables, the IS curve represents the combinations of nominal interest rate i and real GDP Y that are consistent with goods market equilibrium.
It is easy to remember why the IS curve is downward-sloping. First, the IS curve is the response to the question asked above: what happens to aggregate
demand and equilibrium output when the interest rate changes? Second, a higher interest rate reduces private spending on investment, which reduces demand
through the multiplier, and equilibrium output.
Fig. 11.3 The IS Curve
A reduction in the interest rate from i to i′ leads to an increase in investment spending, which in turn leads to an increase in equilibrium output, as shown in the left-hand side panel.
This relation is summarized by the IS curve. For a given change in interest rates, the IS curve is flatter, and the larger is the increase in equilibrium output as measured by the
horizontal distance between A′ and B. The length of A′B in turn depends on: (1) the sensitivity of demand to interest changes, and (2) the multiplier effect, measured by the
14
horizontal distance A′B. A greater sensitivity of desired demand to output and larger values of the demand multiplier lead to a flatter IS curve.
11.3.2 Off the IS Curve

What happens when the economy is not on the IS curve? The short answer is that the IS curve represents the goods market equilibrium condition, so that points
off the curve describe conditions of either excess demand or excess supply. But which is which?
Let us start from point A in Figure 11.3, which is on the IS curve. Now imagine that output increases while the interest rate remains unchanged—for example,
that we move horizontally from A to point C. Because output (which is income) has risen, spending will rise as well, but by less, due to the leakages discussed in
Section 11.2.4. In Figure 11.3, demand at point C falls short of supply—there is not enough demand to absorb output at point C, which is equal to Y´. This
situation of excess supply in the goods market would be reflected in rising inventories and shrinking orders. Similarly, moving vertically up from point A to a
point like D corresponds to an increase in the interest rate at unchanged output. This also leads to excess supply in the goods market because the higher interest
rate reduces aggregate demand.
The IS curve determines two regions in Figure 11.3: (1) above and to the right we observe excess supply in the goods market—inventories are being
accumulated and new orders for product are falling; and (2) below and to the left, we have excess demand in the market for goods and services, and inventories
are run down and orders are growing. At the boundary of the two regions, the IS curve represents those combinations of GDP and the interest rates which are
consistent with goods market equilibrium. The logic of Section 11.2.2 implies that the economy can move temporarily off the IS curve, while firms use their
inventories as a buffer stock, but fairly soon they will adjust their output, returning the economy to goods market equilibrium on the IS curve. For example, starting
at point C in Figure 11.3, the economy will move towards point A as firms react to involuntary inventory accumulation by cutting production. Similarly, from point
D the economy will move horizontally to the left until it reaches the IS curve. More generally, to the right of the IS curve where there is excess supply, goods
market equilibrium requires that the economy moves to the left so that output is reduced to a level compatible with desired demand. Conversely, starting from a
situation of excess demand on the left of the IS curve, the economy will move to the right as firms expand output to meet demand. This is what the Keynesian
assumption implies.
11.3.3 A Key Distinction: Movements Along or Shifts of the IS Curve

A common pitfall is to confuse shifts of the IS curve with movements along it. This is directly related to the distinction between exogenous and endogenous
variables examined in Section 11.2.5. The IS curve describes how the two endogenous variables, the real GDP (Y) and the nominal interest rate (i), are combined to
achieve equilibrium in the goods market, everything else being held constant. ‘Everything else’ is the set of all variables that we treat as exogenous when we draw
the IS schedule. As long as these exogenous variables remain constant, the IS curve stays in place and we can only move along the curve. Any change in any
exogenous variable, however, causes the IS to shift.
Figure 11.4 displays an example when the IS curve shifts. The exogenous change is an increase in public spending G. This is the same experiment already
studied in Section 11.2.4, which showed how an increase in public spending triggers the multiplier process. Note that we implicitly assumed that the interest rate
remains constant. Graphically in Figure 11.4, we start from point A, with output Y and interest rate i. The multiplier process eventually takes us to point B, where
output is Y′ and the interest rate is unchanged. This point lies on the new curve IS′. In the end, the rule is simple: any exogenous change that raises aggregate
demand shifts the IS curve to the right, which quite logically means more output. Conversely, when demand exogenously declines, the IS curve shifts to the left.
Which are the relevant exogenous variables? Fiscal policy is a premier source of shifts in the IS curve. The government is a large player in the macroeconomy,
and changes in government purchases of goods –G (e.g. military procurement or road construction) or services (e.g. the number of civil servants or their pay
packets) exert a significant influence on aggregate demand. Similarly, changes in taxation T alter disposable income available to households or firms, with knock-
on multiplier effects on consumption.
Fig. 11.4 An Exogenous Increase in Aggregate Demand

At unchanged interest rate i, an increase in any of the exogenous components of demand causes an increase in aggregate demand at constant interest rates. At unchanged
interest rate, equilibrium occurs at point B and the new output Y′ level consistent with goods market equilibrium is higher than the initial level Y.
Second, consumption depends on household wealth . Wealth can take many forms, such as land, apartments and houses, financial assets, precious goods
such as jewels, Persian rugs, and art, etc. The value of several of these components can be highly volatile and provoke sharp changes in consumption and thus
the position of the IS curve—at least if they are perceived as permanent or long-lasting, since we know from Chapter 8 that consumers smooth spending over time.
A classic example is the Great Depression of the 1930s, which followed the crash of stock prices on Wall Street. Many recessions in the past century have been
associated with sharp increases of asset prices, increases in consumption spending, and subsequent collapse. The global financial crisis (2008–2009) was driven
by a collapse of overvalued house prices in the United States and some other countries. It serves as an excellent example of how a collapse of wealth can affect
consumption. Box 11.2 gives details.
Third, changes in expectations of the future can have dramatic effects on investment decisions of businesses. Keynes spoke of ‘animal spirits’—business
expectations driven by gut feelings or observing others, as opposed to rational calculations. These expectations of the future profitability of investment—correct
or not—are captured in Tobin’s q. An increase in Tobin’s q implies an increase in investment at an unchanged interest rate, and means that the IS curve shifts
outwards. Similarly, a decline in Tobin’s q would imply lower investment expenditures at any level of interest rates and an inward shift of the IS curve.
Finally, foreign disturbances matter for an open economy. They affect the IS curve through net exports. Net exports are not only a source of leakage, but also
transmit foreign disturbances. Indeed, the net export function (11.8) shows that changes in world activity Y* can generate export-led expansions or recessions. All
other things being equal, an increase in Y* will increase NX and cause the IS curve to shift rightwards. The same applies to the real exchange rate σ = SP/P*, which
determines the economy’s competitiveness. Any change in either the nominal exchange rate S or in domestic and foreign prices P and P* will affect the real
exchange rate σ and shift the IS curve. A real exchange rate depreciation (σ falling) implies a rightward shift of the IS curve; a real exchange rate appreciation
(σ rising) moves the IS curve leftwards.
Box 11.2 The American Dream, An American Nightmare
A home or an apartment is an important acquisition for anyone, and indeed, households hold most of their wealth in their house. They purchase houses on
credit, a long-term loan called a mortgage. Recent changes in regulation of the way lending is given for house purchases around the world, and especially
in the USA, led to new ways to shift income to the present by borrowing more money to purchase the house, and second, to convert additional house value
after the purchase (home equity) into cash today by borrowing against it (using it as collateral). Effectively, this deregulation made it easier to behave like
consumers with access to credit markets, as described in Chapter 8.
Assuming that households ultimately respect their intertemporal budget constraints and repay their mortgages, these changes can make them better off.
Yet in a world of uncertainty it is possible for consumers to borrow too much. Low interest rates in the early 2000s led to excessive borrowing and a run-up
of house prices never seen in US history. The first panel of Figure 11.5 shows how dramatic that increase was. It plots a price index of the average value of
houses in key US cities divided by the consumer price index. Many consumers regarded this increase in wealth as long-lasting and took out loans
guaranteed by their new-found wealth to finance additional consumption.
After 2007, house prices dropped dramatically as owners of second and third homes rushed to sell and insolvent borrowers walked away from their
loans. As property values declined, so did the value of wealth. The basis for consumer loans dried up, while banks toughened lending requirements. The
result was the deepest recession since the 1930s, driven by a drop in consumption. This can be interpreted as an exogenous decrease in demand, as
shown by the second panel. The atypical behaviour of consumption around its long-run trend helps explain the modest recovery of the US economy after
2010.
Fig. 11.5 Consumption and House Prices in the USA, 1975–2015

The first panel plots the Case–Shiller Home Price Index, which measures the evolution of average housing prices in the USA, divided by the consumer price index. It is normalized
to the value of 100 in 1995. The second panel shows the deviation of consumption expenditures, in per cent, from a long-term trend (calculated over the period from 1975 to 2007,
when the financial crisis began). These charts show that consumption is strongly influenced by the valuation of apartments and houses, because dwellings constitute a
significant component of household wealth. They also show the bubble-like housing price increases in the early 2000s that, along with unusually large consumption, led to the
crisis. Consumption has still not returned to its trend from earlier, probably because prices have yet to reach half the levels attained in 2006. Many households are still saving to
service large debts accumulated in the go-go years.
Source: Federal Reserve Bank of St Louis and authors’ calculations.
11.4.1 The Taylor Rule and the TR Curve

After studying the goods market in Section 11.3, we turn to the money market. There, banks interact with households and firms, and the demand for money is
equilibrated with its supply. In the same way as before, we look for levels of real GDP and interest rate which represent equilibrium in the money market. In
Chapters 9 and 10, we saw that a central factor of the money market is the conduct of monetary policy by the central bank. Central banks can try either to control
the quantity of money in the market, or the level of interest rates for short-term lending by banks among themselves. They cannot control both.
Nowadays, central banks conduct monetary policy by setting certain interest rates directly in the money market, and the economy adjusts accordingly.
Effectively, this is the equivalent of a money supply curve. In Chapter 10, we saw that central bank interest rate policy is well-represented by the Taylor rule,
which states that the central bank increases the interest rate relative to the ‘neutral level’ if the inflation rate rises relative to its target rate, or if output rises relative
to its trend level. The neutral—sometimes also called natural—interest rate is the one chosen by the central bank if inflation and output are both on target. The
Taylor rule presented in Section 10.3 stated that the interest rate setting policy of modern central banks can be represented by the following relationship:
(11.11)
where π is the inflation rate, is the central bank’s inflation target, and Y is trend GDP. Recall that i is the neutral—sometimes also called natural—interest rate,
which the central bank would choose if inflation and output were both on target (π = and Y = Y ). The coefficients a and b indicate how responsive the central
bank is to inflation and the output gap, respectively.
The Taylor rule will be simplified in this chapter and the next because we make the Keynesian assumption that prices are constant, i.e. that both inflation and its
target rate are zero .15 Under these assumptions, we can write the Taylor rule in the following simple form:
(11.12)
which says that central bank policy ‘leans against the wind’: it raises the interest rate whenever output Y increases relative to potential output and, conversely,
cuts rates when Y declines.16 Monetary policy conducted by the central bank is centred around ī , the nominal interest rate which the central bank would choose if
the economy were at its equilibrium or trend level of output.
Just as the IS curve shows combinations of interest rate and output consistent with goods market equilibrium, the Taylor rule describes combinations of output
and interest rates which characterize the monetary policy of the central bank. This set of pairs of interest rates and output consistent with the central bank’s
monetary policy are summarized by the TR curve in Figure 11.6. It is upward-sloping because a higher level of GDP, all other things equal, causes the central bank
to raise interest rates. A lower level of GDP prompts interest rate cuts by the central bank. When the economy is at its trend level, the interest rate set by the
central bank is equal to its target rate i . Because the TR curve describes how the central bank behaves, the economy is always located somewhere on it, in
contrast to the IS curve, where we saw that there can be periods of excess demand and supply of goods.
As we saw in Chapters 9 and 10, in order to set the interest rate at its desired level, the central bank uses several instruments in the background. Ultimately it
controls the supply of money using these instruments and allows it to expand or contract to be consistent with the demand for money. Although we do not need
to be concerned with this background action, Figure 11.7 shows what happens below the surface of financial markets. Panel (a) is the Taylor rule, reminding us
that central bank raises interest rates when output rises. Panel (b) shows that the central bank provides the amount of bank reserves that the market demands at
the chosen interest rate.17 Graphically, demand and supply must be equal when the central bank chooses a point along the money demand curve. This is captured
by the right-hand Panel (b) of the figure. In order to steer the interest rate to its chosen level, the central bank must supply the amount of money that corresponds
to the chosen interest rate and the level of output associated with it. This supply of money traces out as the MS curve in Panel (b).
Fig. 11.6 The TR Curve

The central bank sets interest rates according to a Taylor rule. Since inflation is treated as exogenous in the short-run analysis, the level of output plays a larger role for the Taylor
rule. When output rises relative to its trend level, central banks tend to lean against the wind, i.e. raise interest rates, while cutting interest rates in recessions.
Fig. 11.7 The TR Curve and Money Market Equilibrium
When GDP increases, the TR curve implies an increase in interest rates as dictated by central bank policy in Panel (a). At the same time, increased GDP and income lifts the
demand for money; in Panel (b), the demand curve for money has shifted to the right. As a result, the money supply has increased from M/P to (M/P)′. The money market is in
equilibrium with the central bank supplying the money demanded at the higher level of output and interest rates.
11.4.2 Slope of the TR Schedule

The slope of the TR schedule shows how strongly the central bank reacts to the output gap, raising the interest rate when the output gap is positive and cutting it
when the gap is negative. In fact, the strength of response is summarized completely by the parameter b in (11.12). To see this, compare different monetary policies
symbolized by TR curves TR0, TR1, and TR2 in Panel (a) of Figure 11.8. TR2 corresponds to the most vigorous interest rate reaction (points B and F), followed by
TR1 (points C and E). TR0 corresponds to a constant interest rate policy (b = 0).
In Figure 11.8, curve TR2 corresponds to the case when the money supply does not change at all. The central bank refuses to supply any new reserves to the
money market. This type of policy is called monetary targeting. It implies large fluctuations of interest rates in money market equilibrium in response to swings
in GDP, and corresponds to a much steeper TR curve. This is drawn as points B and F in the panels of Figure 11.8.
The Taylor rule describes central bank behaviour well. They decide on the interest rate and the money supply is endogenous. Up until the 1980s, many central
banks did not target the interest rate; instead they sought to control the money supply. This was in line with the view, presented in Chapter 5, that inflation is
driven by money growth. Central banks were tasked with delivering price stability—that is, low inflation—since they were in charge of the money supply. This led
to a different theoretical apparatus, whereby money is exogenously set by the central bank and the interest rate is endogenously determined. This apparatus is
presented in Box 11.3. Money targets have been all but abandoned because they require the demand for money to be stable, a condition which has not been met in
recent years.
Fig. 11.8 The Slope of the TR Curve

Different slopes of the TR curve correspond to different degrees to which central banks lean against the wind and raise interest rates during an expansion of GDP, or cut them as
output declines. As output increases from Y to Y′, different TR curves will imply different interest rates. TR2 depicts the most stringent of monetary policies, involving sharp
increases in response to increasing output. TR1 corresponds to less output-sensitive monetary policy, while TR0 implies that interest rates are held constant. Each TR curve
implies a different supply of money consistent with the interest rate chosen by the central bank. TR0 implies the largest increase in the money supply; TR2 is constructed to imply
an unchanged money supply in response to the increase in GDP.
Box 11.3 When Central Banks Target the Money Supply: the LM Curve
In addition to older central bank practice, the idea that the money supply is the central bank instrument and the interest rate is endogenously determined in
the money market goes back to Keynes. In his seminal book, the General Theory, the macroeconomy is described by two markets: the goods market and
the money market. One year later, University of Oxford economist and later Nobel Laureate J.R. Hicks had already taken up Keynes’ ideas and made curves
out of them: the IScurve maps output and the interest rates that equalize the demand and supply of goods, and his LM curve defines those combinations of
18
output and interest rates when money demand and supply are equal. The resulting IS–LM diagram is the precursor of the IS–TR analysis in this book.
For a constant money supply, this money market equilibrium can be depicted in Figure 11.9. In the left-hand side panel, we display the demand for real
money, Md = k(i)Y which is downward-sloping in output-interest rate space. For a given level of output Y, the demand for money declines when the interest
rate rises, as explained in Chapter 9. We show monetary policy as the vertical line Ms, which represents the central bank’s exogenous real money supply
target Md. The money market equilibrium is at point A where Md =Ms:
(11.13)
and the equilibrium interest rate is i. We next ask if there are other combinations of output and interest rates consistent with money market equilibrium for a
given amount of money supply. Put differently, what happens to money market equilibrium if GDP rises from Y to Y′, holding all else constant? The demand
for money rises as firms and households need more to carry out an increased volume of transactions. The new money demand is shown as Md’.
Equilibrium occurs at point B, where the interest rate is i´. The interest rate is higher because money demand has increased while money supply remains
unchanged. These results are displayed in (b) in the output-interest rates diagram, with corresponding points A and B. We can repeat this exercise any
number of times to find equilibria corresponding to any value of Y. The result is the LM curve.19 It is upward-sloping because when the central bank targets
the money supply, rising interest rates are necessary to restore money market equilibrium in the wake of an increase in GDP (i.e. real income and output),
which raises the demand for money.
Combined with the IS curve, the LM curve can be used to conduct the same analysis as with the IS–TR framework when the money supply is set by the
central bank. We can also see what happens when the central bank changes the money supply, which shifts the Ms line. Now that most central banks set
interest rates directly and behave according to the Taylor rule, the LM curve has become less useful to macroeconomists.
Fig. 11.9 The LM Curve

Panel (a) represents the money market. The vertical money supply line Ms describes the central bank decision. The money demand Md curve is downward-sloping because a
higher interest rate reduces demand for zero-interest yielding cash/demand deposits. It corresponds to a particular level Y of output. Equilibrium occurs at the intersection A of Ms
and Md. If output increases to Y’, the demand for money increases and the corresponding curve is Md’. At unchanged money supply, money market equilibrium occurs at point B.
where the interest rate i’ has increased. Points A and B are transcribed in Panel (b). They trace out the combinations of output and interest rates at which the money market is in
equilibrium for a fixed level of the money supply.
11.4.3 Monetary Policy: Moving Along or Shifting the TR Curve

As with the IS curve, it is important to distinguish between movements along or shifts of the TR curve. The TR curve represents the monetary policy of the central
bank, so it only moves when that policy itself changes . At this stage of the analysis, the central bank’s Taylor rule (11.11) is summarized by two parameters. The
first is the slope of TR curve, given by b. This shows the strength of central bank’s reaction of output deviating from its trend, and corresponds to the underlying
objectives of the central bank in ‘leaning against the wind’ and preventing swings of output. The second key parameter is the central bank’s target interest
rate . This is the desired level of interest rate at a ‘normal’ level of output. For reasons explained in more detail in Chapter 13, i is also sometimes called the
central bank’s estimate of the short-run neutral interest rate.20
Unless i changes, the economy moves along the TR schedule, because this is the central bank policy. The TR curve will change position only when the values
of the parameters listed above change. One possibility is an increase in b which would cause the TR curve to rotate around the point defined by trend GDP Y and
the target interest rate i , as shown in Figure 11.8, say from TR1 to TR2. This would imply that the central bank is more concerned about reducing output
fluctuations.
A second possibility is that the central bank changes its view of the neutral interest rate, for instance reducing it from i to i ´. This would imply a downward shift
of the TR schedule from TR to TR′ in Figure 11.10. At any level of output, the central bank will now choose a lower interest rate; as we will see shortly, this means
a more expansionary monetary policy. Similarly, the central bank could decide to raise interest rates at any level of output. This would imply an upward and
leftward shift of the TR curve.
Fig. 11.10 Shifts of the TR Curve
If the central bank reduces its target rate from to ’, the TR schedule shifts downwards and to the right. When output equals trend level at , the interest rate was set at ;
now it is set at ’. This is an example of expansionary monetary policy. An increase in the target rate to ’’ would represent monetary contractionary policy.
We are now ready to study the macroeconomy when both goods and money markets are in equilibrium. The resulting framework will provide a useful tool for
understanding how output and the interest rate respond to exogenous disturbances. This is an important step forward. Indeed, in Section 11.2, on the basis of an
analysis of the goods market, we had developed a way to understand the behaviour of the output gap, our initial interpretation of business cycles. Section 11.3
derived combinations of output and interest rates consistent with a goods market equilibrium under the Keynesian assumption; Section 11.4 did the same for
money market equilibrium, given that the central bank follows a Taylor rule. We now ask under which conditions both goods and money markets are in equilibrium
at the same time. In effect, we are imposing mutual consistency on the two markets as a condition for macroeconomic general equilibrium. The next step will be to
analyse the effects of changes in exogenous influences on output and interest rates.
11.5.1 Macroeconomic Equilibrium in the IS–TR Model

For the time being, we ignore the exchange rate and the fact that the economy may be subject to financial flows. This is a good characterization of the equilibrium
of the global economy for which the current account is zero, or a large economy subject to controls on movements of financial capital. Net export deficits and
surpluses are financed passively by financial markets. In the next chapter, the objective is to show how things change when we consider an economy open to
active financial flows.
Using the IS–TR apparatus is straightforward. We have seen that the economy cannot stay for long off the IS curve and that it is always on its TR curve. So, for
both goods and money markets to be in equilibrium simultaneously, we require that the economy lies at the intersection of the IS and TR curves. In Figure 11.11,
macroeconomic equilibrium is achieved at that intersection, point A. At this point there is neither excess demand for nor excess supply of goods, nor is there
excess demand for or supply of money, given the central bank’s policy summarized by the Taylor rule.
To help understand this powerful notion of equilibrium, it is useful to remind ourselves what other combinations of output and interest rates would imply. All
other points in the diagram which are off the IS and TR curve represent either disequilibrium in the goods market, an inconsistency of the interest rate with the
Taylor rule, or both. All points above the IS curve imply an excess supply of goods, while all points below it imply an excess demand for goods. Similarly, all
points above the TR curve imply that interest rates are above the level consistent with the central bank’s monetary policy—the central bank can be expected to
take action to reduce them. For all points below the TR curve, the central bank can be expected to raise interest rates.
Fig. 11.11 Macroeconomic Equilibrium

Macroeconomic equilibrium occurs when both the goods and money markets are in equilibrium simultaneously. This occurs at the intersection of the IS and TR curves at point A.
11.5.2 Real Disturbances: Shifts of the IS Curve

We can use the IS–TR framework to understand the effects of exogenous disturbances to the macroeconomy. These disturbances can originate either in the real
side of the economy or from the monetary side of the economy. They may originate at home or abroad. In the IS–TR diagram, disturbances are captured as shifts
of one curve or another. In this section, we consider disturbances to the goods market—shifts of the IS curve. The reasoning applied will be to ask: after the
disturbance occurs, where is the new curve in relation to the original one?
Consider, for example, the effect of the following real disturbance: an exogenous change in government purchases of goods and services, G , say, an
increase from G to G ´. This corresponds to the type of action taken in the wake of the Great Recession in 2008–2009. For example, the government might increase
spending on schools—teachers, computers, or classrooms.21 Starting from IS in the first Panel (a) of Figure 11.12, an exogenous increase in government
purchases makes every point previously consistent with goods market equilibrium now out of equilibrium. In fact, for every point on the old IS curve, aggregate
demand now exceeds output. Logically, the exogenous increase in government spending implies that the new curve should lie to the right of the old IS curve;
there is an exogenous increase in the demand for goods at any level of interest rate, which is met by a greater supply. This is shown as the shift from IS to IS´ in
Panel (a) of Figure 11.12. The new equilibrium occurs at point B. An exogenous increase in demand leads to a higher output and a higher interest rate.
Fig. 11.12 Macroeconomic Disturbances
In Panel (a), starting from point A, an exogenous increase in demand shifts the IS curve to IS′. This takes the economy to point B along the TR schedule, reflecting the central
bank’s raising the interest rate in response to the output expansion. An exogenous decrease in demand shifts the IS curve to IS′′, leading to lower interest rates and output. In
Panel (b), an expansionary monetary disturbance, an exogenous cut in the target interest rate, moves the TR curve to TR′, leading to greater output and lower interest rates; a
contractionary monetary disturbance raises interest rates and reduces output.
This is an interesting result in its own right, but it is just as important to understand what is happening in the background. First, higher demand results in higher
output. After all, the Keynesian hypothesis is that output responds to demand. But there is more to the outcome than that. We know that the IS curve rightward
shift is larger than the initial boost in demand because of the multiplier effect. So we know that the total increase in output results from a succession of impulses
running from more demand to more output, and from more output to more income and therefore yet more demand.
Second, the interest rate has increased because the TR curve is upward-sloping. By assumption, the position of the TR curve has not changed, because the
factors that determine its position remain unchanged. Therefore, we move with the IS curve along the TR curve. As output rises, the monetary authority raises
interest rates. The increase in interest rates tends to reduce demand so output rises less than it would have under a situation of constant interest rates. This
example illustrates how considering the feedback from the goods market to the money market imposes consistency on the overall macroeconomic outcome.
Exogenous shifts in components of real aggregate demand have effects which are qualitatively similar to shifts in government spending. It is possible to identify
a number of factors behind the IS curve which can be the source of exogenous shifts of aggregate demand, given income. Besides government purchases ,
these include:
• taxes (net of transfers)

• household wealth
• Tobin’s q
• the real exchange rate
• foreign income Y*.
To understand the direction of the shift implied by exogenous changes in these variables we merely need to think about the impact they would have on
aggregate demand given output (Y). An increase in taxes (Δ > 0) shifts the IS curve to the left, since this would reduce aggregate demand at every level of
output or interest rates; a tax cut (Δ < 0) shifts the IS curve up and to the right. An exogenous increase in Tobin’s q (Δq > 0) would lift investment expenditures
for any level of output and income, shifting the IS curve to the right; a drop in Tobin’s q (Δq > 0) shifts the IS curve to the left. A rise in the real exchange rate (Δq
< 0) reduces international competitiveness and reduces net exports, depressing aggregate demand and shifts the IS curve to the left, while a decrease (Δσ > 0)
would raise aggregate demand and shift the IS to the right. Finally, an exogenous increase in household wealth Ω causes households to increase their
consumption at all rates of interest, shifting the IS curve to the right, while a decline in wealth causes the IS curve to shift to the left. The last event is highly
relevant for understanding the impact of the financial crisis on the real economy in all countries, but especially those hit by the end of the real estate bubble of the
2000s, as discussed in Box 11.2.
11.5.3 Monetary Policy Disturbances: Shifts of the TR Curve

Now consider another thought experiment. The central bank decreases its estimate of the neutral interest rate. As (11.12) indicates, the neutral rate is the interest
rate that should keep the economy at a zero output gap, and inflation on target in the more general formulation (11.11). This reflects the central bank’s estimate of
what drives demand Y and potential GDP Y . New analyses can lead the central bank to consider that potential GDP has hitherto been underestimated. In that case,
the central bank will revise downward its estimated of the neutral rate interest rate from i to i ´, with i to i ´< i. Its monetary policy will therefore become more
expansionary as it intends to lift actual GDP toward its higher potential level.
This monetary disturbance is tracked in Panel (b) of Figure 11.12. Graphically, the TR curve shifts down to TR′, and the economy moves from point A to
point D. Because it leads to a lower interest rate and higher output, it is called expansionary monetary policy. As the central bank provides more liquidity,
banks can lend to customers at lower interest rates. Why do firms and households wish to borrow more? Because interest rates are lower. At a lower interest rate,
investment spending rises and we move down and to the right along the IS curve. As we do so, the multiplier takes over and consumers start spending more
because their incomes rise. Note, however, that the equilibrium interest rate does not decline by the difference between the neutral interest rates i and i ´. As
output rises from its initial value, the central bank’s desired interest rate rises, even after the initial lowering of the neutral rate. That is the implication of the Taylor
rule.
The TR curve can also shift upwards, meaning that the central bank chooses a higher rate of interest at every level of GDP. When the central bank raises its
estimate of the neutral interest rate, exogenously, for any level of output (and at any phase of the business cycle), this is a contractionary shift in monetary
policy. This policy is captured as an exogenous increase in the neutral interest rate from . In the second panel of Figure 11.12, the TR curve
shifts up to TR″, and the economy comes to rest at point E. The outcome is lower output and a higher interest rate. Why? As the cost of lending rises for the
banks, they pass these higher rates on to their customers and the level of demand drops. As before, the rise of the interest rate in equilibrium is less than the
original disturbance, because the economy moves along a downward-sloping IS curve, leading to lower output and inducing the central bank to moderate its
tougher interest rate stance in the end.
11.5.4 A General Approach

We have given a number of examples of how we can use the IS–TR model to uncover the impact of exo-genous disturbances on output and the interest rate. We
could consider much more complex cases involving combinations of two or more of the disturbances mentioned in the last section. For example, what is the effect
of a simultaneous increase in stock prices and in the neutral interest rate? What might one predict from an exogenous appreciation of the exchange rate combined
with a cut in taxes? It is fairly common for policy-makers to combine or mix policy measures. As we will see in the next few chapters, the net effect of several policy
measures or disturbances taken together sometimes is qualitatively ambiguous. All the same, there is a well-defined strategy for studying these macroeconomic
effects:
Fig. 11.13 Short-term Interest Rates in the UK, USA, and Germany (%per annum), 1990–2016
Nominal interest rates (here: overnight interbank rates) fluctuate over the business cycle. Until the Great Recession, however, they maintained an average value across
fluctuations of several percentage points. Since late 2009, they have hovered close to zero. Because nominal interest rates cannot be reduced below zero, this limits the
effectiveness of monetary policy in severe recessions.
Source: OECD.
• First, we ask which curve is affected by the disturbance. This is where the remarks presented in Sections 11.3.4 and 11.4.4 are relevant.
• Second, we spot the new equilibrium at the intersection of the new IS and TR curves.
Finally, we can go beyond the graphical analysis and interpret the results by tracking down the various channels involved in the response of the economy to the
disturbance.
Box 11.4 Monetary Policy under Duress: The Zero Bound
The Taylor rule assumes that central banks can choose any interest rate they please. The global financial crisis showed that central bankers cannot
always do that. In particular, the nominal interest rate charged to households and firms cannot fall below zero. Furthermore, a negative nominal interest
rate would mean that depositors at banks would actually pay for the privilege. For households and firms at least, it would be more attractive to hold cash
under the mattress. Without additional restrictions on the rules governing the institution of money, a flight to cash would always be an option.22 Thus, the
linear form of the TR curve cannot literally be true. In fact, equation (11.12) implies that when output falls below the value Y (1 – i /b), the interest rate i must
turn negative; following its own rule, the interest rate desired by the central bank has hit a lower bound. If GDP drops precipitously as it did in 2008–9, it may
be simply impossible for central banks to reduce their interest rates as much as their policy preferences would dictate. In the Great Recession, it is
generally agreed that this was because the IS curve had shifted so sharply to the left. Figure 11.13, which plots interest rates in the UK, Germany, and the
USA in the period 1990–2011, shows that room for margin using the traditional instrument of interbank interest rates was exhausted by 2011. We will see
later that since then, interest rates monetary conditions have not improved much, making it necessary to explore a wider range of explanations of the low
nominal and real interest rates we observe. One of them is hinted at in Box 11.5.
Central banks have turned to non-conventional forms of monetary policy already discussed in Chapter 10. The Bank of England, the Federal Reserve, the
Bank of Japan, and the European Central Bank have engaged in massive purchases of both government and private debt to encourage banks to lend. The
zero lower bound on nominal interest rates poses a thorny problem for those who were sure that monetary policy contained enough weapons to fight the
worst recessions.
The IS–TR model can be used to study the effects of any arbitrary demand disturbance coming from the real or the monetary side, in any combination that can
be described in terms of the elements of the model. In the chapters that follow, we will do so extensively. The same procedure applies to: (1) identify which
schedule is affected by the disturbance, (2) identify the new equilibrium, and (3) understand what this means for the goods and money markets, using both panels
in Figure 11.12.
11.5.5 Monetary and Fiscal Policy in the IS–TR model

According to the Taylor rule, the central bank does not decide to be expansionary or contractionary. It merely reacts to economic conditions according to a
behavioural rule. For a given target interest rate, monetary policy is automatically expansionary when the output gap is negative. Conversely, it becomes
contractionary when the output gap is positive. It is natural then to ask: why would the central bank shift its own Taylor rule as it did in the previous section?
Why would it deliberately increase interest rates unconditionally?
The central bank can and does change its behaviour. This is because monetary policy does not occur in a vacuum. New information comes to light each day
that conditions or influences the policy stance of the central bank. For example, the central bank may learn that an adverse demand disturbance is about to occur,
and decide to cut interest rates—or come under political pressure to do so. In that case, the TR curve would shift downwards. According to (11.12), this is the
case when the central bank revises downwards its target interest rate. Similarly, a new forecast of sharply rising future inflation in spite of current moderate
inflation could prompt the central bank to raise its short-term target rate. These types of policy changes do in fact occur, and are best thought of as exogenous
shifts of the TR curve in the IS–TR framework.
The image of monetary policy conveyed by this discussion is a rather optimistic one. As a matter of fact, monetary and fiscal policy have served as important
means of smoothing the business cycle, and a more detailed discussion of these issues can be found in Chapter 16. All the same, the severity of the Great
Recession has challenged policy-makers in fundamental ways and pointed out the limitations of policy. One challenge, already noted in Chapter 10, and explained
in more detail in Box 11.4, is the zero lower bound: the Taylor rule cannot be valid when market rates approach zero. In effect, the TR curve becomes flat as in
Figure 11.14. In Chapter 14, we will see that setting the interest rate in reality means that the central bank needs to keep an eye on a target rate of inflation as well
as have a reasonably good idea of what is going on in the real economy at the same time. Central bankers do not have an easy life these days.
The zero lower bound was discussed in Chapter 10. It can be represented by the flat segment of the TR curve in Figure 11.14. If the economy is at point B,
reductions in the target interest rate have no effect on GDP. In addition, there are two other important consequences: (1) exogenous shifts in aggregate demand
have a larger effect on GDP, because they are no longer cushioned by an upward-sloped TR curve, and (2) the burden for restoring short-run output to previous
levels falls completely on the IS curve (real aggregate demand).
The Great Recession is best thought of as a massive leftward shift in the IS curve, just as shown in Figure 11.14. For Europe in the years 2008–2010, aggregate
demand collapsed for several reasons. As a direct result of the US recession, European exports fell by almost 20% in 2008–2009. Exports fell not only from the US
but from other countries, due to the credit freeze-up and collapse of animal spirits. Tobin’s q fell around the world, signalling a lower attractiveness of investment.
A sharp decline in US asset prices made European households poorer, both directly and indirectly, as banks were threatened. The IS curve shifted to IS´ in Figure
11.14. Because of their high indebtedness, several European governments followed by cutting government spending—exactly the opposite of what would be
called for under normal circumstances. The already dramatic leftward shift of the IS brought interest rates down, even in Europe. But zero proved to be the limit of
monetary policy. Shifting the TR curve out by reducing the target rate would be too little, too late.
Fig. 11.14 The Zero Lower Bound and the Great Recession
The Great Recession meant a dramatic leftward shift of the IS curve. A succession of adverse demand disturbances to IS’, which intersects the TR schedule in its horizontal
segment, corresponding to the zero lower bound. Monetary policy in its conventional forms has little effect on GDP, and must resort to unconventional or non-standard forms.
Many policy-makers have endorsed fiscal policy as a way out of the trap shown in Figure 11.14. Engineering a rightward shift of the IS curve would restore
demand to its previous level and, given the flat TR curve, would not be ‘crowded out’ by increasing interest rates. The question for Europe and for countries in
general is how to do it, under the present circumstances. Box 11.5 gives a review of these issues and poses an interesting question about the sources of those
demand shifts, some of which may not have to do with the Great Recession at all.
Box 11.5 Secular Stagnation Ahead?
The persistence of very low interest rates around the world has led to an intriguing possibility. Could the world become stuck at the zero bound for a long
time? This view has been put forward by Lawrence Summers, an economist at Harvard University and former US Secretary of the Treasury. He observes
that investment rates have slowed in recent decades, and have not fully recovered from the Great Recession. At the same time, global savings have
increased rapidly. In developed countries, life expectancy is rising, social security is limited, and an ageing population wants to put more and more money
aside for old age. Summers also notes that saving rates are very high in emerging economies—China is a prime example—at a time where financial
markets are ever more integrated. An exogenous increase in savings, unmatched by investment, implies a downward shift in the world’s IS curve. This can
also be represented in Figure 11.14 as a shift from IS to IS’. The figure also shows that the TR curve becomes horizontal once the interest rate reaches
zero. The new equilibrium occurs at point B, which is to the left of point A, meaning a rather depressed economy. This is not surprising: if the world saves
much more, it consumes much less and demand shrinks.
The problem is that the central bank is unable to bring the interest rate below zero. It is effectively powerless and cannot move the economy out of its
depressed state. The natural solution would be for governments to dissave, i.e. to run deficits, which would shift the IS curve back up and out. Summers
has pointed out that government investment in infrastructure—such as renovating Kennedy Airport in New York City—has a much higher social return than
the current low rates of interest on US debt. However, many governments are already highly indebted and therefore unable to run adequately large budget
deficits. This is the secular stagnation hypothesis. Since the factors that led to a strong rise in world savings are long-lasting (demography, the
emergence of high-saving economies) the situation may not improve soon. It comes on top of the view that technological change has slowed, as explained
in Box 3.5.
Summary
1 General equilibrium occurs when all markets simultaneously clear. In the short run, macroeconomic equilibrium is characterized by equilibrium in two markets: the market for
goods and services and the domestic money market, which is governed by monetary policy adopted by the central bank.
2 The Keynesian assumption posits that prices are sticky, which implies that output is driven by demand. When prices are constant, inflation is nil—and expected to remain nil—
and the nominal and real interest rates are the same.
3 The consumption function states that private consumption spending depends on disposable income and wealth. The investment function relates investment spending by firms
to the cost of obtaining funds, measured by the real interest rate and Tobin’s q.
4 The primary current account improves when income and output in the rest of the world expand and the real exchange depreciates, and worsens when domestic GDP and
absorption rise at home.
5 An autonomous increase in demand for domestic goods triggers a multiplier mechanism: more demand means more output, and more output means a higher income and hence a
new round of demand increases. The multiplier process is dampened by leakages in the income–demand chain: savings, taxes, and imports.
6 The IS curve represents the GDP levels and interest rates compatible with equilibrium in the market for goods and services. It is downward-sloping because a higher interest rate
reduces domestic demand and output. The IS curve is flatter when demand is sensitive to the interest rate and the multiplier is large.
7 The TR curve describes how the central bank sets the interest rate in order to stabilize inflation around its target and output around its trend. It is upward-sloping because the
central bank generally raises interest rates when output rises. The short-run position of the TR curve is determined by the central bank’s own target interest rate, which is
exogenous in the IS–TR framework.
8 Macroeconomic equilibrium is found at the intersection of the IS and TR curves. This allows us to study how GDP and the interest rate respond to exogenous disturbances.
Key Concepts
recessions
Keynesian assumption
dichotomy
general equilibrium
aggregate supply
aggregate demand
net exports
goods market equilibrium
consumption function, investment function
import, export, net export function
desired demand function
equilibrium GDP
Keynesian demand multiplier
IS curve
excess supply/excess demand
Taylor rule
TR curve
monetary targeting
target interest rate, short-run neutral interest rate
real disturbances, monetary disturbances
expansionary monetary policy
expansionary/contractionary disturbances
secular stagnation hypothesis
Exercises
1 Compare two economies, country A and country B. Country A has a higher marginal propensity to consume from disposable income than country B. Which should have a
flatter desired demand schedule? Show graphically that the steeper the desired demand schedule, the larger is the Keynesian multiplier.
2 Desired demand (DD) is represented by the following simplified function:
Domestic and foreign price levels have been assumed constant and equal to one. Let i = 5% throughout. Initially G = T = 3,000. σ is the nominal exchange rate and
is assumed fixed at one.
(a) Compute the effect on GDP of an increase in T from 3,000 to 3,500. Show your result graphically. What is the value of the (lump-sum) tax multiplier?
(b) Compute the effect on GDP of an increase in G from 3,000 to 3,500. Show your result graphically. What is the value of the government spending multiplier?
(c) Compute the net effect on GDP when both G and T increase by the same amount, from 3,000 to 3,500. Show your result graphically. What is the value of the
balanced budget multiplier? Compare your answers and discuss.
3 Using Box 11.1, show that if the government increases spending but some of that increased spending falls on imports—and that fraction is the same as the marginal propensity
to import (z), that the multiplier is smaller than 1/(1 − c(1 − z)). When is the demand multiplier less than 1? Can it be negative?
4 In reality, tax revenues are not exogenous, but move positively with income (GDP). Suppose that , where t is an exogenous tax rate. Use this tax revenue function to
derive the government spending multiplier in the following simple model:
5 Compare your answer with that when t = 0. Comment on your result. Define the IS curve. Why is it downward-sloping? Show how the IS curve divides the IS–TR diagram into
two regions, such that points off the IS curve correspond to market disequilibrium, which can be characterized as excess demand or supply in the goods market.
6 Derive the IS curve for the economy in Exercise 4, with . Plot it in (Y, i) space.
7 Define the TR curve. Why is it upward-sloping? Explain why the economy cannot be off the TR curve, by definition. Show how interest rates and output respond to situations
in which the economy is away from the crossing of IS and TR curves.
8 Suppose that the Taylor rule of an economy is , where i is measured in percentage points. Plot the TR curve in the diagram with Y on the horizontal axis
and i on the vertical axis. What is the effect of an increase in from 5 to 6 on the TR curve? Solve for an analytical expression for output (Y) and interest rate (i) in an
equilibrium in an economy with this TR curve and the IS curve in Exercise 6.
9 The policy mix describes a combination of fiscal and monetary policy actions. Show graphically how a combination of fiscal and monetary policies could be used to reduce
output without changing the (effective) interest rate.
10 Show what happens to the money supply when demand for money increases exogenously, holding the TR curve constant. Suppose that, in response to an exogenous decline in
the demand for money, the central bank raises its target interest rate. What happens to the money supply? Why do you think the central bank might react in this way?
11 Using Figure 11.10, explain why the TR schedule becomes vertical when b = ∞. (It may help to reason in the following way: consider the effect of a larger b, and then let b
become arbitrarily large.)
12 Show the effect on the TR schedule of an increase in the sensitivity b of the Taylor rule. What is the effect on the interest rate? (Hint: start from a situation where output is at
its trend level.)
13 Consider an increase in demand for goods in the IS–TR model. Using the apparatus of Figure 11.12, show graphically what happens to output, the interest rate, and money
supply when the TR schedule is horizontal. How would you answer the question if the TR schedule is upward-sloping? Interpret the difference.
14 This exercise confirms the argument presented in footnote 3. Ignoring foreign absorption, the NX function is NX(A, σ). Note that absorption is A = C + I + G, so with (11.1)
we have Y = A + NX(A, σ). Show that this implies NX = NX(Y, σ). Use the simple representation of the NX function NX = a − bA − cσ to compute the modified function
NX(Y, σ).
Essay Questions
1 Military spending is considered to be consumption in the GDP accounts. Explain why sharp increases in military spending in the USA and the UK after the terrorist attacks in
2001 increased GDP in those countries in the short run. Would you expect the effect to be larger in the US or the UK, and why? How does your answer compare with the long-
run effects of such spending?
2 Does the Taylor rule summarize everything about monetary policy of a central bank? What do you think determines the slope of the Taylor rule by central banks? Is your
answer likely to be constant over time?
3 There are many ways of conducting an expansionary fiscal policy: raising public spending, cutting income taxes, profit taxes, or VAT. What difference does it make for different
groups in the country? In your discussion, as you take a general equilibrium view of the question, look at the two possible exchange rate regimes.
4 The behaviour of inventories is a closely watched indicator of the state of the economy. Discuss under which conditions it is a valid indicator of things to come. In particular,
does an increase in inventories signal an economic slowdown or a boom? (Hint: inventory changes can be intended or unintended.)
5 ‘Open economies, because they have low multipliers, are likely to be more stable than large economies.’ Discuss.
1 Generally, wealth changes very slowly in response to household savings, so this assumption can be justified for short-run analysis. On the other hand, when asset prices change, wealth
can change suddenly. A good example of this is the consequence of the recent collapse of housing prices in the United States, Spain, and Ireland in the late 2000s. We return to this
point later on.
2 More precisely, firms use the interest rate to evaluate investment projects. This explains not only why Tobin’s q falls when interest rates rise, but also why a firm deciding how to
invest its money would invest less in physical equipment when interest rates are higher.
3 Recall that absorption A is defined as A = C + I + G.
4 This step allows us to eliminate the variable absorption (A) from the analysis. We could carry it explicitly in our characterization of aggregate demand and eliminate it at the end
instead, but the final result is the same. Exercise 14 at the end of this chapter invites you to check this assertion.
5 Macroeconomists have long grappled with the conditions that are necessary and sufficient for a unique value of Y. For an introductory course to macroeconomics, it sufficient to
assume that those conditions are met!
6 Changes in the stock of inventories are treated in the national income and product accounts as investment, irrespective of whether they are voluntary or involuntary. Even when the
economy is out of equilibrium, the identity Y = C + I + G + (X − Z) holds: measured investment is equal to the sum of intended investment and unanticipated changes.
7 This embodies an important assumption that the entire first round of spending is spent on goods produced domestically. In many open economies this assumption is implausible, so
that some of the first round ‘leaks’ immediately in the form of imports. In one of the exercises at the end of the chapter you are asked to show that the multiplier in this case is
(1−z)/(1−c(1−z)) which may not be greater than 1.
8 The overall effect may take several months to complete.
9 Here, however, taxes are exogenous and constant.
10 Using the result in Box 11.1, it is easy to see that a lower c or higher z decreases the multiplier.
11 We treat net taxes as exogenous for simplicity, for the time being. In the WebAppendix to this chapter, more realistically, taxes are allowed to depend positively on income Y.
Indeed, in practice, governments usually set tax rates, and tax revenues tend to rise with output and income.
12 We can add another channel. A rise in q means that stock prices (the price of shares in companies) increase. Since stocks are part of wealth Ω, wealth too increases. This raises
private consumption in line with the consumption function (11.2). Thus wealth can be made endogenous as well.
13 The name of this curve comes from the identity (2.6): I − S = T − G + X – Z, and was proposed by Nobel Prize laureate Sir John Hicks. For simplicity, he assumed government budget
balance (T = G) and no foreign trade (X − Z = 0), so the identity can be reduced to I = S. We draw the IS curve as a line because we do not really know, nor do we need to know, its exact
shape. For a derivation of the IS curve using calculus, see the WebAppendix.
14 The slope of the IS curve is formally derived in the WebAppendix.
15 While Keynesian analysis often assumes inflation equal to zero, this is not necessary for most of the analysis that follows.
16 To simplify the graphical analysis in this chapter and the next, we will consider only absolute deviations from trend output.
17 We saw in Chapter 10 that different central banks employ different ‘battle tactics’ (open market purchases and sales, rediscounting), but they all have the same overall objective—
setting the market interest rate at some target level.
18 Keynes’ idea that the demand and supply for money are the primary determinants of interest rates—as opposed to the demand and supply for loans on banks’ balance sheets—was
revolutionary. His analysis consisted of a couple of equations, and there was little graphical analysis to help. Hicks clothed Keynes’ ideas in mathematical garb, summarizing money
market equilibrium as the equality of a money supply under the control of the central bank and Keynes’ ‘liquidity preference’—the demand for money described in Chapter 9.
19 The name ‘LM’ originates from the fact that, along the curve, the demand for liquidity (L) equals the money supply (M) in equation (11.13). For an explicit derivation of the slope
of the LM curve using calculus, see the WebAppendix to this chapter.
20 In Chapter 14 we will see that in the long run, the neutral interest rate must be equal to the sum of the equilibrium real interest rate r plus the target rate of inflation. In the short run,
central banks can choose the target interest rate to deviate from this long-run condition.
21 We will consider the increase in government purchases as G irrespective of whether it is government consumption or investment.
22 Silvio Gesell, an Austrian monetary economist of the early twentieth century, devised a system of Schwundgeld (vanishing money), which actually reduced the value of banknotes over
time. Many have tried to implement this idea but it has never caught on, probably because there was always a conventional form of cash available as an alternative.
International Capital Flows and
Macroeconomic Equilibrium 12
12.1 Overview
12.3.1 The Interest Rate Parity Condition
12.3.2 The IFM Line
12.3.3 Capital Mobility and Financial Account Restrictions
12.3.4 Exchange Rate Regimes
12.3.5 Preview of What Follows
12.4.1 What is a Fixed Exchange Rate Regime, Exactly?
12.4.2 The Loss of Monetary Policy Autonomy—No TR Curve
12.4.3 Demand Disturbances—Shifts of the IS Curve
12.4.4 International Financial Disturbances—Shifts of the IFM line
12.4.5 A Parity Change
12.5 Output and Interest Rate Determination under Flexible Exchange Rates
12.5.1 Monetary Policy Disturbances—Shifts of the TR curve
12.5.2 Demand Disturbances—Shifts of the IS curve
12.5.3 International Financial Disturbances—Shifts of the IFM Line
12.5.4 Is a Large Economy a Closed Economy?
Summary
The immediate cause of profit-oriented capital movements is an interest-rate differential. The main point is to find out how this interest-rate differential can come about . . . . A
more complete theory of capital movements is an indispensable foundation for the study of the international, or more generally the interlocal, aspects of business cycles.
R. Nurkse1
12.1 Overview
Chapter 11 presented the determination of demand and output when the Keynesian assumption holds and aggregate demand determines output in the short run.
While the IS curve recognizes the existence of trade in goods and services, we ignored international trade in financial assets. International trade in financial
assets is important for a number of reasons. First, every current account deficit must be financed by net borrowing abroad, and every current account surplus
implies net national saving. Second, and more important, trade in financial assets can take on a life of its own, driven by international differences in the rate of
return on comparable investment opportunities.
Opening up the economy to trade in financial assets changes the IS–TR analysis of Chapter 11 significantly. This chapter presents the small open-economy
version of the known as the Mundell–Fleming model.2 This framework is designed for studying the behaviour of small, open economies and is particularly
relevant in a world in which financial markets are integrated and capital is mobile internationally. It holds special significance for Europe, both for countries within
the monetary union and for those outside it.
Central to this chapter is the crucial role of the exchange rate regime. We consider two possibilities. In a fixed exchange rate regime, the central bank
commits to maintaining a certain value of the nominal exchange rate, the value of its currency in terms of other currencies. Second, and alternatively, the central
bank lets the exchange rate float freely, leaving its value to be determined by the market. The exchange rate regime turns out to be crucial to the behaviour of the
economy. For this reason, the two regimes of fixed and flexible exchange rates are studied separately.
The Mundell–Fleming model is relevant for Europe for several reasons. First, all European economies are open, and most are small. Second, for those in the
European Monetary Union, the nominal value of the common currency, the euro, is fixed at 1:1 across member countries, without any actively intervening
national central banks.3 The Eurozone is thus one of the most credible of all forms of fixed exchange rate regimes, since the option of changing the exchange rate
is ruled out—without leaving the arrangement entirely. Third, many nations within and outside the European Union have floating exchange rates—Poland,
Iceland, Norway, Sweden, and the UK to name a few—and have policy options which differ from countries in the Eurozone.
We start by looking at the implications of financial openness. When there is free trade in financial assets without capital controls, a fully financially
integrated economy loses the control of its own interest rate. This limits the room for manoeuvre of the central bank and explains why and in which way the
exchange regime matters so much. We begin with the stark implications of the fixed exchange rate regime, and then contrast these with the case of a flexible
exchange rate regime. Along the way, we develop and refine the Mundell–Fleming model. This is a powerful tool for understanding business cycles, the role of
macroeconomic policies, and details of the interplay between goods markets, domestic financial markets, and international financial integration.
What is a small, open economy, exactly? Small means that the economy has no discernable impact on the rest of the world. Open means that it is profoundly
affected by events that take place beyond its borders. Trade openness—sometimes referred to as current account openness—occurs when goods and services
can be exchanged with only limited impediments through tariffs, quotas, and various other trade barriers. Financial openness—sometimes referred to as financial
account openness, in reference to the balance of payments discussed in Chapter 2—describes the situation when people can lend and borrow freely. Box 12.1
further explains how openness is defined and measured. Most countries can be seen as small and open. Prominent exceptions are Japan, China, USA, and the
European Union taken as whole. These entities are large in absolute size and their trade with the rest of the world is comparatively limited. They can affect
conditions elsewhere in the world but even their impact has diminished in recent years as new economic powers emerge in Asia and South America. As a first
approximation, the ‘small and open economy assumption’ may be increasingly applicable to them as well.
Box 12.1 What is Openness?
In Chapter 1, we presented a first definition of openness related to external trade in goods and services, which is related to GDP by exports and imports.4
Table 12.1 reminds us of how open and how closed some countries are. As in Table 1.2, we present the average of exports and imports as a percentage
of GDP. As one might expect, Table 12.1 shows an inverse relationship between economic size and trade openness: the bigger the country, the less it
needs foreign markets and suppliers. Even so, the two largest entities, the euro area and the USA, are far from being closed.
The second definition of openness involves financial markets. In the last two columns of Table 12.1, financial openness is measured by how much a
country has lent or invested in the rest of the world (its cross-border assets) or how much it has borrowed abroad (its cross-border liabilities), where
‘country’ refers to households, banks, firms, and the public sector. The table shows both assets and liabilities. In many cases, assets and liabilities are
not very different (in the same way that exports and imports can be large and of the same order of magnitude). Financial openness allows borrowers and
lenders to search the world for the best financial deal. Still, some countries, such as the USA, Brazil, or Poland have more external liabilities than assets.
They are called net debtor countries.5 Conversely, Japan and Switzerland are net creditors.
Developed countries tend to be more open than developing countries, both in trade and finance. The globalization process has begun to change this
relationship. Poland, Korea, Brazil, China, and India are among the growing number of so-called emerging market countries in the process of becoming
fully integrated into world financial markets.
Table 12.1 Measures of Openness and Economic Size, 2014 (% of GDP)
Share of World GDP (%) Trade openness Total Assets Total Liabilities
Denmark 0.4 51.0 271.9 229.3
Belgium 0.7 83.5 439.5 388.7
Germany 5.0 42.4 236.2 201.1
Netherlands 1.1 77.2 1000.8 939.7
European Monetary Union 17.2 41.7 194.9 208.2
Poland 0.7 46.8 41.6 103.7
Sweden 0.7 42.7 250.1 257.1
Switzerland 0.9 58.7 600.0 507.9
United Kingdom 3.8 29.3 531.5 554.3
United States 22.4 15.0 140.0 182.2
Brazil 3.0 12.6 31.6 64.5
China 13.3 20.8 61.1 46.6
Japan 5.9 19.3 170.0 104.3
Korea 1.8 48.0 76.4 70.5
Sources: The World Bank, World Development Indicators; Lane and Milesi-Ferretti (2007), updated.
12.3.1 The Interest Rate Parity Condition

We encountered the exchange rate for the first time in Chapter 5. We saw that, in the long run, the nominal exchange rate is driven by the purchasing power
parity principle, a consequence of competition in trade in goods and services. From a short-run perspective, trade in financial assets is more important because
the movements are much larger, as explained in Chapter 7. An alternative parity condition will lie behind the determination of interest rates and nominal exchange
rates. This is the interest rate parity condition. Arbitrage in the financial markets leads to massive and almost instantaneous capital flows as investors try to
take advantage of international differences in investment opportunities.
Consider a typical international investor, for example a large bank, which routinely borrows and lends all over the world. The domestic interest rate is i, while
the international rate of return is i*. Note that we describe i* as a rate of return because, as we will see later, we must take expected movements of the
exchange rate into account. It is not merely the foreign interest rate.
Now suppose that i < i*. One option available to our international investor is to borrow at home, where the interest rate is low, and invest abroad, pocketing
the difference.6 As they do so, they raise the demand for money at home and i will rise towards i*. Possibly, i* will decline towards i because large amounts of
money flow into foreign financial markets, although this is only true if the home economy is sufficiently ‘large’ to affect the rest of the world. This will go on as
long as i < i*. Conversely, if i > i*, money will flow from the rest of the world into our high-interest rate country, pushing i down and i* up. The implication is that
neither i > i* nor i < i* is possible. The only logical way out is i = i*. This is the interest rate parity condition.
Just as the purchasing power parity condition equates price levels, the interest rate parity condition states that returns on similar assets cannot differ
systematically across countries when there is free trade in financial assets—when financial capital is perfectly mobile. In contrast to the purchasing power parity
condition, which takes time to assert itself because good prices move slowly, the interest parity condition is a property which applies even in the very short run.
We saw in Chapter 7 that financial traders—hedge funds, banks, individual investors—are constantly monitoring the whole world, minute by minute, for
arbitrage opportunities. When they find one, they can move huge amounts of money instantly and at little cost. As a result, the working assumption is that the
interest rate parity condition is satisfied at all times, and that any deviation is eliminated immediately. Interest rate parity is simply the equilibrium condition for
international financial markets.
12.3.2 The IFM Line

The interest rate parity condition is depicted as the horizontal international financial markets (IFM) line in Figure 12.1. Here is where the small country
assumption comes in. If the economy is small relative to the rest of the world, the foreign or international rate of return i* is exogenous. It changes for reasons
unrelated to domestic conditions. It follows that it is the domestic interest rate i in the small economy that bears the burden of adjustment whenever the interest
rate parity condition is not satisfied. In what follows, we will assume that i* is constant, unless we explicitly consider a shock, or exogenous change, to its
position.
Fig. 12.1 International Financial Market Equilibrium

When capital can move freely across borders, assets of similar quality should yield the same return. Otherwise, unexploited profit opportunities would exist—borrowing would
occur where interest rates are low and lending where they are high. This is incompatible with the international market equilibrium. Note that i* denotes the return on foreign
assets converted in the domestic currency. For a small financially open economy, i* is exogenous.
As with the IS or TR curves, it is instructive to discuss what happens when the economy is not on the IFM line. Above it, the domestic interest rate exceeds
the international required rate of return. Capital inflows occur as international investors take advantage of higher returns at home. The competitive pressure
resulting from those inflows, which can be massive, promptly drives the domestic interest rate i down to the world level i*. Conversely, below the IFM line,
capital flows out and increasing scarcity drives up the interest rate i back to i* from below.
12.3.3 Capital Mobility and Financial Account Restrictions

So far, we have assumed that capital is freely mobile. In fact, some countries do impose restrictions on their financial accounts. The most extreme type consists of
declaring the national currency inconvertible. It becomes difficult or impossible to convert the local currency into foreign exchange.7 More moderate restrictions
limit or forbid some transactions, like investing abroad or the foreign acquisition of categories of domestic assets. Capital controls, as these restrictions are
generically called, were quite common until the mid-1970s. Figure 12.2 plots an index of financial account liberalization—the absence of restrictions on capital
movements—over the last three decades of the twentieth century. It reflects that developed countries largely abandoned capital controls in the 1980s. The same
process started a decade later among developing countries, but most continue to maintain restrictions on capital flows.
Fig. 12.2 The Evolution of Financial Account Liberalization 1970–2013
The figure presents an index of the degree of ease with which financial assets can be bought and sold internationally, as an unweighted average of developed or developing
countries. The higher the value, the fewer the restrictions imposed on capital movements. The value of the index is normalized so the value for advanced economies in 1970
equals 1.0. While developing countries have relaxed capital restrictions over the past few decades, they continue to maintain significantly stricter controls than developed
countries. Note that the financial liberalization trend stopped temporarily in the mid-1990s when several countries were hit by currency crises. Similarly, another reversal
occurred after the global financial crisis struck in 2008.
Source: The Chinn-Ito Index, http://web.pdx.edu/~ito/Chinn-Ito_website.htm; IMF World Economic Outlook.
With capital controls, the interest rate parity need not hold, because the mechanism described in the previous section cannot operate. Legal restrictions
prevent traders from taking advantage of profit opportunities. Of course, the temptation to circumvent the law is strong and indeed capital controls tend to be
systematically evaded. As long as they are not fully evaded, however, the interest parity condition is likely to be systematically violated.
Why do countries impose restrictions on capital movements? We have seen how deviations from the interest parity condition can trigger massive capital
flows. As we will see later, these massive movements of capital can interfere with the central bank’s conduct of monetary policy, either because the central bank
would like to influence monetary conditions at home, or because it would like to influence the level of the nominal exchange rate. More often than not, countries
abandon capital controls because the forces of international financial integration are simply too powerful.
As we proceed to explore the implications of the interest parity condition, it is important to keep in mind that full capital mobility is not a universal feature. The
Mundell–Fleming framework presented in this chapter is well suited to studying developed countries, but it is less helpful for developing economies. Partly
because there is a wide variety of capital controls, partly because their effects vary depending on the prevalent conditions, there is no simple equivalent
treatment for these countries. If capital controls were watertight, we could simply ignore the interest rate parity condition and work with the macroeconomic
equilibrium described in Chapter 11. As a first approximation, this may be the best way to proceed for some countries.
12.3.4 Exchange Rate Regimes

Because it draws the attention to capital flows, the IFM line forces us to take explicit account of the exchange rate regime. In fact, it is difficult if not impossible to
predict the consequences of macroeconomic shocks without doing so. Because of the enormous variety of regimes, it is convenient to consider two polar cases:
(1) fixed exchange rates, and (2) fully floating exchange rates.
In 2014, the International Monetary Fund (IMF) reported that 107 of its 189 member countries had adopted one form of fixed exchange rates or another. Under a
fixed exchange rate regime, the monetary authorities maintain the value of the exchange rate at a publicly announced parity. This is an official value of the
national currency in terms of another currency, usually the dollar or the euro. For instance, Denmark and several countries in Central and Eastern Europe
(currently, Bosnia and Herzegovina, Bulgaria, Croatia, and Macedonia) peg their currencies to the euro. Most developing countries, including some Latin
American and several African countries, have adopted fixed exchange rate regimes of various types. Taken to the limit, national sharing of a currency—a
monetary union—is the ultimate form of fixed exchange rate regimes. Besides the 19 EU countries of the Eurozone, Andorra, Montenegro, and San Marino have
also adopted the euro.
The other extreme form of exchange rate policy is to allow the exchange rate to float freely. This means that the central bank takes no direct responsibility for
the value of its currency, which is set in foreign exchange markets as explained in Chapter 7. The floating rate regime currently applies to the Eurozone, the UK,
the USA, and Japan, as well as smaller European economies like the Czech Republic, Switzerland, Hungary, Norway, and Poland. It is studied in Section 12.5.
In between these two polar regimes there are a number of versions of ‘managed floating’. Here, the authorities do not commit themselves to a particular parity
but nevertheless attempt to prevent large exchange rate fluctuations. This is the case for Romania and Serbia as well as for many Latin American and East Asian
countries. This regime is a ‘mixed bag’ that does not lend itself to a clean-cut analysis. Depending on the weight put on exchange rate stabilization, it can be
approximated by either a fixed or a flexible exchange rate regime.
12.3.5 Preview of What Follows

The macroeconomic equilibrium studied in Chapter 11 included two markets: the goods market and the money market. We have just added a third: the
international financial market. For each of these three markets, we have established those combinations of real GDP and nominal interest rate which are
compatible with equilibrium. The IS curve deals with the market for goods and services, the TR curve corresponds to monetary policy and the money market,
while the IFM line describes the international financial equilibrium. Each of these curves are drawn ‘all other things equal’, meaning that they do not shift as long
as the exogenous variables associated with that curve remain unchanged.
In the following sections, we extend the Keynesian analysis of the last chapter to include a wide variety of disturbances that small open economies face. The
analysis is thus inherently more realistic and relevant than in Chapter 11, yet it builds carefully on that analysis. In studying each particular scenario, it is
important to ask and answer the following three questions:
(1) What is the exchange regime? The exchange rate regime is crucial for understanding how the economy reacts, because it leads to the elimination of one of the curves used in the
IS–TR analysis. Under a fixed exchange rate regime, the TR curve loses its relevance, while under a flexible exchange rate regime, the IS curve loses its relevance.
(2) Which curve is affected by the disturbance under consideration? An exogenous disturbance affects (at least) one of the three markets and therefore leads to a shift in the
corresponding curve.
(3) Where is the new equilibrium and how should it be interpreted? How and why has the economy moved from the initial to the new general equilibrium?
12.4.1 What is a Fixed Exchange Rate Regime, Exactly?

In a fixed exchange rate regime, a central bank declares a parity for its currency and commits to enforce this exchange rate, usually within margins. We saw in
Chapters 9 and 10 that a central bank that fixes an interest rate must be prepared to supply commercial banks with whatever volume of reserves they demand at
that rate. A very similar reasoning applies in the case of a fixed exchange rate. In fact, the central bank makes a market in its own money; it uses its assets and
liabilities to trade in foreign exchange, just as it would conduct any other open market operation described in Section 10.3. Imagine that, one morning, the
currency appreciates above its declared parity. The central bank is committed to weaken its own currency by selling whatever quantity of it necessary to bring
the exchange rate back down again. Similarly, if the value of the currency fell below its parity, the central bank would have to intervene, purchasing back its own
currency and selling its holdings of foreign currency. These operations are called exchange market interventions, as already mentioned in Chapter 2.
12.4.2 The Loss of Monetary Policy Autonomy—No TR Curve

The similarity with open market interventions goes much further, with a crucially important implication. We have seen in Chapter 9 that a central bank cannot
simultaneously choose the money supply and the interest rate. In an open economy with full capital mobility, the message is even more severe: a central bank
cannot choose both the exchange rate and the interest rate (as would be summarized by its Taylor rule). The only possible interest rate is the one consistent with
the interest parity condition. Because the central bank cannot choose the interest rate in an open economy with full capital mobility, the TR curve has no
meaning! We only use the other two curves, IS and IFM, to study the behaviour of the economy.
Another way of thinking about the loss of monetary autonomy under a fixed exchange rate regime is that a central bank can only choose one point on the
money demand curve presented in Chapter 9 and shown again in Figure 12.3. The central bank faces two constraints: (1) the demand for money, and (2) the
foreign rate of return i* required by the international financial market equilibrium condition.8 The only plausible outcome is point A. There is simply no room for
manoeuvre and the central bank must provide whatever amount of money the market demands. Note that part of the demand for domestic currency may well
originate from abroad.
Fig. 12.3 The Money Market Under Fixed Exchange Rates

Starting from point A, the central bank decides to lower the interest rate from i*, the foreign rate of return, to i. Money market equilibrium occurs at point B. But i < i* is incompatible
with international financial market equilibrium, so capital flows out. In order to support its currency, the central bank must buy it on the foreign exchange market. In that case, the
central bank re-absorbs the money that it had previously created and this money ceases to exist. As a result, the money supply declines and the interest rate rises along D. The
central bank must continue its interventions as long as i < i*. The only position compatible with international financial market equilibrium is point A where i = i*. The amount of
money supplied by the central bank under these conditions is endogenous.
Imagine that the central bank attempted to lower the interest rate to reach point B in Figure 12.3 with interest rate i. To keep things simple, we disregard any
possible impact of this action on output.9 From Figure 12.1, we know that capital flows out when i < i*. Traders borrow at the low rate i and, in order to invest
abroad and obtain the higher return i*, they sell the domestic currency. This weakens the currency on the foreign exchange market. In order to honour its fixed
exchange rate commitment, the central bank must intervene and buy back its own currency.
Now remember from Chapter 9 that money is ‘money in circulation’. As the central bank sells foreign exchange in the open market, it receives domestic money
(its own liabilities) in payment. When it performs this transaction, it withdraws that money from circulation, and the money supply declines . In Figure 12.3, we
move from point B up and to the left along the money demand curve. Capital outflows will proceed and foreign exchange market interventions continue until i =
i*, all the way back to point A. We see that whatever money is supplied on the open market must be promptly removed from the foreign exchange market. This
link can be severed, as explained in Box 12.2, but only temporarily. More generally, any attempt by the central bank to change the money supply or the domestic
interest rate will be undermined by a commitment to uphold the official exchange rate parity.
12.4.3 Demand Disturbances—Shifts of the IS Curve

How does an exogenous change in the demand for goods affect the level of output and the interest rate, the two endogenous variables in our framework? We
examined this question in Chapter 11, but in a context without international capital mobility. Adding full financial market integration changes the rules of the game
significantly.
With a fixed exchange rate and capital freely moving across borders, we know that we only need to consider the IS and IFM curves. We start from point A in
Figure 12.4. To fix ideas, we look at the role of fiscal policy. An expansionary policy, e.g. an increase in public spending or a tax reduction, raises domestic
demand. This analysis would apply to similar exogenous increases in aggregate demand, for example if firms became more optimistic and invested more (Tobin’s
q increases), or if foreign demand for domestic goods were to rise (Y* increases), which would increase exports and improve the primary current account.
Consider an increase in government purchases captured in Figure 12.4 by a rightward shift of the IS curve, from IS to IS′. The economy moves from point A to
point B at the intersection of IS′ and the IFM line. As in Chapter 11, the demand expansion results in an increase in income but now the interest rate remains
unchanged, equal to the foreign rate of return i*. GDP increases by the full extent of the increase in demand implied by disturbance: the increase in government
purchases times the multiplier.
Box 12.2 Monetary Side Effects of Foreign Exchange Market Interventions
A foreign exchange market intervention automatically affects the money supply. Foreign currency purchases by the central bank are paid for with the
domestic currency, which results in an increase in the domestic money stock—a liability of the central bank that rises to match additional assets.
Conversely, sales of foreign currency against the domestic currency result in a decline in both the central bank assets—the foreign assets—and liabilities
—as some of the domestic currency is withdrawn from circulation when purchased by the central bank. This link can be summarized by the following
simple representation of the central bank’s balance sheet:
(12.1) M0 = R + DC,
where DC represents domestic credit—a consolidation of all forms of credit to the rest of the economy—and R measures foreign exchange reserves
owned by the central bank. From (12.1) it can be seen that purchases of foreign currency lead to equal increases in central bank liabilities (M0) and
assets (R), with the opposite outcome when the foreign currency reserves are sold.
This process, which is summarized in Table 12.2 in the case of the sales of foreign currency reserves, is called unsterilized intervention. It explains the
direct automatic link between foreign exchange market interventions and the money supply, which is indeed endogenous when the central bank commits
to maintain a peg. Central banks sometimes attempt to break this automaticity through sterilized intervention. To that effect, the central bank offsets the
foreign exchange market intervention with another operation in the domestic money market. In the case of a sale of foreign currency, it re-injects the
previously purchased domestic currency by spending it on the open market, buying domestic assets, and thus giving back the very liquidity that it withdrew
when it sold its foreign exchange reserves. Through this sterilization intervention, labelled Step 2 in Table 12.2, both M0 and DC increase. Combining the
two steps (see the bottom row), we see that the central bank has raised its stock of domestic assets DC and reduced its stock of foreign exchange
reserves R by the same amount, leaving the money base M0 unchanged. Since the total money supply is a multiple of the base, the sterilized intervention
has in fact cut the link between the money stock and the foreign exchange market intervention.
Table 12.2 Sterilized and Unsterilized Foreign Exchange Market Interventions
Unsterilized interventions Sterilized interventions

M0 R DC M0 R DC
Step 1 – – unchanged – – unchanged
Step 2 + unchanged +
Overall – – unchanged unchanged – +
Experience shows that sterilized interventions can shield the money stock from foreign exchange interventions only for a limited time. It is easy to
understand why. The central bank will not be able to sustain point B in Figure 12.3. As long as i is below i*, capital continues to leave the country, and the
central bank must continue its interventions to maintain the exchange rate at its parity. This cannot go on forever, though, because the stock of foreign
exchange reserves is declining and will be exhausted sooner or later. If the capital flows are large, and they usually are, the reserves can be depleted in a
matter of days, sometimes even hours.
The situation is somewhat different when i is above i*. Sterilized interventions by the central bank lead to an accumulation of foreign exchange reserves,
which can be purchased with the monetary base it creates, effectively without limit. Naturally there must be a catch, and there is. Sterilization now takes the
form of sales of domestic assets, of which the central bank holds only limited amounts. Box 12.3 presents a case when a central bank has been unable
to sterilize its interventions in such a situation.
It might be tempting to conclude that financial openness does not make much of a difference in the context of full capital mobility. This would be completely
incorrect! To see why, imagine the situation in the absence of capital flows, e.g. because of watertight capital controls. In this case, we can ignore the IFM line,
the central bank recovers its ability to carry out an effective monetary policy, and the TR curve is relevant again. The new equilibrium occurs at point C. Note
that, relative to point B, income has increased less and the interest rate has risen. These two effects are related. The interest rate has increased because the
Taylor rule leads the central bank to raise the interest rate to limit output fluctuation (TR curve). Relative to point B, output is lower at point C because the higher
interest rate adversely affects investment spending. This effect reflects the fact that a higher demand reduces, or crowds out, investment spending when it is met
by a higher interest rate. In contrast, under full capital mobility the interest rate does not increase—there is no crowding-out effect—and the expansion is
stronger.
Fig. 12.4 Demand Shocks Under Fixed Exchange Rates
The demand expansion is shown as shifting the IS curve from IS to IS′. The new equilibrium occurs at point B where the goods and international financial markets are in
equilibrium. Without capital mobility, the equilibrium would occur at point C.
There is one demand disturbance that we cannot study, monetary policy. The combination of a fixed exchange rate regime and full capital mobility means that
the central bank has lost monetary policy autonomy.
12.4.4 International Financial Disturbances—Shifts of the IFM line

An international disturbance will usually take the form of a change in the foreign or international rate of return i*. Understanding how a shock to i* affects a small
open economy is an important step to understanding modern macroeconomics, and represents an important example of the interdependence of national
economies.
Consider the case when the foreign rate of return rises exogenously from i* to i*′, because, say, the monetary conditions around the world are becoming
tighter. The result is the upward shift of the IFM line to IFM′, as shown in Figure 12.5. At point A, the initial point of intersection of the IS and IFM curves, the
interest rate is now too low. Capital flows out and the central bank must intervene on the foreign exchange market to prevent a depreciation by buying back some
of its own currency. The money supply declines endogenously and the economy moves to point B. Output declines because the higher world interest rate
reduces aggregate demand.
Fig. 12.5 An International Financial Shock

The increase in the rate of return on foreign assets is captured by the shift from IFM to IFM′. Starting from point A, the new general equilibrium occurs at point B, where output
has declined. Under flexible exchange rates, starting from point A, the exchange rate depreciates. The resulting gain in competitiveness increases the demand for domestic
goods. The IS curve shifts to the right (not shown) until it passes through point C, where the new macroeconomic equilibrium occurs.
This example shows that, under a fixed exchange rate regime, the domestic economy is fully exposed to international financial disturbances. If interest rates rise
worldwide, they must rise at home as well, which is contractionary. The fixed exchange rate regime creates a situation of international monetary interdependence,
which reflects the loss of monetary policy autonomy. When the financial crisis started in the USA in 2007–2008, the world rate of return i*—strongly influenced
by the United States— fluctuated widely. It first rose as investors panicked and asked for higher returns, then it declined as the Federal Reserve Bank, the US
central bank, precipitously cut its own interest rate—one instance when the USA can be seen as a large economy that can pick its interest rate and affect the
whole world. Europe was impacted directly by these gyrations.
12.4.5 A Parity Change

Monetary policy is ineffective as long as the central bank remains committed to a specific exchange rate parity. That does not mean that monetary policy cannot
be used at all. Most countries that adopt a fixed exchange regime allow for adjustments of the exchange rate level. These regimes, called ‘fixed but adjustable’
exchange rates, provide some limited degree of monetary policy effectiveness.
Discrete, occasional changes in the exchange rate level are called revaluations or devaluations, depending on the direction of the change.10 A revaluation
is an increase in the external value of the currency: moving from S to S′ with S′ > S. A devaluation would correspond to S′ < S. For given price levels at home and
abroad (the Keynesian assumption), a change in the nominal exchange rate translates directly into a change of the real exchange rate σ = SP/P*. Herein lie the real
effects of nominal exchange rate changes.
Revaluations and devaluations are conducted by way of an official declaration. The central bank announces the new parity at which it stands ready to buy and
sell the domestic currency against foreign exchange. Consider the case of a nominal devaluation, a lowering of the nominal exchange rate S. The devaluation
lowers the real exchange rate σ, raises the country’s competitiveness, since its goods are now cheaper relative to foreign goods. Exports rise, imports decline, and
the net export account improves. Graphically, the IS curve shifts outwards to IS′ in Panel (b) of Figure 12.6. Output increases as the economy moves from point A
to point B along the IFM line, but the IFM line itself does not shift—the small country assumption implies that i* is exogenous. As would be expected, a
devaluation is expansionary. In the same way, a revaluation is contractionary.
What happens in the money and foreign exchange markets? A parity change is not merely a declaration of intent by the central bank; to be effective, intentions
must be backed by actions. Indeed, a devaluation must be accompanied by a monetary expansion, a revaluation by a monetary contraction. This is shown in
Panel (a) of Figure 12.6. The increase in output found in Panel (b) translates into a larger demand for money. In order to make the new parity ‘stick’, the central
bank must be willing to supply the additional money demanded at the going interest rate imposed by the international financial market equilibrium condition
(point B).
Suppose the central bank didn’t follow through with a monetary expansion, and chose to set the interest rate at some other, higher rate ( i > i*). For simplicity
suppose it simply kept the money supply constant. The domestic interest rate would rise towards point C. This would trigger capital inflows, which would force
the central bank to intervene on the foreign exchange market, buying foreign exchange and paying with domestic currency to prevent the appreciation. If the
intervention is unsterilized, the domestic money supply rises and we end up at point B. Whether it is done through open market operations or foreign exchange
market operations, the result is the same.
Box 12.3 presents a case study of the pitfalls involved in trying to have it both ways. For more than two decades, China experienced an export boom supported
by a steady expansion of investment and its underlying sustainable capacity. As a result, interest rates rose and capital inflows—reflecting a massive trade
surplus as well as foreign direct investment in China—led to upward pressure of the Chinese currency, which the central bank tried to prevent.
Fig. 12.6 A Devaluation

In Panel (b), a devaluation shifts the IS curve out to IS′. The economy moves from point A to point B and output increases. The output expansion raises the demand for money in
Panel (a), which shows that the central bank must allow the money supply to expand to keep the domestic interest rate i equal to the foreign rate of return i*.
We have seen that, when capital is freely mobile, monetary policy autonomy is lost under fixed exchange rates and that monetary and exchange rate policies
are just two sides of the same coin. A more accurate conclusion is that monetary policy can still be carried out, but through realignments—devaluations or
revaluations—i.e. by changing the exchange rate level without abandoning the regime itself. The behaviour of the Chinese central bank described in Box 12.3 can
be better understood in this light.
12.5 Output and Interest Rate Determination under Flexible Exchange

Rates
We now consider the case of a regime of freely floating exchange rates—a description of the situation of the United States, the euro area, the United Kingdom,
Sweden, and some Eastern European countries. The central bank explicitly refrains from intervening in foreign exchange markets. Three conclusions follow
immediately:
(1) The central bank recovers its ability to conduct monetary policy, as summarized by a Taylor rule and represented by the TR curve;
(2) By definition, the central bank gives up the exchange rate policy instrument. The value of the exchange rate is determined by market forces;
(3) Given the stickiness of the price level in the short run, a flexible exchange rate implies that the country’s external competitiveness is endogenous and so is the position of the IS
curve. This last implication is explored in the next section.
Box 12.3 Fixed Exchange Rates in China
In 1995, China began pegging its currency, the renminbi or RMB (officially called yuan), to the US dollar. In doing so, it was able to prevent an appreciation
of its own currency that could hurt its exports, which were growing fast and contributing heavily to an expansion of aggregate demand. Over time, this policy
led to a string of large current account surpluses. Then, recognizing its increasing attractiveness, international investors also began to invest in China,
resulting in large financial inflows. The consequence of this ‘double-surplus’ was to pressure the renminbi toward appreciation, which the People’s Bank
of China (PBOC) resisted by intervening massively on the foreign exchange market, accumulating an unprecedented stock of foreign exchange reserves
of nearly $4 trillion ($4 million millions, or $4,000,000,000,000!). For an uninformed observer, it might be natural to think that the Chinese are trying to ‘buy
up the world’. Yet such rapid growth in reserves is a natural product of the process described in Table 12.2.
Unsterilized intervention means that the money supply will grow, which in the long run means inflation (see Chapter 5). To prevent this, the POBC
undertook to absorb the liquidity effects of its dollar purchases. It began to issue its own debt, with the effect of reducing DC in (12.1)! In fact, it has even
brought DC into negative territory, meaning that the central bank was borrowing from the banking sector and not the other way round, as is normally the
case.11 For a long time, the People’s Bank of China was even making a profit on its sterilization activity!12 By late 2007, however, US interest rates
declined and Chinese rates rose. As a result, the People’s Bank of China has begun to lose money—an unusual outcome for a central bank. Given the
size of its reserves, the losses are significant, probably unsustainable. This may explain why it then started to let its currency appreciate faster.
Fig. 12.7 China: Foreign Exchange Reserves (US$ billion)
China has been accumulating foreign exchange reserves continually since the mid-1990s. In 2014 they reached their peak at $4 trillion, double the level they attained in 2009
—a growth rate of roughly 15% per annum.
Source: IMF, Macrobond China.
Growth hardly slowed after the global financial crisis. Preventing an appreciation led to a further doubling of reserves until 2014, when they peaked (see
Figure 12.7). In the aftermath, the PBOC found itself battling to slow down a currency depreciation, hence the decline in reserves in the figure. Graphically,
this can be readily interpreted in Figure 12.6 as an upward shift of the IFM line as investors require a higher risk premium. Moving to point B, however,
means lower growth, which the Chinese authorities wanted to avoid. China could have reimposed capital controls, but it is committed to ‘RMB
internationalization’, meaning convertibility and more exchange rate volatility. According to the Mundell–Fleming framework, this means moving to point B.
At the time of writing, Chinese foreign exchange reserves continue to decline, the RMB has depreciated a little, capital continues to flow out, and growth
has slowed significantly.
12.5.1 Monetary Policy Disturbances—Shifts of the TR curve

We start by imagining that the central bank pursues a more expansionary monetary policy—in the terminology of Chapter 11. In particular, we study the case
when the central bank is willing to set a lower interest rate at any level of output. The TR curve shifts downwards, as shown in Panel (b) of Figure 12.8. There is
no other exogenous change, so IS and IFM remain where they are. Note that the three curves now intersect each other two by two in three different positions:
points A, B, and C. Which of these points is sustainable as the equilibrium of the economy?
Fig. 12.8 Monetary Policy Disturbances Under Flexible Exchange Rates

Starting at point A in Panel (b), the central bank moves the TR curve to TR′. At point C, the lower interest rate triggers capital outflows and the exchange rate depreciates. The
resulting gain in competitiveness raises demand for domestic goods and the IS curve shifts to the right. It will move all the way to IS′ and full equilibrium is achieved at point B. In
Panel (a), we see that the central bank succeeds at increasing the money supply because money demand shifts outward due to the rise in output found in Panel (b).
Look at point C at the intersection of the IS and TR′ curves. The interest rate has declined, reflecting the easing of monetary policy, which has led to more
spending. Point C, however, is below the IFM line. With i < i*, capital flows out. This time, however, the central bank does not intervene on the foreign exchange
market, and allows the exchange rate to change. Under capital outflow, it will depreciate, as those exiting the currency sell it and buy other currencies. The
resulting real depreciation, given domestic and foreign prices, in turn, means that the economy becomes more competitive. Exports rise, imports decline, the
primary current account improves, and demand for domestic goods increases. Graphically, the IS curve starts shifting to the right. How far will it go? As long as i
< i*, capital outflows continue, the exchange rate continues to depreciate, and the IS curve continues to move rightward. When it reaches the position IS′, the
three curves now pass through point B and all markets are simultaneously in equilibrium. Thus point B shows the total effect of the monetary policy change.
We have just seen that, under flexible exchange rates, the IS curve moves to meet the intersection of the other two curves. This simply reflects the fact that,
being left free to float, the exchange rate is endogenous and its movements affect external competitiveness and demand.
This outcome may be puzzling. Here we have a central bank that adopts a more expansionary policy stance that succeeds in raising output, and yet cannot
change the interest rate because it is tied to the world rate of return. How can that be? The answer is that, with capital mobility, monetary policy does not operate
through the interest rate, but through the exchange rate. The longer answer is given by the money market with the help of Panel (a) in Figure 12.8. As in Panel ( b),
the central bank intends to move down the money demand curve, presumably all the way to point C, and to provide liquidity to the open market along the way.
But capital mobility implies that, as soon as the interest rate declines, money flows out. Any additional money created by the central bank immediately leaks
abroad in the form of capital outflows. Is the economy stuck at points A in both panels? No, because the capital outflows depreciate the exchange rate. The
depreciation, in turn, raises demand and output from Y to Y′. With a higher volume of transactions to finance, money demand shifts out in Panel (a) as the
economy moves from A to B in Panel (b).
The central bank ends up increasing real aggregate demand through an exchange rate only because it brought about a depreciation.13 Still, as previously
noted, it does not succeed in lowering the interest rate. As under a fixed exchange regime, we see that monetary policy and the exchange rate are deeply related.
In the end, an expansionary monetary policy in an open economy under exchange flexible rates operates by bringing about an exchange rate depreciation.
12.5.2 Demand Disturbances—Shifts of the IS curve

An exogenous increase in aggregate demand is shown as the rightward shift of the IS curve from IS to IS′ in Figure 12.9. As in the previous section, the economy
cannot be said to be in general equilibrium as the three curves IS′, TR, and IFM no longer pass through a common point of intersection. Consider first point B, at
the intersection of the IS and TR curves and off the IFM line. With the domestic interest rate exceeding the foreign rate of return, capital flows in and the
exchange rate appreciates. As external competitiveness declines, the current account worsens and the IS curve starts shifting to the left. This proceeds as long as
the interest rate exceeds the foreign rate of return, and until the IS curve is back at its initial position and the economy is back at point A. The demand
disturbance is simply ‘crowded out’ by the negative effect of the exchange rate appreciation on aggregate demand via net exports.
The result applies to any of the exogenous components of real demand we encountered in Chapter 11: expansionary fiscal policies (public spending, taxes),
business optimism or pessimism, or foreign demand driving exports. The general conclusion, under flexible exchange rates, is that an economy cannot
sustainably lift itself up via higher domestic or world demand. In the end, demand impulses are eventually neutralized by exchange rate changes. Output is
insulated from both domestic and foreign demand disturbances. This conclusion also applies analogously to negative demand disturbances, but with the
opposite conclusion. Under flexible exchange rates, an exogenous contraction of aggregate demand will give rise to a beneficial depreciation which ‘crowds in’
net exports, by making exports less expensive and imports more expensive.
Fig. 12.9 A Demand Disturbance Under Flexible Exchange Rates

An expansionary demand shock shifts the IS curve to the right. At point B, however, i > i* and capital flows in, the exchange rate appreciates, external competitiveness declines,
and the IS curve shifts back to its initial position. Demand disturbances are eliminated over time.
The position of the IS curve depends on the real exchange rate σ. With sticky domestic and foreign prices, the real exchange rate is driven by the nominal
exchange rate, which is determined in the foreign exchange market. The exchange rate, therefore, is endogenous, as is the IS curve. More precisely, the exchange
rate changes in such a way that the IS curve shifts to meet the two other curves, TR and IFM.
Under flexible exchange rates, the macroeconomic equilibrium is determined by the TR and IFM curves. In the same way that the TR curve can be ignored
when the exchange rate is fixed—because the money supply is endogenous—the IS curve can be more or less ignored when the exchange rate floats because the
exchange rate is endogenous. Still, the shift of the IS curve required to meet the two other curves tells us what happens to the exchange rate. It appreciates when
the IS curve must move to the left and it depreciates in the opposite case.
It may appear puzzling that fiscal policy—a type of demand disturbance—has no effect on output. Ultimately, the Mundell–Fleming model is a short-cut for
describing effects that may take longer to resolve. It takes time for demand disturbances to work through the economy. This also applies to the effects of
exchange rates on aggregate demand. Another assumption is that any change in the exchange rate affects external competitiveness one for one. If the former
operates faster than the latter, there can be some temporary effect of fiscal policy or demand shocks on output—possibly extending over a year or more. Thus,
while the Mundell–Fleming gives us the bottom line—that the effects of fiscal policy can be undermined by a regime of flexible exchange rates— we also have a
healthy warning that theory should never be taken literally.
12.5.3 International Financial Disturbances—Shifts of the IFM Line

The effect of an increase in returns on foreign assets measured in terms of the home currency (i*) was described in Figure 12.6 for the case of fixed exchange
rates. The same diagram can be used to understand what happens under flexible exchange rates. Graphically, we know that the IS curve will move to wherever it
needs to go through the intersection of the TR and IFM curves. As the IFM line shifts upwards to IFM′, equilibrium moves from point A to point C. The interest
rate must rise to meet the new foreign rate of return and the economy expands. While we can disregard the IS curve to find the new equilibrium, its movement tells
us something important: what happens to the exchange rate? In the present case, the IS curve will have to move to the right, which can only happen because the
exchange rate depreciates.
It may come as a surprise that an interest rate increase could lead to an output expansion. Again, the reason has to do with the exchange rate. Consider point A
immediately after the upward shift of the IFM line. The domestic interest rate is now below the foreign rate of return. This triggers a capital outflow. Since the
central bank does not intervene in the foreign exchange market, the outflow translates into a depreciation of the exchange rate. The country’s external
competitiveness and current account improve and the IS curve shifts to the right until it goes through point C (the new IS curve is not shown).
The increase in the foreign rate of return may be the consequence of a stricter monetary policy stance in the ‘rest of the world’. Think of the rest of the world
as a large economy that trades with us, a small open economy. As the foreign interest rate i* rises, foreign GDP declines. At home, in contrast, the GDP rises, as
we have just seen. A monetary expansion abroad has the opposite effect: a contraction of aggregate demand at home. This is known as the beggar-thy-
neighbour effect, because one country’s expansionary policy comes at the expense of its neighbours. In contrast, under fixed exchange rates, the reasoning
from Section 12.4 shows that foreign monetary policy has the same qualitative effect on foreign and domestic GDPs.
12.5.4 Is a Large Economy a Closed Economy?
This last discussion brings us to an open issue in macroeconomics: how large can an economy be before the Mundell–Fleming framework loses its validity? A
large economy has two main characteristics: (1) its trade represents a small share of GDP, as Table 12.1 shows, and (2) it affects, but is not much affected by,
foreign economies. Most large economies (the USA, the EU, Japan) also happen to be well-integrated financially. The key step taken in the Chapter 12 relative to
the Chapter 11 was to introduce international capital mobility, but if a large economy can affect interest rates abroad, the better framework for thinking about a
large economy is the one presented in Chapter 11. Trade openness does not change the qualitative results of Chapters 11 and 12, but rather affects the size of the
Keynesian multiplier. Financial integration, on the other hand, modifies the analysis because the small-country assumption takes the foreign rate of return i* as
exogenous. Since a large country affects but is not much affected by the foreign rate of return, we could neglect the IFM curve, leaving the framework of Chapter
11, the IS and TR curves.
Can a large, financially open economy really ignore the foreign rate of return i*? Think of the US as a typical large economy. When its central bank, the Federal
Reserve, changes its interest rate, asset holders all around the world reshuffle their portfolios and, in the event, all interest rates are affected, as Figure 12.10
shows. This would leave the impression that the US and its central bank actually fix i*, while other countries are financially too small to have a similar impact on
the US interest rate. At the same time, even the US—a $18 trillion economy in a world with a GDP of roughly $75 trillion—is becoming less central, while
foreigners are holding more and more dollar assets. Market expectations of future central bank policy—good or bad—are becoming more and more important in
determining i*—which implies that even the largest economies are subject to the constraints of international capital flows.
This chapter highlights the importance of the exchange rate regime for determining the reaction of a small open economy to disturbances. This is true irrespective
of whether these disturbances originate in macroeconomic policy decisions or in the behaviour of the private sector. It also shows that the exchange rate regime
is best understood through the prism of monetary policy. Chapter 11 showed that a central bank can control the interest rate or the money supply, but not both.
Now we find that life for the central bank is even more constrained: it can either have its own monetary policy (i.e. managing the interest rate or the money
supply), or fix the nominal exchange rate on the other hand, not both. If it chooses to peg the exchange rate, it gives up its ability to conduct an independent
monetary policy. It can recover policy independence only by letting the exchange rate float. These important results are summarized in Table 12.3. We now
examine what they mean for the choice of an exchange rate regime.
A fixed exchange rate regime represents a commitment to refrain from active use of monetary policy. The TR curve becomes irrelevant in that case because the
exchange rate anchor imposes a tight discipline on the monetary authorities. At the same time, a fixed exchange rate regime leaves the economy vulnerable
to demand disturbances, both domestic and foreign. Graphically, any exogenous shift of the IS curve determines a new equilibrium. Many countries in Europe in
the 1980s (e.g. France and Italy) and in Latin America in the 1990s (e.g. Argentina and Brazil) followed the fixed exchange rate strategy. It can work as long as the
declared parity is not challenged by overwhelming external disturbances, especially capital flow movements generated by changes in i* (shifts in the IFM line).
When this occurs, the exchange regime comes under threat and its credibility as a disciplining device for the authorities is called into question. This credibility
can even be threatened in a monetary union—a question addressed in more detail in Chapter 19.
Fig. 12.10 Long-Term Interest Rates, 1970–2015

The chart displays the evolution of long-term interest rates (bonds of 10-year maturity) in the USA and in the other large developed countries. The levels are often quite different,
for reasons explained later in Chapter 14. Yet month-to-month fluctuations reveal a high degree of coherence. The chart does not say that US interest rates drive the other rates,
only that all rates tend to move together. Statistical evidence does suggest, however, that US rates are the driving force behind the common movements in interest rates over
time.
Source: IMF, International Financial Statistics.
Table 12.3 The Mundell–Fleming Model: A Summary of the Importance of the Exchange Rate Regime
Exchange rate regime

Fixed exchange rates Flexible exchange rates
Impact of disturbances and policy on GDP:
IS curve disturbance (animal spirits, real exchange rate, wealth; fiscal policy) Increase No effect
TR curve disturbance (changing target interest rate) No effect Increase
IFM line disturbance (change in foreign rate of return) Decrease Increase
Impact on available policy instruments:
Exogenous monetary policy instrument Exchange rate Interest rate
Endogenous monetary policy instrument Interest rate Exchange rate
When the exchange rate floats freely, monetary policy independence is preserved. The TR curve describes how the central bank conducts that policy. The
economy as a whole is shielded from real demand disturbances. Graphically, the IS curve endogenously shifts to meet the TR and IFM curves. Yet the exchange
rate may fluctuate strongly in response to international financial disturbances, upsetting external competitiveness in one direction or another and confronting
firms with rapidly changing circumstances. An exchange rate appreciation, for instance, hurts exporters as much as a depreciation benefits them.
Countries that operate under a freely floating exchange rate regime usually possess sufficient economic and political stability to entrust the central bank with
the task of delivering price stability. This is the case of the USA, the euro area, the UK, Sweden, and several other countries. Yet it is common for countries to
change exchange rate regime in the light of new challenges. Box 12.4 tells the story of Switzerland, which threw in the towel of floating rates after a bruising
appreciation that followed the financial crisis, and hitched the Swiss franc to the euro—before switching yet again. Other countries have abandoned fixed
exchange rates, having been burned by speculative attacks against a previous fixed regime. This is the case in Argentina, Brazil, Chile, Russia, and many others.
Many countries let their exchange rate float because inflation is so high that any peg would quickly lead to overvaluation and crisis.14 We revisit these issues
again in Chapters 14 and 19.
Box 12.4 The Challenges of the Swiss Franc
Ever since the end of the Bretton Woods fixed exchange rate system (1971), Switzerland has put monetary policy autonomy at the top of its priorities
because it regards price and financial stability as vital for its global financial and banking sector. Letting the Swiss franc flow freely has been the logical
conclusion. This changed abruptly in September 2011 when its central bank decided to effectively peg the franc to the euro. Then, in January 2015, it
created another surprise when it abandoned the arrangement. Since then, while officially floating again, the franc has been under the tight supervision of
the central bank, a case of managed floating.
Why did Switzerland change course? As Figure 12.11(a) shows, following the global financial crisis of 2008 and the Eurozone crisis that started in 2010,
the Swiss franc had appreciated by almost 50%. An appreciation of such a great magnitude threatened exports while spurring massive imports. The
surge was driven by inflows of capital seeking a safe haven, a paradoxical consequence of Switzerland’s economic and financial stability. Viewed from
Switzerland, the IFM line was constantly shifting downward as foreign yields were declining. The situation is shown in Figure 12.12. Starting from point A,
under a flexible exchange rate regime, the economy moves to point B, where the exchange rate has appreciated, which is hurting external competitiveness
and results in a fall of GDP (the underlying IS curve, not shown, shifts to the left to pass through point B). For a while, the central bank tried to limit the
exchange rate surge by conducting massive exchange rate interventions, and its reserves quadrupled, as seen in Figure 12.11(b). The capital inflows
were too large for the interventions to be sterilized, as discussed in Box 12.2. Indeed, the figure shows that the money supply was increasing at a
breathtaking speed, threatening price stability.
The September 2011 announcement was credible and effective and the exchange rate stabilized at a level significantly below the previous peak. In
Figure 12.12, this can be seen as an attempt to shift to point C.15 Yet, enforcing the exchange rate limit requires the Swiss bank to keep purchasing
foreign exchange and, indeed, the foreign exchange reserves rose by another 50% in the months following the announcement. In early 2015, it transpired
that the European Central Bank would bring its interest rate into negative territory, a further downward shift of the IFM line, which renewed capital inflows.
Fearful of having again to absorb more reserves, the Swiss central bank gave up the peg. It simultaneously brought its interest rate to –0.75%, an
unprecedented step that reflected how far down the IFM line had moved. Figure 12.11(a) shows that the exchange rate shot up again and that the central
bank had to keep intervening in the foreign exchange market.
Fig. 12.11 Swiss Franc/Euro Exchange Rate and the Swiss Money Supply, 1999–2015
Switzerland has faced significant capital inflows as a result of the global financial crisis. Since 2009, the Swiss National Bank, the central bank, has acquired significant
foreign exchange reserves to slow down the appreciation of the Swiss franc (Panel (a)). As a result, total official reserves have risen spectacularly, with significant effect
on both high-powered money M0 as well as the broader narrow money measure M2, as shown in Panel (b). It is noteworthy that total official reserves of the central bank
exceed M0, due to foreign central bank holdings of Swiss francs as well as SNB repurchase agreements with those central banks.
Source: Swiss National Bank.
This back-and-forth on the exchange rate regime illustrates how dependent a small open economy is on events that take place outside its borders. It
also shows that no exchange rate regime provides complete protection from adverse disturbances. Moving to point B in Figure 12.12 implies accepting an
appreciation that hurts exports and raises imports. Moving to point C requires foreign exchange market interventions that may overwhelm the central bank,
leading to an unacceptable increase in the money supply.
Fig. 12.12 An Interpretation of the Swiss Case
The deterioration of financial conditions abroad is captured by a downward shift of the IFM line. Starting from point A, under an flexible exchange rate the economy moves to
point B, while a fixed exchange rate would take it to point C. In the latter case, however, the central bank must intervene on the foreign exchange market to enforce the peg.
Summary
1 The Mundell–Fleming model is the logical extension of the IS–TR model to an economy which is open for trade in financial assets as well as goods and services. It describes
the simultaneous equilibrium of three markets: (1) the market for goods and services, (2) the domestic money market, and (3) the international financial market.
2 Like the IS–TR model of Chapter 11, the Mundell–Fleming model adopts the Keynesian assumption that prices are sticky. Under these circumstances, output is determined
by aggregate demand.
3 When a country’s financial markets are well integrated into world markets, the domestic interest rate is tied to worldwide conditions. Under conditions of complete capital
mobility, a third equilibrium condition requires that the domestic interest rate be equal to the world rate of return. This is the interest rate parity condition. This condition
does not hold in the case of limited international capital mobility—e.g. because of capital controls. The IFM line summarizes the interest rate parity condition.
4 With all three curves, it is possible to study the general macroeconomic equilibrium, when all three markets are simultaneously in equilibrium.
5 When the exchange rate is fixed, demand disturbances affect domestic GDP. Committed to upholding the declared exchange rate parity, the central bank cannot however
conduct an autonomous monetary policy. The TR curve becomes irrelevant and the equilibrium is described by the intersection of the IS and IFM curves.
6 When the exchange rate is freely floating, the central bank regains monetary policy autonomy, but the exchange rate necessarily becomes endogenous. Its movements affects
the economy’s competitiveness and, therefore, the position of the IS curve. The exchange rate movements are such that the IS endogenously moves to meet the TR and IFM
curves. The fact that the IS curve does not determine the equilibrium outcome means that the economy is shielded from demand disturbances.
7 Monetary and exchange rate policies are just two sides of the same coin: the central bank can peg the exchange rate or it can conduct an autonomous monetary policy, but it
cannot do both at the same time.
8 The choice of an exchange rate regime involves trade-offs. Different countries choose different regimes, and often adapt their regimes depending on the circumstances.
Key Concepts
Mundell–Fleming model
exchange rate regime
capital controls
interest rate parity condition
foreign or international rate of return
international financial markets (IFM) line
general equilibrium
exchange market interventions
sterilized and unsterilized interventions
foreign exchange reserves
monetary policy autonomy
revaluation, devaluation
realignments
appreciation, depreciation
beggar-thy-neighbour
exchange rate anchor
Exercises
1 Consider a central bank in an economy that has floating exchange rates and sets interest rates according to the Taylor rule. Show graphically what happens when the central
bank decides to reduce the target interest rate from to . State clearly the steps involved.
2 Consider the same central bank in the previous question, but this time under a fixed exchange rate regime. What is the outcome? Trace the steps of adjustment carefully. What
do you conclude about the effect of monetary policy shocks in a fixed exchange rate regime?
3 Show the effect of a demand shock under a fixed exchange rate regime: the economy moves from point A to point B in Figure 12.4. Explain why point C is not sustainable
and why the economy must move to point B.
4 In 2011–2012, many southern European countries which use the euro undertook drastic measures to rein in their budget deficits, in particular tax increases and spending cuts.
It is often claimed that such austerity measures are counterproductive, especially in a fixed exchange rate regime. Explain why this is the case. How would your answer
change if these countries had their own currencies and flexible exchange rates?
5 Show the effects of an increase in the foreign rate of return under fixed and flexible exchange rates. What happens to the money supply in each case?
6 In a situation of economic or political instability, a standard reaction of citizens is to exogenously reduce their demand for money. What happens then to the interest rate,
income, and the money supply under a fixed exchange rate regime? When the central bank obeys a Taylor rule? Explain also what happens to the nominal exchange rate.
7 Suppose the UK left the European Union, and that this move meant that UK exports to the EU faced the common external tariff. As a result, UK exports drop sharply. (Its
imports would also decline, but assume for the sake of the exercise that the decline of exports is larger.) Explain the effect of this ex-ogenous decline in aggregate demand on
GDP, interest rates, and exchange rate in the UK, given that the central bank lets the pound float. How would your answer change if the UK had a fixed exchange rate with
the euro?
8 What is the effect of an exchange rate revaluation in a fixed (but adjustable) exchange rate regime?
9 What is the effect of tax cut under a fixed exchange rate regime? Under a floating exchange rate regime?
10 Figure 12.11 shows that in troubled times, safe-haven currencies like the Swiss franc can rise sharply. Concerned international investors seem willing to accept a lower rate
of return than in normal times. Evaluate the impact of becoming a safe-haven small open economy with (1) flexible exchange rates, and (2) fixed exchange rates. What can the
central bank do?
11 Consider a small open economy in a monetary union (i.e. a fixed exchange rate regime). Trace through the consequence of a price level decline at home, holding the foreign
price level constant. How would your answer change if foreign prices declined in the same proportion?
12 Answer the same question but under a flexible exchange rate regime (so there is no point in talking of devaluation). Answer the question when (1) the central bank keeps the
nominal money supply constant, and (2) when it follows a Taylor rule. In each case, what happens to the exchange rate?
13 Imagine that the Danish central bank prevents its exchange rate from appreciating by selling 100 million kroner on the exchange market and buying euros. Use the balance
sheets of the central bank and of commercial banks presented in Chapters 9 and 10 to describe carefully this foreign exchange market operation. Use the same balance sheets
to describe an operation which sterilizes the effect of the foreign exchange purchase.
Essay Questions
1 International financial markets mean that the domestic interest rate is determined abroad. Why does this mean that monetary policy is abandoned, but only in the absence of
capital controls? Why are capital controls nevertheless considered a bad idea?
2 As their name suggests, beggar-thy-neighbour policies have a bad reputation. Discuss why this may be the case and what can be done to limit the perceived drawbacks.
3 ‘Adopting an exchange rate anchor is a mixed blessing.’ Comment.
4 Even within a monetary union, it is still possible to have a real depreciation or appreciation, but it involves changes in the price level at home or abroad. Why is this difficult
to analyse using the IS–TR-IFM framework?
5 Emerging market countries typically have to offer a risk premium when they borrow. This means that their domestic interest rates are higher than those abroad. What is the
impact of a sudden loss of trust? Discuss why this could happen. What are the policy options in each case?
1 Ragnar Nurkse (1907–1959) was an Estonian economist who taught at Columbia and Princeton. He also worked at the League of Nations from 1934 to 1945 and was a
leading expert of his day on questions of international capital flows, fixed exchange rates, and economic development.
2 The Mundell–Fleming model is named after the Nobel Laureate Robert Mundell (1932–) from Columbia University and the Briton J. Marcus Fleming (1911–1976), who worked at
the IMF.
3 As of February 2017, the Eurozone consisted of Austria, Belgium, Cyprus, Estonia, Finland, France, Germany, Greece, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the
Netherlands, Slovenia, Slovakia, Spain, and Portugal.
4 Remember the goods market equilibrium condition states that output Y is equal to C + I + G + X − Z.
5 Remember that ‘liabilities’ are used in the broad sense to include all inward foreign investment, including not only debt obligations and government borrowings, but also portfolio and
foreign direct investments.
6 In order to do this type of financial deal, one needs to have collateral, i.e. assets which can be pledged in case the deal fails. Large international investors have no difficulties doing
this.
7 In this case, black markets usually flourish.
8 The money demand curve was described formally in Chapter 9 as M/P = k(i)Y, with k′ < 0. Holding income Y constant, an increase in interest rates reduces the demand for real
balances; holding the interest rate i constant, an increase in income Y increases the demand for real balances.
9 An exercise at the end of the chapter asks you to verify that the main conclusion stands when we take into account the fact that a lower interest rate should raise output. However,
since monetary and financial reactions are much faster than goods market changes, the simplified presentation is a good approximation to what happens in practice.
10 When the exchange rate floats, exchange rate changes are called appreciation or depreciation, see Section 12.5.
11 In (12.1) we must have DC < 0 when the value of foreign exchange reserves exceeds the money base (R > M0).
12 Interest rates were lower in China than in the USA, so the central bank could obtain a higher return on dollar-denominated reserves than it was paying on its yuan-denominated debt
to the commercial banks.
13 In practice, given the speed of reaction of financial markets, the economy does not really go through point C.
14 Recall that the real exchange rate is defined as SP/P*. With S fixed, if P rises faster than P*, the real exchange rate appreciates. These situations will be addressed in Chapters 13 and
16.
15 A more precise depiction of the situation would show the IFM line shifting further down so that point B is the new departure point toward point C further down on the new IS curve
that passes through point B.
Output, Employment,
and Inflation 13
13.1 Overview
13.2 General Equilibrium with Flexible Prices: The Neoclassical Case
13.2.1 From the Keynesian Short Run to the Neoclassical Long Run
13.2.2 Supply-Determined Output in the Long Run
13.2.3 Implications for the Long Run
13.3.1 A. W. Phillips’ Discovery
13.3.2 Okun’s Law and a Supply Curve Interpretation
13.3.3 Hard Questions about the Phillips Curve
13.4 Accounting for Inflation: The Battle of the Mark-ups
13.4.1 Prices and Costs
13.4.2 The Battle of the Mark-ups: A Simple Story
13.4.3 Productivity and the Labour Share
13.4.4 Cyclical Behaviour of Mark-ups
13.4.5 More on the Underlying Rate of Inflation
13.4.6 Completing the Picture: Supply Shocks
13.5.1 The Phillips Curve Rehabilitated
13.5.2 Underlying Inflation and the Long Run
13.5.3 Aggregate Supply
13.5.4 Factors that Shift the Phillips and Aggregate Supply Curves
13.5.5 From the Short to the Long Run
Summary
When the demand for a commodity or service is high relative to the supply of it we expect the price to rise, the rate of rise being greater, the greater the excess demand. Conversely when the demand is low relatively to the supply we expect the price to fall, the rate of fall being greater, the
greater the deficiency of demand. It seems plausible that this principle should operate as one of the factors determining the rate of change of money wage rates, which are the price of labour services.
A. W. Phillips7
13.1 Overview
A cup of tea in London that cost 5p in 1965 goes for £1.80 or more in 2016. The baguette in Paris which cost 40 centimes (one centime = one hundredth of the old French franc) in 1965 fetches 13 times more in terms of euros 50 years later. Prices seem to grow
relentlessly. Yet inflation is not just about changes in the price of tea in Britain and bread in Paris. Inflation measures the rate of increase of the price level, i.e. the number of units of pounds or euros paid for a bundle of goods. It is easy to see that the inflation rate
simultaneously measures the rate at which money loses its value in terms of those goods. The phenomenon does not stop there. When the price level rises, the level of nominal wages tends to rise too. Wages rise partly due to rising productivity, but they also
chase prices, which in turn chase wages. And the nominal exchange rate seems to be engaged in the same kind of race. Somehow, all nominal variables seem to leapfrog each other.
Whether very low or excruciatingly high, inflation is a key feature of modern economies. It was not always so. From time immemorial to the middle of the twentieth century, prices were pretty much trendless. Continuing inflation is a relatively new event in the
history of humankind, and coincides with the emergence of modern central banks and fiat money. This should not come as a surprise once we remember the monetary neutrality principle. Ultimately, central banks and their ability to create money are the source of
inflation.
Until now we have studiously avoided talking about inflation. We discussed it in Chapter 5 when we established the principle of monetary neutrality, but then we relegated it to the faraway long run. In Chapters 11 and 12, we explicitly ruled it out by adopting
the Keynesian assumption that prices are sticky. These were useful steps that allow us now to focus on this central feature of modern economies. In this chapter, we focus on the ‘medium run’. Think of the medium run as a halfway-house between a short run that
corresponds to the Keynesian view and a long run described by the neoclassical perspective presented in Chapter 5. After sufficient time has elapsed—in the long run—all prices are able to adjust and the economy is governed by monetary neutrality. In the
medium run, prices move but not sufficiently to allow all markets to clear. We start by clarifying these issues.
Then we look at the Phillips curve, a very simple description of the inflation phenomenon based on data and observation. We will see that the Phillips curve is useful for helping us think about the short-run behaviour of inflation. At the same time, the curve
appears unstable and seems to vanish at times. So we embark on a detective-like investigation of the puzzle of the ‘disappearing Phillips curve’. This leads us to track down the various reasons why prices rise. We start by asking who sets prices (firms do) and
why (mostly to cover their production costs). As we study production costs, we return to the wage bargaining process first encountered in Chapter 4, and find that wage negotiators worry about prices of goods, for different reasons. We end up facing the
apparently circular conclusion that prices drive wages and wages drive prices, in a sort of race between employers—who want high profits—and employees—who want high wages. The outcome of the analysis is an accounting of the factors that add up to a full
explanation of inflation. This analysis helps us solve the Phillips curve puzzle. It also allows us to derive the aggregate supply curve, which will be teamed up with the aggregate demand curve developed in the next chapter to complete the system of inflation
and output determination.
13.2 General Equilibrium with Flexible Prices:

The Neoclassical Case
13.2.1 From the Keynesian Short Run to the Neoclassical Long Run
In Chapters 11 and 12, we made extensive use of the Keynesian assumption that prices are sticky. We justified this assumption by asserting that prices move slowly in normal times. In the short run, we said, ignoring inflation is an easy way to make things simple.
When thinking about the long run, however, we need to make the diametrically opposite assumption that prices are fully flexible. This is the neoclassical assumption against which Keynes rebelled in the 1930s. He well knew the limits of his own assumption,
but then famously wrote that ‘in the long run, we are all dead’. In a parallel fashion, this section relaxes the Keynesian assumption and presents the view espoused by neoclassical economists (‘neoclassicals’ for short) writing long before and after Keynes. While
some Keynesians and neoclassicals still fight it out, most economists now agree that the Keynesian assumption is an acceptable way to think about the short run, and the neoclassical assumption is the right way to think about the long run.1 This section,
therefore, studies the long run, but we also look at how things change over time, from the short to the long run.
Our taxonomy of exogenous versus endogenous variables can help understand the difference between the two perspectives. In the Keynesian view, the price level is exogenous; output is endogenous and responds one-for-one to demand. In the neoclassical
view, it is the price level that is endogenous and plays the equilibrating role in the goods market. In principle, we have already examined the neoclassical case in Chapter 5 when we established the monetary neutrality principle. From that chapter, recall the
Cambridge equation:
(13.1)
where, for the moment, we treat the parameter k as constant.3 We asked what happens when the money supply increases, say by 10%. In the short run, when prices are sticky, the only way for the money market equilibrium condition (as explained in Chapter 9) to
hold is for real GDP to increase by 10%. This is the Keynesian case. If prices are flexible, however, equilibrium can be achieved with a 10% increase in the price level, without any change in real GDP. This is the neoclassical case and its underlying result is
monetary neutrality.
If the Keynesian assumption is valid in the short run and the neoclassical assumption describes the long run, then it must be the case that an increase in the money supply—which is engineered when central banks cut the interest rate—is followed first by a rise
in output that will be eventually dissipated through an increase in the price level. This dynamic adjustment path for output and prices is shown in Figure 13.1.
Fig. 13.1 From the Short to the Long Run
A one-off increase in the money supply has short and long effects which are qualitatively very different. The short run corresponds to the Keynesian assumption: prices are sticky and output moves to respond to demand. The neoclassical assumption describes the long run, when prices
fully adjust and monetary neutrality implies that output is unaffected.
This can be expressed in terms of rates of change. The inflation rate is the rate of change in the price level, defined as a rate: ΔP/P. Equation (13.1) implies that the price level is given by:
(13.1ˊ)
So for a constant k, the inflation rate, the growth of the money supply ΔM/M, and the growth rate of output ΔY/Y, the following relationship must hold:
(13.1˝)
where µ is the rate of growth of money and g is the rate of growth of real GDP. This is the ‘rate of change’ version of the classical quantity equation. Later, it will help us to think about the long run.
Why would the money supply increase without bound? This question does not have an obvious answer, because central banks have long abandoned the practice of setting monetary targets and in fact prefer to use their power in money markets to set short-
term interest rates. It turns out that this power does not come for free. Setting the interest rate necessarily means that a central bank must be ready to supply all the liquidity the economy demands at that interest rate. As a corollary, it must accept the rate of growth
in the money supply which results from that interest rate. How the money supply evolves endogenously is explained in more detail in Box 13.1.
13.2.2 Supply-Determined Output in the Long Run

Why does price flexibility lead to a situation where demand adjusts to supply? Let us start with the supply side, that is, the long-run output and labour market outcomes studied in Chapters 3 and 4. Figure 13.2(a) displays the production function. Panel (b) shows
the demand for labour implied by the marginal productivity of labour—as determined by the position and the slope of the production function, and the supply of labour—as shaped by labour market institutions. The resulting equilibrium at point A in Panel (b)
indicates how much labour is utilized. The corresponding point A in Panel (a) shows how much is produced and supplied. In the classical long run, output equals its trend value (which is generally growing over time), employment equals , and the
equilibrium rate of unemployment is : These points are labelled as A in both panels of the Figure 13.2. We can also see in Panel (b) the equilibrium real wage, the ratio of nominal wages W to the price level P.
It is useful to examine an outcome under the Keynesian assumption, that is when wages and prices are sticky. We studied this situation at length in Chapters 11 and 12. In particular, we look at a situation when demand weakens to Firms will respond by
cutting production and move to point B in Panel (a) in Figure 13.2. Panel (b), however, shows that there is no market equilibrium corresponding to this reduction of production. If wages and prices are sticky, the outcome will be at point B in Panel (b).7 What is
remarkable about point B is that it is neither on the labour demand nor on the labour supply schedules. The distance BA measures involuntary unemployment, the extent of disequilibrium seen by employees. Firms are unhappy too, since being off the labour
demand curve means that profits are not maximized. The paradox is that the real wage is at its ‘correct’ level, the one corresponding to point A. The following analysis would also hold if the real wage would change to take us to B′ (higher than the equilibrium
value) or B″ (lower than its equilibrium value).
Box 13.1 Where Does the Money Growth Come From?
Chapter 5 showed that the long-run price level and rate of inflation were related to the money supply and its growth rate. Yet in Chapter 9, we saw that central banks do not set growth rates of monetary aggregates. Rather, monetary policy means targeting
interbank money market interest rates. In practice, central bank interest rate policy is well-represented by the Taylor rule, which automatically implies a long-run rate of money growth. Because it supplies the money demanded given its chosen interest
rate, the rate of money growth follows the target rate of inflation. In other words: ‘Every central bank gets the money growth and inflation rate that it chooses.’
It is easy to see this by re-examining the Taylor rule, which shows how central banks set interest rates:
(B13.1)
In the long run, inflation is equal to target inflation and so it must be the case that The fact that the nominal interest rate is stable means that the condition for equilibrium in the money market
is:
(B13.1)
In the long run, real output growth g = ΔY/Y is determined by real factors (the dichotomy assumption) and does not depend on money growth. Since the interest rate is constant, k is constant. Equation (B13.2) is simply another version of (13.1′); the logic of
(13.1′′) implies that the rate of nominal money growth is equal to the sum of the long run real growth rate and the central bank’s inflation rate target:
(B13.3)
While money growth is neutral, it is not causal for inflation. Instead, the money supply grows endogenously to reflect the underlying trend growth in the economy (g) as well as the inflation target ( ) of the monetary authority. The difference between
(13.1”) and (B13.3) is deeper than simply moving variables on different sides of the equation. It establishes the long-run difference between money targeting, whereby the central sets µ, which determines π, and interest targeting, where the central bank
sets π, which determines µ. In both cases, dichotomy implies that, in the long run, g is exogenous with respect to monetary policy.
In the long run, the neoclassical principle implies that nominal wages and prices will adjust—rise or fall—until equilibrium is restored in both product and labour markets. This means returning to point A. How will they move? The underlying guiding principle is
that prices rise when demand for goods is strong, and fall when it is weak. Similarly, wages rise when demand for labour is strong, and decline when demand for labour is weak. In Figure 13.2, firms facing weak demand would reduce their prices. If the nominal
wages (W) remain sticky, real wages (W/P) rise—a falling price level (P) means a higher purchasing power of wages. As we move from B toward B′, involuntary unemployment gets even larger. Over time, as nominal wages become flexible, they would decline.
Fig. 13.2 Output and the Labour Market in the Long Run
Long-run output Y is produced by firms using long-run labour input –L (point A in Panel (a)) and the equilibrium of labour demand and collective labour supply (point A in Panel (b)). Point B depicts a Keynesian situation in which aggregate demand is insufficient and less than trend level of
output, even though the real wage is at its long-run level. (Points B′ and B″ show initial situations in which the real wage is too high or too low, respectively.) From point B, it would be sufficient simply to raise aggregate demand; in the other cases, the real wage must fall or rise via some
appropriate combination of nominal wage and price changes. In the neoclassical case, price reductions and wage cuts will be necessary to restore equilibrium when demand is low. If exchange rates are flexible, a depreciation can help restore demand to long-run equilibrium at point A.
Are we just going endlessly up and down from B to B′ and back? The answer is no, because the decline in goods prices will stimulate demand for goods and this will bring the economy back to point A in both panels. How exactly will this come about? Let us go
back to the IS–TR analysis of the previous chapter. If the exchange rate is flexible, the Taylor rule implies that a declining price level will stimulate lower interest rates, which in turn brings about a nominal and ultimately real depreciation, a process that will continue
until demand is restored to its initial value. If the exchange rate is fixed, the decline of the domestic price level also implies a real exchange rate depreciation, which lifts foreign demand and reduces the demand for imports. Either way, price and wage flexibility will
bring the economy back to point A, with lower price and wage levels but at the initial real wage level.
13.2.3 Implications for the Long Run

If the neoclassical assumption describes the long run, it follows that demand disturbances only have a temporary impact on output and employment. Thus, we start from the Keynesian short run where supply passively adjusts to demand and gradually moves to
the long run as prices and wages respond to prevailing conditions. Eventually, the economy is back to full equilibrium or, equivalently, GDP returns to its trend level and equilibrium unemployment (as defined in Chapter 4) prevails again. The supply side
represents what the economy’s resources (physical capital, human capital, and the state of technology) can produce and how much income can be earned. This notion that demand disturbances predominate in the short run, only to be eliminated in the long run,
nicely fits the idea of business cycles represented in Figure 11.1, and implies that price adjustments gradually substitute for quantity adjustments (see Figure 13.1). The task of the following sections is to examine carefully how prices and wages move in response
to temporary disequilibria in the goods and/or labour markets.
13.3.1 A. W. Phillips’ Discovery

The short-run Keynesian assumption and the long-run neoclassical assumptions are fairly easy to deal with: prices either do not change at all, or they fully adjust. The medium run, as represented in Figure 13.1, is considerably messier. Prices do move, but only
part of the way in any given period of time. The Keynesian assumption always had the major disadvantage of leaving inflation unexplained by assuming it away. Even die-hard Keynesians conceded at the high point of their influence that they had no clue how to
incorporate inflation in their model. They referred to it as the ‘missing equation’. The search for this equation turned up the Phillips curve, a negative relationship observed to hold between inflation and unemployment, the twin ‘bads’ of macroeconomics. It is
represented in Figure 13.3.
Fig. 13.3 The Phillips Curve in Theory
The Phillips curve implies a negative relationship, or tradeoff, between unemployment and inflation: you can get less of one bad (unemployment) only by accepting more of the other (inflation). When discovered, the Phillips curve was seen as representing a menu of options from which
governments could choose. For example, it could keep unemployment down (point A) at the cost of some inflation, or could limit inflation (point B) but only by accepting higher unemployment.
In the late 1950s, A. W. Phillips plotted the annual rate of growth of nominal wages, i.e. wage inflation, against the rate of unemployment in Britain during the period 1861–1957. He found a remarkably robust negative correlation, which was confirmed for a
number of other countries. 5 Figure 13.4 plots real Phillips curves—using the rate of price inflation instead of wage inflation—for the UK and the average of 16 advanced economies for the period 1921–1973 (excluding war years). While far less clean than the
stylized version (a number of outliers correspond to exceptional events), empirical Phillips curves—those based on real data—are consistent with the negative relationship shown in stylized form in Figure 13.3.
The Phillips curve found widespread acceptance at the time for two main reasons. First, it provided the missing theoretical relationship that complements and indeed completes the Keynesian analytical framework. Second, even as its durability was increasingly
questioned, it remained a practical tool for policy-makers. They can aim at low unemployment but they must accept substantial inflation (point A in Figure 13.3), or they may prefer low inflation but at the cost of high unemployment (point B), or any intermediate
situation. Once they decide, they can use their instruments (monetary and/or fiscal policy) to reach the chosen point on the Phillips curve.6 This policy choice came to be known as the Phillips trade-off.
Fig. 13.4 The Phillips Curve in Reality

In reality the Phillips curve was less stable than depicted in Figure 13.3. The two panels depict the evolution of inflation and unemployment in the data for the United Kingdom in the period 1888–1975 and for an average of 16 countries 1921–1973 (excluding 1939–1949). While movements in
northwest to southeast directions do dominate the year-to-year movements, there are notable deviations.
Sources: Maddison (1991); Mitchell (1998). Unweighted average of observations for Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Italy, Japan, the Netherlands, Norway, Sweden, Switzerland, the UK, and the USA. For some years, some countries are missing.
13.3.2 Okun’s Law and a Supply Curve Interpretation

The Phillips curve relates inflation to unemployment. This would not be enough, however to establish a positive, aggregate relationship between inflation and output. Another stylized fact, known as Okun’s Law, performs this function. Okun’s law is a negative
relationship between output and unemployment.7 More precisely, Okun ’s Law links the unemployment gap, the distance between actual and equilibrium unemployment as defined in Chapter 4, to the output gap, the distance between real GDP and its long-
run trend, as illustrated in Figure 11.1. Figure 13.5 plots estimates of the unemployment and output gaps for Germany as an example of how unemployment and output gaps systematically move in opposite directions. The logic is intuitive: when the economy grows
fast and output is above trend, demand for labour is strong, and unemployment falls below its equilibrium level.
A stylized representation of Okun’s Law is shown in Figure 13.6. Formally, Okun’s Law can be represented as follows:
(13.2)
where and
The unemployment gap Ugap is defined as the difference between the current unemployment rate U and its equilibrium level . Similarly, the output gap is the per cent deviation of real GDP Y from its potential level .8 Okun’s Law means that when output
is above trend, the output gap is positive and the unemployment gap is negative; when output is below trend, as in Figure 13.2, the output gap is negative and the unemployment rate is above trend. The parameter h captures the response of unemployment to GDP
fluctuations, measured as the output gap.
Fig. 13.5 The Output Gap and Unemployment in Germany, 1970–2016

The output gap measures the distance between real GDP and its trend and is computed in per cent of trend GDP. The unemployment gap is the difference between actual and equilibrium GDP, both measured in per cent of the labour force. When business conditions vary, firms adapt the
supply of goods and services. To that effect, they adjust their demand for labour. For example, when the economy goes into a recession, firms employ fewer workers and the unemployment rate rises. This explains the systematic opposite moves—the negative correlation—of the output
and unemployment gaps. Note that, since 2012, the two gaps appear to move together, but this may reflect Germany’s Hartz reforms, which increased the effective supply of labour after the mid-2000s.
Notes: Trend real GDP and equilibrium unemployment are estimated by the European Commission.
Sources: European Commission, AMECO on line.
Fig. 13.6 Okun’s Law in Theory

Okun’s Law implies that, when the economy slows down, unemployment increases; when output rises relative to trend, unemployment declines.
Fig. 13.7 Okun’s Law in Reality
Okun’s Law is not identical across countries. The two panels show how Okun’s Law accounts for changes over time in the unemployment rate (in per cent) against percentage output changes, for quarterly data in the USA and Germany for the period 1971–2010. (This is slightly different
from the depiction in Figure 13.6.) In the USA, a 1% increase in output is associated with roughly a one-third percentage point decline in the unemployment rate. The response of output to an unemployment change in Germany is about a fifth of that in the USA. Okun’s Law cannot explain all
the variation in unemployment but it does explain a fair amount. It accounts for much more overall variation of unemployment in the USA than it does in Germany, where the largest changes over the decades have been associated with structural change.
Sources: OECD.
Figure 13.7 plots quarterly changes in the unemployment rate against quarterly changes in output for the USA and Germany. The negative relationship is very clear in both countries, but the slope of the Okun relationship is not equal. In Germany, an increase in
growth of 1% yields a much lower reduction in unemployment than in the USA. Among other things, this reflects labour laws that make it more difficult to shed workers in Germany. Of course, Okun’s Law does not explain all the fluctuations in unemployment,
which in reality are affected by many other factors.
When we combine the negative relationship between inflation and the unemployment rate from the Phillips curve with the negative relationship between the unemployment rate and the output gap from Okun ’s Law, we obtain the positive relation ship between
the output gap and inflation shown in Figure 13.8. This relationship is called the aggregate supply curve. Section 13.4 provides a complete interpretation of the supply curve, but the logic behind it is straightforward. The supply curve answers the following
question: under which conditions will an increase in aggregate demand lead firms to supply more output, and employees to work more to produce that extra output? The short answer is: inflation must increase, to boost wages and profits.
The positive slope of the aggregate supply curve may remind us of supply curves we have seen in microeconomics. While this similarity may be intuitive, it is misleading. In microeconomics, the horizontal axis corresponds to the production of a single good,
and the vertical represents that good ’s particular price. Here we relate aggregate output (real GDP) on the horizontal axis and the overall price index—or here its rate of increase, which is the inflation rate—on the vertical axis. This distinction highlights the
difference between micro- and macroeconomics.
Fig. 13.8 The Aggregate Supply Curve

Combining the Phillips curve and Okun’s Law, we obtain the aggregate supply curve. The curve says that inflation rises when output increases.
13.3.3 Hard Questions about the Phillips Curve

Weak foundations
Not long after its discovery, the interpretation of the Phillips curve was perceived as standing on weak theoretical legs. In the late 1960s, Milton Friedman and Edmund Phelps independently attacked the Phillips curve implication of a permanent trade-off between
inflation and unemployment or output.9 They both asked the following question: how could the rate of change of nominal variables, such as wages or prices in pounds or euros, be related to real variables, such as employment, unemployment, and output in the
long run? If the monetary neutrality principle is valid in the long run, then rates of change in the price level and other nominal variables should be unrelated to the real economy. 10 In this case, they noted, the Phillips curve can only be a temporary phenomenon.
Only if workers and firms suffer from money illusion—i.e. if they act on increases in their own prices or wages whilst ignoring nominal price changes all around them—will they permanently raise their labour supply and real output.
To understand the critique of Friedman and Phelps, it is helpful to think about the long-run behaviour of labour markets and output as presented in Section 13.2. The principle of monetary neutrality asserts that the economy is dichotomized in the long run: the
real and nominal sectors of an economy stop influencing each other. If the long run is defined as when output level is on its trend growth path and unemployment is at its equilibrium rate , then the rate of inflation is uniquely determined by monetary policy,
not by the level of output or the unemployment rate. These arguments were presented in detail in Box 13.1. Graphically, if unemployment returns to its equilibrium level in the long run, the long-run aggregate supply curve and the Phillips curve must be
vertical, as displayed in Figure 13.9. This conclusion is in agreement with the neoclassical view, developed in Section 13.2, which implies that output is determined by the supply side in the long run.
Wobbly evidence
The critique of Friedman and Phelps was largely ignored in the late 1960s, but it proved to be right on target in the 1970s when, as Figure 13.10 shows, the Phillips curve simply vanished. Quite spectacularly, in the mid-1970s and early 1980s, both inflation and
unemployment started to rise, a phenomenon which came to be known as stagflation. Stagflation was strictly incompatible with the Phillips curve and its trade-off. A number of consequences followed. Policy-makers, now grappling to beat back inflation, became
sceptical of ‘Keynesian activism’. They started to put greater emphasis on long-run monetary neutrality as a guiding principle for monetary policy; not surprisingly, money targeting became a popular policy approach at central banks. Macroeconomics also
underwent a profound transformation, which can be thought of as a massive effort to rehabilitate the Phillips curve.
Fig. 13.9 The Long Run
In the long run, unemployment is at its equilibrium rate and output is on its trend growth path. Both the Phillips curve and the aggregate supply curve are vertical. Inflation is determined by the rate of money growth or by the central bank’s target rate of inflation, independently of output or
unemployment.
Over nearly a century, the inverse relationship between inflation and unemployment had seemed relatively robust. Why did it suddenly break down in all countries, and at about the same time? The systematic breakdown of the Phillips curve is now much better
understood than four decades ago, and is back in fashion, although in a fundamentally reconstructed form.
The challenge is to explain both the existence of a Phillips curve and its disappearance, as well as the striking similarity between different countries’ experiences. Nearly everywhere inflation and unemployment increased sharply, at first around 1973–1974, and
then around 1979–1980, precisely at the time of sharp increases in the price of petroleum, the so-called oil shocks. Oil prices increased threefold in 1973–1974 and then doubled again in 1979–1980. Interestingly, in between the oil shocks, and after the second oil
shock, Phillips curves re-emerged, each time further above and to the right of the previous one. Two examples of these episodes are presented in Figure 13.10 and further dissected in the following section.
13.4 Accounting for Inflation: The Battle

of the Mark-ups
The disappearance and re-emergence of the Phillips curve at particular points in time is important information that will help us understand the determinants of inflation over the decades. To account for the fate of the Phillips curve, we will follow a pragmatic,
‘divide and conquer’ strategy. Breaking inflation into its most important components will help us better account for its immediate sources and we will be better equipped to think about it. We will start by thinking about the reasons firms change their prices.
Fig. 13.10 Phillips Curves: Recent Experiences in the Euro Area and the UK
The Phillips curve broke down at the end of the 1960s. In the euro area, abrupt movements in inflation in the early and late 1970s coincided with the two oil shocks. Afterwards, the reduction of inflation in the 1980s is compatible with the traditional Phillips curve, as are the periods 1990–
1994 and 1997–2007. In the UK, the old Phillips curve survives over 1960–1967 until the upward jumps in 1973–1975 and 1979–1981 that coincided with sharp increases in oil prices—the so-called ‘oil shocks’. As in the euro area, three different Phillips curves can be identified: during the
deflation period of the 1980s, over 1986–1993, and in 1994–2007.
Sources: OECD, Economic Outlook.
13.4.1 Prices and Costs

Price setting
The aggregate price level is the average of hundreds of thousands of individual prices, but who sets those prices of goods and services, and how?11 Looking at price tags in stores and price lists in catalogues, you may correctly conclude that producers and
retailers decide on the prices of their goods and services. Obviously, each producer prefers higher to lower prices, but their enthusiasm is held in check by the need to attract enough buyers. This is the process that we will now study. Box 13.2 sets out a few
principles. It explains how competition limits the ability of firms to set prices, and that firms try hard to reduce these limitations. It describes firms as setting prices as a mark-up above nominal production costs, aiming at a margin consistent with maximizing profits.
Fig. 13.11 The Price of Oil, 1950–2015
The price of oil is generally quoted in US dollars; in this figure, it has been converted into euros. The first two oil shocks (1973– 1974 and 1979–1981) and the two next ones (1999–2001 and 2005–2012) are clearly visible. In 1986 and 2015, a favourable oil shock occurred—a sharp
decline of oil prices.
Sources: European Commission, AMECO on line.
Costs
Now that we understand the price mark-up—the relationship of price to cost—it will make sense to look at those production costs, called average or unit costs, measured in pounds or euros.12 Unit cost (UC) is simply the total cost of production divided by the
number of units produced. It is convenient to break down unit costs into two main categories: labour and non-labour costs:
Unit costs in euros
= total costs in euros/number of units produced
= unit labour costs + unit non-labour costs.
For the economy as a whole, labour costs are the single largest component of unit production costs. Table 13.1 shows that at the macro level, the share of labour costs (wages, benefits, plus contributions to social insurance) ranges from 50% to 70% of value
added in developed countries, and is usually higher in labour-intensive services than in capital-intensive industries.13 For this reason, we will focus on labour costs for the next few sections. We will consider non-labour costs later, in Section 13.4.6.
Box 13.2 When and How Firms Change Prices
Economists speak of perfect competition in goods markets when firms are unable to set prices. A good example is the fruit grower who sells his apples in the town market. With many other sellers around, the farmer cannot set his price very far from
some average price for apples of the same variety. If he raises his price just a few pence or cents, he loses all his customers; if he lowers his price, he will sell everything, but regret the forgone profits. Producers of standard products, such as agricultural
products or raw materials, have relatively little choice. We say that they have no market power.
Yet most firms do have market power, and some firms have a lot of it. In fact, they go to extreme lengths to establish market power because this helps them achieve higher profits. This is called product differentiation, making their product special and
different from those of competitors. They do so through design (similar cars always differ in many subtle ways), through advertising and marketing to win consumer loyalty (some people like Volkswagen, others Renault), or product quality and reputation.
The payoff is a higher willingness to pay by their customers, and a greater degree of market power. Firms can charge higher prices above the bare minimum that they need to survive, and still keep a large share of their customers.
Mark-up pricing is a simple description of how firms with market power set prices. Obviously, when setting a price, a firm wants to cover its production costs, and then it tries to do more. The margin over the cost of producing one more good, the
marginal cost, is the mark-up. The mark-up depends on the sensitivity of the particular product’s demand to price—its demand elasticity. If the market is highly competitive, the demand elasticity is near-infinite, because customers have lots of
alternatives: any price increase will drive customers to other sellers. In that case the mark-up is zero and the price is equal to marginal cost. This is the case of perfect competition. If competition is weak, because the firm has been able to differentiate its
good and build a ‘niche’, or because other firms cannot enter the market, the demand elasticity is low and the mark-up will be higher.
Table 13.1 Wage Share of Value Added by Country and Selected Industries, 2014
Total Economy Manufacturing Construction Finance & Insurance Basic Metal Industries Wholesale & Retail Trade
Belgium 57.1 63.5 50.5 44.4 80.2 56.3
Czech Rep. 44.4 45.9 40.6 33.3 53.1 51.5
Denmark 60.5 57.1 76.4 47.5 84.1 71.0
Germany 56.5 62.9 61.4 64.5 69.2 69.0
Italy 44.4 60.5 41.6 40.0 65.9 42.3
Japan 52.1 54.2 71.0 49.9 46.8 53.6
Netherlands 55.1 54.9 61.8 43.4 59.2 52.9
Poland 42.2 45.7 32.0 36.7 49.7 28.8
Spain 51.8 57.1 50.1 51.0 70.5 56.5
USA 55.2 46.9 63.5 57.2 62.1 62.0
Source: OECD STAN Database. Basic metal industries for Germany and Spain are for 2013.
Labour costs and wages

A firm’s total labour costs are simply the product of the number of hours worked in the firm (L) and the average gross hourly labour cost (W). Gross hourly labour costs include not only wages and salaries per hour, but also paid vacations, direct labour taxes,
social security contributions, and other benefits paid by employers on behalf of their workers. (In many European countries, these additional costs can be nearly equal to the net pay received by the worker.) A country’s aggregate unit labour cost is computed in
the same way. It is simply the ratio of total labour costs WL, also called the wage bill, to total real output, i.e. real GDP Y:
(13.3)
Nominal unit labour costs must be distinguished from the real labour cost of producing a unit of real GDP or, equivalently, the ratio of the nominal wage bill to nominal GDP:
(13.4)
The real unit labour costs measure the share of GDP (PY) that goes to labour (WL).
13.4.2 The Battle of the Mark-ups: A Simple Story

Prices as mark-up on labour costs
Mark-up pricing means that firms set the price of goods as much as they can above their nominal unit costs. We have agreed to ignore non-labour costs, so we now consider that firms will aim for a price as much above their nominal unit labour costs (WL/Y) as it
takes to maximize its profits. This can be written as follows:
(13.5)
where θ ≥ 0 is the price mark-up. For instance, if the mark-up is θ = 0.3 = 30/100, the price is set 30% above nominal unit labour costs.14
Wages as a mark-up on prices
Firms mark prices above labour costs, but what determines labour costs and nominal wages? A good starting point is Chapter 4, which notes that wages are set through negotiations between employees and employers. These negotiations deal with two different
issues.
First, the employees want to secure as large an income as possible. Yet both wages and salaries as well as employment must originate in value added (GDP) of ongoing firms. Thus, while labour is likely to aim at as large a labour share of output as possible, as
we saw in Chapter 4, its efforts are limited by firms’ demand for labour.
Second, wage negotiators can only bargain over nominal wages. They do not know for sure what tomorrow’s price level will be. Typically, wage agreements cover a period of one or more years, and the future evolution of the price level is unknown. Naturally,
employees want to protect their nominal wages from inflation. Employers normally agree to incorporate the likely evolution of the price level in wage settlements but worry about overestimating it and paying their workers ‘too much’—paying workers too much
means making low profits or even losses. This is why the expected evolution of the price level, which is incorporated into the wage agreement, is a central part of negotiations. We denote the resulting expected price level by Pe.
In the end, both sides bargain directly over the nominal wage W and indirectly over the expected labour share WL/PeY. A simple description of the outcome of the negotiations is that wages determine unit labour costs as a mark-up γ over the expected price
level:
(13.6)
where is the ‘normal’ share that wage negotiators accept as representing an appropriate division of income in normal times, i.e. when the demand for labour is neither too strong nor too weak.15 The wage mark-up γ shows how the agreed-upon share varies over
the cycle. The mark-up can be positive or negative, and it is zero on average. In some years, γ > 0, and the agreed-upon wage is above its normal level. In other years, γ < 0 and the wage is lower.
Putting it all together
We have described prices as a mark-up over unit costs, agreed to ignore all non-labour costs and therefore to focus on unit labour costs, and found that the expected labour share is a mark-up over its normal level. This in turn means that wages are a mark-up over
prices. In the end, prices depend on wages and wages depend on expected prices.
Isn’t this a bad case of circular reasoning? Quite to the contrary, this description captures the fundamentally conflictual relationship between employers and employees as they divide up the GDP cake. It has earned the title ‘ battle of the mark-ups’.16 The
price mark-up, equation (13.5), sets the price level as a mark-up on wages, while the wage mark-up, equation (13.6), sets wages as a mark-up on the (expected) price level. Firms increase profits by reducing labour costs relative to the prices that their products can
fetch. Employees increase real wages by pushing them up relative to expected prices.
Note carefully the distinction: actual prices depend on wages, while wages depend on expected prices. The apparent circularity really links actual prices to expected prices, as can be seen by combining (13.5) and (13.6), resulting in:
(13.7)
We see that the circular process has an anchor: Pe, the level of prices expected by wage negotiators to prevail over the course of the contract. We will later find that this expectation is one central determinant of inflation in the medium run.
13.4.3 Productivity and the Labour Share

Why doesn’t productivity show up in the price setting equation (13.7)? One might think that, as productivity increases, unit cost and prices will fall. But that would be a partial equilibrium mode of thinking. In normal times, wage determination will reflect the
productivity of workers and claw back part of the productivity in higher pay.
Box 13.3 The Wage Mark-up and the Labour Share
Equation (13.4) defines the labour share as We have characterized wage negotiators as proceeding in two steps. First, they agree—in a process that can in fact be highly conflictual—on a ‘normal’ level
of this share:
where is the corresponding ‘normal’ real wage. Once this is done, wage negotiators will recognize that the labour market is tight (strong demand because the economy is booming) or weak (in a recession). They then agree—an
equally controversial step—on the expected real wage (W/Pe) as a markup γ on the normal wage, delivering equation (13.6).
What drives the normal labour share? This is a complex social issue. Here we just note that the share remains constant where workers get their appropriate or ‘fair’ share of productivity gains. To see this, note that the labour share can be rewritten as
(W/P)/(Y/L). Using the rate of change rule presented in Box 5.3, we see that the percentage increase in the labour share is:
The labour share is constant when real labour costs (W/P) grow at the same rate as labour productivity (Y/L).
Here it is useful to take a step back and recall one of the important stylized facts from Chapter 3: labour productivity (Y/L) and real wages (W/P) tend to grow systematically over time. Furthermore, recall that equation (13.4) defines real unit labour cost (or the
labour share) as WL/PY. When real wages grow at the same rate as labour productivity, real unit labour costs remain constant, as shown formally in Box 13.3. Put differently, when productivity is rising, real wages can increase at the same rate without raising
production costs. This is the base case, which defines the ‘normal’ labour share . The battle of the mark-ups studies fluctuations in real unit labour costs that occur when real wages temporarily depart from this principle.
The base case is precisely what we expect to see following the principles developed in Chapter 4 and one of the stylized facts concerning economic growth discussed in Chapter 3. Indeed, Figure 1.3 in Chapter 1 showed that, in four countries over half a
century, labour shares fluctuate from year to year, but with little discernible trend. This is just another way of saying that technical progress continuously generates higher incomes, and gross hourly wages (including all employee costs) and hourly productivity
track each other pretty well.
13.4.4 Cyclical Behaviour of Mark-ups

Now we turn to the determinants of the two mark-ups. The brief answer is that they tend to move over business cycles. To see which side would ‘win’ the battle of the mark-ups, we need to separate out the two, distinctly different mark-up decisions.
Start with the price mark-up θ. The price mark-up is like an operating profit margin—the excess of revenues over variable costs. Firms naturally would like profits to be as large as possible, but have to worry about competition.17 Competition increases in good
times, when there is ‘more money in the marketplace’. While firms may try to raise price mark-ups when demand is high, high business volumes increase total profits so much that new firms enter and steal competitors’ customers. The effect here is uncertain.
The wage share mark-up γ is the outcome of wage bargaining at the level of collective negotiations, but could also reflect individual outcomes of a more competitive labour market. During boom periods, rising employment generally improves the bargaining
position of unions and workers in general, which is reflected in a higher wage mark-up. In addition, firms may offer higher real pay to motivate employees to work harder or longer hours, or even to encourage others to join the labour force.18 More likely than not,
the wage mark-up is—or is likely to be—procyclical.19
Combining these two observations, it follows that the product (1 + θ)(1 + γ) of the two mark-ups that appears in equation (13.7) is procyclical. It tends to push the actual price level above its expected level in boom times, and to pull it down in slumps. This
procyclical outcome of the battle of the mark-ups is the foundation of the Phillips curve. It is also consistent with evidence on the countercyclical nature of the wage share—WL/PY tends to decline in booms and rise in recessions.
We now use (13.7), which describes how the price level P is set, to conclude that the inflation rate, the rate of increase ΔP/P of the price level, is driven by two factors: (1) the combined change of the two mark-ups, and (2) the expected inflation rate. Using the
symbol π to stand for the rate of inflation, and πe to represent the expected rate of inflation, we represent this result as:20
(13.8)
We are now just two short steps away from reaching the Phillips curve. First, we have seen that the combination of the aggregate mark-ups moves with the business cycle. It rises when the real GDP Y moves above its trend , i.e. when the output gap Ygap is
positive, and it declines when the gap becomes negative. Alternatively, we can use Okun’s Law to describe mark-ups as rising when the unemployment rate U declines below equilibrium –U, i.e. the unemployment gap Ugap is negative, and declining in the
opposite situation.
The second step recognizes that wage bargaining and price setting do not occur all at once, but are staggered over time. It is simply too costly for firms and workers to discuss wages all the time; after all, there ’s other work that needs to be done! Similarly, firms
are not in a position to change their prices every day. Customers expect a certain stability in pricing, and may even have entered contractual agreements to guarantee that stability. As a result, while decisions are taken rationally, they are often based on older
information. There is no single expected inflation rate πe, but rather a complex mix of many expected rates, some relevant for the present, some coming from the past and looking to the future. For the aggregate economy, we can capture this inertial inflation in the
economy as , which we call the underlying inflation rate. As we will see, this is not only a realistic detail, it also turns out to be crucial for understanding the Phillips curve.
We have now reached our final destination. The process captured by equation (13.7) is turned into a relationship between actual and something which resembles expected inflation. Indeed, taking into account the cyclical behaviour of the combined mark-ups,
the relationship (13.8) can be written as:
(13.9)
The first part of the equation is an augmented aggregate supply curve. The second is an augmented version of the original Phillips curve. Both are augmented by consideration of underlying inflation.
The relationship (13.9) offers a simple but very effective accounting of what drives inflation. It asserts that inflation is driven by two main forces:
(1) Underlying inflation, that is what people expect or have expected inflation to be, now and in the future.
(2) Cyclical conditions, with wages and prices tending to rise faster in years of rapid economic growth and slowing down in harder times.
The positive parameters a and b simply describe how the mark-ups jointly respond to cyclical fluctuations, represented alternatively by the output and unemployment gaps. They say that a high level of activity leads to higher mark-ups.
13.4.5 More on the Underlying Rate of Inflation

In the last section we introduced a new concept, the underlying rate of inflation.21 Because this concept is so important, we will spend a little more time discussing it here.
Recall that the expected price level in equation (13.6) is one important component of wage bargaining. It describes how wage negotiators expect the price level to evolve over the course of wage contracts under discussion. In that sense, the underlying inflation
rate is forward-looking. In practice, it also includes a backward-looking component for several reasons. First, workers may use past behaviour of inflation to forecast the future. This may be a clever way of learning what the inflation rate is doing, or it may reflect a
cheap alternative to hiring some economist to do the job. Second, cost-of-living clauses in labour contracts may actually mandate increases in nominal wages based on past inflation. Third, inflation forecasts are inevitably inaccurate ex post, and workers and firms
will try to correct their past mistakes—if they can.
Consider the following examples of the last mechanism at work. Suppose a worker’s wage contract incorporates a nominal wage increase of 4% in the year to come, reflecting the expectation of 3% inflation plus a 1% gain in labour productivity. After the fact,
productivity indeed does rise by 1% but inflation turns out to be 5%! Workers, having lost 1% of purchasing power, will feel ‘cheated’ of not just 2% inflation forecast error, but also by the fact that their employer made off with 1% productivity. They are likely to
try to recover these losses in negotiations for the following year.
Conversely, suppose inflation is only 1% after the year has passed. Then real wages are likely to have risen too fast and profits squeezed. Employers will no doubt use this fact to fight wage increases in the future. Correcting for past forecast errors is an integral
part of wage negotiations, and indeed this is why the logic of the purely forward-looking price level Pe in (13.8) has to be extended with a combination of forward- and backward-looking considerations. This is what underlying inflation is designed to capture.
Forecast errors are unavoidable. When inflation is reasonably low, the errors are small and the combination of catch-up and forward-looking forecasts encoded in the underlying inflation rate is acceptable. When inflation is substantial, say, around 20% per
year, forecast errors can easily be much higher. In periods when inflation is high, it also tends to be more variable, leading to even larger forecast errors. This is why various mechanisms designed to cope with high inflation tend to emerge when the price level is
rising fast. For example, explicit or implicit indexation clauses can link wages automatically to the evolution of the consumer price index. Another response is to reduce the contract length for wages and prices to shorter periods, like a quarter, or a month. In
periods of extremely high inflation, wages and prices may even be reset weekly or daily! Avoiding such situations is why price stability, i.e. low inflation, is so desirable.
13.4.6 Completing the Picture: Supply Shocks

It is now time to look at the non-labour costs of production, carefully ignored until now. These costs correspond to the other factors of production—e.g. capital and land—as well as to intermediate inputs such as unfinished goods, primary commodities, and
energy. In general, these costs follow the general trend of final goods prices. Their effect on inflation is then adequately reflected in the underlying inflation rate, adding nothing new to the analysis.
For this reason, these costs can be neglected most of the time. Now and then, however, special circumstances arise when non-labour costs do not behave so innocuously and significantly affect the short-run evolution of inflation, independent of the state of
the economy, the market power of workers and firms, or what they have expected inflation to be. These events are treated as exogenous because they happen elsewhere or are special in the sense that they are not explained by the framework that we develop.
Because they affect the costs of production, they are called supply shocks.
The leading example of a supply shock is a sharp increase in the price of oil, commonly referred to as an oil shock. Petroleum is a primary energy source and is an especially important raw material for the electrical generation, transport, chemical, and construction
sectors. These sectors supply all sectors with inputs, so changes in oil prices have an important impact on production costs across the economy as well as on household budgets. Oil shocks have occurred at various points in time during the 1970s, early 1980s,
and mid-2000s. Figure 13.11 gives a historical perspective. Another example of a supply shock is a steep depreciation, which renders imported good prices more expensive when evaluated in the domestic currency. Supply shocks may also be favourable. Oil prices,
for instance, fell very fast in late 2008 after having risen considerably in 2007. A sharp appreciation cheapens imports and for an open economy can have a significant negative impact on the short-term evolution of inflation, even while possibly hurting
international competitiveness (raising σ) at the same time.
In the end, we think of non-labour costs as generally following underlying inflation π∼ , but need to remember that supply shocks can and will occur—the sharp energy price increases seen in 2007–2008 are a clear reminder. To that effect, we allow for an
occasional occurrence of a supply shock, a catch-all for exogenous disturbances affecting production costs. We denote this supply shock term by s and extend the aggregate supply curve and Phillips curve (13.9) in an important way:
(13.10)
(13.11)
As long as non-labour production costs are in line with the underlying rate of inflation, the supply shock term is zero. Occasionally, in unmistakable circumstances, it can be positive (an adverse supply shock) or negative (a favourable supply shock). On average it
should be nil.
Beyond changes in non-labour costs, two other types of supply shocks can be important. First, we assumed that the labour share is constant. A quick glance at Figure 1.2, however, shows that it is approximately trendless, but certainly not constant.
Occasionally, it rises or declines independently of the normal mark-up fluctuations, a process that can last several years. Such changes are usually the outcome of socio-political events that strengthen or weaken trade unions or employers’ associations. Changes
in the labour share can be interpreted as supply shocks. An exogenous increase of the labour share, for example, adds to labour costs and can be treated as s > 0.
Second, supply shocks can be caused by the government. An increase in certain types of taxes (for example, the value added tax) or prices of regulated utilities can raise the inflation rate even though the underlying rate and labour market conditions are
constant. Many taxes paid by firms relate directly to production and affect the final selling price—value added or excise taxes, profit taxes, establishment and property taxes, and so on. Other costs imposed by governments are implicit, but may have a significant
impact (e.g. labour, environmental, or consumer protection regulations).
13.5.1 The Phillips Curve Rehabilitated

The inflation-accounting framework summarized by equation (13.11) effectively solves the puzzle of the Phillips curve. We can explain its persistence over decades, its instability over particular periods of time, and the impression that it shifts now and then. The
original Phillips curve claimed that inflation depends only on the level of unemployment. The inflation account shows that cyclical labour market conditions do indeed matter, but so do underlying inflation, equilibrium unemployment, and occasional supply
shocks. For a Phillips curve to be visible, these latter factors must be stable so that changes in inflation are mostly driven by cyclical conditions. When these factors are not stable, the Phillips curve seems to vanish. This is precisely what happened during the
1970s, when price and commodity shocks became a major source of instability (see Figure 13.11). As inflation rose, underlying inflation rose as well and became more variable, reflecting rapidly changing expectations. The Phillips curve’s demise reflects the
emergence of underlying inflation and supply shocks. In addition, equilibrium unemployment rose in many countries during this period, for reasons explained in Chapter 4. These additional determinants of inflation, above and beyond cyclical factors, have helped
to rehabilitate the Phillips curve.
How should we think of the Phillips curve if it proves to be unstable? Much as with the IS and TR curves, the answer is to keep clearly in mind which variables are endogenous and which ones are taken as exogenous. As long as the exogenous variables remain
constant, the curve does not move. When they change, the curve will shift and apparently ‘vanish’. This is the way the reconstructed Phillips curve explains the apparent puzzle. The modern Phillips curve is sometimes referred to as the expectations-augmented
Phillips curve. This name emphasizes the central role of underlying inflation. Yet, it is more than that; it also recognizes that equilibrium unemployment –U may change. It also allows for supply shocks.
Fig. 13.12 The Expectations-Augmented Phillips and Aggregate Supply Curves
By definition, point A represents the case where actually observed inflation π is at its underlying rate and where unemployment U is at its equilibrium rate and output Y is at its trend value . When unemployment is low and output high, actual inflation is above the underlying inflation
rate (point B). When unemployment is higher and output is lower than their respective equilibrium levels, actual inflation is below underlying inflation (point C).
Locating the Phillips curve requires a proper understanding of the underlying determinants of inflation. According to (13.10) and (13.11), when supply shocks are zero ( s = 0) and when actual unemployment equals its equilibrium level and output is on trend
the unemployment and output gaps are nil and actual inflation equals the underlying inflation rate . The pair serves as an anchor which pins down the position of the Phillips curve at any point in time.
This situation corresponds to point A in either chart of Figure 13.12. Point A uniquely determines the position of the Phillips or aggregate supply curves. The other points on the curves simply follow from allowing the unemployment rate to vary around its
equilibrium level, holding underlying inflation constant and setting the supply shock to zero. This is the key intuition from the old Phillips curve that inflation varies with the business cycle. For instance, at point B unemployment is below equilibrium and output is
above trend, so the demand pressure pushes inflation above its current underlying rate. Conversely, point C corresponds to the case where inflation is below its underlying rate because the unemployment rate is above equilibrium and output is below trend. In the
left-hand chart of Figure 13.12, this traces a downward-sloping schedule that resembles the old Phillips curve shown in Figure 13.3. In fact, it is the same but with a crucial difference: its position is determined by point A, that is, by the underlying inflation rate and
equilibrium unemployment .22
13.5.2 Underlying Inflation and the Long Run

Underlying inflation captures the rate of inflation agreed upon during wage negotiations, both now and, for some bargaining parties, at points in the past. It has both a backward-looking (it reflects old forecasts of inflation today, as well as efforts to fix past errors
of inflation forecasts used in previous contracts) and a forward-looking component (what will inflation be in the future). Ultimately, it must be related to the actual rate of inflation. How this is so in the short run is considered in the next chapter. In this section we
deal with the long run. We already saw in Section 13.3.3 that the long-run Phillips curve is vertical. We now look at how this comes about.
As negotiators consider the amount of inflation to be factored into wage settlements, they strive to guess it accurately. Of course, employees have an incentive to overstate the underlying rate of inflation, but employers have precisely the opposite incentive. If
there were no uncertainty and both sides always knew ex ante what inflation would be over the lifetime of the wage contract, underlying and actual inflation would just be equal when the unemployment and output gaps are nil. Uncertainty therefore means
forecasting (or guessing) underlying inflation. More often than not, the guesses are wrong. Yet, the principle of rational expectations from Chapter 6 means that wage negotiators do not make systematic forecast errors. Although forecasts are almost always
incorrect, the errors are largely unsystematic and average to zero: in some years inflation is overestimated, in others it is underestimated.
These observations have two important implications. First, there must be a link between actual inflation π and underlying inflation . Underlying inflation must track, albeit imperfectly, actual inflation. The backward-looking component implies that underlying
inflation lags behind actual inflation, but the forward-looking component implies that underlying inflation leads actual inflation. The interplay of both components is bound to be rather murky. There exist statistical techniques to do so, however, and they are
routinely used by economists.
The second implication relates to the long run. The rational expectations hypothesis means that actual and underlying inflation cannot remain different in any systematic way. Any discrepancy between π and ˜π must be temporary. Obviously, over the years to
come, we expect booms and recessions, possibly as consequences of supply shocks, positive and negative. We cannot really guess what the situation will be five or ten years from now. The best bet is the agnostic one, that there will be no shocks and that the
economy will not be far from trend, with actual unemployment equal to underlying unemployment. For this reason, it makes sense to define the long run as the situation where the economy is back on its trend. This means that the Phillips curve is vertical. From
equation (13.10), we know that when s = 0 and U = , then π = . In a way, the long run is atemporal; the backward- and forward-looking components of underlying inflation have worked themselves out.
Thus, for the third time, we conclude that the long-run Phillips curve is vertical. We first encountered this result in Section 13.2.2 when we found that long-run price flexibility implies that output is supply determined. Next, in Section 13.3.3, we argued that long-
run neutrality—which really follows from long-run price flexibility—implies a vertical Phillips curve. Now, we see that a vertical Phillips curve is also a consequence of the battle of the mark-ups, once we admit that there cannot be any permanent deviation between
actual and underlying inflation.
The vertical long-run Phillips curve only tells us that the actual and underlying inflation rates are equal. It leaves entirely open the question of what these rates are. In the next chapter, we bring back the demand side of the economy to pin down long-run
inflation, actual and underlying. In fact, we already know that it is the inflation target pursued by the central bank. Here again, the central bank may not always hit its target but it would be a grave failure if it never reached it. We will see that it takes time for
underlying and actual inflation to catch up with each other and stabilize at whatever rate monetary policy allows for. Views vary about how quickly this happens, and herein lie some of the most fundamental controversies in macroeconomics, already encountered
in Chapter 11 and to be studied further in Chapter 16.
The vertical Phillips curve carries a crucial implication: there cannot be a long-lasting trade-off between unemployment and inflation. Demand policies cannot move the actual unemployment rate permanently away from its equilibrium level. But the equilibrium
level can very well shift over time, e.g. as labour markets undergo structural changes. Indeed, one of the implications of Figure 13.10 is that unemployment equilibrium must have shifted over the last decades. It massively increased in the euro area over the 1970s
and 1980s—shifting the Phillips curve to the right—and then declined in the late 1990s. In the UK, the decline of the equilibrium unemployment rate started earlier and has been more pronounced—a legacy of Mrs Thatcher’s supply-side policies aimed at
weakening the bargaining power of unions, lowering unemployment benefits, and reducing the equilibrium unemployment rate.
13.5.3 Aggregate Supply

All of the previous reasoning applies to the aggregate supply curve shown in the right-hand chart in Figure 13.12 since it corresponds—via Okun’s Law—to the Phillips curve. Its position is determined by the output trend and the underlying inflation rate, and it
shifts when any of the exogenous variables ( , , or s ) changes.
The aggregate supply curve conveys two important messages:
(1) In the short run, as GDP fluctuates about its trend growth path, the actual rate of inflation can differ from the underlying rate. In the absence of supply shocks, output and inflation move in the same direction.
(2) In the long run, GDP must return to its growth path, regardless of what the inflation rate is. Real forces determine the growth of real activity.
The long-run aggregate supply schedule is vertical. It will, however, shift continuously to the right since, as a consequence of long-run economic growth, trend output keeps rising.
13.5.4 Factors that Shift the Phillips and Aggregate Supply Curves
The original position of the Phillips and aggregate supply curves in Figure 13.13 is determined by point A. This point corresponds to the initial underlying inflation rate, to the equilibrium unemployment rate, and trend GDP. Thus, the underlying inflation rate and
the equilibrium unemployment rate or trend GDP are taken as exogenous when we draw the curves. The fact that they determine the position of the curve, giving us the location of point A, provides two reasons why the curves can shift. The first is a change in the
underlying inflation rate: an increase in underlying inflation shifts the curves up. The second reason is that equilibrium unemployment and trend GDP may change. The supply shock s is the third exogenous variable that shifts the curve.
There is no presumption whatsoever that either underlying inflation or equilibrium unemployment are constant over time. If either changes, point A moves, and so does the whole Phillips curve. This implies that, potentially, there exists an infinity of Phillips
curves, corresponding to the infinity of values that the underlying inflation rate or the equilibrium rate of unemployment can take. It just so happened that, over the hundred years surveyed by Phillips, the underlying rate of inflation and the equilibrium rate of
unemployment did not change much, so there seems to have been just one Phillips curve.23 This changed in the early 1970s when unemployment became much more stable—as the result of active Keynesian policies—while more volatile commodity prices became
the dominating factor affecting inflation.
Shifts in the equilibrium unemployment rate occur occasionally. Table 13.2 presents some estimates of the equilibrium unemployment rate obtained by asking what rate would keep the Phillips curve unchanged when the actual and underlying inflation rates are
equal and in the absence of a supply shock.24 The table shows that the equilibrium rate rose in most countries in the 1970s and then declined in some countries two or three decades later, as the result of labour market reforms. This is one reason for the rightward
shift of most Phillips curves in the 1970s, and for the subsequent leftward shift in those countries that managed to bring unemployment down in the 1990s or 2000s. Two examples are presented in Figure 13.10.
Trend output continuously rises as the outcome of long-run growth. To avoid dealing with a curve that constantly moves to the right, in later chapters we will draw the aggregate supply curve with the output gap on the horizontal axis since it does not follow
any trend.
Table 13.2 Equilibrium Unemployment Rates
1970 1980 1990 2000 2010 2015

Austria 2.1 2.5 3.7 4.1 4.4 4.5
Belgium 2.8 6.4 7.9 8.0 8.0 8.0
Denmark 2.0 5.5 6.7 5.5 6.0 6.3
Finland 3.6 4.9 7.4 10.2 8.0 7.4
France 2.1 5.4 8.2 8.6 8.7 9.2
Germany 0.9 4.3 7.2 7.6 6.7 4.9
Hungary 6.4 10.2 8.4
Ireland 5.8 9.9 13.8 8.4 10.0 10.7
Italy 5.8 6.0 8.9 9.3 8.0 9.1
Japan 1.6 1.8 2.4 3.9 4.1 3.8
Netherlands 2.9 5.9 7.5 4.9 5.2 5.9

Norway 1.6 2.1 4.3 3.6 3.3 3.3
Portugal 2.4 7.0 6.0 6.3 10.4 11.7
Spain 4.3 7.9 14.3 13.6 16.8 18.6
Sweden 2.8 3.2 5.1 7.0 7.5 7.5
Switzerland 0.6 1.7 3.2 3.9 4.0
United Kingdom 2.3 7.6 8.8 6.2 6.4 6.0
United States 4.9 7.3 5.9 5.6 5.6 5.4
Finally, the supply curve shifts in the presence of supply shocks. If these shocks raise the cost of production, the Phillips and aggregate supply curves shift upwards. If the increase is temporary, the curves will eventually return to their initial positions, unless
underlying inflation also increases in the meantime. Oil shocks are a prime example of sudden increases in non-labour costs. The shocks of 1973–1974, 1980, and the early 2010s also played a role in the shifts visible in Figure 13.10.
13.5.5 From the Short to the Long Run

We have salvaged the old Phillips curve by replacing it with two curves: (1) a short-run expectations-augmented Phillips curve, which resembles the old one but can move, and (2) a long-run Phillips curve, which is always vertical. To these two curves correspond,
via Okun’s Law, a short-run and a long-run aggregate supply curve. These results are brought together in Figure 13.13. In both panels, point A describes the long-run situation where actual and underlying inflation are equal, there is no supply shock (s = 0), and
actual unemployment is at its equilibrium level or output is equal to its trend. Note that point A belongs to both the short- and the long-run curves. As just noted, it represents the long-run situation, but it also determines the position of the particular short-run
curve when the underlying inflation rate is . In other words, the only position on the short-run curve that is sustainable in the long run is point A. Any other position is temporary, as we now illustrate.
To that effect, starting from point A, imagine a demand expansion designed to reduce unemployment and shift the economy to a point like B. The short-run trade-off means less unemployment and more output, but also more inflation. This is because the new
inflation π2 exceeds the underlying rate , which was used in previous wage negotiations. Sooner or later, there will be another round of wage negotiations, which will recognize that inflation has increased. Let us assume that they adopt π2 as the new underlying
rate. Graphically, this means that the short-run Phillips and aggregate supply curves shift upwards, passing through point A′, which corresponds to the new underlying inflation rate . (The equilibrium rate of unemployment and trend
output are assumed constant throughout.) This shift worsens the unemployment–inflation trade-off, since any level of unemployment now requires a higher rate of inflation. If the authorities react by picking point C on the new curves, both unemployment and
inflation will rise, while output will decline. Yet point C is not permanently sustainable either, since inflation is now even higher than the newer underlying inflation rate . So, once again, the curves must eventually shift further up in both panels of Figure
13.13. We can imagine a succession of increasingly desperate policy reactions triggering further shifts in the short-run Phillips and aggregate supply curves, which will eventually drive unemployment and output back to their equilibrium positions at a point like Z
on the long-run curve.
Fig. 13.13 From the Short to the Long Run

For a given underlying rate of inflation, the economy can move from point A to point B, with lower unemployment and higher output but at the cost of higher inflation. This trade-off is not permanent, however. When underlying inflation rises to track higher actual inflation, the short-run Phillips
curve shifts up. In the long run when the actual and underlying inflation are equal (for example, point Z), the trade-off vanishes.
The fact that only one point on every Phillips curve is stable in the long run led Friedman and Phelps to predict that the historical curves would vanish if the authorities tried to exploit the short-run trade-off, for example by moving from point A to point B. In the
end, they predicted, the economy would end up at point Z in Figure 13.13. That the Phillips and aggregate supply curves are bound to shift whenever we move from the long-run point means that the inflation–unemployment trade-off cannot be seen as more than
a short-run relationship that may be worse than it initially appears. This last observation calls for caution when pursuing macroeconomic policies. We return to this theme in the next chapter.
Summary
1 The Phillips curve was once considered to be an adequate description of the supply side and the inflation process. Its message was that a permanent trade-off existed between unemployment and inflation. Output could rise to meet an increase in demand but would, in the process,
generate a higher rate of inflation.
2 Okun’s Law states that the output gap (the deviation of real GDP from its trend) and unemployment (as a deviation from its equilibrium level) systematically move in opposite directions. It translates the downward Phillips curve into an upward aggregate supply curve.
3 From the late 1960s to the late 1980s, Phillips curves around the world vanished. Contrary to the notion of an inflation–unemployment trade-off, inflation and unemployment both rose in the mid-1970s and early 1980s, a phenomenon called stagflation.
4 Accounting for inflation starts with the study of how firms set their prices. The result is mark-up pricing, i.e. setting prices as a mark-up over production costs.
5 Production costs are separated into two broad categories: labour and non-labour costs. Labour costs rise when wages—and related costs—increase faster than labour productivity. They often represent the most important source of cost changes.
6 Nominal wages are also set as a mark-up on the nominal price level.
7 Wages are set through negotiations that acknowledge three main factors: (1) underlying inflation, (2) productivity gains, and (3) the state of the business cycle, which largely reflects the relative bargaining strength of employees and employers.
8 Wage contracts attempt both to catch up on past inflation and to protect wages from future inflation. Underlying inflation captures both these backward- and forward-looking aspects.
9 Inflation accounting describes the actual rate of inflation as responding to: underlying inflation, demand pressure in the goods and labour markets, and occasional supply shocks.
10 The inflation accounts explain both why a Phillips curve could exist for a century, and then vanished in the light of mounting inflation in the 1960s and early 1970s, the consequences of the two oil shocks of 1973–1974 and 1979–1980, and of a rising equilibrium unemployment rate.
11 In the long run, unemployment returns to its equilibrium rate. Equivalently, real GDP cannot permanently deviate from the productive potential of an economy. In the long run, the Phillips curve and aggregate supply schedules are vertical. The economy is dichotomized, growth and
real rigidities determine the GDP and unemployment, money growth determines inflation, and there is no trade-off between inflation and unemployment.
12 The Phillips curve describes the supply side and can be transformed into an aggregate supply curve using Okun’s Law. The supply curve says that, for increased output to be supplied, inflation increases because production—mainly labour—costs rise faster than anticipated, or than is
reflected in underlying inflation.
13 The positions of the Phillips and aggregate supply curves are determined by the underlying rate of inflation and, respectively, equilibrium unemployment rate and trend GDP. Any change in any one of these variables leads to shifts in the short-run curves. Supply shocks, whether
positive or negative, also shift the curves.
Key Concepts
Phillips curve
aggregate supply curve
neoclassical assumption
wage inflation
Phillips trade-off
Okun’s Law
output gap, unemployment gap
money illusion
stagflation
oil shock
market power
product differentiation
mark-up pricing
average or unit costs
wage bill
battle of the mark-ups
underlying inflation rate
indexation
supply shocks
long-run Phillips curve
Exercises
1 Figure 13.1 describes how real GDP and the price level react over time to a one-off increase in the money stock. What happens to nominal GDP?
2 Using the IS–TR analysis developed in Chapter 11 and 12, consider an exogenous decline in demand, for example a reduction in public spending. Show graphically that either exchange rate flexibility or price flexibility is enough to achieve the neoclassical result that output remains
constant at its trend level.
3 A Phillips curve is represented by the following relationship: , where s is a supply shock term, on average equal to zero. Draw the Phillips curve when = 7% and underlying inflation = 3% and 6%. Okun’s Law is
. Draw the aggregate supply schedule when = 10,000.
4 Using the conditions of the previous exercise, show the impact of supply shocks s = 5%, and s = –2%. Separately show the effect of raising to 10,200. Why is it argued that improving the performance of the supply side of the economy is good for both inflation and employment?
5 In the 1980s, when inflation in Brazil exceeded 1,000% a year, all prices and wages were indexed week by week. Why was indexation not only popular but generally perceived as vital? If inflation were exactly 1,200% over a year, what is the implied weekly inflation rate (assuming
weekly compounding for simplicity). What is the effect of missing the inflation rate by 1% per week over the course of three months?
6 What could be the effect of an increase in value added taxation (VAT) on inflation? Of an increase in corporate profit taxes? Of an increase in personal income taxes? State your assumptions carefully.
7 Using Exercise 3, consider the effect of a reform of labour market institutions which reduces equilibrium unemployment from 7% to 5%. Explain the effect on inflation in the short term and in the long term.
8 Suppose a government underestimates the equilibrium rate of unemployment and attempts to reduce the unemployment rate below the equilibrium rate by stimulating aggregate demand. Show the likely outcome of such a policy using the short- and long-run Phillips curves.
9 The supply curve is given by the following relationship: Initially, π = and s = 0. Then the government increases demand and raises output and keeps it at this level until
inflation reaches the rate of 4%, at which stage they let the output gap return to zero. Assume that the underlying inflation rate ˜π adjusts each period by half of the difference between its previous value and the previously observed inflation rate (for example if we currently have π = 6%
and = 4%, then next period the underlying rate will be 5%). There is no supply shock so s = 0 throughout. Compute period after period the actual and underlying inflation rate until the economy returns to a long-run equilibrium. What conclusion do you draw?
10 Between 2000 and 2007 the price of oil, quoted in US dollars, increased by 92.4%. Quoted in euros, it ‘only’ increased by 52.8%. How can you explain the difference? Explain why a ‘hard currency’ is frequently considered a weapon against imported inflation.
Essay Questions
1 Imagine that you are back in the 1960s, when the Phillips curve was believed to be stable. How, do you believe, were politicians arguing about where it is best to be on the curve? What mistake would each side of the debate be likely to make?
2 Why is the battle of the mark-ups important for understanding the origins of inflation? Discuss the significance of each of these assumptions: (1) imperfect or monopolistic competition among firms; (2) labour unions or collective bargaining; (3) constant labour productivity.
3 Why do some economists plead for policy measures that increase the flexibility of wages and prices? What can the counter-arguments be?
4 ‘That the long-run Phillips curve is vertical means that governments should never attempt to reduce unemployment with expansionary policies.’ Comment.
5 Under what conditions could a sharp depreciation only temporarily raise inflation?
1 The son of a Kiwi dairy farmer, A. W. Phillips (1914–75) started out as an apprentice electrician working in an Australian mine, then left for Britain via China and Russia in the late 1930s. After a tour of duty in the Second World War and time spent as a prisoner of war, he studied at the London
School of Economics and became a lecturer and later professor there. Phillips is remembered not only for his curve relating unemployment to rates of wage change, but also for the Moniac, a complex hydraulic representation of macroeconomy, on display in the Science Museum in London.
2 This intellectual debate is presented in Chapter 20.
3 In the analysis of Chapter 9, this would correspond to the case in which the interest rate is constant—as it is in the steady state.
4 We ignore the possibility that firms hoard rather than dismiss unessential labour. In fact, firms often avoid shedding their employees in a brief recession if they can afford to do so. In this case they would not be ‘on their labour demand curve’, they would employ more labour than strictly optimal.
5 Phillips was not the first to discover the Phillips curve. The American economist Irving Fisher published a paper in the International Labour Review of 1926 in which he unearthed a similar relationship in the USA.
6 This view of a trade-off was echoed by Helmut Schmidt, the ex-chancellor of West Germany, who stated in a newspaper interview in 1978 that he would prefer 5% inflation to 5% unemployment. How times have changed!
7 The law is named after the US economist Arthur Okun (1928–1980). In his research, he showed that a 1% decline of the US unemployment rate was associated with a 3% increase of GDP above trend. Referring to equation (13.2), this implies a value for h of one-third.
8 Note that the unemployment gap and the output gap are written in different ways. This is a technical detail. In Figure 13.5, both gaps are shown in percentage terms. Since the unemployment rate is already measured as a percentage of the labour force, its difference from equilibrium unemployment
is measured in percentage points. For the output gap to also be in per cent, we compute the difference between GDP and its potential level—both measured in the local currency—as a proportion of potential GDP.
9 Both men received the Nobel Prize in recognition of their critical analysis of the Phillips curve, as well as for other lifetime achievements. Milton Friedman, the Chicago iconoclast, received the Nobel Prize in 1976. Edmund Phelps of Columbia University was Nobel laureate in 2006. It was
Edmund Phelps who first formulated the famous golden rule of economic growth discussed in Chapter 3.
10 In his address to the American Economic Association in 1967 Friedman argued that ‘there is always a temporary trade-off between inflation and unemployment, there is no permanent trade-off. The temporary trade-off comes not from inflation per se, but from a rising rate of inflation’, Friedman
(1968: 10).
11 Chapter 2 presents various ways of measuring the price level.
12 In theory, the relevant concept is the marginal production cost, the cost of producing another unit of output. In practice, marginal costs are difficult to measure, so we approximate them by unit costs. This is an acceptable approximation under conditions of constant return to scale in all factors.
13 In interpreting these numbers, it is important to remember the distinction between value added and turnover or total sales, which was stressed in Chapter 1. As a percentage of total sales, wage shares are much lower because total turnover in an economy includes the costs of intermediate goods. The
figures reported in Table 13.1 net out payments for intermediate inputs produced by other firms.
14 Formally, .
15 Box 13.3 provides a formal explanation.
16 The battle of the mark-ups approach to understanding inflation has found empirical support in OECD countries in groundbreaking work by researchers at the London School of Economics: Richard Layard, Steven Nickell, and Richard Jackman, among others.
17 A natural tendency is for competitors to agree on large mark-ups. Such collusion is usually illegal and competition authorities seek out and prosecute anti-competition agreements.
18 This reasoning is developed in detail in Chapter 4.
19 The wage share tends first to fall in an expansion, catching up with a significant lag and continues to rise past the peak of the usual cycle, as can be seen in a Burns–Mitchell diagram.
20 For the mathematically interested reader: using the results of Box 5.3, (13.7) provides the rates of change of all variables, with the following approximation: . Assuming that the
normal labour share is constant, equation (13.8) follows, where Δ(mark-ups) is formally equal to .
21 Sometimes the underlying rate is simply called ‘inflationary expectations’, relying more on the interpretation that it incorporates the anticipated inflation of firms and workers. Others use the term ‘core inflation’—which is sometimes also used in the business press to describe the more stable
inflation rate for non-food, non-energy items.
22 This is why the equilibrium rate of unemployment is sometimes called the NAIRU: the nonaccelerating inflation rate of unemployment. At point B inflation accelerates above its underlying rate. At point C it decelerates. Only at point A is inflation stable.
23 There are good reasons for this. The period corresponds to the time of the gold standard and the Bretton Woods system, both of which constrained inflation from rising too much and kept underlying inflation in check.
24 This is precisely the NAIRU as explained in footnote 22.
Aggregate Demand and
Aggregate Supply 14
14.1 Overview
14.2.1 Aggregate Demand in the Long Run
14.2.2 The Short-Run Aggregate Demand Curve
14.2.3 Movements Along versus Shifts of the AD Curve
14.2.4 The Complete System
14.2.5 Fiscal Policy and Demand Disturbances
14.2.6 Monetary Policy and Realignments
14.3.1 Nominal versus Real Interest Rates: The Fisher Equation
14.3.6 Monetary Policy
14.4.1 Lags and Time Horizon
14.4.2 Supply Shocks
14.4.3 Demand Disturbances
14.4.4 Disinflation
Summary
Money influences only monetary variables and not real variables in the long run. The problem is ‘how long is long?’ The ‘Keynesian’ answer embodied in the concept of the
Phillips curve was ‘too long to matter!’: the ‘monetarist’ rejoinder was ‘shorter than the Keynesians think!’; extreme rationalism provides the answer ‘too short for anything else
to matter!’—answers that no one concerned with either the history or the practice of stabilization policy is likely to accept.
Harry G. Johnson1
14.1 Overview
This chapter is the watershed of the textbook. It combines the demand side of the economy—the Mundell–Fleming model of Chapter 12—with the supply side—
the response of output and inflation consistent with the plans of firms and workers developed in Chapter 13—into a single unified framework. This workhorse of
modern macroeconomics is known as the AS–AD model.
Until now, aggregate demand was analysed under the assumption that prices are sticky. The task in this chapter is to deal with inflation.2 The central result will
be the downward-sloping curve AD displayed in Figure 14.1: the higher the inflation rate, all other things being equal, the lower aggregate demand. The upward-
sloping aggregate supply curve AS has already been derived in the previous chapter. In a market economy demand equals supply, so the position of the
economy is described by the intersection of the AD and AS curves.
In this analysis of aggregate supply, we identified two AS curves: one for the short run and one for the long run. This distinction is fundamental. In the short
run, there is a trade-off between output (or unemployment) and inflation. In the long run, the supply curve, shown in Figure 14.1 as LAS, is vertical and the trade-
off has disappeared—monetary factors have no impact on real economic variables, e.g. real GDP, unemployment, or the real exchange rate. We will see that a
similar conclusion applies to aggregate demand, but it is horizontal in the long run, as displayed in Figure 14.1 as LAD.
Ultimately, the inflation rate is set by exogenous forces: either by inflation in the rest of the world in the case of fixed exchange rates, or by the monetary
authority in the case of flexible exchange rates. In the short run, inflation and demand are closely related, and understanding these interactions and linking the
economy’s short run to the long run is a key function of the AD–AS framework. The chapter concludes with several examples of the AS–AD model’s usefulness.
Fig. 14.1 Aggregate Demand and Aggregate Supply, Short Run and Long Run
The complete description of the macroeconomy comes in three steps. In the short run, the AD and AS curves allow us to understand the impact of changes in the exogenous
variables. In the long run, the dichotomy principle produces the LAD and LAS curves. The medium run describes how we move from the short to the long run.
As in Chapter 12, we will distinguish sharply between regimes of fixed versus flexible exchange rates under conditions of capital mobility. The reason is that
the AD curve is derived from the IS–TR–IFM framework, which operates very differently according to the exchange rate regime. In particular, monetary policy
operates through exchange rates, and is lost when the exchange rate is fixed, while fiscal policy is undermined by exchange rate movements under a flexible rate
arrangement. This framework also works differently in the case of a closed economy, without trade and capital movements. In that case, we can simply ignore the
IFM line and both fiscal and monetary policies work, exactly as in Chapter 12.3
We know from Chapter 12 that, when capital is mobile, a country that fixes its exchange rate to another currency loses its ability to pursue an independent
monetary policy. The central bank has no choice but to set its own interest rate at the world level. The TR curve is irrelevant, so we will again ignore it. This
means that shifts in aggregate demand will only arise because of shifts in the IS or IFM curves.

It is always a good idea to start with the long run. We showed in Chapter 5 that relative purchasing power parity (henceforth: PPP) implies that the real
exchange rate (σ) is constant, and that this is a good rule of thumb for thinking about the long run.4 With σ = SP/P*, the fact that S is fixed and σ is constant in
the long run means that the domestic inflation rate (π) must equal the foreign inflation rate (π*):
If the nominal exchange rate is constant, domestic inflation must be equal to foreign inflation in the long run. Put differently, a fixed exchange rate regime rules
out permanent differences between domestic and foreign inflation. If they could diverge permanently, the real exchange rate would appreciate or depreciate
without limit. For example, if domestic inflation exceeded foreign inflation permanently despite a constant nominal exchange rate, the resulting real appreciation
would make the economy increasingly uncompetitive and worsen the net export account with no end in sight. This cannot be a long-run equilibrium.5
This restriction is represented in Figure 14.1 as the horizontal long-run aggregate demand (LAD) line. It is a demand-side restriction, because any permanent
deviation would eventually lead to unsustainable current account deficits or surpluses.
Fixing the exchange rate means importing inflation from the country whose currency is used as a peg. The exchange rate becomes an anchor for monetary
policy. Figure 14.2 shows how this anchor has worked for Denmark, which has fixed its exchange rate since the mid-1980s, first to the deutschmark, and then to
the euro.
Fig. 14.2 Inflation in Denmark and the Euro Area, 1992–2016
Since the mid-1980s, the Danish central bank has committed to a fixed exchange rate regime. It first pegged the value of the currency, the krone, to the currencies of the
European Monetary System (primarily the deutschmark). After 1999 it pegged the krone to the euro. The figure shows that Danish inflation has remained relatively close to
inflation in the euro area. Because PPP does not hold in the short run, the relationship does not hold exactly every year. Yet over the period displayed in the figure, average
inflation was 2.0% in Denmark, only slightly below the overall euro area rate of 2.1%.
Sources: IMF, World Economic Outlook Database.
From Chapters 5 and 12 we know that, when the exchange rate is fixed, imported inflation drives the money supply. Indeed, foreign inflation eventually
determines inflation at home. Since economic agents are interested in the real value of money, as explained in Chapter 5, their demand for nominal money grows
one-for-one with the inflation rate. Under fixed exchange rates, the central bank is committed to satisfy whatever demand for money is forthcoming, and will
therefore let the money supply grow along with inflation. When the exchange rate is fixed, it is money growth that adjusts to (foreign) inflation, rather than
inflation adjusting to money growth. Box 14.1 formally derives the endogenous growth rate of money under fixed exchange rates.

Once prices are allowed to move, the real exchange rate σ can fluctuate and its movements directly affect aggregate demand. Our framework for thinking about the
short run is the IS–IFM framework of Chapter 12, which assumed that the price level was constant. Now we must adapt it to allow for inflation.
To do this, let us ask what happens when the inflation rate changes, all other things being equal. In Panel (a) of Figure 14.3, we start from general equilibrium at
point A. We assume that this is a stable equilibrium, at least as long as domestic inflation π is equal to foreign inflation π*. Now suppose that the domestic rate of
inflation rises from π to π′—holding the foreign rate of inflation π* constant, so π > π*. In this situation, the real exchange rate will appreciate. Our
competitiveness is eroded, the net export account worsens, and demand for domestic output declines. Graphically, the IS curve shifts to the left, say to IS′. The
new equilibrium occurs at point A′, the intersection of the new IS curve and the IFM line.6
Now suppose instead that inflation declines to π″, so π″ < π*. Competitiveness would improve, the real exchange rate would depreciate, and the IS curve would
shift outwards to IS″. The new equilibrium—after a period of one year, say—would be described by point A″. Connecting points like A, A′, and A″ in Panel (b) of
Figure 14.3, we trace the aggregate demand curve AD. The curve is downward-sloping because rising inflation weakens the country’s external
competitiveness, which reduces domestic and foreign demand for domestic goods. It represents aggregate demand because movements along the curve are
induced by shifts of the IS curve, the goods market equilibrium condition under the Keynesian assumption that supply passively responds to, and therefore
merely equals demand. It is a short-run curve because, as long as domestic inflation differs from foreign inflation, demand continues to change (so that a year
later, say, output would have moved further away from Y in Figure 14.3, further reducing demand and flattening the demand curve). The long-run demand curve,
instead, accepts the implication from PPP that domestic inflation is equal to foreign inflation. As already noted, it is the horizon line LAD whose position is
determined by the foreign inflation rate π*. We will see shortly how we move to the long run.
Box 14.1 The Real Exchange Rate and Money Growth Under a Fixed Exchange Rate Regime
A regime of fixed exchange rates imposes restrictions on domestic inflation, monetary policy, and the growth rate of money. This box formally explains
these restrictions. It rests on three key concepts already developed earlier in the book. The first one is PPP, the second one is money market equilibrium,
both of which are presented in Chapter 5, and the last one is the interest parity condition introduced in Chapter 12.
Let’s start with PPP, which implies that inflation is the same at home and abroad when the nominal exchange rate is fixed. For a small economy, this
means that the domestic inflation rate is equal to the foreign rate:
(14.1)
Next, the demand for money was introduced in Chapter 9 as k(i)Y. Finally, the interest parity condition under fixed exchange rates implies that i = i*.
Combining this condition with the money demand, we have:
(14.2)
Under a fixed exchange rate regime, the central bank supplies all money that is demanded. Equation (14.2) implies that the nominal money supply is
determined by demand and equal to k(i*)PY. Using the arithmetic principle presented in Box 5.3, this means that the nominal money growth rate is:
(14.3)
where g is the trend growth rate of GDP. Using (14.1) and recognizing that the rest of world target inflation to achieve a long-run rate of π*, we finally see
that the domestic central bank must allow money to expand at the growth rate given by:
(14.4)
Students will see similarities between this result and equation (B13.3) of Box 13.1. This is not a coincidence. Under fixed exchange rates, the role of target
inflation of the central bank is replaced by the foreign rate of inflation, π*.
As in previous chapters, it is essential to distinguish between movement along the AD curve and shifts of the curve itself. The rule is always the same: the curve
shifts when a relevant exogenous variable changes. Since the AD curve is nothing more than a summary of the IS–IFM framework, the list of endogenous and
exogenous variables is similar to those already identified in Chapter 12. The difference, of course, is that the price level and the inflation rate are now
endogenous.
This also means that any exogenous variable that shifts the IS curve also shifts the AD curve. For example, starting from initial inflation rate π, an increase in
government spending is represented in Figure 14.6(a) by a shift from IS to IS″. As long as inflation remains unchanged, in Panel (b) the corresponding point is
B. The new demand curve which passes through B must lie to the right of the initial curve, as shown in Figure 14.4. This reasoning applies to all exogenous
variables studied in Chapter 11: government purchases , net taxes , household wealth , the exogenous component of Tobin’s q (‘animal spirits’), and foreign
income Y*. Missing in that list is the real exchange rate, which is now endogenous, because it depends on the evolution of domestic prices—which are no longer
fixed—relative to foreign prices.
Fig. 14.3 The Aggregate Demand Curve Under Fixed Exchange Rates
Starting from inflation π* at point A, an increase in the rate of inflation to π′ reduces the country’s external competitiveness. The IS curve shifts leftward in Panel (a). The
resulting decrease in demand is reported in Panel (b). Conversely, a reduction in inflation to π˝ improves competitiveness, shifts the IS curve rightward, and aggregate demand
increases. The aggregate demand curve is downward-sloping.
Changes in exogenous variables that affect aggregate demand shift the AD curve, rightwards when aggregate demand rises, leftwards when it declines.
Conversely, the AD curve stays in place when these variables remain constant. Any change in other variables implies that we move along the AD curve. Let us
now examine how and when this happens.
Fig. 14.4 Shifts in the Aggregate Demand Curve

Exogenous changes in demand which shift the IS curve also shift the short-run aggregate demand curve in the same direction. The outward shift shown here could be caused
by an increase in government spending ( ), reductions in taxes ( ), a rise in Tobin’s q, an increase in foreign GDP (Y*), or an increase in household wealth (Ω).

In Figure 14.5 aggregate demand and supply are brought together. The demand side comes in two parts: (1) the downward-sloping short-run aggregate demand
curve AD, and (2) the horizontal long-run LAD line, which reflects the endogeneity of money in fixed exchange rate regimes and the dependence of long-run
domestic inflation on the foreign inflation rate. The supply side, derived in Chapter 13, also comes in two parts: (1) an upward-sloping short-run supply curve AS,
and (2) the vertical long-run line LAS. The position of the AS curve depends on the underlying inflation rate, . The supply side in the long run dictates that
actual and trend GDP are equal (Y = ), which requires that actual and underlying inflation be equal as well , at the point at which the AS and LAS
curves intersect. On the demand side, the long-run domestic inflation rate equals the foreign rate . In other words, in the long run, the economy
stabilizes when GDP is on its trend path, inflation is the same as abroad, and the underlying inflation is in line with actual inflation. In Figure 14.5, the two long-
run curves intersect at point A.
Fig. 14.5 Aggregate Demand and Supply Under Fixed Exchange Rates
In the long run, output is at its trend growth level, the output gap is zero, and the inflation rate is equal to the underlying rate, as well as the foreign inflation rate. The short run is
determined by the AD and AS curves. The figure depicts a situation of long-run equilibrium in which all four curves intersect.
Note that we have changed the horizontal axis from previous chapters. Before, we tracked real GDP (Y) along this axis. We now represent the output gap Ygap =
(Y − )/ . This rescaling is important. We learned in Chapter 3 that trend GDP grows over time in most countries. To study the level of GDP would require a
continuous rightward shift of the LAS line, which would be quite cumbersome and, more importantly, would divert our attention from the focus of this chapter,
which is the origin and propagation of business cycles.
Figure 14.5 characterizes a long-run equilibrium in which the two short-run curves—the AS and AD curves—also pass through the long-run equilibrium
point A. In the following sections, we will study the short-run equilibrium as it evolves over time and distinguish it from the long-run position. In doing so,
we explain how the economy moves from the short to the long run.
14.2.5 Fiscal Policy and Demand Disturbances

Short run
We now track down the effects of an exogenous demand disturbance. One common example is a fiscal policy expansion—an increase in government
purchases (Δ > 0) or a reduction in net taxes (Δ < 0).7 We assume that, initially, at point A, the economy is in long-term equilibrium: output Y is at its trend level
, and actual (π) and underlying ( ) inflation are both equal to the world inflation rate π*. In the background of the demand-side analysis (IS–IFM), the domestic
interest rate is equal to the world rate of return (i = i*), and foreign inflation is equal to foreign central bank’s target inflation rate
The expansionary demand disturbance is depicted in Figure 14.6(a) as the rightward shift of the IS curve to IS′, which moves the AD curve to AD′ in the same
direction in Panel (b). The new curve AD′ shows the short-run effect of fiscal policy, say after one year. At point B, output has increased—as would be the case
under the Mundell–Fleming framework—but inflation has also risen, which was previously ignored by assumption. The rise in inflation is due to the upward-
sloping aggregate supply curve. The combination of a fixed nominal exchange rate and an inflation rate higher than abroad (P is increasing faster than
P*) implies that the real exchange rate appreciates. External competitiveness is eroding and the net export account is deteriorating. Thus rising
inflation reduces the impact of the demand disturbance. This is precisely why the AD curve is downward-sloping. Had inflation remained unchanged, as when
working with the IS–IFM fixed-price assumption in Chapter 12, competitiveness would have remained unchanged, and the outcome would have been at point B′
in both panels, along the IS′ curve—constant inflation with a larger increase in output. The loss of competitiveness, however, has shifted the curve leftward to
IS″. The horizontal distance between B and B′ is a measure of the inflation-induced deterioration of net exports.
Long run
The long run is governed by three central restrictions:
First, the government budget constraint rules out permanent fiscal expansions. For the initial shift in aggregate demand to be a long-run expansion, it must occur in each period.
For fiscal policy to be sustainable and consistent with a steady state, on the other hand, the public debt must be stabilized, as explained in Chapter 6. The fiscal expansion
implies that the public debt is rising, which is not sustainable. To reach a new long-run situation, the expansionary policy must eventually be reversed. When this is done, the
aggregate demand curve will return to its initial position AD.8
Second, output must return to its trend and the economy will stabilize along the LAS line. The logic here is that any non-zero output gap implies, by construction, that
underlying and actual inflation differ, which is not sustainable indefinitely, either.
Third, inflation cannot deviate from the foreign inflation rate for very long, if the exchange rate is to remain fixed. Thus, the economy must return to the LAD line.
The conclusion is that in the long run, the economy must return to point A, exactly where it started. The effect of a fiscal expansion is transitory because a
fiscal expansion cannot be permanent if the government’s budget constraint is to be satisfied.
The medium run

To summarize, we started from point A in Figure 14.6, moved to point B, and eventually moved back to point A. The actual path taken by the economy from the
immediate short run at point B to the long run at point A can be reconstructed using the observations just made. We already know that the budget constraint will
force the government at some point to reverse gears and either cut spending or raise taxes. When this will happen is a political decision—it could depend on the
timing of elections, for instance—and we cannot say much more about it. At any rate, the AD curve must eventually shift back to its initial position.
We can say more about the behaviour of the AS curve. Remember that its position is determined by underlying inflation ( ), which is assumed initially—say,
at time t = 0—to equal foreign inflation π*. In Chapter 13, we saw that the underlying inflation has backward-looking and forward-looking components. The
backward-looking component reacts to actual inflation conditions, ‘catching up’ with current inflation. Now note that at time t = 1 when the economy has moved
to point B, actual inflation is higher than the initial underlying inflation rate at t = 0. Inevitably, wage negotiators—whose judgements determine the underlying
inflation rate when nominal pay increases were agreed—will recognize that the current (time t = 1) inflation rate is higher than it used to be assumed. They will
naturally agree to increase nominal wages faster, in effect raising underlying inflation.
For the sake of the argument, suppose underlying inflation is simply equal to the inflation rate in the previous period. Then the AS curve would shift to AS′,
which intersects the LAS curve at the inflation rate observed at point B. The new (time t = 2) short-run curve AS′ cuts the LAS line at the underlying
inflation rate corresponding to the height of point B. This means that at time t = 2, the economy moves from point B to point C, at the intersection of the AD′ and
AS′ curves in Panel (b). From B to C, inflation rises further, but now, GDP is declining: inflation higher than foreign inflation implies further erosion of external
competitiveness and a deeper deterioration of aggregate demand. Behind this is a further leftward shift of the IS curve (not shown) as we move up along AD′.
This is an instance of stagflation.9
Fig. 14.6 Fiscal Policy Under Fixed Exchange Rates

A fiscal expansion shifts the IS and AD curves to the right, to IS′ and AD′. However, domestic prices rise, so the short-run outcome is point B instead of B′, implying an increase
in real GDP accompanied by higher inflation. The resulting decline in competitiveness implies that the IS curve only shifts to IS″. As underlying inflation catches up with actual
inflation, the AS curve shifts to AS′, which corresponds to a higher underlying inflation rate, equal to the inflation rate prevailing at point B. If the government does not change its
fiscal policy stance, the new equilibrium occurs at point C, where inflation has again increased above the underlying rate, leading to further shifts of the AS curve. If the
government cancels the fiscal expansion, the aggregate demand curve moves back to or even below AD, and the new short-run equilibrium is at point D, where actual inflation is
now lower than underlying inflation. The AS curve starts shifting to the right and will do so until it returns to its initial position and the long-run equilibrium is restored at point A.
We are not yet back to the long-run equilibrium, however, if only because fiscal policy is not yet corrected. At point C, the output gap is still positive and the
new current inflation rate exceeds underlying inflation (remember that along AS′, underlying inflation is equal to the inflation observed when the economy was at
point B). It is just a matter of time until underlying inflation rises again, pushing the AS curve farther upwards and to the left above AS′. In that case, stagflation
continues as the economy moves up along the AD′ curve. Eventually, though, the government will have to reverse the fiscal expansion.
So let’s now imagine that the government cancels its fiscal expansion at time t = 2 when the economy is at point C. The fiscal policy correction has a
contractionary effect and the AD curve moves from AD′ back to AD (assuming a complete policy reversal compatible with the long run). In that case, at time t =
3, the economy moves from point C to point D in Panel (b). Even though the AD curve is back to its initial position, output is now below trend. The reason is that
the AS curve has shifted to AS′ because underlying inflation has increased. The higher inflation hurts external competitiveness. In Panel (a), the corresponding
IS curve (not shown) is therefore to the left of the initial one (IS) since fiscal policy is back to its initial stance but external competitiveness has been worsened by
inflation.
Point D is on the AS′ curve, which corresponds to the underlying rate of inflation at time t = 2 at point C. But since point D is below point B, inflation is lower
than the underlying rate. The next round of wage negotiations will recognize that inflation is ebbing and underlying inflation will also decline. The AS curve will
shift down below AS′ and the economy will move down along the AD curve from point D in the direction of point A. At the intersection of AD and the new AS
curve (not shown), which corresponds to an underlying inflation rate equal to the actual rate observed at point D, inflation has again declined below underlying
inflation. This prompts a new reappraisal of underlying inflation, a new downward shift of the AS curve, and a continuing movement along the AD curve. The
process will continue until the AS curve has returned to its initial position and the long-run equilibrium is achieved at point A.10
Table 14.1 Tracking Movements in Figure 14.1
Time Event Movement Equilibrium

0 Initial situation Point A
1 Expansionary policy AD shifts to AD ′ Point B
2 Underlying inflation catches up with inflation at point B AS shifts to AS ′ Point C
3 Expansionary fiscal policy rescinded/loses impact Back to initial AD Point D
4 Underlying inflation catches up with inflation at point D AS′ shifts to the right Point E (not shown)
5 Underlying inflation catches up with inflation at point E AS shifts to the right Point F (not shown)
6 Underlying inflation catches up with inflation at point F, etc. AS shifts to the right Point G (not shown)
Long run Underlying inflation has caught up with actual inflation Back to initial AS Point A
To summarize, the acknowledgement of inflation requires that we reason in three steps:

1. the immediate short run, described by the amended Mundell–Fleming (IS–IFM) framework embedded in the AD curve and by the short-run AS curve;
2. the long run, described by the long-run AD and AS curves;
3. the medium run—the transition from the short to the long run—a drawn-out process driven by successive shifts of the short-run AD and AS curves.
The details of the curves and their movements with reference to Figure 14.6 are summarized in Table 14.1.
Until now, we have emphasized the role of the backward-looking component of underlying inflation. What is the role of the forward-looking component?
Anticipating the future evolution of inflation, wage negotiators will reduce the lag between underlying and actual inflation and the AS curve will adjust faster,
which will speed up the return to departure point A. We will see in Chapter 16 that this dimension depends strongly on the credibility of the anticipated policy
measures which lie behind inflation expectations.
The incorporation of inflation is a crucial amendment of the macroeconomic model. It modifies some of the conclusions that we reached in Chapters 12 and 13.
In particular:
First, a demand disturbance does not just move output, it also changes the inflation rate as we move along the AS curve. Along a given supply curve, an increase in demand
raises the inflation rate, while a decline in demand lowers inflation.
Second, bringing inflation into the picture naturally leads us to think beyond the current period. This was already apparent with the Phillips or AS curve because we need to
think about the evolution of the underlying inflation rate. On the demand side, we were led to think about the government’s budget constraint, which was ignored in Chapter 12
as we strictly focused on the short run. This led us to recognize that fiscal policy is inherently temporary.
Third, a fiscal expansion is partly undermined by the increase in inflation that it generates. The resulting loss of competitiveness—when the nominal exchange rate is fixed—
reduces net exports and world demand for our goods.
This explains why the countercyclical use of fiscal policy is much less popular than it used to be in the heyday of Keynesianism, before the old Phillips curve was
replaced by its rehabilitated version, augmented by forward- and backward-looking elements of expectations of inflation.
14.2.6 Monetary Policy and Realignments

A key lesson from Chapter 12 is that it is impossible to carry out an autonomous monetary policy when the exchange rate is fixed. But we indicated that exchange
rate parities can be and are changed, at least on occasion, since most fixed exchange rate regimes are considered ‘adjustable’. We now show how exchange rate
realignments allow for some limited role of monetary policy.
A devaluation, for instance, means reducing the nominal exchange rate. How is this done in practice? Formally, the central bank simply announces the new
parity and follows through with whatever is needed to make this happen. This means that a monetary policy expansion—achieved either on the open market or
through foreign exchange market interventions—will be necessary to increase the supply of domestic currency such that its price, the exchange rate, declines as
intended. More precisely, more money implies a lower interest rate and capital outflows, which weaken the exchange rate. This in turn raises demand and the
interest rate increases back up to the global return implied by the interest parity condition, the IFM line.
For simplicity, the initial situation at point A in Panel (a) of Figure 14.7 is assumed again to correspond to a long-run equilibrium, with actual and underlying
inflation both equal to foreign inflation. A nominal depreciation translates into a gain in external competitiveness, so that declines immediately
when S is reduced. As a result, the IS curve shifts rightward to IS′. In the background, the target interest rate is cut as already mentioned. Recovering the
monetary policy instrument is precisely the role (and some would say the goal) of exchange rate depreciations or appreciations. These are fleeting moments when
the central bank can recover monetary room to manoeuvre.
The short-run equilibrium is reached at point B. In Panel (b), the demand expansion is shown as the shift of the aggregate demand curve from AD to AD′, and
the corresponding outcome is represented by point B. As is now becoming customary, we find that an output expansion does not come for free, it is
accompanied by rising inflation. This, in turn, reverses and undermines partly the expansionary effect of the devaluation, which is captured by the leftward shift
of the IS curve from IS′ to IS″ in Panel (a).
Point B does not represent a long-run equilibrium because it is neither on the LAD nor on the LAS curves. It lies on the short-run AS curve with position
determined by initial underlying inflation (equal to the world rate of inflation at the outset), but now at point B inflation has risen. As a consequence, underlying
inflation increases and the AS curve will shift up. On the other hand, rising inflation means that the domestic price level rises faster than the foreign price level.
As a result, the real exchange rate keeps appreciating, net exports decline, and the IS curve further shifts leftwards in Panel (a). The economy will return to point
A, after a period of inflation above the world level. During the transition back to point A, the inflation differential progressively undoes the real depreciation
achieved through the initial devaluation.11 At point A, this effect is complete: the competitiveness benefit from the initial depreciation has been entirely offset by
the inflation differential with the country to whose currency the domestic currency is pegged.
Fig. 14.7 Devaluation
Starting from a long-run equilibrium, a devaluation causes the IS curve to shift to IS′, but rising inflation undoes some of the real depreciation, so the net effect is a shift to IS˝,
corresponding to point B in both panels. Output expands, but not as much in the original Mundell–Fleming model. Ultimately, higher inflation will lead to increasing underlying
inflation, which will eventually lead to pressure to devalue again, or a painful period of lower inflation and a negative output gap to restore competitiveness.
Fig. 14.8 Expansionary Monetary Policy Under a Fixed Exchange Rate Regime
A devaluation causes domestic inflation to rise above the foreign inflation rate. The real exchange rate, initially reduced by the devaluation, starts appreciating again until it
returns to its pre-devaluation level. The central bank may devalue the nominal exchange rate again, which immediately depreciates the real exchange rate and triggers a new
period of high inflation and real appreciation. In the end, monetary policy results in higher inflation and a chain of successive devaluations.
This is yet another case of long-run monetary neutrality. We knew all along that the long-run equilibrium is at point A, so the question was what would take us
there. In the end, if we start from long-run equilibrium, all real variables must return to their initial values, which applies to the output gap and to the real exchange
rate. Inflation rate, too, must eventually return to the world level. The important conclusion is that devaluations or appreciations only have temporary effects.
Another possibility, however, is for the central bank to devalue again. If it does so, the temporary competitive advantage will again be gradually eroded by the
inflation differential, leading to another devaluation, etc. In this limited sense, monetary policy independence is restored under a fixed but adjustable exchange
rate regime. Through a succession of devaluations, the central bank can keep temporarily pushing real GDP above trend, but it will have to accept a permanently
higher inflation rate. Figure 14.8 illustrates this path of the real exchange rate over time. It will depreciate abruptly at the time of each devaluation, only to
appreciate again afterwards. In the end, this strategy merely allows a country to opt for a different inflation rate from the one that prevails in the country to which
the currency is pegged. Box 14.2 shows how such an arrangement existed between France and Germany for nearly two decades. Of course, if exchange rate
devaluations become very frequent, the distinction between a fixed but adjustable exchange rate and a floating regime becomes blurred.
Box 14.2 Conflict and Coexistence with Different Inflation Rates: France and Germany
France and Germany have long existed side by side with very different views about inflation. Germany, still remembering the devastating hyperinflation of
1922–1923, was committed to low inflation, while France was more interested in using monetary policy to boost short-run growth. Yet as neighbours with
deep trade and financial relations, France and Germany were unwilling to allow the exchange rate to fluctuate from day to day according to market forces.
Their solution was to peg their exchange rates to each other, first informally in the 1970s, then formally following the launch of the European Monetary
System in 1979, and then irrevocably after the adoption of the euro in January 1999. The peg was adjustable and frequent realignments did take place
between the currencies, with the franc being regularly devalued vis-à-vis the deutschmark.
Figure 14.9 shows the evolution of the real exchange rate between France and Germany during this period. It is the real-life version of Figure 14.8.
Occasional nominal depreciations of the franc vis-à-vis the deutschmark, shown as sharp declines of the real exchange rate, are subsequently undone by
gradual real franc appreciation due to higher French inflation, until the next devaluation. The figure also shows that, after a severe crisis in 1993, the real
exchange rate stabilized. This reflects France’s ‘franc fort’ policy. In its preparation for monetary union, France gave up monetary policy independence,
renounced boom-and-bust devaluation cycles, and gradually managed to achieve an inflation rate virtually identical to Germany’s. This can be seen by the
fact that the real exchange rate between the two countries has hardly changed at all in since the early 2000s.
Fig. 14.9 The Real Exchange Rate, France vs. Germany, 1975–2011
The nominal exchange rate between the French franc and the German mark was fixed but adjustable during the period 1975–1998. France’s greater tolerance for inflation can
be seen in recurrent devaluations of the franc vis-à-vis the German mark. Each devaluation was followed by a real appreciation, until the next devaluation corrected the
cumulated exchange rate misalignment. Following German reunification, the situation was reversed: the bilateral French–German exchange rate depreciated (from the French
perspective) as German inflation surged in the early to mid-1990s. Since monetary union in 1999, devaluations have been ruled out—meaning that lower inflation in Germany than
in France since then has led to a real appreciation of the French–German real exchange rate.
When exchange rates are flexible, monetary policy and the nominal exchange rates swap roles. Monetary policy is exogenous and under the control of the central
bank. The nominal exchange rate, in contrast, is no longer exogenously fixed, but is determined by market forces and is endogenous. The AD–AS analysis turns
out to be remarkably similar to the fixed exchange rate case—with clear differentiation between short, medium, and long run—but there are fundamental
differences. In particular, we will see that the aggregate demand curve is negatively sloped, but for different reasons.
14.3.1 Nominal versus Real Interest Rates: The Fisher Equation

To understand how aggregate demand works when exchange rates are flexible, it is crucial to remind ourselves of the distinction between the nominal interest
rate and the real interest rate, already taken up at many places in the book. Because we have ignored inflation up to now, we could also disregard the
distinction between these two rates. But in a world with inflation, this would be a mistake. In Chapter 5, we defined the real interest rate (r) as the difference
between the nominal interest rate (i) and the expected rate of inflation (πe):
(14.5)
This relationship is known as the Fisher principle or Fisher equation.12 In the presence of inflation, future repayments are made in money that will be worth
less tomorrow than today. This loss of purchasing power of money tomorrow can be seen as a.pngt to present-day borrowers or, more precisely, it means that the
real cost of borrowing is less than the nominal cost, by an amount equal to the inflation rate between today and the future repayment.
The Fisher equation can be rewritten as:
(14.6)
Nominal rates are observed—they are used when specifying terms of loan contracts—but real rates can only be inferred from their definition (14.5). This requires
knowing what expected inflation is at the time of the loan contract but expectations are not observable. This is why real interest rates are generally not observable
either. Eventually, we will know what inflation will have been during the life of a contract and can always compute the ex post real interest rate—that is, after the
fact (see Figure 14.10). However, this ex post real interest rate (i − π) will differ from the ex ante real rate (i − πɛ), whenever actual inflation differs from what it was
expected to be (π ≠ πɛ). As a result, someone will be disappointed. If inflation turns out higher than expected, the ex post real rate is lower than ex ante
anticipated, which is good news for borrowers and a source of disappointment for lenders. Conversely, if actual inflation ends up lower than expected, lenders
will have earned more than they were asking for.
In Chapter 11, we stressed that aggregate spending (consumption and investment) is driven by the real interest rate while the demand for money is driven by
the nominal interest rate—the opportunity cost of holding it. Indeed, cash bears a zero nominal interest rate—no one gets paid for holding money. So, when
comparing return on money with return on other assets, we had to use the nominal rate. The Fisher equation shows that, ex post, the real rate of return on holding
money is r = 0 − π = −π, the negative of the inflation rate. This explains why inflation is often called a tax on those who hold money.
Fig. 14.10 Nominal and Real Interest Rates on French Government Debt, 1998–2011
Panel (a) shows the nominal and real interest rates on French government bonds, called OATs (obligations assimilables du Trésor). The rates of interest, or yields, on these
bonds are determined by the forces of demand and supply. One type of OAT compensates the owner for inflation, and its yield reflects the market’s assessment of the real
returns on financial assets. The other pays in purely nominal terms (in euros).The difference between the nominal and the real interest rates is a measure of the inflation rate
expected by market participants, which is shown in Panel (b). Until the global financial crisis in the fall of 2008, the long-run real interest rate was stable at around 2%, and the
implied expected inflation about 2%. Since then, however, real bond yields and the expected inflation rate have both dropped sharply. For a brief time, the markets appear to
have expected deflation—a period of negative inflation—in France.
Sources: Agence France-Trésor, Reuters.

In Chapters 11 and 12, the behaviour of central banks was formalized as the Taylor rule:
(14.7)
The output gap Ygap defined as , is the percentage difference between actual and potential or trend GDP. The inflation gap π gap is equal to π– and
represents how far the current inflation diverges from the central bank’s target rate . Recall that parameters a and b—which are both positive—express the
overall monetary policy objectives or ‘tastes’ of the central bank. A high value of a relative to b would represent an ‘inflation-fighting’ central bank. In contrast, if
b is large relative to a, this would signify that the central bank is more interested in reducing the output gap than keeping inflation close to its target.
The target or neutral interest rate is the rate chosen by the central bank when inflation and output are both on target—meaning that both inflation and
output gaps equal zero . What affects this choice? The Fisher equation gives the answer: the sum of the long-run real interest rate and the target inflation rate.
Let’s take a careful look at each of these components.
The neutrality principle tells us that the long-run real interest rate is given by the real economy. It is the return available from investments after adjusting for
inflation or, according to Chapter 8, the marginal productivity of physical capital. It is driven, therefore, by technology, the availability of labour, and other
factors, all of which are beyond the control of the central bank.
All that remains is the long-run inflation rate. Since the central bank ultimately controls the inflation rate, it can logically only be the inflation target as
embedded in the Taylor rule. It follows that the natural interest rate is equal to the economy’s real rate plus the inflation target:
(14.8)
A central bank that follows this logic will see to it that the inflation rate ultimately returns to its target rate, so π- , while our analysis of the AS curve establishes
that, in the long run, the economy will ultimately return to trend, so . This means that, by choosing its inflation target, a central bank
determines the long-run aggregate demand curve as the horizontal line LAD shown in Figure 14.12.
The LAD line under flexible exchange rates resembles that in the case of a fixed exchange rate regime, but its rationale is different. When the central bank opts
for a fixed exchange rate regime, long-run inflation is driven by foreign inflation; under a flexible exchange rate regime, it is set by the target inflation rate
embedded in the Taylor rule.13 The common feature is that long-run inflation and the position of the LAD line are always determined by the monetary policy
strategy. This is a key implication of the dichotomy principle.

The similarity between fixed and flexible exchange rate regimes carries over to the short-run aggregate demand curve. Yet the logic behind this curve is
fundamentally different under flexible exchange rates. Now, the AD curve is downward-sloping because a higher inflation rate leads the central bank to raise
interest rates—to pursue contractionary monetary policy. To see this, we again use the Mundell–Fleming model, now under flexible exchange rates—the TR–
IFM framework—remembering that the position of the IS curve is endogenous when the exchange rate floats freely.
We need first to adapt the way we use the Taylor rule to a world of variable inflation. In Chapters 11 and 12, the inflation rate was treated as exogenous and set
equal to zero for simplicity, so the Taylor rule only involved a relationship between the interest rate and output. With inflation, we must recognize that the
interest rate responds both to changes in output and in the inflation rate—that is the thrust of the Taylor rule. This means that inflation is now taken as
exogenous in the TR–IFM diagram; when it changes, the TR curve shifts.
Take A in Figure 14.11(a) as a starting point, a situation of short- and long-run equilibrium, so that inflation has long been at a level targeted by the central
bank. In order to find out the shape of the short-run AD curve, we study the effect of changing inflation, ‘all else constant’, including monetary policy as
summarized by the central bank’s Taylor rule, given its target inflation rate and neutral interest rate .
Fig. 14.11 The Aggregate Demand Curve Under Flexible Exchange Rates
The figure shows the effect of an increase in the rate of inflation on aggregate demand. Starting at point A with an inflation rate π, inflation rises to π′. The higher inflation rate at
the unchanged target rate of inflation prompts the central bank to raise nominal interest rates, shifting the TR curve upward in Panel (a). Demand is reduced (point A′), hence the
downward-sloping curve in Panel (b).
Suppose the rate of inflation were to rise from π to π′. The Taylor rule states that the central bank reacts by raising the nominal interest rate—by the amount
a(π′ π) for any level of output. This means that the TR curve shifts upwards to TR′ in Panel (a). The new equilibrium point A′ shows that the effect of a higher rate
of inflation is a decline in output. This reduction is brought about by a real exchange rate appreciation, which reduces the demand for goods—the leftward shift
of the IS curve (not shown).14 The exchange rate appreciation is a direct effect of monetary policy tightening in the face of higher inflation. The move from A to
A′ when inflation rises from π to π′ is reported in Panel (b) to obtain a downward-sloping short-run aggregate demand curve AD.
To summarize, the short-run aggregate demand curve is downward-sloping under both fixed and flexible exchange rates, but for different reasons. When the
nominal exchange rate is fixed, a higher inflation rate reduces demand through a loss in external competitiveness. Under flexible rates, higher inflation triggers an
interest rate hike by the central bank via the Taylor rule. Competitiveness declines too, but now because the nominal exchange rate S appreciates. The common
feature is the role of the real exchange rate in both regimes.

The AD curve shifts when any of the exogenous variables in the TR–IFM model changes. Under flexible exchange rates, the real exchange rate adjusts and the IS
curve shifts endogenously to meet the intersection of the TR and IFM curves. This means that factors that affect the IS curve only do not have any impact on
the AD curve. We already saw in Chapter 12 that under flexible exchange rates, fiscal policy fails to move the IS curve significantly because its effects are offset
by the reaction of the exchange rate. This applies to all aggregate demand shocks, including animal spirits and foreign output.15
The AD curve does shift, however, when either the TR or the IFM curves do. The TR curve represents the monetary policy strategy. The central bank changes
its interest rate when the inflation rate changes, and this explains the slope of the AD curve. Changes in the inflation rate, therefore, take the economy along the
AD curve. The remaining reasons why TR shifts would actually move the AD curve are changes in the Taylor rule itself. This includes either changes in the
central bank’s target inflation rate ( ) or in the preferences of the central bank, as represented by parameters a and b in (14.7).16 For example, a higher target
inflation rate would result in an upward shift of the AD (and LAD) curve. Changes in parameters a and b, instead affect the slope of the AD curve. If the central
bank reacts strongly to inflation changes, a higher a for instance, the AD curve becomes flatter. This is because an increase in inflation induces a larger increase
in the interest rate by the central bank, and with it a stronger real appreciation and a sharper decline in output.
The position of the IFM curve remains determined by the foreign rate of return i*, so changes in this parameter can also affect aggregate demand. As we will
see, long-run changes require the consent of the central bank or the government more generally, because they ultimately involve the long-run target rate of
inflation . Yet in the short to medium run, i* can move when international monetary policy becomes tighter (i* rises) or looser (i* falls), or when market
expectations of future exchange rates change. This latter possibility will be discussed in detail in Chapter 15.

Figure 14.12 presents the complete system under flexible exchange rates. It includes the now-familiar short- and long-run aggregate supply curves as well as the
short- and long-run aggregate demand curves. The figure displays a long-run equilibrium: actual output is equal to trend output—a zero output gap—as required
by the supply side, and inflation is equal to target inflation as required by the demand side. The PPP principle, presented in Chapter 5, further allows us to infer
the evolution of the exchange rate in the long run. The long-run rate of change in the exchange rate compensates for the difference between the domestic and
foreign inflation rates: ΔS/S = π* - π. If the foreign rate of inflation is below the domestic rate (π* < π), the exchange rate is depreciating ( ΔS/S < 0); if the
domestic rate is lower, (π* > π), it is appreciating (Δ S/S > 0).17
Fig. 14.12 Aggregate Demand and Supply Under Flexible Exchange Rates
In the long run, output is at its trend growth level (the output gap is zero) and the central bank’s target inflation rate determines the rate of inflation (the height of the LAD line). The
figure depicts long-run equilibrium when the short-run aggregate demand and supply curves pass through the same point as the long-run curves.
14.3.6 Monetary Policy

Central banks which follow a Taylor rule systematically set the interest rate relative to the natural or long-run target rate when inflation or output depart from their
target and trend levels, respectively. In that sense, the rule-based monetary policy does not change, but the rule itself may be changed. It can be the adoption of
a new target rate of inflation, or a changing view about the natural real interest rate or the equilibrium trend level of output, or a different sensitivity to the
inflation or output gaps (parameters a and b). Here we consider the case when it raises its inflation target from to ′. This means that the neutral interest rate
which is the sum of an unchanged real interest rate and the new target rate of inflation , will also increase permanently.
Long run
A permanently higher target rate of inflation means that the LAD line shifts up to LAD′ in Figure 14.13(a). The dichotomy principle implies that the real side of the
economy is left unaffected, so LAS is unchanged. In the long run, therefore, the economy will move from point A to point C. The vertical distance AC
corresponds to the increase in the target inflation rate.
Fig. 14.13 Monetary Policy Under Flexible Exchange Rates

Starting at point A in panel (a), a monetary policy expansion—here shown as a permanent increase in the central bank’s target inflation rate—shifts the AD curve rightwards to
AD′. At the same time, the LAD curve also shifts to LAD′, reflecting the permanently higher target inflation rate and intersecting the LAS curve at point C. In the long run, the
economy settles at point C, with GDP equal to trend output and the increase in inflation equal to the increase in the target inflation rate. Short-run equilibrium occurs at point B,
where actual inflation exceeds underlying inflation. Thereafter underlying inflation increases to its long-run level and the economy moves from point B towards point C.
Short run
The increase in the inflation target implies that the central bank will attempt to reduce the nominal interest rate, producing an exchange rate depreciation and
therefore an increase in aggregate demand. This is depicted in Figure 14.13(a) by the rightward shift from AD to AD′. In the background, the increase in shifts
the TR curve down and to the right in Panel (b) of Figure 14.13. The economy moves from point A to point B in both panels. Overall, output rises (and
unemployment declines) and inflation increases, but by less than the change intended by the central bank.18
The medium run
The transition from short to long run takes the economy in steps from point B to point C in Panel (a). At point B, where output is above its growth trend level, the
actual rate of inflation exceeds the underlying rate. What happens during the transition—and therefore the details of the trajectory—depends on the behaviour
of the underlying rate of inflation. To the extent that it is backward-looking, underlying inflation is sluggish. Initially, the AS curve does not move and the
economy moves to point B. Over time, underlying inflation begins to track actual inflation, the short-run AS curve shifts upwards, and the economy moves from
B towards C, along the curve AD′. As in the case of a fixed exchange rate regime, along the path from B to C, actual inflation exceeds underlying inflation. As
underlying inflation catches up, actual inflation rises, and output declines along AD′. Indeed, the rising inflation rate leads the central bank to tighten its stance
according to the Taylor rule. In Panel (b), this response to inflation is captured by an upward shift of the TR curve, as explained in Section 14.3.3. This will go on
as long as inflation continues to rise. The TR curve must continue to move up until it passes through point C, the new long-run equilibrium.
In summary, an expansionary monetary policy—described here as an increase of the inflation rate target—raises output and inflation in the short run. In the
long run, the effect falls entirely on higher inflation with no effect on output—the neutrality result is confirmed under flexible exchange rates. In the short run, the
backward-looking component of underlying inflation creates the non-neutrality needed for an output effect, while the forward-looking component tends to make
neutrality more likely to hold in the shorter run. The role of underlying inflation receives closer scrutiny in Chapter 16.
The complete AS–AD framework provides macroeconomists with a key tool for studying real-life events and answering important macroeconomic questions. We
proceed with three aims: (1) to illustrate the principles developed earlier, (2) to develop familiarity with the framework, and (3) to study and understand historical
developments of general interest.
14.4.1 Lags and Time Horizon

We start by briefly providing indications on the duration of the short, medium, and long runs, an issue already discussed in Chapter 1. Here we link this
discussion with the AS–AD and IS–TR–IFM apparatus. The question is: how long does an isolated disturbance need to work its way through the system?
Naturally, the economy isn’t served a single disturbance on a platter, but is constantly subjected to small and large ones, which move it away from its long run.
Despite this fact, an orderly discussion of the short, medium, and long run is useful—it gives us a time horizon for understanding the effects of disturbances and
more important, policy measures.
The short run corresponds to the Mundell–Fleming model, which is based on the Keynesian assumption of price stickiness. It lasts as long as the short-run
AS curve remains roughly in place. It takes about one to two years for demand disturbances to affect output, and sometimes up to three years for inflation to
react. At this point, inflation triggers changes in the underlying rate of inflation, the AS curve starts shifting, and we move to the medium run.
The medium run—the transition from the short to the long run—lasts from two to five years. This is when the short-run AS curve begins to shift as the
backward-looking component of underlying inflation starts catching up with actual inflation. The shorter the transition, the faster underlying inflation catches up.
This depends therefore on the relative contributions of its backward- and forward-looking components. As previously noted, if the forward-looking component
dominates, the short-run AS curve quickly reaches its final position—especially if prices and wages are not sticky. Wage and price indexation speeds up the
transition. In countries where inflation is very high, formal or informal indexation schemes are usually in place and the long run occurs very rapidly. We return to
this important issue in the next section.
The long run is defined as the horizon over which the dichotomy asserts itself. This means that we look beyond the business cycle horizon. Although two
business cycles are never exactly alike, experience shows that they generally last five to eight years.
14.4.2 Supply Shocks

Supply shocks occur when conditions of production change suddenly, with an impact on production costs and the evolution of inflation. Supply shocks come in
many different forms, but share the common feature that they invariably create difficulties for policy-makers, who are ill-equipped to face the consequences. This
is because traditional demand-side policies are ineffective in dealing with supply shocks. In addition, supply-side policies are complex, slow in generating
tangible results, and often politically unappealing.
The simplest example of an adverse supply shock is the sudden loss of human or physical factors of production resulting from natural disasters or wars. This
leads immediately to medium- or long-run loss of production potential (think of the Fukushima disaster in Japan or earthquakes in Turkey). Supply shocks can be
favourable as well, e.g. an acceleration of technological advances. The often-cited information technology revolution (described in Chapter 3) that started in the
mid-1990s is a recent example. Previous major episodes include the invention of electric generation and transmission, automobiles, and plastics. The discovery of
natural resources or a sustained surge in capital investment (think of the Solow model) are other examples of favourable supply shocks—possibly very persistent
ones.
Oil shocks are the best-known supply shock, with the instances of 1973 and 1979 representing major turning-points in twentieth-century post-war economic
history. They marked the end of the rapid growth performance of most European countries, and were followed by markedly higher inflation and unemployment
rates. Japan and the USA were also badly affected. The AD–AS model was developed largely in response to events of the 1970s, just as the IS–LM model was a
response to the Great Depression.
A short-term policy dilemma
We saw in Chapter 13 that supply shocks shift the aggregate supply curve. The increase in production costs is passed on by firms in the form of price increases
at any given level of output. This was represented by the exogenous shock s in the aggregate supply equation:
(14.9)
When the shock is unfavourable, i.e. when s > 0, the short-run aggregate supply curve shifts upwards from AS to AS′, as shown in Figure 14.14. The move from
point A to point B is a case of stagflation, i.e. declining real growth and rising inflation. If the relative price increase is a one-off event, the AS curve will shift back
to its initial position.19 This is optimistic, however. While the economy is at point B, workers unexpectedly face higher prices. Quite likely, they will demand
higher nominal wages and, if they succeed, the backward-looking component of underlying inflation rises. Such second-round effects are the reason why, even
after the commodity price increase has been absorbed (when s goes back to zero), the AS curve is unlikely to shift back quickly or completely. The answer will
depend on the behaviour of underlying inflation.
Fig. 14.14 An Adverse Supply Shock

An adverse supply shock shifts the AS curve up to AS′. The economy will suffer stagflation as it moves from point A to point B. If the authorities decide to avoid a fall in output
and a rise in unemployment, they can adopt expansionary demand-side policies and move the economy towards a long-run equilibrium at point C. This occurs at the cost of a
permanent increase in the central bank’s target rate of inflation. If, in contrast, they choose to fight inflation, they can adopt contractionary demand-side policies—a lower target
inflation rate—and aim at point D. Here the cost is a deep recession.
Stagflation is a serious policy challenge for governments. One approach is to soften the blow to output and unemployment by adopting an expansionary
demand policy (monetary or fiscal, depending on the exchange rate regime). Aiming at point C, and shifting the aggregate demand curve to AD′ in Figure 14.14,
hastens the return to trend growth but at the cost of higher inflation. Another approach is to prevent inflation from ever rising, so that underlying inflation
remains under control. This calls for a prompt contractionary policy reaction, shifting the short-run aggregate demand curve back until it goes through a point
like D. This reaction deepens the recession but, once the shock has worked itself through (and s = 0), the aggregate supply curve moves back to AS and the
restrictive demand policy may be lifted to return to point A. The nature of the dilemma should be clear: the authorities can either aim at maintaining output and
employment, but at the cost of higher inflation, or they can prevent a sharp inflationary impact, but at the cost of a low output and high unemployment. The
reason behind this dilemma is also clear: macroeconomic management policies are demand-side policies and they are ill-adapted to deal with supply shocks.
The exchange rate regime
The reaction of underlying inflation is decisive for determining the outcome of the policy response to a supply shock. Underlying inflation tends to increase
because of its backward-looking component, but what about the forward-looking component? The answer ultimately hinges on which long-run equilibrium is
expected to be reached. If agents believe that policy-makers are aiming at point D in Figure 14.14, the forward-looking component is likely to support this policy.
If wage negotiators are convinced that inflation will be kept under control, they see the jump to point B as strictly temporary and keep underlying inflation at the
pre-shock level. Once the shock is over, the aggregate supply curve promptly returns to AS and the economy’s trajectory will be from A to B and back to A. If,
instead, wage negotiators expect an accommodating policy that aims at point C, underlying inflation will rise and shift the AS curve to AS′, even after the shock
has passed. The trajectory will be from A to B and beyond, higher and to the left of B along the new AD′ curve. However, since the output gap is negative,
underlying inflation is above actual inflation, so the AS curve will eventually shift back towards AS′, even though the one-off supply shock is over. The
economy winds up at point C.
Under flexible exchange rates, it is the central bank that determines the position of the LAD line. By their choice of an inflation target, the monetary authority
can choose the long-run inflation rate and decide whether point A or point C will be eventually reached. This is not the case with a fixed exchange rate regime
where the position of the LAD line depends on the ‘foreign’ inflation rate. In the presence of a severe supply shock, a fixed exchange rate regime can be
maintained only among like-minded countries which have compatible views of how they will react. In Europe, for instance, the oil shocks of the 1970s and 1980s
seriously strained the European Monetary System as different countries adopted different strategies. The adoption of a common currency, which floats freely,
means that this decision is now in the hands of the European Central Bank (ECB). Even with a common currency, policy disagreements concerning the correct
response to inflation remain. Following the rise of oil prices in 2003–2004, and again in 2007–2008, the ECB has been criticized by some governments as being too
tight and by others as being too lax. Since the financial crisis, inflation has been subdued and other concerns have become dominant, but the issues regarding
the management of expectations and credibility will always be waiting in the wings for the next supply shock to arrive.
Lessons from supply shocks
Three general lessons can be drawn. First, an adverse supply shock is bad news. It depresses growth while raising unemployment and inflation at the same time,
contradicting the Phillips curve trade-off. Second, traditional demand management policies are not useful for dealing with an adverse supply shock. When the
aggregate supply curve moves up and to the left, demand management cannot deal with both inflation and output. Policy-makers must choose between accepting
the shock as an increase in inflation or as a drop in output with higher unemployment. The appropriate response should be supply-side policies, aiming at
bringing back the aggregate supply curve as soon as possible to its initial position. This is not easy. The best hope is to manage the forward-looking component
of underlying inflation and to try to ‘disconnect’ the backward-looking component. This requires a clear and credible signal from the authorities that they will not
accommodate the shock. Third, the exchange rate regime becomes crucial. A fixed exchange rate can be maintained only among countries that adopt the same
policy mix.
Fig. 14.15 An Adverse Demand Disturbance
An adverse demand disturbance is represented by a leftward exogenous shift of the short-run aggregate demand curve. The economy moves from point A to point B. In
principle, the government has instruments at its disposal—monetary or fiscal policy, or both—which could restore the AD curve to its original position.
14.4.3 Demand Disturbances

In principle, exogenous demand shifts are easier for policy-makers to contend with. Some of these disturbances are the direct result of macroeconomic policy
actions, fiscal or monetary policy depending on the exchange rate regime. Other examples are exogenous events, like the global financial crisis which began with
the bankruptcy of the US investment bank Lehman Brothers, and was followed by a freezing-up of credit markets and a worldwide demand slowdown. Another,
earlier example of a positive demand disturbances was German reunification, the source of an unexpected demand surge in central Europe during the first half of
the 1990s.
Figure 14.15 shows the consequence of an adverse demand disturbance as a leftward exogenous shift of the short-run aggregate demand curve from AD to
AD′. The economy moves from point A to point B: inflation declines and output falls below its trend level. In principle, the government has the required
instruments at its disposal—monetary or fiscal policy, depending on the exchange rate regime—that could restore the AD curve to its original position.
When the global economic crisis broke up in 2008, it was well understood that demand-side policies were needed to cushion the blow and prevent a remake of
the Great Depression of the 1930s. Many central banks promptly cut their interest rates, bringing them very close to zero, then hit the lower bound, as can be
seen in Figure 14.16. After a decade dedicated largely to bringing down inflation in Europe, the ECB displayed some reluctance to give up those achievements
again. Yet, broadly similar monetary policies limited the risk of potentially disruptive large exchange rate movements among freely floating currencies. In the
IS-TR-IFM framework, the downward movements of the TR and IMF schedules, representing respectively domestic and foreign monetary policy actions, were
broadly similar.
Fig. 14.16 Monetary Policies 2007–2015

Central banks promptly cut interest rates when the financial crisis gathered steam starting in 2007 and culminating with the collapse of the investment bank Lehman Brothers in
September 2008. While the US Federal Reserve and the Bank of England brought their interest rates to almost zero, and kept there for many years, the ECB reacted less
forcefully, only to go all the way to zero by late 2014.
Sources :Federal Reserve Board, Bank of England, ECB.
Fiscal policy, which does not have long-run inflation effects as noted earlier, was also used, but fears of crowding out through the current account among
countries with flexible exchange rate regimes led a G20 Summit to call upon all countries to adopt expansionary fiscal policies. Furthermore, joint expansion could
reduce demand leakages through trade, because increasing imports by all countries imply increasing exports overall—and thus increase the fiscal multiplier. Two
years later, large increases in public debt in the US, the UK, Germany, Japan, and many other countries brought this policy stance to an end. In the Eurozone,
public debts became so large in some countries (Greece, Ireland, Portugal) that the financial crisis mutated into a public debt crisis. The end of coordinated
monetary and fiscal policies expansion highlighted the differences between fixed and flexible exchange rate regimes. Box 14.3 compares the situation of Ireland, a
member of the Eurozone, with Iceland, which had its own currency and floating exchange rates during the crisis.
14.4.4 Disinflation
How to deal with a high rate of inflation already long in place? We know that high and persistent inflation is the consequence of excessive monetary growth that
the central bank has chosen or been forced to choose. The cure must be to implement a lower target inflation rate, raising the interest rate and, therefore, slowing
down money growth.20 How a policy of disinflation—a successful reduction in the rate of inflation—is ultimately implemented depends on the exchange rate
regime.
Under flexible exchange rates, the central bank can choose its inflation rate target, so the solution to high inflation is technically simple, but often very painful.
Figure 14.17 shows why. We start from point A, which we take to be a long-run equilibrium. Thus we assume that the currently high inflation rate is indeed the
central bank’s target—we are on some original LAD line which is not drawn—and that actual and underlying inflation are equal—we are on the LAS line. If the
central bank exogenously reduces the target inflation rate well below the current rate of inflation, LAD shifts downward to LAD′. In the TR–IFM model (not
shown), the TR curve shifts up and to the left, the nominal and real exchange rates appreciate, the current account worsens, and demand declines. This is
captured by the leftward shift of the short-run aggregate demand curve from AD to AD′. The short-run effect of this disinflationary policy corresponds to point
B: inflation declines but so does output, and unemployment rises. At point B actual inflation is below underlying inflation so the latter will be revised downwards
and, over time, the short-run aggregate supply curve will shift downwards until it reaches the position AS′. At point C, a new long-run equilibrium is reached and
the disinflation is complete. The cost has been a period of negative output gap and high unemployment, which may extend over several years as noted in Section
14.4.1.
Under a fixed and adjustable exchange rate regime, high and lasting inflation is only possible if the exchange rate is regularly depreciated, as explained in
Section 14.2. Bringing inflation down requires a change in monetary policy. If the fixed exchange rate regime is to be retained, this means doing away with chronic
depreciations. In that case, we know that the LAD line is set by the foreign inflation rate, so it is essential to peg the exchange rate to the currency of a country
where inflation is suitably low. The peg becomes the anchor that will deliver disinflation. Initially, inflation is higher at home than abroad. If inflation moves
slowly, fixing the nominal exchange rate means that the real exchange rate appreciates as long as domestic exceeds foreign inflation. In the IS–IMF model (not
shown), the IS curve shifts to the left. The resulting decline in aggregate demand is represented by the leftward shift of the short-run aggregate demand curve
from AD to AD′ in Figure 14.17. Then the logic is the same as under the flexible exchange rate regime. Underlying inflation must decline in light of lower realized
inflation rates and the economy will eventually settle at point C on the long-run aggregate demand LAD′, which corresponds to the lower foreign inflation rate.
Here disinflation also requires a period of negative output gap and high unemployment.
Box 14.3 Ireland v. Iceland: Vulnerable Islands in a Global Financial Tsunami
Iceland and Ireland are small, wealthy, and very open economies which experienced strong real growth in the 1990s and pursued an aggressive course
of international financial integration in the late 1990s and early 2000s. Previously backwater banks in the 1980s rose to become star international players
in the early 2000s. In the ten years from 1995 to 2005, demand deposits in Iceland—a country of 320,000 people—grew from $6.8 billion to $65.8 billion!
In Ireland growth was similar, from $33.1 billion to $163.5 billion. More frequently than not, these bank deposits belonged to foreign households, foreign
firms, or foreign financial institutions. Icelandic banks went as far as to open subsidiaries in the UK and in the Netherlands, offering higher interest rates
than the local competition; by all appearances, these banks were running Ponzi schemes (see Box 6.6). The Icelandic banks eventually collapsed and
defaulted on their foreign depositors. In the case of Ireland, the banks financed a housing boom similar to America’s. When house prices started to
decline, many borrowers defaulted on their banks because the loans now exceeded the value of their houses. The banks were distressed and some had
to be taken over by the state.
The financial crisis was accompanied by a sharp decline in credit and plummeting stock prices (Tobin’s q falling), which led to collapses of investment
spending and aggregate demand in both countries. At the same time, the IFM curve shot up because the rest of the world sharply raised the interest rate i*
at which it would be willing to lend to these countries (Iceland was even shut off from foreign borrowing, hitting the vertical part of the risk-return curve in
Figure 7.3). Just as the IS–TR–IFM model predicts, the demand shock hit GDP hard. In the AS–AD diagram, the AD curve shifted sharply to the left. Table
14.2 shows that both countries experienced similar cumulative declines in real GDP (in Iceland, about 10.4% in the period 2009–10; in Ireland, 12.1% in
the period 2008–10). Unemployment skyrocketed and government budget deficits swelled to double-digit percentages of GDP as tax revenues fell in line
with incomes.
Despite the remarkable similarity in the nature of the demand shock hitting the two economies, the reaction could not have been more different,
highlighting fundamental differences between fixed and flexible exchange rate regimes. Ireland, a member of the Eurozone, had no monetary policy
option. and had little choice but to bite the bullet. Iceland, on the other hand, let its currency depreciate—from 2007 to 2009 the dollar value of the krona
dropped by 50%. Predictably, with such a boost to competitiveness, the recession was over more quickly there. Its current account, while swinging more
wildly than in Ireland, contributed decisively to the Icelandic recovery. In order to benefit from foreign demand, Ireland had to boost its competitiveness,
which called for deflation (falling prices). Given the slope of the AS curve, this required a deep recession, lasting long enough to durably bring the
underlying inflation rate down.
Both governments had to recapitalize their failed banks. Iceland chose to expropriate foreign depositors, thus considerably reducing the costs.
Additionally, the central bank could lend money to the government. As a Eurozone member, Ireland could not treat foreigners differently from domestic
stakeholders. The government was even strongly encouraged to fully protect all depositors and lenders. The costs were huge: in a few months, the
government’s debt rose by more than 30% of GDP, reflecting a record budget deficit that was not even expansionary as the money—not provided by the
central bank—was used to avoid losses by domestic and foreign bank depositors and bondholders. By the end of 2010, Ireland was in a full-blown debt
crisis.
In terms of Figure 14.15, both countries initially went from A to B. Expansionary monetary policy brought the Icelandic AD curve back up, while fighting the
debt crisis through subsequent fiscal contraction pushed the Irish AD curve further to the left. In addition, in Iceland, the huge devaluation acted as a
supply shock as foreign goods became more expensive. Inflation rose sharply in Iceland, and became negative in Ireland, as can be seen in Table 14.2.
The sharp reduction in Irish inflation was central to restoring Ireland’s competitiveness evident in the most recent years.
The interesting questions are: how long does it take to move from point B to point C in Figure 14.17? And how much output is lost along the way? The output
cost of disinflation is lower the faster the AS curve comes down. That, in turn, depends on the speed at which underlying inflation adapts to a declining
inflation rate. The backward component slows down the speed at which the AS curve shifts, while the forward-looking component accelerates the adjustment. In
periods of disinflation, therefore, it would be helpful to give more weight to the forward-looking component, possibly even shutting down the backward-looking
component. The backward-looking component depends on wage- and price-setting institutions, an issue examined in Box 14.4. The forward-looking component is
often referred to as the ‘psychological’ nature of price- and wage-setting, but it can be influenced by policy institutions.
Table 14.2 Iceland and Ireland: Key Economic Indicators, 2004–2015
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Real growth (%) Iceland 8.2 6.0 4.2 9.5 1.5 –4.7 –3.6 2.0 1.2 4.4 2.0 4.0
Ireland 4.4 6.3 6.3 5.5 –2.2 –5.6 0.4 2.6 0.2 1.4 5.2 7.8
Unemployment rate (%) Iceland 3.1 2.6 2.9 2.3 3.0 7.2 7.6 7.1 6.0 5.4 5.0 4.0
Ireland 4.5 4.4 4.5 4.7 6.4 12.0 13.9 14.6 14.7 13.0 11.3 9.4
CPI Inflation (%) Iceland 3.2 4.0 6.7 5.1 12.7 12.0 5.4 4.0 5.2 3.9 2.0 1.6
Ireland 2.3 2.2 2.7 2.9 3.1 –1.7 –1.6 1.2 1.9 0.5 0.3 0.0
Exchange rate (USD, 2007 = 100) Iceland 91.3 101.9 91.7 100.0 72.8 51.8 52.4 55.2 51.2 52.4 54.9 48.9
Ireland 90.7 90.7 91.6 100.0 106.7 101.4 96.7 101.5 93.8 96.9 96.8 81.5
Investment rate (%of GDP) Iceland 24.6 29.2 36.0 29.8 26.0 14.9 13.9 15.6 16.1 15.5 17.4 19.2
Ireland 27.2 30.4 31.9 29.2 24.6 20.3 17.5 17.7 19.3 18.1 20.3 23.3
Current account (%of GDP) Iceland –9.8 –15.8 –23.3 –14.0 –22.8 –9.7 –6.6 –5.3 –4.2 5.7 3.7 4.2
Ireland –0.6 –3.4 –3.5 –5.4 –5.7 –3.0 0.6 0.8 –1.5 3.1 3.6 4.5
Source: IMF World Economic Outlook, OECD.
Fig. 14.17 Disinflation

Disinflation moves the economy from point A to point C. Using demand-side policies implies the use of contractionary monetary or fiscal policies which move the aggregate
demand curve from AD to AD′ and the LAD line to LAD′. The short-run equilibrium at point B explains why disinflation is usually painful: it requires a period of low output and high
unemployment. Long-run equilibrium is achieved at point C when the short-run aggregate supply curve has shifted to AS′. The speed of this shift depends on the time required by
underlying inflation to catch up with lower actual inflation.
Box 14.4 Wage Negotiations: The Time Dimension
In most European countries and in the USA, wage negotiations are staggered over a year or more. One wage negotiation takes the previous one into
account, and may even anticipate the next one. Employees do not want to be outdone by their colleagues, and employers do not want the competition to
undercut their labour costs. In contrast, in Japan wage negotiations are synchronized. They take place every year at roughly the same time, the so-called
‘spring offensive’ (shunto). Each industry opens up bargaining, but closely monitors the state of play elsewhere. When one bargain is struck, it sets the
trend and all the others follow quickly. For a time, wage negotiations in Northern Europe were centralized and therefore highly synchronized. Even when
they are staggered, some negotiations are trend-setting: they result in similar agreements later on and sometimes even trigger readjustments to
previously reached ones, thus injecting a dose of synchronization. With wage staggering, aggregate nominal wages (the average of all nominal wages)
move slowly, which delays the return to equilibrium unemployment. The AS curve will move relatively slowly. With full synchronization, average nominal
wages are stable between negotiations, and then jump. This implies a more rapidly-moving AS curve. The implications for the economy are profound.
Either a quick return to the equilibrium unemployment rate if the real wages are set right, or a prolonged departure if they are set incorrectly.
The situation is different in economies where inflation is high and has been so for a long time. There it is common to have mandatory or mutually
agreed indexation schemes for wages. Brazil was particularly advanced in this regard, indexing virtually all nominal prices, including house rents,
corporate balance sheets, taxes, and public utilities rates. Such indexation schemes can often reduce the staggering considerably, with the same effect
as an increase in synchronization of wage-setting.
Although wage indexation removes some costs of high inflation to households and firms, it has serious adverse side-effects. First, indexation generally
perpetuates any real wage gain achieved. This gives an incentive to any group of wage-earners to be the first to demand higher wages. The result is that
all groups rush to be first, as much to protect themselves as to achieve a head start. Second, indexation reduces both public and government support for
anti-inflation policies. This is why Germany, after its famous hyperinflation in the 1920s, made indexation illegal. Third, indexation makes disinflation
costlier in terms of unemployment. When inflation is on the way up, nominal wages trail behind prices: real wages are reduced and labour demand is
robust. When inflation is on the way down, wages indexed on past inflation trail actual inflation: real wages rise, firms’ profits are squeezed, and
unemployment rises. This is why most European countries with legal or simply widespread indexation clauses eliminated them in the 1980s, much
against the will of trade unions. Fourth, indexation eliminates downward real wage flexibility as real wages are at least constant unless there is a sharp
burst of inflation. The lack of flexibility can cause unemployment when an adverse supply shock occurs.
Wage negotiators may have different incentives when formulating their expectations. It is good bargaining tactics for workers to argue that inflation is and will
remain high, while employers tend to predict declines in the rate of inflation. Jointly, however, employers and employees have an incentive to be as close as
possible to target, for errors may be costly in terms of competitiveness and profitability. As they aim at disinflation, policy-makers have a strong interest in
convincing wage negotiators that inflation will surely decline, since this will accelerate the downward movement of the AS curve. One solution is to credibly use
the exchange rate as an anchor. To do so, the authorities must demonstrate that they will not let the exchange rate depreciate again. This is why a number of
countries have adopted hard pegs, a variety of fixed exchange rate arrangements that makes it politically costly or even illegal to devalue.21 If the exchange rate
is not fixed, it is the credibility of the central bank as an inflation-fighter that becomes crucial. This is why a number of countries have given formal independence
to their central banks, instructing them to aim at price stability. Many independent central banks have adopted the inflation-targeting strategy as already
described, publicly announcing the inflation rate they intend to achieve and explicitly and publicly linking their actions to the target, and staking their credibility
on their success in achieving the target.
Summary
1 The macroeconomy is analysed as the interplay of aggregate demand and aggregate supply. This framework emphasizes the distinction between the short run and the long
run, when output returns to its trend growth path.
2 Under fixed exchange rates, inflation is restricted to be equal to foreign inflation in the long run. Under flexible rates, long-run inflation is determined by the target inflation
rate.
3 The short-run aggregate demand curve is downward-sloping. Under fixed exchange rates, an increase in inflation above the foreign rate erodes external competitiveness and
reduces demand for domestic goods. Under flexible exchange rates, an increase in the inflation rate relative to the central bank’s inflation target prompts an increase in the
interest rate. This in turn results in a nominal and real exchange appreciation with a contractionary effect on aggregate demand.
4 Only in the flexible rate regime can the monetary authority determine the long-run inflation rate. Under fixed exchange rates, some monetary independence is possible, but
only by repeated devaluations or revaluations.
5 Under fixed exchange rates, fiscal policy can affect aggregate demand and output. The effects of a fiscal policy action are temporary, however. The change in spending or cut
in taxes which leads to the shift in the AD curve cannot be sustained indefinitely. In the long run, the government’s budget constraint prevents a permanently expansionary
fiscal policy.
6 A fiscal expansion initially raises the output level at the cost of a higher rate of inflation. Over time, as underlying inflation rises and the unavoidable retrenchment of fiscal
policy occurs, demand returns to trend output.
7 Under fixed exchange rates monetary policy is ineffective. This is also the case for fiscal policy under flexible exchange rates.
8 Under flexible exchange rates, a monetary policy expansion—the consequence of an increase in the central bank’s target inflation—initially raises output and inflation. Over
time, inflation continues to increase, ultimately sowing the seeds of the next recession, as the central bank is forced to raise interest rates and return output back to its trend
growth path.
9 An adverse supply shock simultaneously lowers output and raises inflation. Demand management policies are ill-equipped to deal with a supply shock. They may cushion
the fall in income at the cost of more inflation, or reduce the inflationary impact at the cost of a deeper fall in output and more unemployment.
10 Disinflation requires a permanent reduction in the target inflation rate when the exchange rate is floating or sticking to the peg under a fixed exchange rate regime. It can be
costly in terms of lost output and above-equilibrium unemployment.
11 The faster underlying inflation adjusts, the lower the costs of disinflation. This calls for adopting credible institutions that can convince wage negotiators that the disinflation
policy is ‘serious’.
Key Concepts
AS–AD model
purchasing power parity (PPP)
aggregate demand curve
LAD line
short-run versus long-run equilibrium
demand disturbance
stagflation
nominal interest rate, real interest rate
Fisher principle/equation
target/neutral interest rate
indexation
disinflation
output cost of disinflation
hard pegs
inflation-targeting strategy
Exercises
1 Use the AD–AS model to trace the short-run effect under a fixed exchange rate regime of: (1) a one-off increase in taxes; (2) a one-off decrease in government spending; (3) a
one-off decrease in animal spirits (Tobin’s q). What are the long-run effects?
2 Use the AS–AD model to study the short-, medium-, and long-run effect of a permanent decrease in the inflation target of the central bank. Assume that underlying inflation
in the current period is simply equal to inflation which is observed in the previous period. Contrast your answer with the alternative assumption that underlying inflation
overshoots inflation in the previous period.
3 Consider the following numerical version of the AS–AD model:
Assume the real interest rate is equal to 2, the target rate of inflation is 3, and s = 0. Use a spreadsheet or other computer program to characterize the steady
state as a sequence of values of inflation, output, and interest rates which are stable. Then calculate the dynamic evolution of the economy in response to
(a) a permanent increase in the target inflation rate to 5.
(b) a permanent decrease in the target inflation rate to 2.
(c) a supply shock of s = 5 for one period.
(d) a supply shock of s = 5 for four periods.
4 Consider the AD–AS model where the economy is not in long-run equilibrium, in particular assume there is a negative output gap (that is, the economy is in a recession).
Describe the adjustment under fixed exchange rates if there is no government intervention. Contrast your answer with that under flexible exchange rates.
5 Now consider an economy in a fixed exchange rate regime in which the output gap is positive—a booming economy—and domestic inflation exceeds foreign inflation.
Describe the adjustment if there is no intervention.
6 If supply shocks predominate, what can you predict about the direction of co-movement between inflation and the output gap? If demand shocks are more important? Under
which conditions would you expect to observe a Phillips curve?
7 Use the AD–AS and IS–TR–IFM frameworks to study the short- and long-run effects of an increase in foreign inflation under a fixed exchange rate regime.
8 Under fixed exchange rates, use the IS–TR–IFM and AD–AS models to analyse the effects of a combined tight fiscal policy and expansionary monetary policy.
9 Under flexible exchange rates, use the IS–TR–IFM and AD–AS models to analyse the effects of an expansionary fiscal policy and tight monetary policy.
10 Trace the effects of a favourable supply shock, such as a sudden decline in oil prices using the AS–AD framework. Suppose the government wants to take advantage of this
event to bring about a permanent decrease in inflation. Under which exchange rate regime is this possible, and how could the government achieve its goal?
11 A government wants to use monetary policy under a flexible exchange rate regime to keep actual GDP above its trend growth rate for ever. In the AS–AD diagram, show
graphically the consequences of such a policy.
12 Assume that underlying inflation is entirely forward-looking and that expectations are forward-looking and equivalent to perfect foresight. What are the effects of fiscal
policy (under fixed exchange rates) and monetary policy (under flexible exchange rates) on output and inflation? Consider both cases of expansionary or restrictive policies.
13 Central banks that adopt the inflation-targeting strategy usually publish forecasts of future inflation, thus implicitly or explicitly signalling what they plan to do in the
future. It is argued that being able to convince the public of their intentions greatly enhances the effectiveness of central bank actions. Use the AD–AS framework to explain
why this might be the case.
14 Show under what condition the Taylor rule (14.7) implies that an increase in inflation leads to a higher ex post real interest rate, defined as i − π. Is this a desirable feature of
the rule? Why or why not?
Essay Questions
1 ‘Expansionary demand policies are based on fooling people.’ Comment.

2 Why does adding the supply side partly undermine the usefulness of demand management policies?
3 Why are supply-side policies more appealing than demand-side policies. What kind of policies can you imagine which might stimulate the supply side of the economy?
4 ‘Sluggish expectations are helpful when inflation is rising but troublesome when inflation is declining.’ Evaluate this assertion and name possible policy implications.
5 Suppose you are in charge of monetary policy in a developing country and the price of primary commodities (food, oil) rises sharply. What will be the consequence of
pursuing a flexible exchange rate regime with a fixed target rate of inflation? Are there disadvantages to following a fixed exchange rate regime?
1 Harry G. Johnson (1923–1979), a Canadian, was a professor of international trade and monetary economics at the University of Chicago and was known both for his dry wit and a
wet whistle. Among his most important contributions to macroeconomics is the monetary approach to the balance of payments, which lies behind our understanding of the workings
of monetary policy under fixed exchange rates.
2 The book has been concerned until now with extremes: Chapter 5 dealt with the flexible price case, while Chapters 11 and 12 assumed constant prices. The real world must be
somewhere in between, with price levels and inflation moving but much more slowly than the lightning speed of the neoclassical model.
3 This issue was discussed in Section 12.5.4.
4 As stressed in Chapter 5, constancy of the real exchange rate implied by relative PPP should not be confused with absolute PPP or the law of one price. For countries of similar
development, such as Germany, the UK, and France, relative PPP is fairly reliable. It tends to break down when comparing countries with significantly different levels of GDP per
capita, especially when one is growing much faster than the other over a longer period. These issues are discussed in more detail in Chapter 15.
5 Within the euro area, exchange rates are fixed in the most extreme way possible: a single currency. In part, the origins of the crisis can be found in persistently different inflation
rates, e.g. higher in Greece and Spain than in Germany. In the long run, this is not sustainable.
6 Point A′ is just a snapshot taken during a contractionary process that continues as long as inflation is higher at home than abroad. As long as domestic inflation exceeds the foreign
rate, the real exchange rate will continue to appreciate. To keep things simple, we do not elaborate on this aspect.
7 A contractionary fiscal shock would be a tax increase or a reduction in government spending on goods and services.
8 In fact, the public debt will have risen in the meantime and must be paid for by a permanently higher primary budget surplus. This requires that the AD curve shifts back beyond its
original position. We overlook this additional complexity. It is acceptable to do so if the fiscal expansion does not last long enough to increase the debt–GDP ratio. In the course of
the global financial crisis, this detail appears to have been neglected. We return to this important issue in Chapter 17.
9 In Chapter 13 we saw that the instability of inflation was the downfall of the old-fashioned Phillips curve—or its mirror image, the AS curve. By understanding the dynamics of
underlying inflation, it is possible to understand why the Phillips curve shifted so much.
10 In fact, the economy will need to move temporarily below point A in Panel (b), because a period when inflation is lower at home than abroad is required to bring the real exchange
rate back to its original initial level. Only if inflation is below π* for some time is it possible for a real depreciation to occur, competitiveness to be restored and the IS curve to return
to its initial position.
11 The careful reader will ask: what gets us back to point A in Panel (b)? Here we would need to use the observation in footnote 8: during all the time when π > π*, the curve AD
becomes increasingly flat, year after year. In the end it will coincide with LAD. In the intervening periods, it will be necessary for π < π* for a time, so the real exchange rate can
depreciate to its original level.
12 Named after Irving Fisher, the Yale economist, already referred to in several earlier chapters.
13 If the central bank chose to control the money supply instead of following a Taylor rule, the same conclusion would result. The inflation rate in the long run would be given by the
chosen rate of money growth minus an adjustment for trend growth in the demand for real money balances. Indeed Box 14.1 shows that, in the long run, the inflation rate is the
difference between the money growth rate and the GDP growth rate.
14 At this stage, you may need to return to Section 12.5 of Chapter 12 to remind yourself why the exchange rate appreciates.
15 This is a strong statement, which may be modified to acknowledge that the deterioration of net exports implied by the exchange rate changes (sometimes called ‘external crowding
out’) can often take several months to occur.
16 The dichotomy principle implies that the real interest rate is determined exclusively by the real economy. Changes in the level of the real interest rate, while they may affect the
Taylor rule, are not considered in the analysis that follows.
17 Short-run movements in the exchange rates can be inferred in a straightforward way from the TR–IFM apparatus.
18 We ignore an issue of further complexity. The Fisher principle implies that, eventually, the nominal interest rate must rise, which would shift the IFM curve up. Ignoring this aspect
does not affect the general argument.
19 A supply shock, such as a one-off increase in oil prices, directly affects the price level, not its rate of increase, unless these prices continue to rise. Normally, once they have
reached a new higher level, the impact is passed on into higher goods prices. While the level of these prices remains higher, inflation is no longer directly affected by the shock.
20 By the Taylor rule, a decrease in –π, holding all other things constant, implies an increase in the nominal interest rate.
21 This is the strategy adopted, for example, by Argentina in 1991 and by Bulgaria in 1997. In both cases, it worked, although Argentina’s arrangement collapsed in 2001, for other
reasons (chiefly, large budget deficits in provinces). By then, however, inflation had turned negative!
The Exchange Rate
15
15.1 Overview
15.2.1 Main Characteristics
15.2.2 Instruments
15.2.3 Triangular Arbitrage
15.3.1 Covered Interest Rate Parity
15.3.2 Uncovered Interest Rate Parity
15.3.3 Risk Premia
15.3.4 Real Interest Rate Arbitrage in the Long Run
15.4.1 The Exchange Rate as an Asset Price
15.4.2 Implications of the Interest Rate Parity Condition
15.4.3 An Apparent Contradiction and its Resolution
15.4.4 The Fundamental Determinants of the Nominal Exchange Rate
15.5.1 The Long Run and the Primary Current Account: A Review
15.5.2 Equilibrium Real Exchange Rate and Primary Current Account in the Long Run
15.5.3 The Fundamental Determinants of the Real Exchange Rate
15.5.4 How to Think About the Equilibrium Real Exchange Rate?
15.6.1 The Equilibrium Real Exchange Rate as an Anchor
15.6.2 The Bridge to the Long Run: the Parity Conditions
15.7.1 Volatility and Predictability
15.7.2 Currency Crises
Summary
Suppose a man climbs five feet up a sea wall, then climbs down 12 feet. Whether he drowns or not depends upon how high above sea-level he was when he started. The same
problem arises in deciding whether currencies are under- or over-valued.
The Economist, 26 August 1995
15.1 Overview
In one sense, the foreign exchange rate market is just another financial market, a platform for trading one currency against another, as was already discussed
briefly in Chapter 7. Like other assets, currencies are durable and their value today depends on the values expected to prevail tomorrow. But this market deserves
special attention for several reasons. First, the price of foreign currencies, the exchange rate, matters not just for financial transactions, but also for trade; in the
short run, it determines a country’s external competitiveness (as we saw in Chapter 11). Second, monetary authorities often fix, or at least stabilize, the exchange
rate, and this influences the effectiveness of fiscal and monetary policy (as explained in Chapter 12). To achieve that end, they intervene on the foreign exchange
market. Third, the dynamics of the exchange rate affect the interest rate, a channel that had been ignored thus far. It is now time to ask more carefully how the
exchange rate, both nominal and real, is determined when it is endogenous and not fixed by the monetary authorities.
We start in Section 15.2 with a description of foreign exchange markets. Present virtually all over the world in the form of computer terminals, they are
dominated by one currency, the US dollar, with a growing role for the euro and a handful of lesser vehicle currencies. The amounts traded are staggeringly large.
We then think again about the link between interest rates at home and abroad. The link was presented briefly in Chapter 12 but plays a crucial role in the entire
textbook. It deserves more attention. In Section 15.3, we show how to make it more precise. The results are two interest rate parity conditions that directly
involve the exchange rate—current and anticipated. The analysis presented in Chapter 12 will be modified to take account of this more precise view of
international interest rate parity.
Interest rate parity conditions can explain what drives the nominal exchange rate—a volatile asset price subject to both short-run fluctuations and long-term
swings. As in previous chapters, we start by separating the short and the long run. The short-run analysis, presented in Section 15.4, is similar to that of Chapter
7, treating the exchange rate as an asset price. As such, it is driven solely by financial considerations and market expectations. In stark contrast, the long-run
analysis developed in Section 15.5 focuses on the real side of the economy in the form of the national intertemporal budget constraint. The exchange rate is part
of a long-run mechanism enabling national economies to repay their long-run obligations, if they have any, or encouraging more spending, should they have
positive net international positions. This leads us to the concept of an equilibrium real exchange rate.
In summary, the long-run analysis explains the real exchange rate while the short-run analysis explains the nominal exchange rate. All that remains is to bridge
the two outcomes—to show how the short run is consistent with the long run, which it must be. The result, presented in Section 15.6, is an internally consistent
treatment of the exchange rate. The last section ties up some important loose ends involving the implications of principles developed in this chapter, including
the regular occurrence of currency crises.
15.2.1 Main Characteristics

The remarkable thing about the foreign exchange market is that it does not exist in a physical sense, and hasn’t for decades. Instead, hundreds and thousands of
traders sit in front of computer terminals around the world and round the clock and swap currencies with each other. Furthermore, they do not trade in banknotes,
but rather in the liabilities of commercial banks—bank deposits—and of other financial institutions. As financial intermediaries, they act on behalf of their
customers, who need currencies for trade or financial transactions. Most often, they do it for their colleagues in the same financial institution, who buy and sell
assets denominated in various currencies, or for their own trading accounts when they spot a fleeting deviation from the no-profit condition, just as explained in
Chapter 7.
A characteristic of the exchange markets is that they are dominated by a few currencies, called vehicle currencies. Until recently, the US dollar was the
vehicle currency, present in one side of nearly half of all transactions. Table 15.1 shows that its importance has diminished somewhat. The euro is the second
most important international currency. Other currencies play minor roles, although the growing importance of emerging market economies is reflected in the
shares of their currencies.
Like other asset markets, the foreign exchange market is designed for wholesale trades involving standardized transactions of large size. This feature reinforces
the domination of the vehicle currency and means that most exchanges operate via the US dollar, as Table 15.2 shows. This means that, for instance, a trader who
wants to change Polish zlotys into Indian rupees will first change the zlotys into dollars and then the dollars into rupees. This gives rise to triangular arbitrage, as
explained later in Section 15.2.3.
Table 15.1 Shares of Currencies in Foreign Exchange Market Transactions
2001 2004 2007 2010 2013

US dollar 44.9 44.0 42.8 42.4 43.5
Euro 19.0 18.7 18.5 19.5 16.7
Yen 11.8 10.4 8.6 9.6 11.5
Sterling 6.5 8.2 7.4 6.4 5.9
Australian dollar 2.2 3.0 3.3 3.8 4.3
Swiss franc 3.0 3.0 3.4 3.2 2.6
Chinese renminbi 0.0 0.0 0.2 0.4 1.1
Emerging market currencies 1.9 2.2 3.2 4.2 6.7
Source: Bank for International Settlements (2015), www.bis.org/publ/rpfxf10t.htm.
The volume of foreign exchange traded each day is breathtaking. Table 15.3 shows that total transactions on an average day in 2010 amounted to almost 4
billion (million million = 1012) US dollars. This daily turnover represents about one-quarter of annual GDP of the USA and more than four times that of the
Netherlands. The growth of this market is even more dizzying: since 1992, turnover has increased by almost 400%—an annual compound growth rate of 7.8% per
annum, compared with world GDP, which grew in US dollars since 1992 by only 2.6% annually.
Table 15.2 Shares of Currency Pairs in Foreign Exchange Market Transactions
2004 2007 2010 2013

US dollar/euro 28 27 28 24
US dollar/yen 17 13 14 18
US dollar/sterling 13 12 9 9
US dollar/Australian dollar 6 6 6 7
US dollar/Swiss franc 4 5 4 4
US dollar/Canadian dollar 4 4 5 3
US dollar/Swedish krona 0 2 1 1
US dollar/other 13 15 11 21
Euro/yen 3 3 3 3
Euro/sterling 2 2 3 2
Euro/Swiss franc 2 2 2 1
Euro/other currencies 2 2 3 3
Other currency pairs 2 3 2 4
2004 2007 2010 2013
Source: Bank for International Settlements (2015), www.bis.org/publ/rpfxf10t.htm
15.2.2 Instruments
The basic deal involves immediately changing one currency for another one. The price at which this is done is called the spot exchange rate. An alternative is
to agree now to do the transaction at some specified date in the future. The corresponding price is the forward exchange rate. Current forward contract
maturities range from ‘tomorrow’ to one year. When two traders agree to a forward contract, they commit today to transact at a future date using an exchange rate
which may be different from the prevailing spot rate when the future rolls around. The no-profit condition implies that traders will want to avoid consistently
paying too much, and so forward and spot markets are intimately linked as shown in detail later. Another important type of deal is a swap, a formal agreement by
one trader with another to a simultaneous spot and a forward transaction, one long (a purchase) and one short (a sale). An example would be a sale today of €5
million for dollars and a purchase agreed today for €5 million in three months’ time. (The purpose of such transactions will be made clear shortly!). Swaps
represented more than half of all foreign exchange market transactions.
Table 15.3 Average Daily Foreign Exchange Transaction Volume (US$ billion)
1992 1995 1998 2001 2004 2007 2010 2013

820 1190 1490 1239 1934 3324 3981 5345
Source: Bank for International Settlements (2015), www.bis.org/publ/rpfxf10t.htm
15.2.3 Triangular Arbitrage

Whether we like it or not, the world of foreign exchange trading is on the dollar standard. Because the US dollar is most frequently used on one side of foreign
exchange transactions, direct exchange rates between most pairs of currencies are implicitly set by the dollar rate of each currency in the pair. The question is
whether these cross rates are mutually consistent. The no-profit condition presented in Chapter 7 implies that this is indeed the case—taking due account of the
bid–ask spread, of course. Recall the example illustrated in Figure 7.4. If the exchange rate of the Danish krone (DKK) vis-à-vis the euro is 8—it takes 8 krone to
buy 1 euro—and if one euro is worth $1, the krone–US dollar exchange rate must be (DKK 8/€)/($1/€) = 8 DKK/$.
This outcome is an example of triangular arbitrage presented in Chapter 7. Suppose the three exchange rates were not consistent with each other, e.g. that the
DKK/$ exchange rate is 7 instead of 8. This means that the krone is overvalued vis-à-vis the dollar, or undervalued vis-à-vis the euro. A trader could immediately
use, say, DKK7 million to acquire $1 million (ignoring the bid–ask margins, which would have to be subtracted as a ‘transaction cost’), then sell the dollars into
euros to acquire €1 million, which they would finally use to buy back DKK via the Euro/DKK rate (8 DKK/€) to obtain an instant profit of (8–7)/7 = 14.3%! Note
that this operation entails minimum risk, and just a few mouse clicks at the terminal. Indeed, the trader could execute these orders simultaneously. A trading desk
at a bank could even use borrowed funds—a so-called leveraged trade—so that the bank’s own capital isn’t even necessary. Understandably, it would be folly
to pass up such a ‘money pump’, so dozens of other traders will join the fray, forcing the exchange rates to move virtually instantly: massive purchases of the
dollar would lift its value, while massive sales of the euro would lead to a depreciation. Such pricing inconsistencies—even much smaller ones—would be self-
eliminating, because they offer such large profits.
In Chapter 12, we noted that full capital mobility implies that domestic and foreign returns must be aligned. This is yet another application of the no-profit
condition of Chapter 7. We called i the interest rate on domestic bonds and i* the ‘return’ on foreign bonds, stressing that it was evaluated in the domestic
currency, i.e. from the point of view of the home country. We will now look at what this means in more detail by quoting both interest rates in the currency of the
respective countries involved. This will lead to two versions of the condition which directly address the effects of an exchange rate change on the rate of return
abroad: (1) covered interest parity, an example of riskless yield arbitrage, and (2) uncovered interest parity, which explicitly accounts for risk.
15.3.1 Covered Interest Rate Parity

The covered interest rate parity (CIRP) condition is the result of arbitrage between domestic and foreign currency returns in the absence of risk-taking.
Consider the following example. An investor based in the Netherlands can obtain an annual rate of interest i on Dutch government bonds issued in euros and i*
on British Treasury bills issued in pounds .1 Both investments are of very low and comparable risk, and have a maturity of one year. Yet they are not exactly
equivalent because the value in euro of the sterling return may change over the investment period, thereby altering the total return measured in euro for the
sterling-denominated investment. Let the sterling price of one euro at the beginning of the year be St . (In August 2016, 1 euro cost about 0.85 UK pounds.) This
is the spot exchange rate. Selling one euro at the beginning of the year, a Dutch investor would obtain St pounds that she can invest in the UK bond to receive (1
+ i*)St at the end of the year. This is a known return in pounds sterling, since both the interest i* and exchange rate St are known at the beginning of the year.
When comparing the two investments—the Dutch bond versus the UK bond—the investor will naturally want to convert those pounds back into euros. The
hitch is that she faces foreign exchange rate risk: the exchange rate that will prevail at the end of the year is unknown at the beginning of the year. Financial
markets offer the possibility of eliminating the risk by signing a forward contract at the beginning of the year to sell the total value of her end-of-period
investment, (1 + i*)St pounds, for a certain euro exchange rate today. The corresponding forward exchange rate, denoted by Ft , is known and agreed upon at the
beginning of the year. Thus the investor can be certain that for every pound invested this way, she will receive (1 + i*)St /Ft euros at the end of the year. Because
the forward contract eliminates all exchange risk, the foreign investment is said to be covered or hedged.
For a Dutch investor, the benchmark for judging the attractiveness of the UK Treasury bill investment is a Dutch government bond, which yields (1 + i) for
every euro invested at the end of the year with no exchange risk. Arbitrage guarantees that the two returns must be equal. The result is the covered interest rate
parity (CIRP) condition, which can be formally written as:
(15.1)
A useful approximation of the CIRP condition can be found using informal arguments. Let the interest rate on one-year Dutch bonds be i = 3% and for one-year
UK Treasury bills i* = 4%. The UK bills look more attractive, but the investor is no fool. She knows that the 4% interest is earned in pounds, so she consults the
financial page of the newspaper and looks up the forward exchange rate of the pound in terms of euros, the price at which she could secure the value of her
future interest and principal today. Assume that it is equal to the spot rate so that Ft = St . Effectively, pounds and euros are identical and the UK investment is
much more attractive. In fact, the situation violates the no-profit condition. For it to be satisfied, the higher interest earned on UK bills must be offset in some
way.
Suppose the forward value of the pound is less than its spot value. This means that the pounds paid out at the end of the year will be worth less in terms of
euros. This forward discount implies a c apital loss for our investor. For example, if the loss is of 3%, the return from the pound investment will be the 4% UK
interest rate less the 3% capital loss, thus offering a total return of 1% when evaluated in euros. This is less than the 3% offered in euros by the Dutch bond.
This, too, is incompatible with the no-profit condition. Of course, myriads of British investors will have seen that investing in Dutch bonds is more profitable than
UK Treasury bills. Given the interest rates, the no-profit condition is satisfied when the capital loss on the UK investment implied by the forward rate exactly
offsets the difference in interest rates, i.e. when it is exactly 1%.
More generally, the Dutch investor compares the domestic interest rate i on one hand, and the total return on the foreign investment, on the other hand. The
return on the foreign investment brings an interest rate in pounds of i*, but this must be converted back into euros. If the forward value of the pound in euros is
less than its spot value—the pound is at a discount vis-à-vis the euro—our investor suffers a capital loss: her pounds one year ahead are worth less than today.
Equivalently, the euro is at a forward premium vis-à-vis the pound. Measured as the interest rate in percentage terms, the forward premium of the euro is (Ft −
St )/St . Arbitrage requires that both investment strategies yield the same return:
(15.2)
A positive premium means that the forward value of the euro in terms of pounds is higher than its spot value, i.e. that the euro is ‘stronger’ forward than spot. In
that case, the euro interest rate i can be lower than the sterling interest rate i* and yet the two investment strategies yield equal returns.2
15.3.2 Uncovered Interest Rate Parity

Instead of entering into a forward contract, our investor can buy sterling at spot rate St , invest in the UK bond for one year, and wait until the end of the year to
buy back euros at the then-prevailing spot exchange rate St +1. As explained before, this investment strategy is risky because St +1 is unknown at the beginning of
the year. What appeared to be a good deal may turn out to be disappointing if the euro appreciates vis-à-vis sterling, leaving the investor with fewer euros than
expected when the sterling proceeds are sold. Of course, the euro can depreciate unexpectedly, and the deal will turn out to be terrific. Leaving the foreign
investment open, or uncovered, or unhedged, the investor takes a risk, and this will involve a risk premium, as predicted in Chapter 7.
To examine the new situation, we start by assuming that the investors are risk neutral, meaning that they do not worry about risk. Then the risk premium is
zero. The strategy is the same as before, except that the pound investment will not be sold at the end of the year using the forward rate Ft agreed upon earlier, but
at the then-prevailing spot exchange rate St +1. Going through the same reasoning, the expected return in euros from one euro invested in pounds is
,the end-of-year exchange rate as expected at the beginning of the year. A one-year investment in euros still yields (1 + i). The no-profit
condition for a risk-neutral investor implies that both returns are expected to be equal. This gives the uncovered (UIRP) condition, which can be written
formally as:
(15.3)
which can be approximated as:

(15.4)
.
The UIRP asserts that rates of return are equalized across countries once expected exchange rate changes are taken into account. On the left-hand side of (15.4)
we have the one-year euro interest rate; on the right-hand side we have the pound interest rate less the expected capital gain or loss from changes in the
sterling–euro exchange rate, expressed in percentage terms, for the same one-year maturity. Expectations cannot be observed directly, but the UIRP provides an
implicit measure of what the market expects. If interest rates in the euro area are lower than in the UK (i < i*), then the euro must be expected to appreciate, and
the expected rate appreciation of the euro is given by the difference in observed interest rates:
To be indifferent between euro-denominated assets with a lower interest rate than pound-denominated assets, investors must be compensated by an expected
capital gain. If euro rates are higher, the UIRP implies that sterling is expected to appreciate vis-à-vis the euro.
15.3.3 Risk Premia

Empirically, the UIRP does not hold most of the time. The interest rate spread between, for example, the euro and the pound does not predict future movements in
the exchange rate very well. But this shouldn’t be too surprising. Expectations of a depreciation or appreciation would explain the interest rate spread only if
investors are risk neutral. If investors instead are risk averse, the UIRP need not hold at all. Risk aversion also plays an important role in connecting interest rates
and exchange rates internationally.
The uncovered interest parity condition can be modified to accommodate risk aversion. Our Dutch saver has the choice between remaining safely invested in
euro-denominated assets or assuming the risk associated with British assets, i.e. assets denominated in the UK currency. 3 The no-profit condition implies that
both investment strategies must be equally desirable, but the British investment must also include a premium that makes up for the risk of choosing foreign-
currency assets (from the Dutch investor’s perspective, there is a risk of loss when converting to the home currency). If ψt represents the risk premium, the no-
profit condition requires:
(15.5)
Turning things around, we can define the risk premium as the deviation from the uncovered interest rate parity condition, which can vary over time:
(15.6)
We looked at the situation from the point of view of a Dutch investor who seeks to be compensated for the risk of holding pound-denominated assets. At the
same time, British investors will hold euro-denominated assets. They face the same type of exchange risk, but in the reverse direction. All things being equal,
they too would require a premium on euro-denominated assets, i.e. a negative ψ. So, euro area-based investors want a positive risk premium, UK investors want it
negative. What is the end result? As always, the answer is given by the market. In the end, the risk premium will be such that it exactly balances all these
demands and supplies from the UK, euro area, and elsewhere. As these demands and supplies vary, so does the risk premium. In fact, the risk premium is known
to be volatile and usually small because for any UK investor taking the action described, we are likely to find a euro area-based investor doing the same.
Clearly, the risk premium is a complicated phenomenon whose full treatment is beyond the scope of this textbook. 4 All we need to understand is that the
existence of a risk premium ψt (which can be positive or negative) means that we should not expect the uncovered parity condition to hold exactly. It will continue
to hold, but only after addition of a (possibly quite volatile) risk premium.
15.3.4 Real Interest Rate Arbitrage in the Long Run

The UIRP conditions, with and without a risk premium, link nominal interest rates at home and abroad and embody the tight linkages implied by international
financial integration. Does the arbitrage argument extend to the real interest rate, which is decisive for intertemporal decisions? Intuitively, one might expect an
arbitrage opportunity to arise if real interest rates—with similar risk characteristics—differed significantly across countries. As shown in Box 15.1, it turns out
that the purchasing power parity condition (PPP) does imply that real interest rates at home and abroad will be equal. While the nominal interest parity condition
holds in the short run, PPP only holds in the long run and, therefore, so does the real interest rate parity condition.5
Box 15.1 The Real Interest Rate Parity Condition
This real interest parity condition follows from the UIRP condition, e.g. in the form of (15.4). Rearranged, this means that the interest rate
differential is equal to the expected exchange rate appreciation of the domestic currency. But we saw in Chapter 5 that relative purchasing power
parity (PPP) equates the inflation differential to the rate of appreciation of the currency in the medium to long term:
(15.7)
If forecasts of inflation at home and abroad are consistent with PPP, the definition of the real interest rate and along with
UIRP (15.4) and relative PPP (15.7) imply:
(15.8)
This relationship is called the international Fisher equation. Based on relative PPP, it is a proposition for the medium to long run. As such, it is a useful
benchmark for evaluating long-term foreign investment strategies, as it implies that the real rate of interest should be largely the same in all countries and
is independent of the evolution of exchange rates.
15.4.1 The Exchange Rate as an Asset Price

Panel (a) of Figure 15.1 presents day-to-day changes of the nominal €/£ exchange rate during 2015. Sharp changes are often followed by movements of similar
magnitude in the opposite direction. The variability of nominal exchange rates is often remarkably high, with daily changes of almost ±1% per day commonplace.
(A daily change of 1% corresponds to an annual compounded return of more than 3,000%!) As we saw in Chapter 7, this pattern is typical of asset price
movements. In the short run, exchange rates are hard to distinguish from other asset prices. They are forward-looking, driven by market expectations of future
values.
On the other hand, longer-term fluctuations seem of a different nature. Panel (b) of Figure 15.1 shows quarterly changes of the average exchange rate since the
introduction of the euro in 1999. The movements are larger than in the day-to-day case, but there are fewer changes which are abruptly reversed. Annual
exchange rate changes are even less volatile. In Chapter 5 we saw that the real exchange rate is the relative price of domestic and foreign goods. We also have
learned in our analysis of macroeconomic fluctuations that prices tend to be sticky. This observation suggests that, in the long run, the exchange rate should
look rather like the relative price of goods than the price of an asset. It should be governed by more than the simple logic of financial markets discussed in
Chapter 7 and in Section 15.3. The exchange rate is a ubiquitous economic variable which has two distinct faces: it is an asset price, but is also a relative price of
goods. In the short run, it is the former feature that prevails, in the longer run, the latter feature dominates, in a way consistent with the stylized facts about
exchange rates summarized in Box 15.2. The exchange rate acts as a bridge between the nominal and real sides of the economy, and different patterns over the
short and the long run provide the key that we need to understand this bridge. In this section, we look at the exchange rate as a relative asset price, focusing on
the short run. Later, in Section 15.6, we think about the exchange rate as the relative price of goods, focusing instead on the long run, as we did with the economy
in previous chapters. As we do so, we will see how these two views are consistent with each other.
Fig. 15.1 Fluctuations of the Euro/Sterling Pound Exchange Rate
Day-to-day variability of the nominal exchange rate is considerable. Sharp changes in one direction are frequently undone on the following day. In quarterly data, more
systematic, and even larger movements become evident as noisy movements are averaged away.
Source: European Central Bank.
15.4.2 Implications of the Interest Rate Parity Condition

Thinking of the exchange rate as an asset price leads us back to the interest rate parity conditions. Its volatility suggests that its movements are largely driven by
market expectations of the future. Indeed, exchange markets are continuously absorbing and assessing news regarding political conditions, releases of economic
data, and pronouncements by government ministers, central bankers, bank analysts, prominent businessmen, gurus, etc. After the fact, much of this ‘news’ will
be amended, made more precise, or disavowed, if not actually proved wrong. In the meantime, however, it exerts a powerful influence on the evolution of the
exchange rate. This provides an explanation of the pattern shown in Figure 15.1: news—both genuine facts and rumours—move the exchange rate up one
moment, and down the next. The importance of ‘news’ helps us to make sense of Stylized Facts 1 and 2 described in Box 15.2.
Box 15.2 Mussa’s Stylized Facts and the Asset Behaviour of Exchange Rates
In 1979, Michael Mussa, professor at the University of Chicago at the time, assessed the first half-decade of floating exchange rates after the end of the
Bretton Woods system.6 His observations, which remain true today, can be summarized in the following stylized facts:
1 On a daily basis, changes in floating foreign exchange rates are largely unpredictable.
2 On a month-to-month basis, over 90% of exchange rate movements are unexpected, and less than 10% are predictable.
3 Countries with high inflation rates have depreciating currencies, and over the long run the rate of depreciation of the exchange rate between two countries is
approximately equal to the difference in national inflation rates.
4 Countries with rapidly expanding money supplies tend to have depreciating exchange rates vis-à-vis countries with slowly expanding money supplies. Countries with
rapidly expanding money demands tend to have appreciating exchange rates vis-à-vis countries with slowly expanding money demands.
5 In the longer run, the excess of domestic over foreign interest rates is roughly equal to the expected rate of appreciation of the foreign currency. On a day-to-day basis,
however, the relationship is more tenuous.
6 Actual changes in the spot exchange rate will tend to overshoot any smoothly adjusting measure of the equilibrium real exchange rate.
7 The correlation between month-to-month changes in exchange rates and monthly trade balances is low. On the other hand, in the longer run, countries with persistent
trade deficits tend to have depreciating currencies, whereas those with trade surpluses tend to have appreciating currencies.
Fig. 15.2 The UIRP Condition Repeated Over and Over Again
As a financial asset price, the nominal exchange rate is inherently forward-looking. All versions of UIRP impose tight restrictions on the relationship between interest rates at
home and abroad, and the current and expected future exchange rate as shown in the upper chain. This relation is recursive, which means it also applies at future dates as
well. If traders understand this, successive applications of the UIRP imply that the current exchange rate is a function not only of current, but also of future expected future
interest rate differentials—plus some longer-run terminal condition as shown in the lower part.
The UIRP condition is therefore the right point of departure, since it directly involves the expected future exchange rate. Ignoring the risk premium for the
moment, we can start with (15.3) and note that it links the current interest rates it and it * at home and abroad, the current exchange rate St , and its expected value
next period . (To be more precise, we will now index the interest rate explicitly in each period using the time subscript t.) We can reinterpret the UIRP
condition as saying that the current exchange rate St depends on the current interest rates it and it * and on the exchange rate expected to prevail next
period. This logical implication of the UIRP condition is shown in the leftmost blue arrow in the upper part of Figure 15.2.
This way of representing the UIRP condition is interesting because it shows that we can think of the current spot exchange rate St as determined by domestic
and foreign interest rates and by the market’s current expectation of next period’s exchange rate . Like all asset prices, the nominal exchange rate is forward-
looking. What happened before is irrelevant—bygones are bygones and the exchange rate is not tied to its past. It is totally free to jump to any level warranted
by current or expected future conditions. Especially important is the implication that any increase in raises St . In other words, an appreciation anticipated to
occur in the future will show up immediately in the current exchange rate.
Can we say anything more about ? Next period, the situation will be the same. The UIRP condition will still be valid next period, so St + 1 will be explained
by the interest rates it + 1 and at home and abroad, and by the expected future exchange rate in period t + 2, This is shown in the second blue arrow
of Figure 15.2, upper part.
Now we can see how the markets reason. Today’s exchange rate is explained by the interest rates it and it * and by the expected exchange rate . As they
try to guess they are naturally led to link it with their expectations of and .7 This means that today’s exchange rate is driven by current
and next period’s interest rates and by the exchange rate expected to prevail two periods ahead. Naturally, will also be driven by the same factors, which
will bring and into the picture. Once we move in this direction, there is no end in sight! The last arrow of Figure 15.2 stops the clock n periods
ahead, letting the reader imagine how far away t + n may be. This so-called recursive reasoning is common in macroeconomics, and is made more formal in Box
15.3. It provides equation (15.10), the exact formula suggested in the last row, which will be used in what follows.
Box 15.3 Iterating Forward the UIRP Condition
Start with the UIRP as characterized by equation (15.3) and rewrite it so that it now explains the current exchange rate in the current period t:
(15.9)
If the same condition also holds in the next period t + 1, we have:

This means that, as seen from today’s perspective, our best guess of is based on our best guesses of interest rates in period t+1 and of the
exchange rate in t+2, St+2:
Having recognized this, we can substitute in (15.9) with the next period version and, dropping the superscript ‘e’ for the interest rates, we find:
which can in turn be repeated again and again, ad infinitum. After performing the iteration n times we have:
(15.10)
This is the precise formulation of the symbolic link shown in the lower part of Figure 15.2.
As with stock prices, the current exchange rate reflects all current and future interest rates at home and abroad, and its own long-run value. This expression
shows just how important expectations of the future are for the present. Even events far into the future can have a large impact on today’s exchange rate. This is
why exchange markets—and asset markets in general—are so concerned with information. Even remote future events affect the present.
While not yet providing a complete theory of exchange rate determination, this analysis shows that relative conditions in national money markets are essential
for understanding how macroeconomic conditions influence nominal exchange rates. For instance, it shows how the anticipation of future tight monetary policy
at home (i is expected to rise) can lead to an appreciation today (S increases). This is an observation commonly made when interpreting movements of the
nominal exchange rate.
15.4.3 An Apparent Contradiction and its Resolution

We have a remarkably consistent account of the effect of future monetary policy on the current exchange rate. Yet a contradiction seems to lurk in the deep
corners of Figure 15.2. Suppose that the interest rate it at home rises unexpectedly. According to (15.9), (15.10), or Figure 15.2, the exchange rate St should
immediately rise, i.e. appreciate. Yet, the UIRP condition (15.4) states that a higher interest rate at home should be associated with an expected depreciation of our
currency. How can this be?
Because both conclusions are based on the same UIRP condition, it cannot be a contradiction: both have to be true! But to resolve this paradox, it is important
to make clear what is given and what is free to adjust. Given the foreign interest rate, uncovered interest parity links the domestic interest rate it to the expected
rate of appreciation . Turning this relation on its head, the level of the exchange rate St , is related to the interest rate it , given the expected
future level of the exchange rate, . Similarly, in (15.10) or Figure 15.2, we take as given all future domestic and foreign interest rates and , the expected
future exchange rate far off into the future. Thus, there is no real contradiction, only a confusion of future exchange rate levels and rates of appreciation. In the
moment when the news arrives, there is a jump adjustment to allow for the UIRP to hold in future periods.
Fig. 15.3 An Increase in the Domestic Interest Rate

When the domestic interest rate (i) rises above the world interest rate, uncovered interest rate parity requires an expected exchange rate depreciation. Given an unchanged
expected long-run nominal exchange rate, the exchange rate must appreciate now in order to generate that expected depreciation later.
It is possible to show this diagrammatically in Figure 15.3. We have seen that a higher interest rate it is accompanied by an expected depreciation over the
future according to (15.4), and by a higher exchange rate St , according to (15.9) or (15.10). In effect, these two conclusions must both hold simultaneously. As the
domestic interest rate rises it above the world rate i*, the current exchange rate St appreciates immediately, but temporarily so. Indeed, if it is expected to
depreciate next period or sometimes in the future, we can now have both a current appreciation and a future expected depreciation. Put differently, the exchange
rate must appreciate enough now to be consistent with the expectation of depreciation in the future. These shifts reflect the phenomenon of exchange rate
overshooting: following a tightening of monetary policy, the exchange rate initially must appreciate above its long-run level, in order to depreciate in an
anticipated way over the periods to follow, a condition which follows from the interest parity condition. Box 15.4 provides more details on the macroeconomic
aspects of the overshooting result.
Box 15.4 Interest Rate Parity and Short-Run Open Economy Macroeconomics
In the IS–TR–IFM model of the small open economy, the international financial markets integration ( IFM) line imposed consistency of local money market
interest rates with the return available on foreign assets measured in the local currency. The UIRP condition (15.4) makes the meaning of this return
precise: it is a foreign interest rate i* less the expected rate of exchange rate appreciation. As a result, the position of the IFM schedule depends on the
expected rate of change of the exchange rate. It will shift whenever the exchange rate is expected to change. In the case of a fixed exchange rate regime,
nothing changes as long as the chosen exchange rate peg remains credible and is not expected to change. In that case, and the parity
condition confirms that i = i*. When the exchange rate is floating, however, this simplification cannot really be defended. But it turns out that the key results
of Chapter 12 still hold; in fact, the UIRP acts like a magnifying glass, reinforcing the short-run potency of monetary policy under fixed exchange rates. This
box sketches this effect graphically; a full treatment can be found in the WebAppendix.
Consider the case of a monetary expansion—the central bank lowers its target interest rate i –. This is shown by the rightward shift of the TR schedule
in Figure 15.4 where the economy is initially at point A. In Chapter 12, we reasoned that at point B, the economy is below the IFM line: the interest rate is
now too low, capital flows out, and the exchange rate depreciates. This increases the country’s external competitiveness, the current account improves,
and the IS curve shifts to IS′. The new equilibrium was at point C. Under flexible exchange rates, the IS curve is endogenous and moves to meet the TR′
and IFM schedules. If we now recognize that the exchange rate depreciation is likely to be anticipated, the interest parity condition implies that the
domestic interest rate must be higher for investors to be indifferent between investing at home and abroad. As a consequence, the IFM line too is
endogenous to changes in expected exchange rate changes. In the present case, it shifts up to IFM′. Graphically, this implies that the IS curve will have to
shift until it passes through point D, the intersection of TR′ and IFM′. The interest rate parity condition causes monetary policy in the short run to be more
expansionary than in the original case—D lies to the right of C. The initial shift from IS to IS′ corresponds to the depreciation caused by capital outflows
that reflected the vertical distance from point B to the original IFM line. The interest parity condition shifts IFM to IFM′; however, this increases the distance
from B and calls for an even larger depreciation. The expected future depreciation driving the shift from IFM to IFM′ immediately causes an additional
depreciation over and above the one identified earlier. This extra depreciation further shifts the IS schedule from IS′ to IS′′. Finally, as we move from B to C
and to D along the TR′ schedule, the central bank, which initially relaxed its policy stance, is induced by expanding output—due to booming exports as
following the exchange rate depreciation—to raise the interest rate over and beyond where it initially was.
There is one slight logical problem with point D, however. It cannot be a resting place, since at any point along IFM′, the exchange rate must continue
depreciating! This is logically inconsistent with the fact that i should be i* again. In fact, once point D is reached, capital outflows and the depreciation will
cease. As the expected depreciation slows, the IFM line moves from IFM´ back down again—again, a consequence of UIRP. The exchange rate
depreciation can only stop when the IFM line has once again returned to its initial position at point C. The conclusion is that, by ignoring the interest rate
parity in Chapter 12, we miss the temporary trip from point B to point D and back. In the background, we also ignore the shift from IS′ to IS′′ and back.
Fig. 15.4 Monetary Policy Under Floating Exchange Rates

Starting at point A, a monetary policy expansion is captured by the downward shift of the Taylor rule schedule from TR to TR′. At point B, the lower interest rate triggers
capital outflows and the exchange rate depreciates. The IS curve shifts to IS′ to reflect improved external competitiveness. But, if expected, the exchange rate depreciation
raises the domestic interest rate according to the UIRP condition and the IFM schedule shifts up to IFM′, which triggers a further depreciation and the shift from IS′ to IS″ until it
passes through point D.
15.4.4 The Fundamental Determinants of the Nominal Exchange Rate

The ‘cumulated’ interest parity condition (15.10) states that the exchange rate is determined by present and future interest rates and that it is anchored by its
long-run value. The exchange rate fundamentals, therefore, are those variables that can affect the current and future domestic and foreign interest rates as well
as the long-run exchange rate. Domestic and foreign economic conditions—as captured by the IS–TR and AD–AS frameworks—drive domestic and foreign
interest rates. The fundamentals thus include present and future monetary and fiscal policies at home and abroad. They also include those factors that affect the
exchange rate in the long run. We now examine these factors.
15.5.1 The Long Run and the Primary Current Account: A Review
We now turn to the second role of the exchange rate: the relative price of domestic goods in terms of foreign goods. This notion is captured by the real exchange
rate σ = SP/P*. For given prices at home P and abroad P*, the nominal exchange rate S determines the real exchange rate. For the time being, we take these prices
as given as part of the general macroeconomic picture in the short run. Later, we return to this issue in more detail.
The nation’s intertemporal budget constraint was introduced in Chapter 6. It states that a nation cannot borrow beyond its means and that accumulated assets
can and should be eventually spent. In present-value terms, the country meets its external constraint when the current and future primary net export deficits
match the initial net asset position of the country (or the surpluses must at least match the initial debt). In the simplified two-period framework, this statement was
written formally in Chapter 6 as (imposing equality):
(15.11)

where F1 is the net external position of the country at the beginning of the period. F1 is positive when the country was a net lender previously, and negative
when the country is a net debtor.
As long as it meets its intertemporal budget constraint, the country is free to choose the pattern of its primary accounts over time. This degree of freedom
evaporates in the second and ‘last’ period. Tomorrow’s primary account must match the accumulated net external position. As in Chapter 6, ‘tomorrow’ is a
metaphor for the long-run steady state when, on average, short-run fluctuations simply cancel out. Then the primary current account must be such that, by the
end of period 2 (the proverbial ‘end of time’) the country repays its accumulated debt, or spends its assets, principal, and interest, inherited from period 1. We can
rewrite (15.11) as:
(15.12)
where F2 is the net external position of the country at the beginning of the second period. A positive asset position ( F2 > 0) would allow the country to run a
deficit in the second period, while a position of external indebtedness (F2 < 0) will requires a surplus. The size of F2, the net foreign asset position of a country
going into the second period, is given the budget constraint applying to period 1:
A nation’s foreign asset position will be larger in the second period if its foreign asset position ( F1) is already large to start with, and if its current account
surplus in the first period (PCA1) is large. In a symmetric way, indebted countries that run first period deficits will inherit the burden of running a primary account
surplus in the second period. Higher interest rates will magnify these effects.
15.5.2 Equilibrium Real Exchange Rate and Primary Current Account in the Long Run
The requirement that the long-run primary current account be consistent with the country’s external budget constraint determines the equilibrium, or long-
run, real exchange rate. The link between the primary current account and the real exchange rate was already studied in Chapter 12, when we described the
net export function. Because net exports are a dominant component of the primary current account, it is useful to extend the net exports function to encompass
the primary current account surplus and its dependence on the real exchange rate. Just as net exports, the primary current account deteriorates when the real
exchange rate appreciates. Other factors, such as current and foreign GDP, also affect the primary current account, but they are mostly cyclical and less relevant
for the long run.
For this reason, we write the primary current account function simply as PCA(σ, …). The dots remind us of these other factors. The key relationship to
remember is that PCA is a negative function of the real exchange rate σ. The primary current account function is depicted in Figure 15.5 by the downward-sloping
schedule. The diagram shows that, everything else held constant, each level of the current account is associated with a particular level of the real exchange rate.
If the current account is to be in surplus, the real exchange rate must be rather low to create the export demand and reduce the import demand. Similarly, a current
account in deficit will require an appreciated exchange rate, all else being given.
Fig. 15.5 The Primary Current Account Function

The primary current account function shows the relation between the real exchange rate and the primary account balance. It is downward-sloping because a lower real
exchange rate leads to an improvement in the primary current account, holding all else equal.
Next, we return to the national budget constraint (15.12), which requires that the primary current account be consistent with the solvency condition. In the two-
period parable, this means that in the second period (the long run) the primary current account must be equal to the negative of the country’s net foreign asset
position, F2. When F2 is negative, it means a primary current surplus is necessary to pay off that debt. If F2 is positive—the country is a net creditor, or has a
positive investment position—the country can (and should) run down its accumulated assets via a deficit. Since this long-run constraint is independent of the
real exchange rate, it is shown as the vertical schedule –F2 in Figure 15.6. For net international creditor countries (F2 > 0) this schedule lies to the left of the zero
point, since –F2 < 0; they can live beyond their means in the future. Net international debtors will lie to the right of zero; F2 < 0, so –F2 > 0 and the schedule –F2
will lie to the right in the figure. Such countries need more net exports to satisfy their intertemporal budget constraints.
Fig. 15.6 The Equilibrium Real Exchange Rate
Long-run equilibrium requires that the future primary current account be consistent with the net asset position, F2. To point A corresponds the equilibrium real exchange rate .
A less favourable net asset position F′2 < F2 implies a lower primary current account deficit (or a larger current account surplus) in the future. The vertical schedule is shifted to
the right. If indebted (F2 < 0), the country must have a greater future primary current account surplus to service the greater external debt and the long-run real exchange rate ′
must be lower to generate that surplus. If the country is already a net creditor (F2 > 0), a lower level of net foreign assets today implies a lower deficit tomorrow consistent with
the national budget constraint.
For the budget constraint to be satisfied in period 2, the economy must be at the intersection of the primary current account schedule and this vertical line. At
point A, the equilibrium exchange rate can be read off the vertical axis. Quite simply, the downward-sloping primary current account schedule shows the real
exchange rate level necessary in the long run for a country to obey its intertemporal budget constraint and honour its international obligations. This is the
equilibrium real exchange rate .
As it represents a long-run tendency, the equilibrium real exchange rate is unlikely to correspond to the observed real exchange rate at any point in time, if only
because it fluctuates widely. This is because the real exchange rate is a function of a highly volatile nominal exchange rate S and two slowly-moving price levels:
σ = SP/P*. When the real exchange rate is above its equilibrium level, we say that it is overvalued; in the opposite case it is undervalued. In the long run,
however, it must return to its equilibrium value to ensure that the budget constraint is not violated. Market forces ultimately lead the real exchange rate to its
long-run equilibrium value.
15.5.3 The Fundamental Determinants of the Real Exchange Rate

The conclusion that the net external position drives the equilibrium real exchange rate, through the nation’s budget constraint is a very powerful one. Yet it is a
long-run result, meaning that the time required to get there can take several years or longer. The concept of an equilibrium real exchange rate looks beyond
transitory phases of over- and undervaluation to focus on the steady state. Using two metaphors from the nautical world, the equilibrium real exchange rate is the
lighthouse that shows where the real exchange rate is headed, but it is also an anchor which ultimately pins down the real and, as we will see, the nominal
exchange rate in the long run.
Graphically, the result that the equilibrium real exchange rate is determined by the PCA schedule and the initial net foreign asset position is deceptively simple. It
is possible, however, to go beyond the curves and ask which variables, known as the fundamental determinants, or fundamentals for short, are responsible for
the evolution of the equilibrium real exchange rate. What follows is a list of the most important fundamentals.
Non-price competitiveness
Let us think harder about competitiveness. Competitiveness is all about relative prices—i.e. the real exchange rate—but ultimately about many other things in the
background. To start with, our competitiveness has a great deal to do with how well-designed and original our goods are. A bottle of French champagne is
unique, and can fetch a high price before consumers switch en masse to similar sparkling wines, which may be good or even better but do not carry the glamour
of the original. Similarly, German cars have a reputation of reliability that gives them the edge, just as British bankers have built up the formidable City of London
network, or Italian manufacturers have developed a knack for style and design. A country’s exports may suddenly be à la mode (think of fads in tourism). When a
country becomes fundamentally more competitive, its primary current account improves for any level of the real exchange rate. Graphically, in Figure 15.7, the
PCA schedule shifts upward and to the right, to PCA′. This gain in non-price competitiveness—defined as relative attractiveness of our goods holding the
real exchange rate constant—translates into a higher equilibrium exchange rate ′. This result can be explained in two ways. First, because they have become
more attractive, domestic goods can be sold at a higher relative price. Second, starting at the old equilibrium rate , i.e. at unchanged relative price, more goods
can now be sold, which will result in a larger primary current account than required by condition (15.12). In the long run, this surplus will have to be eliminated by
a real appreciation, which makes our goods more expensive and their goods cheaper.
Fig. 15.7 The Equilibrium Exchange Rate and Non-Price Competitiveness

A gain in non-price competitiveness is captured by the rightward shift of the PCA schedule to PCA′. The equilibrium real exchange rate rises from to ′.
A good example of a rapid improvement in non-price competitiveness is the transformation process in Central and Eastern Europe. As the Berlin Wall was
dismantled in 1989 and the Soviet Bloc crumbled, these countries moved from central planning to the market economy. Starting with antiquated productive
equipment and a dilapidated public infrastructure, they quickly started to catch up with the West. As they climbed up the export quality ladder, their non-price
competitiveness quickly improved. Unsurprisingly, their real exchange rates have been continuously appreciating, as can be clearly seen by the trends depicted
in Figure 15.8.
Natural resources
The Netherlands in the 1960s and the UK and Norway in the early 1980s were lucky enough to discover large reserves of gas or oil under the North Sea. In a
matter of a few years, they could reduce their import bill or even become net exporters of fossil fuels.
Fig. 15.8 Real Exchange Rates, 1995–2015

Countries that experience catch-up growth tend to experience appreciating equilibrium exchange rates at the same time. One explanation for this is that their non-price
competitiveness—the attractiveness of their products in world markets—has improved. In contrast, developed countries tend to have more stable real exchange rates over the
long term. This tendency is borne out in by real exchange rates of the Czech Republic, Estonia, Slovakia, and Poland after their EU accessions.
Source: IMF, International Financial Statistics.
To understand the effect of such natural resource discoveries, it is useful to break the primary current account into two components, oil and non-oil, and we
can write PCA = PCAnon-oil + PCAoil. Before the discovery, PCAoil was substantial and negative, as is the case in most oil-importing countries. Afterwards it
became nil or even positive. If the primary current account was already at its long-run level before the discovery, the consequence of the discovery is to move
PCAoil to a large surplus, leading to a sharp real appreciation. The discovery of natural resources has much the same effect as an improvement in non-price
competitiveness. This is well captured by the rightward shift of the PCA schedule in Figure 15.7. As before, it implies that the equilibrium real exchange rate has
appreciated. What happens next is told in Box 15.5.
Net external position

We have seen that the net external investment position is another fundamental determinant of the real exchange rate. All other things being equal, the more
positive the external investment position, the more appreciated the equilibrium exchange rate will be. Similarly, indebted countries will require depreciated real
equilibrium exchange rates to generate resources for debt service.
Box 15.5 The Norwegian Cure for the Dutch Disease
When oil was discovered in the UK’s North Sea in the mid-1970s, the sterling real exchange rate soon appreciated by more than 30%. Ironically, the result
was a shift away from ‘traditional’ industry (other traded goods besides oil) towards services, with massive plant closures and unemployment as
industrial workers could not be immediately absorbed by the oil industry or the service sector. This striking process of de-industrialization—a
straightforward consequence of the intertemporal budget constraint—is called the Dutch disease, ever since it was first diagnosed in the 1960s when gas
was discovered in the Netherlands’ side of the North Sea.8 The Dutch disease explains why resource-rich countries typically find it more difficult to
develop and maintain an internationally competitive industrial sector. The problem can be seen today in contemporary Russia. This need not be a long-
term concern. Does it make sense to produce industrial goods when Mother Nature provides easy exports? Yet in the short run, the adjustment can be
painful.
To save its industry from the Dutch disease, Norway established the Government Petroleum Fund in 1990 (renamed Government Pension Fund—
Global in 2006). As described by the Ministry of Finance:
Central government’s net cash flow from petroleum operations is transferred in its entirety to the Government Pension Fund Global via the state budget,
whereas the guidelines stipulate that only the expected real return on the Fund should be returned to the budget for general spending purposes. In this
way the Fund serves as a long-term savings vehicle to let the Norwegian Government accumulate financial assets to help cope with future expenditures
associated with the ageing of the population…. At the end of 2005 the Fund amounted to close to NOK 1400 billion or 99 per cent of mainland GDP (i.e.
excluding petroleum activities). Projections indicate that the Fund will grow to 180 per cent of mainland GDP at the end of 2010. Given the present
guidelines for fiscal policy, the Fund is expected to reach a level of around 250 per cent of mainland GDP in the years after 2030, before starting to
gradually decline.9
The Fund’s resources—which in 2016 totalled roughly €750 bn—must be invested abroad. This provision is designed to protect Norway from the Dutch
disease. Since much of oil income does not enter Norway, there is no pressure on the exchange rate to appreciate. As a result, the account is in perennial
surplus. This would make no sense in the long run, but for Norway the long run is when oil reserves will be exhausted. In the meantime, keeping the
current account in surplus prevents de-industrialization. Norway’s central bank sums up the situation well: ‘The Petroleum Fund makes it possible … to
avoid abrupt shifts in the industry structure, such as we have seen in many other countries with substantial revenues from natural resources, and
contributes to sustainable business and industry in the long term.’10
To see how the net external position affects the equilibrium real exchange rate, we return to Figure 15.6. Point A describes a situation of long-run equilibrium:
when the external debt position is F2, depicted in the figure as a net debt so –F2 > 0, the equilibrium real exchange rate is . Now consider the same country, with
the same PCA schedule, but with a net asset position F2′ that is even more negative (is more indebted) than F2. To this debt corresponds a new vertical schedule
which lies to the right of the original one, and a new equilibrium real exchange rate ′ that is lower than . We see that the larger the net external debt, the more the
real exchange rate must depreciate in the long run. Conversely, an improvement in the net external position will be associated with a real exchange appreciation.
The net external position of a country is a legacy of its history and evolves slowly. It may also take many years for the actual exchange rate to respond to
changes in its equilibrium value. When the exchange rate is misaligned, meaning that it is not at its equilibrium level, its net external position must be changing.
Indeed, we know that when the current account is in surplus, the country lends to the rest of the world, and it borrows when the current account is in deficit. In
the long run, the real exchange rate is at its equilibrium level and we have PCA2 = −F2 so that the current account is CA2 = PCA2 + F2 = 0. Figure 15.6 shows that
if the exchange rate were overvalued, so that σ > , relative to point A, we would be up and to the left along the PCA schedule. This would imply PCA2 < −F2 and
therefore a current account deficit. As a net borrower, it lives beyond its means. Conversely, if σ < , the exchange rate is undervalued and the country’s net asset
position improves as it runs a current account surplus.
Box 15.6 What Goes Down Must Come Up, and Conversely
Figure 15.9 shows that the primary current accounts of Ireland and Greece deteriorated continuously after the creation of the euro in 1999 (Greece joined
two years later, in 2001). As mentioned in Chapter 7, housing prices rose sharply in Ireland after the mid-1990s, encouraging people to borrow and
participate in the bonanza. In Greece, both private sector and government borrowed heavily after interest rates declined to low German levels, an
implication of the interest rate parity principle.11 As we know, the primary current account represents the country’s net lending. Increasingly heavy
borrowing in Ireland and Greece translated into deepening PCA deficits. At the same time, as Eurozone members, both countries had lost their own
currencies and nominal exchange rates. Could they indefinitely hide behind the euro and defy their budget constraints?
Obviously not. Either they would have to stop borrowing and start saving to service their debts, or the real exchange rate would have to depreciate, or
both. With a fixed nominal rate vis-à-vis the rest of the Eurozone, a real depreciation can only be achieved by lowering domestic prices P relative to the
foreign price level P*, i.e. less inflation for a sufficiently long time. In the event, both adjustments were necessary. At the beginning, the financial world
simply ignored the growing imbalances. Once the world slipped into the Great Financial Crisis, badly battered markets became highly nervous. They
quickly spotted the unsustainable situation of Greece and Ireland and sold their bonds, effectively bringing lending to both countries to a sudden halt. In
effect, the financial markets enforced the budget constraints. All of a sudden, Greece and Ireland had to service their debts. To cushion the shock, they
asked for external help from other Eurozone countries and the IMF. Official loans were provided but under strict conditions, one of them being to cut
spending, raise taxes, and reduce borrowing.
Fig. 15.9 Primary Current Accounts and Real Exchange Rates, Greece and Ireland 1999–2015
Similar patterns of adjustment were experienced by Greece and Ireland after the onset of their financial crises (2008–2009). Gaping current account deficits were swiftly
pushed towards surplus, reflecting both a collapse of GDP in both countries and a real depreciation based on both consumer prices and unit labour costs. In both cases, the
exchange rate supports the adjustment process which enforces the national intertemporal budget constraint.
Less spending had two effects. First the current accounts improved, as can be seen in the upper panel of Figure 15.9. Second, the economies went into
a deep recession as domestic demand declined. As predicted by the AD–AS framework, inflation declined and the real exchange rate depreciated. The
lesson here is that the budget constraint, and its implications for the primary current account and the real exchange rate, is an iron law for the long run.
The exchange rate regime cannot change this fact; it only modifies the way budget constraint asserts itself. Indeed, fixed exchange rates may give the
impression that the long run is very far away. As the late MIT economist Rudiger Dornbusch once said: ‘The crisis takes a much longer time coming than
you think, and then it happens much faster than you would have thought.’12
This might seem worrisome. Could debts spiral out of control? If the exchange rate is allowed to play its corrective role, the answer is no. But examples abound
of countries such as Indonesia and Nigeria, which were made fabulously wealthy by oil discoveries, but then squandered their wealth and even became highly
indebted as a result. More recently, countries in southern Europe have seen their current account deficits grow without a corrective depreciation. Box 15.6 takes a
look at the real exchange rate adjustment in Greece and Ireland after the financial crisis and gives insights into problems ahead for the euro.
15.5.4 How to Think About the Equilibrium Real Exchange Rate?

In practice, disturbances large enough to produce significant changes in the equilibrium real exchange rate are rather rare events. When considering countries of
a comparable level of economic development, it is a good rule of thumb to view the equilibrium real exchange rate as stable or even constant. This implies relative
PPP, which we already encountered in Chapter 5, and possibly even the absolute version of PPP, which is much stronger and says that price levels across
countries are comparable. That PPP is a powerful rule of thumb also means that we must be always on the watch for the possibility that a major change is under
way. We have just seen the main fundamentals that determine the equilibrium real exchange rate, so we now know why it may change.
An important case in point is the catch-up process that poorer countries experience during a successful take-off growth phase. We observed in Figure 15.8
that the real exchange rates of the Central and East European countries appreciated steadily since they embarked on their transformation process that began in
the early 1990s. This is an excellent example of a general phenomenon known as the Balassa–Samuelson effect. It is described in Box 15.7.
Box 15.7 The Balassa–Samuelson Effect13
Experienced travellers invariably notice that wealthier countries are systematically more expensive than poorer ones. This is not merely a perception, but a
hard fact (see Table 15.4). Economics can help us understand why. Consider a relatively poor country. Most wages will be much lower than in a rich
country because labour productivity is low. As we know from Chapter 3, catching up with the richer country occurs when a less-developed country
accumulates capital, imports advanced technology, and generally becomes more productive. As this happens, we expect standards of living, and
therefore local wages, to grow. Now work backwards in time from when catch-up will have occurred and wages will have caught up with those of the most
advanced economies. For this to happen, wages must grow faster during the catching-up period. As purchasing power rises, so do demand and prices.
To grow faster in poor rather than in rich countries and eventually converge to the same levels, prices must start from a lower base. It is for this reason that
in January 2016 the famous Big Mac burger cost $6.44 in Switzerland, $4.93 in the United States, $3.22 in Portugal, and only $1.77 in South Africa!
This phenomenon is documented more systematically in Table 15.4, which shows price levels and GDP per capita in a sample of countries around the
world, as measured in 2009. For ease of comparison, both prices and GDP are indexed relative to the USA. For example, in 2009, Argentine GDP per
capita was only about 30% of US levels, and prices in the same currency were 46% lower. In rich Norway, GDP per person was 24% higher, and the price
level was 56% higher than in the USA. The association between income and price levels is unmistakable, although not perfect.
15.6.1 The Equilibrium Real Exchange Rate as an Anchor

We now reconcile the short-run ‘financial market’ and the long-run or ‘relative price of goods’ views of the exchange rate. Each perspective is rooted in its own
compelling logic. The short-run view was encoded in Figure 15.3 or equation (15.10). Both link the current nominal exchange rate to the expected future exchange
rate far into the future, with a bridge being the intervening expected future rates of interest at home and abroad. Of course, no one knows what the future will be,
but markets that set the exchange rate nevertheless have to make a guess. Between now and the distant future, many things will happen. There will be expansions
and recessions, current account surpluses and deficits, and much more. Yet, a good bet is that, in the long run, the relative-price-of-goods view will prevail
because a country cannot forever escape its budget constraint, and, just as important, the real exchange rate will play a role in enforcing that budget constraint.
Over- and undervaluation do occur, but not forever.
Returning to Figure 15.2 or equation (15.10), we can now think of the very long-run nominal exchange rate expected far into the future with its long-run level.
As n gets larger, we are getting closer to expectations of the long-run nominal exchange rate . This long-run concept—which will never be observed but is in the
back of all our minds—is the anchor which pins down the nominal exchange rate, not only in the long run, but also in the short run. This is the missing link
between the short- and long-run views of the exchange rate. In the next section, we show how to bridge the gap between a short-term volatile asset price and the
long-run one, given by monetary and real conditions over periods of 10 years or more.
Table 15.4 GDP per Capita and Price Levels, 2014 (USA = 100)
GDP per Capita Price Level GDP per Capita Price Level
Europe South America
Austria 87.3 107.2 Colombia 24.5 59.2
Belarus 33.3 44.2 Bolivia 12.1 47.1
Belgium 79.5 109.0 Brazil 29.1 73.8
Croatia 39.6 62.3 Chile 40.4 65.8
Czech Republic 57.1 62.5 Peru 21.9 54.6
Denmark 83.4 133.3 Venezuela 30.2 70.9
Finland 74.5 122.5
France 72.0 108.6 Asia
Germany 84.9 103.0 Bangladesh 5.7 34.8
Greece 49.2 80.7 China 24.2 57.5
Hungary 45.9 56.0 India 10.4 27.7
Ireland 90.4 110.0 Indonesia 19.3 33.2
Italy 64.9 99.3 Israel 61.7 110.4
Luxembourg 180.2 118.4 Japan 67.0 98.8
Norway 120.1 148.3 Korea 61.1 83.8
Poland 46.2 56.8 Pakistan 8.8 27.4
Portugal 52.6 76.9 Saudi Arabia 95.2 46.9
Russia 42.1 55.4 Singapore 151.5 68.0
Slovak Republic 51.9 65.3 Sri Lanka 20.3 34.2
Spain 61.6 88.4
Sweden 82.9 130.0 Africa
Switzerland 109.0 143.8 Central African Republic 1.1 60.3
Turkey 36.2 53.1 Chad 4.0 47.0
United Kingdom 73.6 115.1 Congo, Dem. Rep. 1.4 59.3
Côte d’Ivoire 6.0 47.4
North America and Oceania Egypt 11.5 32.0
Canada 82.5 111.5 Ethiopia 2.7 38.2
Mexico 31.7 59.6 Kenya 5.4 46.0
United States 100.0 100.0 Nigeria 10.8 54.2
Australia 84.1 135.0 South Africa 23.9 49.7
Note: Data for Venezuela are from 2012.
Source: World Bank World Development Indicators.
15.6.2 The Bridge to the Long Run: the Parity Conditions

We now want to connect the day-to-day nominal exchange rate St and the long-run nominal exchange rate as it enters Figure 15.2 or equation (15.10). Yet, the
long-run equilibrium exchange rate found in Figure 15.6 is defined in real terms. How do we bridge these gaps?
In fact, we already know the answer. The real determinants of , the long-run (or equilibrium) real exchange rate, were discussed in Section 15.5. If we think of
as the ‘doubly-deflated’ long-run nominal exchange rate, then . The price levels are those which prevail at home and abroad, respectively, in the long run.
These price levels will be the result of cumulated inflation between now and then. Chapters 5 and 13 inform us how to think about inflation in the long run, so we
know how to think of P and . It only remains to use the definition of the real exchange rate to pin down the long-run nominal exchange rate, which can be
rearranged as .
Thus, the long-run equilibrium nominal exchange rate is driven by the long-run equilibrium real exchange rate and by the long-run price levels at home and
abroad. Anything that raises the long-run equilibrium real exchange rate also raises the long-run equilibrium nominal exchange rate, holding the evolution of
domestic and foreign inflation constant. Given a long-run equilibrium real exchange rate, the long-run equilibrium nominal exchange rate is higher, the higher is
the foreign price level (due to higher cumulative inflation) and the lower the domestic price level (due to lower cumulative inflation).
The bridge between short and long run helps us understand Stylized Facts 3 and 4 in Box 15.2. Higher domestic inflation, relative to foreign inflation, will drive
higher relative to —and lead to a more depreciated . This has direct consequences for the current nominal exchange rate St (Stylized Fact 3). Recalling that
inflation and money growth go hand in hand in the long run, we can now understand Stylized Fact 4 as well.
The last step is to replace with in (15.10). The result, for n very very large, is:
(15.13)
We have now completed the link between the current nominal exchange rate and its long-run value. Indeed, this expression indicates that the current nominal
exchange St depends on three sets of fundamentals:
(1) The path of future domestic and foreign interest rates—therefore all relevant economic conditions affecting the interest rate (as captured by the IS–TR framework): monetary
and fiscal policies, foreign demand, etc.
(2) The real exchange rate that is required to meet the nation’s intertemporal budget constraint—therefore the foreign debt and the country’s competitive position.
(3) The level of prices at home and abroad far into the future—therefore present and future inflation, both at home and abroad.
The list of fundamentals is long because of the ubiquitous role of the exchange rate in the relative price of goods and assets. In light of this, it is easy to
understand why exchange markets react to a very broad range of indicators. Yet, the link between the budget constraint and the current exchange rate is also
long—it passes through the real equilibrium exchange rate , as well as the long-run nominal rate —and is possibly obscured by the long string of domestic and
foreign interest rates that appear in (15.13). This is the central point of Stylized Fact 7.
15.7.1 Volatility and Predictability

The short-run volatility of the exchange rate, visible in Panel (a) of Figure 15.1, is a direct consequence of (15.13). We have stressed the very large number of
events that may potentially affect the current exchange rate via all future interest rates and the anticipated long-run nominal exchange rate . As expectations are
constantly changing, the exchange rate reacts. Over time, much of this noise will cancel or dissipate, as Panel (b) illustrates. Gradually the long-run fundamentals
that drive come to dominate (Stylized Facts 1 and 2).
Stylized Fact 1 makes another important point. An implication of (15.13) is that the current nominal exchange rate reflects what we think we now know about the
future. What makes tomorrow or next year different from now? The answer is the arrival of new information. This means that the exchange rate changes between
one period and the next as a result of new information, which is unpredictable by definition—if we could predict the arrival of new information, it wouldn’t be new
information anymore! This is why exchange rate movements are largely unpredictable.
The unpredictability of short-run fluctuations is sometimes cited as a proof that economists do not really understand the exchange rate. The preceding
observation explains why this argument is wrong. Quite to the contrary, understanding the mechanism that determines exchange rates necessarily implies that it
must be unpredictable! Those who make predictions must implicitly claim that they know things that the market does not know. This may be the case now and
then, but it is very unlikely to be the case systematically.14
15.7.2 Currency Crises

A particular form of volatility is a currency crisis.15 One example, among many, is the crisis that affected Southeast Asia in the late 1990s. Figure 15.10 shows the
evolution of Korea’s exchange rate. Korea was then one among several East Asian ‘dragons’ that had been growing at close to 10% per annum over more than a
decade. It was hailed as a success story, which led to the country’s joining the OECD, the club of developed countries. Despite this success, the Korean
currency, the won, collapsed in 1997 and the economy plunged into a deep recession. While it has recovered, its growth still pales in comparison to its
spectacular growth performance in the first half of the 1990s. In fact, Korea has become a mature OECD economy with world-class corporations. Yet, a decade
later, the won again lost some 40% of its value in the weeks that followed the Wall Street crisis of September 2008.
Fig. 15.10 Korea’s Exchange Rate (index 100 = 2005 average)

The exchange rate is measured as US$ per Korean won. In two months at the end of 1997, the won lost half of its value. In a similar fashion, the financial crisis of 2008–2009 hit
Korea, an exporting economy, rather hard. But growth has returned since then.
Source: IMF.
This example illustrates the fact that currency crises occur frequently and not just in badly managed countries. Most often, though, they are the consequence
of the authorities’ efforts at maintaining an overvalued exchange rate. The most typical case involves a fixed exchange rate regime combined with rapid domestic
inflation. In Chapter 14, this was shown to be an unsustainable situation. For a given level of the equilibrium real exchange rate , the corresponding equilibrium
nominal exchange rate will decline as the domestic price level increases. If the central bank continues to intervene heavily in the foreign exchange market to keep
the actual nominal exchange rate S fixed, the currency becomes increasingly overvalued. As selling pressure mounts, foreign exchange reserves quickly decline
and, eventually, the central bank must throw in the towel. A brutal adjustment follows.
The case of Korea is more interesting. Prior to the 1997 Asian crisis, there was no serious inflation and the current account was approximately balanced,
suggesting at face value that the exchange rate was not overvalued. Yet, several countries in the region had overvalued currencies. This was the case, in
particular, of Thailand, which faced a currency crisis in June 1997. At that stage, the markets started to wonder whether neighbouring countries were not in a
similar situation. The collapse of Korea’s currency is an instance of a self-fulfilling crisis.
The interpretation of crises follows directly from Figure 15.2 or equation (15.10). This relationship implies that today’s exchange rate St is driven by its expected
value in the future, . If markets expect to fall, Figure 15.2 predicts that the current rate will fall immediately. When the nominal exchange rate is kept fixed at
an overvalued level, it is enough for markets to expect a depreciation at some point in the future for the depreciation to occur now. This means that crises can be
triggered by abrupt changes in expectations of faraway outcomes.
And this is exactly what happens when the exchange rate is overvalued. As long as the central bank continues to defend the fixed parity Sfix which is
overvalued relative to the long run (i.e. ), the market may continue to believe that . When the stock of reserves becomes depleted, the market knows that the
central bank cannot maintain the fixed parity Sfix and that the exchange rate will have to be devalued to its equilibrium level. In that instant, the expectation
switches to . This expectation can be based on a simple extrapolation of how long the reserves will last, or may even anticipate that the market may clean out the
central bank sooner than that.
A self-fulfilling crisis can be triggered by a change of view about the equilibrium exchange rates . This change of view can be brutal because it is panic-driven.
Even a small probability of a depreciation implies potentially large losses from holding assets denominated in the currency in question. Even if most market
participants do not believe that a depreciation is in the making, a rumour that it could be is sufficient to convince them to sell their assets denominated in that
currency as a precaution. Large sales—a capital outflow—can depress the exchange rate and can even validate the expectation that the currency might be
devalued. This leads to more panic sales and eventually to a full-blown crisis.
Such crises were common in the time of the European Monetary System, the fixed-exchange rate system that preceded the European Monetary Union.16 The
2008 currency crisis in Korea is another example of a self-fulfilling crisis. At that time, like a decade earlier, the Korean macroeconomy (inflation, external balance)
was healthy but many firms and some households had borrowed in dollars and euros to spend or invest in wons. This created a vulnerability: if the won were to
depreciate sharply, Korean borrowers would find it difficult or impossible to honour their debts. This would lead to widespread bankruptcies and further defaults
on foreign borrowings. At a time of exacerbated market anxiety, investors in Korea panicked and sold their holdings of wons, which produced the vicious circle
that they had initially feared.
The main conclusion is that volatility and crises are part and parcel of foreign exchange markets, in fact of all financial markets. Exchange rates and asset prices
are driven by expectations which can shift rapidly. Self-fulfilling crises happen when expectations, whether justified or not, are ratified ex post by the panic that
they generate. Volatility and misalignment of the exchange rate is, however, not merely a problem of an asset market as we have seen: it also affects the
functioning of the macroeconomy and the well-being of millions of households and firms.
Summary
1 Foreign exchange markets are a worldwide network of traders who constantly swap currencies and assets denominated in different currencies. They deal in large amounts and
promptly arbitrage away any potential source of profit. This ensures consistency among all bilateral cross-rates.
2 The no-profit condition, a characteristic of efficient financial markets, ensures that the covered interest rate parity condition is satisfied. When capital is internationally
mobile, a higher domestic interest rate is matched by a forward exchange rate premium.
3 In the presence of risk, the no-profit condition leads to an uncovered interest parity condition. When capital is internationally mobile, an expected exchange rate depreciation
should be compensated by a higher domestic interest rate, and conversely, up to a risk premium.
4 If both uncovered interest parity and purchasing power parity conditions hold, real interest rates are equalized worldwide. Since PPP holds at best in the long run, real
interest rate equalization is only a long-run proposition.
5 Exchange rates are forward-looking variables. An implication of the uncovered interest parity condition is that today’s exchange rate is driven by today’s interest rates and by
the expected exchange rate next period.
6 Today’s exchange rate is linked to present and future interest rates at home and abroad, and to a long-run spot nominal exchange rate. The link takes the form of a chain of
present and future uncovered interest parity conditions. It is in the nature of an asset market that all that is currently known about the future is reflected in today’s value of
the exchange rate. Changes in the exchange rate occur primarily because new information arrives, including revisions of expectations about the future.
7 In the long run, the real exchange rate must ultimately be consistent with the country’s intertemporal budget constraint. The constraint means that past international
borrowing must be serviced by future current account surpluses. Conversely, international lending in the past allows a country to eventually dissave in the future via current
account deficits. Consistency with the intertemporal budget constraint is a central determinant of the equilibrium real exchange rate.
8 The equilibrium real exchange rate appreciates when the net external position improves or when the economy becomes more competitive for reasons such as the discovery of
natural resources or gains in non-price competitiveness.
9 Purchasing power parity occurs in the long run when there is no change in the country’s net external position (when the current account is balanced on average) or non-price
competitiveness. Economies that catch up by improving their productivity see their equilibrium real exchange rates appreciate. The resulting Balassa–Samuelson effect is a
key reason that PPP is violated.
10 The long- and short-run views of the exchange rate are not inconsistent. Real factors that drive the long-run real exchange rate are present in today’s nominal exchange rate via
the exchange rate expected in the long run. Thus the long-run exchange rate is the anchor that guides the path of future expected exchange rates and relates it to the current
value.
11 The fundamental determinants of the exchange rate include all the variables that affect the equilibrium real exchange rate and the domestic and foreign interest rates that link
the current exchange rate to its long-run equilibrium level. Domestic and foreign inflation, which determine the link between the equilibrium real exchange rate and its nominal
level, are also among the fundamentals.
12 The forward-looking aspect of the exchange rate explains its volatility and even the occasional occurrence of currency crises.
Key Concepts
interest rate parity

currency crises
vehicle currencies
spot and forward exchange rates, swaps
covered interest rate parity (CIRP)
hedged, covered
forward premium/discount
capital gains and losses
open position (uncovered, unhedged)
uncovered interest rate parity (UIRP)
international Fisher equation
exchange rate overshooting
fundamentals
equilibrium, long-run, real exchange rate
primary current account function
overvalued, undervalued
non-price competitiveness
misaligned
Balassa–Samuelson effect
self-fulfilling crisis
Exercises
1 Suppose that you could buy and sell US dollars for $1.25/EUR and pounds for 0.75£/EUR, but that at the same time the dollar–pound rate was 1.65$/£.
(a) What strategy would you pursue to take advantage of this ‘money pump’? What would be the likely effect of the market’s recognition of its existence?
(b) Question (a) ignores the existence of a bid–ask spread. How would your answer change if the bid–ask spreads were: $1.24–1.26/EUR, 0.74–0.76£/EUR, and 1.64–
1.66$/£?
2 The interest rate in Austria is 3% while it is 10% in Hungary. What does this imply for the forward premium/discount of the forint, the Hungarian currency, vis-à-vis the
euro (which is used in Austria)?
3 Consider the same interest rates as in the previous question. What would you do if you were 100% sure that the forint/euro exchange rate would remain constant, perhaps
because there was a fixed exchange rate regime? Would this necessarily be a violation of the no-profit condition? Explain.
4 Suppose the Polish currency, the zloty (PLN), is trading at 0.225 EUR/PLN, and market participants anticipated that Poland would fix its exchange rate to the Euro in one
year at a rate 0.25 Euro/PLN. What do you think will happen to the exchange rate today? How would it help to know interest rates in Poland and in the Eurozone?
5 For decades, interest rates in Switzerland have been lower than in Germany (by 1 to 2%) while the exchange rate between the Swiss franc and the German mark first, and now
the euro, has remained stable. What does this mean for the interest rate parity condition? How could you explain this phenomenon?
6 Within the euro area, interest rates on government bonds have become quite different recently. For instance, on 21 December 2011, interest rates on 10-year government
bonds were: 3.35% in Germany, 8.23% in Ireland, 12.99% in Portugal, 6.74% in Italy, and 31.29% in Greece. Interpret these numbers. Is a no-profit condition being
violated?
7 Since 1970, Australia’s current account has almost always been in deficit, sometimes to a considerable extent. The real exchange rate has lost very little, some 10%. Is it
surprising? How can this be explained?
8 Figure 15.10 shows that Korea’s nominal exchange rate has considerably appreciated since the 1997–1998 crisis described in Section 15.8.2. What are possible explanations
for this appreciation?
9 ‘The equilibrium real exchange rate remains constant when the actual real exchange rate remains durably close to its equilibrium level.’ What is the logic of such an assertion?
Is this always true?
10 Facing market pressure for an appreciation of the Swiss franc, the Swiss central bank lowered its interest rate. Why?
Essay Questions
1 ‘The advantage of a fixed exchange rate regime is to greatly reduce exchange rate uncertainty.’ Comment.
2 True or false: ‘Financial institutions tend to be for flexible exchange rates, while firms in manufacturing and traded output in general tend to be in favour of fixed exchange
rates.’
3 ‘The fact that currency crises can be self-fulfilling is greatly disturbing because it suggests that any country can see its money hugely devalued, for no good reason.’ In
commenting on this statement, carefully establish the difference between self-fulfilling and other types of crises.
4 Equations like (15.10) suggest that the market must look at horizons so far out that they seem to make little sense. Use the logic of the yield curve, presented and explained in
Chapter 7, to explain the similarity between the two ways of reasoning and provide an interpretation of how the markets think about the very long run.
5 Concern has been raised that the discovery and extraction of oil in Kazakhstan and other central Asian countries might have adverse effects on their economies. What might be
the reasons for these concerns? What effects would you predict the discoveries to have on the real exchange rate and the long-term current accounts of these countries?
1 We will use ‘bond’, ‘bill’, and ‘note’ interchangeably here. A Treasury ‘bill’ is a pure discount bond of very short maturity, whereas notes or bonds generally involve a
periodic coupon payment. A government bond or note that is close to maturity is for all practical purposes like a Treasury bill.
2 The WebAppendix further develops the concept of the forward premium. Equation (15.2) can be derived directly using straightforward mathematics. Simply take natural logarithms
of both sides of (15.1) to obtain ln(1 + i) = ln(1 + i*) − ln(Ft/St) and recall that as a first approximation: ln(1 + x) ≈ x. We can then write ln(1 + i) ≈ i, ln(1 + i*) ≈ i*, and ln(F/S) = ln(1
+ (F − S)/S) ≈ (F − S)/S.
3 The payoff in pounds sterling is risky because the investor lives in a region which does not use the pound as legal tender, so that exchange rate risk is equivalent to nominal price
risk, which means consumption risk. For the investor residing in the UK, a euro investment would be risky for the same reason.
4 Finance theory (the Capital Asset Pricing Model or CAPM) states that in equilibrium, the risk premium is determined by the correlation between the asset’s return and the return
from the world portfolio.
5 We discuss purchasing power parity in Chapter 5. The relative version is employed here. The Fisher equation, which relates real and nominal interest rates explicitly, was introduced
in Chapter 14.
6 We discuss the Bretton Woods system in more detail in Chapter 19. It is sufficient to know that exchange rates during this period (1946–1971) were fixed by an international
agreement and were very difficult to change.
7 Note that i , i , i , etc. are actually not known as of now (i.e. in period t ), so in fact we are talking about expectations and we should add the ‘e’ subscript. To simplify the
t +1 t +2 t +3
analysis and keep notation tidier, we dispense with this precision.
8 The WebAppendix presents the phenomenon in more formal detail.
9 ‘The Norwegian Petroleum Sector and the Government Pension Fund—Global’ by Secretary General Tore Eriksen, Ministry of Finance, www.regjeringen.no.
10 Norges Bank, www.norges-bank.no.
11 Within the Eurozone, nominal depreciation is impossible, so nominal interest rates can only reflect risk premia in the member countries or the probability of exit from the
monetary union and the adoption of an independent (and presumably depreciating) currency. During the crisis, membership in the Eurozone became less certain and defaults were seen
as possible, leading to significant interest rate differentials for government bonds of different member countries.
12 Quoted by Paul Krugman in <http://economistsview.typepad.com/economistsview/2014/10/dornbuschs-law.html>.
13 The Balassa–Samuelson effect is named after Bela Balassa (1928–1991), a Hungarian-born economist at Johns Hopkins University, also known as a gourmet and author of a
confidential restaurant guide, and after Paul Samuelson (1915–2009), the 1970 Nobel laureate from MIT. A formal derivation is available in the WebAppendix.
14 This argument holds for all asset prices determined in efficient markets.
15 This is a brief presentation of the currency crisis phenomenon. A more extensive analysis is provided in Chapter 19.
16 The functioning of the European Monetary System in the period 1979–1999 is discussed in detail in Chapter 19.
PART V
Macroeconomic Policy
in a Global Economy

19 The Architecture of the International Monetary System
20 Epilogue
What can a government, or several governments acting together, do to improve upon macroeconomic outcomes? Having
worked through two-thirds of this book, the answer to that question might seem obvious to many readers, but the right
answers are in fact never always clear. While governments are obviously always ‘larger’ than individual economic agents,
it is important to ask: Can they really always influence their environment in predictable ways? Should they? Will they?
The following four chapters deal with different aspects of these questions. Chapter 16 is concerned with the potential
for aggregate demand management policies, i.e. government actions which steer the economy’s path closer to its
potential by moving the AD curve. In Chapter 17, particular aspects and issues of fiscal and monetary policies are taken
up in detail, regarding the stability, credibility, and political and economic sustainability of such policies. Since moving
only one of the two curves may not be appropriate in all situations faced by policy-makers, it is important to discuss
supply-side policy as well. In Chapter 18, we confront the issues raised in Chapter 3: how can countries improve their
standards of living and their rates of growth?
Chapter 19 examines the scope of policy and policy coordination in the international financial arena. For citizens of the
world, this question has achieved enormous importance through the globalization of capital markets but, more urgently, in
light of the recent banking and credit crises. For Europeans, it has raised questions of an existential nature—what does it
mean to be a European? Are Europeans willing to give up some element of their economic and national sovereignty to
guarantee the survival of their common currency? Finally Chapter 20 concludes with a short history of economic thought in
the eight decades since the publication of the General Theory by John Maynard Keynes.
Demand Management
Policies 16
16.1 Overview
16.2.1 Equilibrium or Disequilibrium: That is the Question
16.2.2 Persistence of Expectations and Underlying Inflation
16.2.3 The Costs of Inflation
16.3.1 Frisch, Slutsky, and the Modern View of the Business Cycle
16.3.2 Uncertainty, Policy Lags, and the Friedman Critique of Demand Policy
16.3.3 Politics and Political Constraints on Demand Management Policy
16.4 Beyond Controversies: the Synthesis
16.4.1 The Evidence is in the Middle
16.4.2 Macroeconomic Policies Face Limits
16.4.3 The Synthesis
16.5 The Great Recession and Demand Management: New Challenges or Old Dilemmas?
16.5.1 The Policy Challenge
16.5.2 The Policy Response
16.5.3 Diagnosis and Lessons for the Future: Fiscal Policy
16.5.4 Diagnosis and Lessons for the Future: Monetary Policy
Summary
Policymakers are unable to predict with great confidence even how (or how quickly) their own actions are likely to affect the economy. In short, if making monetary policy is like
driving a car, then the car is one that has an unreliable speedometer, a foggy windshield, and a tendency to respond unpredictably and with a delay to the accelerator or the brake.
Ben Bernanke1
16.1 Overview
Writing in 1930, John Maynard Keynes yearned for the day when economists ‘could manage to get themselves thought of as humble, competent people on a
level with dentists’. Since Keynes and the Great Depression, policy-makers have hoped that macroeconomic management of the business cycle would be no more
complicated than getting a tooth filled. In this chapter, we explore the theoretical possibility of demand management policy, and address a number of factors that
limit what macroeconomic policy can achieve. In the 1970s, attempts to manage the macroeconomy ended with high levels of inflation, unemployment, and public
indebtedness. The painful medicine administered in the early 1980s to combat those problems left a bitter taste and little stomach for experimentation with policy
activism.
The global financial crisis that erupted in 2008 presented a fundamental challenge to policy-makers: How to avoid a replay of the Great Depression of the
1930s? The success of those efforts is mixed at best, especially in Europe. Figure 16.1(a) shows the growth of GDP since 2005 in four of the more successful major
economies, three of which are in Europe. There, the 2009 recession was deep but short-lived; this might be seen as a sign of success. Yet even there, nearly a
decade later, growth is not what it used to be. Panel ( b) shows that several other larger European countries have barely recovered their pre-crisis GDP levels. A
third group of countries in Panel (c) have experienced years of Great Depression-style decline. While Ireland is now undergoing a strong recovery, Greece has
suffered a cumulative output loss of more than 25%.
Is this lacklustre performance a signal that economists are still a long way from routine dentistry? Or is it rather a signal that the patient didn’t follow the
dentist’s advice? Naturally, economists are being blamed, and they blame policy-makers as well as each other. After all, why did economists fail to predict the
financial crisis? Well, for a start, some did. Then, we know from Chapters 7 and 15 that financial markets are naturally volatile and, when they are efficient, their
movements are inherently unpredictable.
These are questions that do not have universally accepted answers. In fact, these are not even new controversies. Many go back to the seminal controversy
between neoclassical and Keynesian economists that started when Keynes launched his celebrated attack on the economic orthodoxy of his time. Over the
decades, waves of new arguments have been brought forward by each side of the debate, only to be undermined by new counter-arguments from the other side.
Yet despite ever-increasing sophistication, the same familiar themes continue to dominate the discussion. First, how quickly do goods and labour markets adjust
to shocks? Second, how do agents form expectations and how quickly are these embedded in their decisions? Third, given that inflation is usually a short-run by-
product of demand policy, what are its costs to households, businesses, and society as a whole? These questions comprise the first section of this chapter.
Fig. 16.1 Real GDP Levels (2005 = 100)
Real GDPs are normalized to 100 in 2005. For comparison purposes, all figures use the same vertical scale.
Sources: Economic Outlook,OECD, June 2016.
The next section deals with the inherently random nature of business cycles and the feasibility of demand management policy. Even if the thorny issues of
scope and credibility of managing demand are resolved, economists must admit that they are not gods. Despite their ability to make pronouncements on
television and advise governments, economists are not oracles either. The prevalence of random influences in the economic system at any point in time will
prevent them from predicting future outcomes with perfect precision, even if they have the correct model in hand. This fundamental uncertainty—shocks as we
have called them—cannot be forecast, by definition. Few people anticipated the oil shocks in the 1970s, and no one foretold the fall of communism and the Iron
Curtain in the late 1980s, the development of the internet, terrorist attacks in 2001 and 2015, or the current banking and sovereign debt crises. These events have
all proven to be portentous in the economic sense, heralding large adjustments of the economy, but were wholly unanticipated by economic actors.
The chapter continues by asking about the nature of the business cycle. Are the shocks which set an economy in motion primarily real, or do they come from
the monetary or financial sphere? Do they originate from the demand or the supply side? While it is not easy to settle this, it is possible to assess particular
episodes, much as a doctor diagnoses a patient with many contradictory symptoms. Economic policy, like medicine, is often as much an art as it is a science.
Finally we ask how demand management fared in dealing with the Great Recession, a great challenge indeed.

16.2.1 Equilibrium or Disequilibrium: That is the Question
Since the publication of The General Theory in 1936, a fierce debate has raged between the Keynesians, who had embraced the Keynes’ way of thinking about
the macroeconomic short run, and those who tend to defend the classical framework of flexible prices and rely on the inherent self-corrective nature of the
economic system. The latter group, known as the neoclassicals, stressed the role of money and monetary developments while de-emphasizing the usefulness of
fiscal policy. Early debates centred on arcane issues such as the slopes of curves, but the contemporary debate focuses to a large extent on the question of
nominal price and wage rigidity. While there is near-universal consensus that prices and wages are sticky, disagreement persists about the extent of that
stickiness and its relevance for shaping the evolution of output.
Fig. 16.2 The Neoclassical Case

An aggregate demand expansion—an increase in the target inflation rate under flexible exchange rates or expansionary fiscal policy under fixed exchange rates—moves the
economy from point A to point B in the short run. In the case that a higher target inflation rate has been chosen, the economy lands at point C in the long run. If deviations from
trend y – are short-lived, the move is actually from point A to point C, which is not really helpful. It is more desirable to aim at point D, where inflation is low.
The debate is neatly illustrated in Figure 16.2. Starting from point A, an expansionary demand policy (an increase in the target rate of inflation under flexible
exchange rates, or increased spending and/or tax cuts under fixed exchange rates) takes the economy first to point B. Under flexible exchange rates, the output
effect is dissipated and the economy ultimately comes to rest at point C, where the effect is entirely absorbed by a permanently higher rate of inflation and
exchange rate depreciation. Under fixed rates, the economy returns to A; only if the world simultaneously chooses a higher inflation rate consistent with C will the
long-run inflation rate be higher. Regardless of the regime, the central question for the usefulness of demand policy depends on how much time the economy
spends at B.
The path taken from B to C depends on the speed at which the short-run AS curve shifts. This in turn depends on how the underlying inflation rate π ∼ reacts
to actual inflation. As discussed at length in Chapters 13 and 14, the underlying rate of inflation reflects the forward-looking expectations of wage- and price-
setters, who are free to set prices at the present, the expectations of future inflation embedded in wages and prices set in earlier periods but not free to change,
and elements of contracts which are simply based on past inflation, such as wage and price indexation. Quick adjustment means a rapid transition from Ato C, in
which case policy only creates inflation, at best with a temporary boost to GDP. If the economy spends most of the time on its long-run aggregate supply
schedule (or curve) LAS anyway, it follows that a superior solution is simply to aim at point D, where inflation is low. In this view, the move from A to D is very
rapid and low inflation can be achieved at little or no output cost. This is the neoclassical perspective.
Keynesians tend to see the economy as spending a lot of time away from its LAS schedule, usually to the left of it—i.e. below potential. As a result, they
frequently endorse activist demand policies aimed at returning it to that long-run potential. They are interested in avoiding situations like the one
corresponding to point A in Figure 16.3, where output is below its trend level, and unemployment is above its equilibrium level. By construction, actual inflation is
below its underlying rate at point A. If nothing is done, the underlying inflation will decline over time and the short-run AScurve will shift downwards until point
C is reached. In the Keynesian view, this adjustment may take a very long time, during which unemployment is high and output is lost. This means frustration
among the unemployed, and inefficiency as productive resources remain underutilized. The preferred solution for Keynesians is to pursue an expansionary
policy, shifting the aggregate demand schedule out to AD ' and bringing the economy quickly to point B.
The disagreement boils down to a simple question: can the economy stay away from the vertical aggregate LAS schedule for a long time? Put differently, how
quickly does the short-run AS shift to bring the economy to rest on the vertical long run aggregate supply curve? Neoclassicals argue that these shifts occur too
fast for policy to make much difference, or that policy is wasteful or unnecessary because GDP is never far from its growth trend. They assert that the forward-
looking component of underlying inflation dominates the backward-looking component. Keynesians disagree. It is remarkable that so sharp a divergence about
the desirability of policy actions arises from an apparently narrow dispute concerning what seems to be a point of detail. Indeed, it is possible to agree completely
about the analysis of demand and supply—as presented in Chapter 14 and recalled here—and yet disagree on every aspect of economic policy simply because
there is some doubt about the speed at which the AS curve shifts.
16.2.2 Persistence of Expectations and Underlying Inflation

The speed at which the AS curve shifts depends on the relative importance of the forward-looking component of underlying inflation. This raises the question of
whether this component ultimately correctly anticipates developments that will determine the future inflation rate. For some time, Keynesians argued that people
learn slowly about events and therefore adjust their expectations very gradually. Put differently, the forward-looking component of underlying inflation was
thought to be driven by the past and not to contain information that fundamentally differs from the backward-looking component.
Fig. 16.3 The Keynesian Case
When the economy is below trend at point A, waiting for prices to adjust to reach point C may take a long time. An expansion of aggregate demand can return the economy to full
employment at point B.
Neoclassical economists achieved a great victory when they argued that the only consistent way of thinking about expectations is to treat them as rational.
The rational expectations hypothesis discussed in Chapter 6 posits that people do not make systematic errors. They may occasionally underestimate future
inflation, and then overestimate it, but on average they get it right. If they do, and if the forward-looking component of underlying inflation dominates, then the
AScurve will move fast. Two implications follow. First, on average, the economy will always be close to, if not exactly spot on, the LAS schedule. Second, the
only departures from the LAS schedule will be the result of random expectation errors, which does not provide much room for policy to play a useful role. Indeed,
in that case policy must rely entirely on surprises and errors. If surprises can be engineered now and then, they cannot be systematic, simply because there is no
such thing as a systematic surprise. As for expectation errors, they can only be exploited by policy-makers if they know better than the public at large. If that is
the case, one obvious and democratic solution would be for the government to share that knowledge with the general public.
The rational expectation hypothesis is intellectually compelling, not because it sees people as cold and heartless automatons, but rather because it rules out
systematic (and possibly downright stupid!) mistakes on the part of economic agents. For instance, it is hard to accept that workers and firms would consistently
underestimate inflation when it is rising systematically, because so much is at stake. In every country in the world, banks and economic consultants produce and
sell forecasts to firms, trade unions, the media, and even the government. Their services are valued if they are right, or at least not systematically wrong. And, as
they all compete for the same customers, it is reasonable to expect that those that are systematically wrong would eventually go out of business.
Modern Keynesians are usually willing to concede this point, but with two responses. First, they note that underlying inflation consists not only of expected
inflation. Even if all agents perfectly anticipate the future, they may have signed nominal contracts in the past that lock them into prices and nominal wage
increases, which were in turn based on rationally expected inflation in those earlier periods. Thus, when inflation deviates from these older expectations, there will
be some interval of time when agents can do little or nothing about their errors. During that interval, which can be long-lasting, the economy deviates from trend
output. Second, the speed at which the short-run AScurve moves depends on the speed at which actual inflation—i.e. prices—reacts to expected inflation. Once
again, price stickiness is at the heart of the debate.
This discussion is far from simply academic. The question of how long economies can deviate from their long-run potential is a central practical issue for reform
and stabilization programmes. For many decades the problem of high inflation plagued developing countries. It is often difficult to establish a credible anchor for
agents’ expectations. Put differently, when inflation has been more than 10–20% per year for many years, underlying inflation tends to reflect stubborn
expectations that the central bank and the government can do no better. In such cases, a fixed exchange regime, a so-called hard-peg, may convince the
population that the government intends to change its ways. The most extreme version of this idea is to join a monetary union. Box 16.1 describes the failure and
success of Argentina und Bulgaria, which both chose an exchange rate anchor to pin down inflationary expectations and undermine the forward-looking
component of underlying inflation, even at the cost of giving up monetary autonomy, the cost of having fixed exchange rates as stressed in Chapter 12.
16.2.3 The Costs of Inflation

Figure 16.3 illustrates an important drawback to the Keynesian policy response of demand management. It is accompanied by higher inflation. Unless ‐
contractionary policy is pursued with the same vigour as expansionary policy, demand management policies can easily take on an inflation bias. Some politicians
argue that this is an acceptable price to pay. Couldn’t lower joblessness and higher output outweigh the inconvenience of permanently higher inflation? The cost
of inflation is another source of disagreement between Keynesians and neoclassicals. Inflation is certainly undesirable, yet it is surprisingly difficult to identify
and quantify those costs. Later we present a list of potential reasons why inflation bothers economists and possibly reduces people’s well-being.
Box 16.1 The Exchange Rate Anchor: Argentina and Bulgaria
An exchange rate anchor is potent medicine in economic policy, but it can be tricky to administer. It can work beautifully, but it can backfire if taken for too
long and with inappropriate supporting measures. Argentina offers an example of a success story that ended very badly. Throughout its modern history,
this country has been crippled by high and ever-rising inflation rates, reaching 10,000% per annum in 1990. Powerful vested interestsacting in the
background ensured that the central bank could not or would not impose the required discipline. In early 1991, the Menem Government passed the
‘Convertibility Law’, which tied the peso to the US dollar at a remarkably visible rate of 1:1. Figure 16.4(a) shows that inflation quickly declined. By 1993, it
was about the same as in the USA, and stayed that way for the rest of the decade.
The Argentine cure, however, was directed too much at the symptoms and not enough at the root causes of the inflation disease. In the late 1990s,
individual provincial governments began to run large budget deficits that added up to current account deficits for the nation as a whole.2 Mounting losses
of foreign exchange reserves raised concerns among international investors, and soon after-wards capital outflows added to the loss of reserves. In late
2001, facing acute economic and political instability, the government resigned. Its successor immediately abolished the Convertibility Law. The exchange
rate promptly lost two-thirds of its value, output fell by more than 10%, and inflation was on the rise again.
In contrast, Bulgaria represents a success story, driven by an intense desire to enter the European Union. After the collapse of Communism and central
planning in the early 1990s, the country was plagued by acute political instability. Inflation soared and the value of the leva, the local currency, virtually
collapsed. By 1997, its value was less than one-hundredth of what it was in 1991. A year later, the exchange rate was pegged to the German mark—and
later to the euro. Inflation promptly declined, as Figure 16.4(b) shows. Despite considerable difficulties posed by the recent financial crisis, Bulgaria
remains committed to a fixed exchange rate regime.
Fig. 16.4 Exchange Rate Anchors in Argentina and Bulgaria, 1992–2010

In Argentina, the exchange rate vis-à-vis the dollar was fixed at the one-for-one parity in 1991. The resulting constraint on monetary policy led to a massive decline in the
inflation rate, which even turned negative for several years. When the peg was abandoned in late 2001, the exchange rate collapsed and, soon afterwards, inflation soared.
After a period of punishing inflation, Bulgaria fixed its exchange rate in 1998. As Panel (b) shows, inflation promptly disappeared.
Source : IMF, International Statistics.
Income and wealth redistribution

Inflation has important redistributive effects—it shifts income and wealth from some groups of the population to others. To start with, inflation often distorts
relative prices. When prices rise rapidly, even small differences in rates of increase of particular goods or of wages can lead to dramatic relative price changes. If
labour unions are powerful, real wages will tend to stay ahead, which hurts firms’ profitability, eventually deterring investment and harming growth. Those on
fixed incomes and with limited political clout, such as pensioners or social benefit recipients, do not keep up. Real exchange rates tend to swing widely when
inflation is high. This shifts income between local and foreign producers, and between the local producers of traded and non-traded goods since exchange rates
exert strong influence on traded good prices but leave non-traded good prices relatively unaffected. Box 16.2 presents a truly spectacular example.
In addition, inflation redistributes wealth. When inflation comes as a surprise, it changes the real value of nominal assets such as money, conventional
government and corporate bonds, and bank accounts. In contrast, real assets—such as housing and land, durable goods, foreign exchange, and precious metals
—tend to maintain their value in the face of changing inflation. When actual inflation lies significantly above the underlying or expected level, real interest rates
can become negative, and wealth is redistributed from lenders to borrowers. Hyperinflations, in particular, can leave a legacy that survives many generations:
almost a century after the terrible hyperinflation of the early 1920s, Germans still consider inflation as an absolute evil.
Box 16.2 Switzerland: The Safe Haven
As the financial crisis hit the world, panicked investors looked for safe havens where they could park their wealth. Despite the near-failure of its largest
bank, Switzerland continues to represent a haven of stability and security. Figure 16.5 shows that the Swiss franc started to appreciate vis-à-vis the euro in
late 2008. By September 2011, when the Swiss National Bank decided to peg the exchange rate, the franc had appreciated by more than 40%. As a result,
foreign imported goods became cheaper, which forced domestic competing firms to also cut prices. By contrast, the prices of non-traded goods are
sheltered from foreign competition. The figure shows that while housing rents—houses are the ultimate non-traded good—kept increasing along the
previous trend, the overall consumer price index (CPI) abruptly stabilized and even declined. Producers of traded goods complained bitterly, while firms
producing non-traded goods and services felt little or no pressure at all. Foreign firms exporting to Switzerland also profited to the extent that they could
maintain Swiss franc prices unchanged and pocket much of the 40% revenue increase in their own currencies.
As described in Box 12.4, the Swiss central bank came under heavy pressure to prevent further appreciation by adopting a one-sided peg, allowing the
Swiss franc to depreciate but not to move below 1.20 franc per euro. Committed to preventing any further appreciation, the Swiss National Bank intervened
heavily in the foreign exchange market. Even though it sterilized its interventions, the central bank grew increasingly alarmed by rapid increases in its
foreign exchange reserves. In January 2015, it abandoned its one-sided peg, which led to a new round of exchange appreciation. The CPI responded by
declining steadily while rents kept increasing. Once again, the central bank faced a chorus of bitter complaints from the domestic retail sector and
exporters.
Fig. 16.5 The Swiss Franc and Prices for Traded and Non-traded Goods, 2007–2016
Source: Swiss National Bank and Swiss Federal Statistical Office.
Uncertainty and the value of price signals

In a market economy, prices play a central role as signals. They tell producers what and how to produce, consumers what and how to consume, whether to save,
etc. In particular, the relevant signal is the relative price of one particular good measured in terms of another—not the price of goods in terms of pounds, euros, or
dollars. Understanding these relative price signals is crucial to the efficiency of a market economy. It is here that high inflation can inflict great damage on the
ability of market economies to function. This is because high rates of inflation usually are associated with more variability of inflation. The more variable inflation
is, the less confidence firms and households have that nominal price movements they observe represent changes in real economic opportunities. Firms and
households underreact to true relative price signals, because it is difficult to recognize those signals when all nominal prices are changing and do so in a highly
variable manner—they find it hard to distinguish movements in the price level from changes in relative prices.
Fig. 16.6 Variability of Inflation, Real Wages, and Unemployment: OECD Countries, 1960–2010
For the past 50 years, higher average rates of inflation in the OECD economies have been strongly associated with more volatile inflation rates (correlation coefficient: 0.95).
More variability of inflation rates is also accompanied by more volatile real wage growth (correlation= 0.46) and more volatile unemployment rates (correlation = 0.43), but these
linkages are weaker.
Sources: IMF; OECD, Economic Outlook; Bureau of Labor Statistics.
Convincing evidence for such distortionary effects of inflation can be found in Figure 16.6. Panel (a) documents the positive link between inflation level and
variability. Panels ( b) and (c) show that more variable inflation is positively associated with more variable real wages and unemployment. Nominal fluctuations
have real effects. When the efficient functioning of the price mechanism is attenuated, overall productivity declines, eventually resulting in lower growth and
higher unemployment. Indexation—the contractual linking of individual prices or wages to the overall rate of inflation—can worsen matters because it tends to
freeze relative prices and to lock in real rigidities. This further dulls the reaction of economic agents to relative price signals.
The value of money

While indexation may hurt the efficiency of the price system, it is often seen as one way to prevent income and wealth redistribution due to inflation. This is
because richer and better educated individuals are better equipped to protect their wealth, while the poor often have no obvious mechanism to prevent their real
wages and social transfers from eroding. If all nominal values—all prices and wages, but also nominal interest rates, asset prices, and the exchange rate—are
indexed, the losers are simply those who hold money which, by definition, is not indexed. The losses suffered when the value of money declines steadily—more
frequent trips to the bank—may seem trivial. Yet examples of hyperinflation in history show that the consequences can be highly disruptive. Box 16.3 describes
these costs and explains the concept of an ideal or ‘optimal’ rate of inflation.
16.3.1 Frisch, Slutsky, and the Modern View of the Business Cycle
An important aspect of demand management policy is that it be feasible—that the central bank or the finance ministry can actually execute it, and do a good job at
it. An important controversy between Keynesians and neoclassicals concerns the ability of governments to conduct policy effectively. The most important
reason for scepticism is the nature of the cycle itself.
Economists’ initial optimism about demand management policy was driven by a belief that the economy was more or less deterministic, meaning that the
present depends mechanically on the economic history of the past. Business cycles do in fact exhibit important regularities and seem to follow an internal logic.
Throughout the history of economic thought, economists have sought mechanisms by which each cycle sows the seeds of its successor, so that cycles go on
reproducing themselves, just like the rising and falling of the tides. This would be the case if the economic system constantly generated forces that successively
speeded it up and then slowed it down. It turns out that it is quite possible to imagine how this can come about. And it would be easy to imagine how a dose of
demand policy, properly administered, could cancel the wave to produce a flat growth path.
This optimistic view was badly shaken by the discovery that cycles can be generated from purely random, or stochastic, factors. This stochastic view of
business cycles can be traced to the work in the late 1920s and early 1930s by the Russian Eugen Slutsky (1880–1948) and the Norwegian Ragnar Frisch (1895–
1973).3 To understand this important discovery, consider an economy which starts away from its steady-state equilibrium path but returns to it over time—for
example, the growth model of Chapter 3. The question to ask here is: how did it get off that path in the first place? Slutsky and Frisch reasoned that economies are
constantly subject to random disturbances, or shocks. These might come from demand (e.g. alternating optimistic and pessimistic entrepreneurial animal spirits or
the consumer mood, or policy actions) or supply (e.g. exceptionally bad or good crops and natural disasters; important inventions or discoveries like the steel
furnace, railways, electricity, or computers, as well as minor ones; social unrest). The list of potential shocks is endless. Like drops of rain generating ripples on a
lake, these shocks constantly buffet the economy, moving it away from its previous position.
Box 16.3 Optimal Inflation
The late Milton Friedman argued that the optimal rate of inflation should be negative—a deflation, when the price level falls steadily. More precisely,
Friedman argued that the optimal inflation rate should equal the negative of the real interest rate. By the Fisher principle, the nominal interest rate should
be set to zero.4 Friedman’s radical argument rested on a principle that production of a good should be expanded as long as the marginal benefit, i.e. what
consumers and firms are willing to pay for it, exceeds the marginal cost. This principle can be readily applied to money produced by a central bank.
Chapter 9 showed that the opportunity cost of holding money is the nominal interest rate. The cost of creating money turns out to be virtually nil (the
production of even a €100 banknote only costs the central bank or minting company a few cents). In theory, the state can make its citizens better off by
providing them with the public good of money, which costs virtually nothing to make, at no cost. Naturally, if it did so, interest rates would have to be very
low. As the argument goes, they should be zero.5
In principle, Friedman was right. Yet for a number of good reasons, Friedman’s prescription has never been taken up in practice. During most of the
2010s, inflation has been virtually nil throughout Europe, and even negative in some countries, and few have found this to be a desirable state of affairs.
Why not? To start with, deflation means that goods and services will be cheaper tomorrow than today—why buy anything now when prices will be lower
next year? This pattern has long been visible in Japan, as it suffered a prolonged slump since the early 1990s. Second, it is hard to distinguish between
price declines due to generalized deflation from price declines of some prices of goods that reflect sharp productivity advances, such as in the case of
many electronic goods. Finally, it is a fact of life that almost all goods and services are taxed, so why should money be treated any differently? Inflation is a
tax on money holdings, and has a role to play alongside all other taxes. In fact, in many countries where tax collection is weak, inflation is one of the very
best sources of tax revenues that can be tapped. For these countries, it may even be efficient to have significant inflation. In any case it is hard to make the
case for zero inflation or deflation in practice.
As already noted in Chapter 10, many central banks around the world have adopted inflation targeting, meaning that they publicly state and commit to
some inflation rate as an objective of policy. The underlying presumption is that inflation is acceptable when people stop worrying about it. Table 16.1
presents their targets for 2016. Typically, these central banks aim at inflation rates around 2%. This is also the inflation rate which the European Central
Bank associates with ‘price stability’.
Table 16.1 Inflation Targets Set by Inflation-Targeting Central Banks in 2016
Australia Brazil Canada Chile Colombia Czech Republic Hungary

20%–3% 4.5% ± 2% 2% ± 1% 3% ± 1% 3% ± 1% 2% ± 1% 3% ± 1%
Iceland Israel Rep. of Korea Mexico New Zealand Norway Peru
3% 1%–3% 2% 3% ± 1% 2% ± 1% 3% 2% ± 1%
Philippines Poland South Africa Sweden Switzerland Thailand United Kingdom
3% ± 1.0 2.5% ± 1% 3%–6% 2% less than 2% 2.50% ± 1.5% 2%
Sources: National central banks.
These shocks, often referred to as impulses, change the demand or supply conditions in the economy. Once randomly disturbed, the economy embarks on a
deterministic adjustment described in the previous section, until the occurrence of the next shock. The impulse-propagation mechanism transforms, or
cumulates, these random impulses into irregular cyclical oscillations. Even though the response to individual shocks is damped, old oscillations are constantly
being supplemented by new ones corresponding to more recent shocks. In the vision of the business cycle shown in Figure 16.7, it is sufficient to accept the view
that an economy is regularly buffeted by endless series of shocks, and never really settles down to its stationary state. The rest of this chapter follows on this
lead, the cornerstone of modern business cycle theory.
Fig. 16.7 The Impulse-Propagation Mechanism
Random shocks hit the economic system, which produces business cycles. The cycles result from the averaging or cumulation of these random disturbances over time.
How likely is it that purely random impulses working their way through the economy actually generate the kind of behaviour corresponding to our earlier
stylized facts? Can shocks which continuously move the economy away from its steady state and are followed by its internal dynamic actually explain business
cycles? Box 16.4 recalls how this question received an affirmative answer at the dawn of the computer era. Figure 16.8 shows how the AS–AD model can
transform random shocks into realistic cycles. In Panel (a), 200 purely random shocks drawn over time are plotted.6 Such observations—truly random impulses—
could represent a succession of unforeseeable events, large or small. Panel (b) shows what happens when these shocks are ‘filtered’ by the propagation
mechanism implied by a version of the AS–AD model described more precisely in Box 16.5. The result is a succession of artificial business cycles which resemble
those of, say, the UK data presented in Figure 1.5. Although the impulses themselves are not cyclical at all, the transformed variable exhibits irregular, periodic
movements which resemble business cycles encountered in reality.
Fig. 16.8 Impulses and Propagations: An Example

Panel (a) depicts 200 observations of a random variable, εt. Panel (b) displays data from Panel (a) after they were transformed, or ‘filtered’, by the formula Y t = 1.3 Y t−1 −0.4
Yt−2+ ε t, starting from a given Y0 and Y −1. This ‘filter’ has the ability to generate realistic-looking data series, working only with purely stochastic influences. It is the solution to
one particular version of the AS–AD model studied throughout this book.
The impulse-propagation mechanism is now the dominant way of thinking about business cycles, because it accords well with the stylized facts. Box 16.5
explains in detail how the AS–AD model provides a natural framework for thinking about how cycles arise. Since they are random, the shocks typically generate
cycles of different sizes and magnitudes. If many of these impulses are related to permanent technological innovations, not only will they trigger cycles, but in the
long run they will also cumulate into a process of long-run growth. This is consistent with the observation that in the long run, the growth process (the
accumulation of positive shocks) dwarfs business cycles (the reaction to individual productivity and other shocks).
Box 16.4 Computers, Scientists, and Business Cycles
Although the path-breaking work of Slutsky and Frisch on random events and cycles took place in the 1930s, hard evidence that random influences can
generate cyclical behaviour resembling actual economies was not forthcoming until the late 1950s, when computers became widely available to
researchers. Frank Adelman, a nuclear physicist, and his wife Irma, an economist, studied an economic model developed by Lawrence Klein, the 1980
Nobel laureate from the University of Pennsylvania, and Arthur Goldberger of the University of Wisconsin. (This model consisted of 25 equations
summarizing the most important macroeconomic relationships in the US economy.) First, the Adelmans found that the Klein–Goldberger model could not
generate a cycle on its own because the fluctuations died down after a few years. Yet, when continually perturbed by random shocks, it generated data that
had the same statistical properties as actual US business cycles. They concluded:
Ever since the path breaking article of Frisch on the propagation of business cycles, the possibility that the cyclical movements observed in a
capitalistic society are actually due to random shocks has been seriously considered by business cycle theories. The results we have found in
this study tend to support this possibility…The agreement between the data obtained by imposing uncorrelated perturbation upon a model which
is otherwise non-oscillatory in character is certainly consistent with the hypothesis that the economic fluctuations experienced in modern, highly
developed societies are indeed due to random impulses. (Adelman and Adelman, 1959: 620)
The Adelmans simulated, or solved, their model 100 years into the future on an IBM 650 and were proud that ‘computations for one year could be
made during an operating time of about one minute’. Now that simulations take seconds or less on a laptop, this approach has become routine.
Much effort has gone into improving and enlarging models and the algorithms used for simulations. More recently, Robert E. Lucas of the
University of Chicago and 1995 Nobel Prize laureate in economic sciences, made this research strategy explicit:
Our task as I see it … is to write a FORTRAN [a programming language] program that will accept specific economic policy rules as ‘input’ and
will generate as ‘output’ statistics describing the operating characteristics of time series we care about, which are predicted to result from these
policies. (Lucas, 1980: 709–10)
In the last decade, the economics profession has taken Lucas’s challenge to heart. Rather than isolated policy interventions, central banks and
finance ministries are more likely to evaluate the macroeconomic effects of policy rules—Taylor rules for monetary policy, and tax and
stabilization rules for fiscal policy, for example.
16.3.2 Uncertainty, Policy Lags, and the Friedman Critique of Demand Policy
For policy, fundamental uncertainty behind the business cycle is bad news. It necessarily implies that governments will never be able to completely iron out the
cycle. The mere fact that markets fail to adjust to shocks quickly due to nominal rigidities does not imply that governments can offset them effectively. The same
uncertainty or incomplete information that prevents markets from making the best possible use of available resources may also hamper government actions.
Furthermore, private actors in the economic system such as firms and wage negotiators have a strong incentive to discover their mistakes promptly and take
corrective action. It is not obvious that governments face the same incentives or, even if they do, that they can react as quickly.
Box 16.5 Impulse-Propagation Mechanisms in the AS–AD Model
Chapter 14 showed how the AS–AD model can help us understand the determinants of output and inflation in the short and long run. It can also be used
to study business cycles. The point of departure is to identify the shocks as factors that shift either the AS or AD curves. Demand shocks shift the AD
schedule, while supply shocks affect the position of the AS curve. Both demand and supply shocks can be positive or negative. Positive shocks, for
example, move the relevant schedule rightwards; negative shocks shift them to the left. Lags in economic relationships are the central source of dynamics
in the AS–AD framework. Lags exist for various reasons: slow responses of demand to income and of supply to demand imply that the AD schedule reacts
gradually to demand disturbances. Consumers smooth spending relative to income, and adjustment costs may slow the response of investment to new
opportunities. The supply side may also be a source of persistence, or dependence on the past, if underlying inflation catches up only gradually with past
inflation.
As an example, consider a permanent increase in the target inflation rate of the central bank operating in a system of flexible exchange rates.7 Assume
that underlying inflation is equal to last period’s inflation and start with an economy already in a stationary state at point A. (This is a simplifying
assumption which abstracts from other cycles already under way that would interfere with the one under study.) The cycle triggered by this disturbance is
tracked in Figure 16.9. The increase in the target inflation rate leads to a reduction in the nominal interest rate and, therefore, an increase in aggregate
demand, which will materialize gradually. Initially, this takes the economy to point B in Panel ( a), with an increase in both output and inflation. Owing to
demand lags, the initial shift of the AD curve to AD' represents only an initial, partial response. For a time, the AD schedule will continue to move rightwards
towards AD''
The AS curve will also shift upwards to AS' as underlying inflation gradually catches up with the actual inflation rate. As this happens, the economy
moves towards point C. With actual output still in excess of its trend level, the AS curve keeps rising. At point D along AS'' inflation is higher than the target
inflation rate. The Taylor rule leads the central bank to raise its interest rate and actual inflation then starts to decline, followed by underlying inflation and a
downward shift of the AS schedule to AS'', which is drawn to pass through point E, the final resting place where inflation has reached its target. It is
probable that, along the way, the AS curve will initially move below AS'' and then back above it in successive moves towards AS''. The adjustment is thus
characterized by oscillations—the ‘loops’ shown in Panel (b), with alternating periods of output above and below equilibrium level, i.e. business cycles.8
Fig. 16.9 The Propagation Framework in the AS–AD Model

A demand shock, for example the adoption by the central bank of a higher inflation target, triggers cyclical movements as the result of the interplay of lagged effects. In the
long run, the economy moves from point A to point E, with possible fluctuations in the intermediate period.
Indeed, macroeconomic economic policy must contend with a number of lags. First, a recognition lag hinders policy-makers’ ability to intervene effectively.
This is simply the time needed to discover that some policy intervention is required. As the previous section suggests, there is good reason to believe that
fundamental uncertainty will always be present, and governments will be just as surprised as private agents. Next, governments need time to formulate policy.
This is called the decision lag, and is usually associated with the details of parliamentary democracies. Depending on the government structure, this can be
coupled with an implementation lag, as ministries must originate, and parliaments must pass, legislation. Even if implemented quickly, policies need time to
produce their effects. This is especially true of monetary policy. It can take several quarters before the easing of money market conditions and depreciating
exchange rates have an impact on real activity. To make matters even worse, the duration of this effectiveness lag is quite variable, ranging from a few months
up to two or three years.
Fig. 16.10 Lags and Demand Management Policy
The blue line shows the business cycle arising from fluctuations in aggregate demand in the private sector. A demand management policy correctly implemented would smooth
out the fluctuations (curve A). If there are significant effectiveness lags, the government may need to enact these measures well in advance of the turning points in the cycle. If
instead the government reacts passively, it may simply reinforce the cycle (curve B).
How should governments react to long and variable lags in economic policy-making? One possible reaction is to do nothing at all! This is what Milton
Friedman argued in the 1950s. According to the Friedman critique, illtimed policy interventions may actually worsen the cycle. Under the best of
circumstances, there is no guarantee that policies will be implemented in time, or achieve the stated objectives. If the government has no information advantage
over the private sector, it might even do more harm than good. The best policy that governments can pursue is a steady hand, rather than making matters worse
by adding uncertainty.
Figure 16.10 illustrates the point. The blue line represents the path of output subjected to business cycle fluctuations that would arise if the economy were just
left to itself. A perfectly thought-out and implemented policy would begin to stimulate the economy just when it is nearing a peak so that, given the various lags
described, its effects come into play just at the time it is needed. Similarly, it would turn restrictive just when the trough is passed, so as to moderate the strength
of the upturn. Ideally, the outcome would be curve A. Now add uncertainty about the recognition, decision, implementation, and effectiveness lags. In the worst
case, we face the risk of causing outcome B. The stimulus planned for the downturn affects the economy exactly when it is coming out of the recession, while
restraint comes into play just when the economy has peaked. Here, policy makes things worse!
Box 16.6 Leading and Lagging Indicators
Central banks and ministries are in the business of forecasting GDP. National budgets depend on tax revenues, and tax revenues depend on economic
activity. Not only do governments need to get ahead of the curve when making policy, they also need to communicate overall economic trends to the public.
In Chapter 1 we learned that many variables move together over the cycle, or are coincident (e.g. consumption, investment, and interest rates). Yet if one
variable systematically leads or lags another, this can help policy-makers anticipate turning points in the cycle and reduce the bite of the Friedman critique.
The Burns–Mitchell diagrams in Figure 16.11 display the cyclical behaviour of a number of frequently tracked macroeconomic variables, averaged over
eight OECD economies.9 Recall that the horizontal axis of the Burns–Mitchell diagram indicates the number of quarters before (−) or after (+) the peak of
the reference cycle for GDP. The vertical axis generally displays the value of an index which takes a value of 1.0 on average over the reference cycle (in
some cases we display percentage deviation from an average value). The diagrams show average tendencies, and not perfect regularities. Evidently,
some variables tend systematically to peak ahead of output, while others systematically trail behind it. For example, stock prices, real money balances, the
primary current account, the real exchange rate, vacancies, and inventory investments are all examples of leading indicators. They tend to predict the
emergence of recessions and expansions. Variables which attract the most attention in the press such as employment, unemployment, and inflation tend
to be lagging indicators. Investment, short-term interest rates, and capacity utilization are examples of coincident indicators, although in some countries
they provide a warning of a quarter or two before a downturn begins.
Our knowledge of the macroeconomy can account for these regularities, and you are asked to do so in the exercises at the end of this chapter. Financial
variables are useful leading indicators because they are forward-looking. Share prices generally decline three to four quarters in advance of a downturn.
Since real stock prices are a measure of Tobin’s q, financial markets anticipate the next phase of the cycle and correctly expect poorer profitability during
economic downturns. The real money supply declines or stops growing six months to a year in advance of the onset of a recession. The real exchange
rate appears—on average—to depreciate sharply five to six quarters before the onset of a recession. It even appears to overshoot, and starts to
appreciate, just after the business cycle peak is reached.
At the same time, the problem may not always be as severe as Friedman’s critique makes it out to be. Returning to Keynes’ metaphor, we might say that if the
dentist is called early enough and allowed to act decisively, he may prevent the most serious of toothaches. He might also be able to prescribe some medicine
which may take effect automatically and reduce the pain and swelling next time a molar must be pulled. For example, ministries can prepare tax cuts or increases in
advance, in order to cut down on the implementation and effectiveness lags. Governments can be given greater budgetary authority to reduce the time associated
with the decision lag. The adoption of automatic stabilizers is an effective way for legislators and policy-makers to get around the machinations of politics.
Automatic stabilizers, which are presented in Chapter 17, slow an economy down in a boom phase and stimulate it in recessions. The most important aspect is the
automatic part. They need few or no political decisions and can thus evade the political wrangling that may impose a decision lag on an otherwise well-informed
government.
Fig. 16.11 Burns–Mitchell Diagrams, Macroeconomic Indicators in Eight OECD Economies,
1970–2016
Burns–Mitchell diagrams display the average behaviour of key macroeconomic variables centred around the peak or a trough of well-defined business cycles across several
countries. The zero point corresponds to a cyclical peak. For more detailed explanations of the Burns–Mitchell diagram, see Chapter 1.
Sources: OECD; IMF; authors’ calculations.
In order to conduct monetary policy more effectively and independently, central banks have also entered the forecasting business, either generating their own
‘in-house’ predictions or employing outside consultants to help. In practice, national policy-makers use the many indicators of business conditions discussed in
Box 16.6. For example, real stock prices and real money balances are relatively reliable leading indicators, and anticipate downturns by roughly four to six quarters.
The yield curve, capacity utilization, changes in inventory stocks, investment spending plans, and real exchange rates can sometimes prove useful in the
forecasting game, offering somewhat less lead time. The very fact that the lead time of these indicators is frequently less than a year suggests that keeping ahead
of the cycle is difficult.
16.3.3 Politics and Political Constraints on Demand Management Policy

Until now, we have described governments as well-meaning entities that care about the nation’s welfare. Alternatively, we could see governments as distinct
economic agents run by politicians who care about getting elected and staying in power. Policy then becomes endogenous to both economic and political
circumstances, and, reciprocally, economic circumstances are partly shaped by policy actions. The loop is closed and provides interesting, if not always
encouraging, insights into the role and function of demand management policies. Demand management policies may not be employed to stabilize the economy,
even if they are effective. What’s more, their abuse by politically motivated governments can even be a source of business cycles.
If governments care about remaining in office as long as possible, it is natural to expect them to use macroeconomic policies to boost their re-election
prospects. Incumbent governments will try to make sure that the economy is booming and employment is high on election day. Naturally, this might soon be
followed by inflation, so governments will often tighten their policies after being re-elected. Panel (a) of Figure 16.12 shows the prediction from this view of
electoral business cycles.
A different view is that, when in power, a political party pursues a partisan agenda. This agenda can be motivated by ideology or by the coalition of interests
that back the party. The partisan business cycle view predicts that changes of party in power result in policy changes that generate their own cycles. It is
customary to describe left-leaning governments as being Keynesian and right-leaning governments as being neoclassical. Then, if a left-leaning government
replaces a right-leaning government, we should expect to see GDP growth initially increase and unemployment decrease, at the expense of rising inflation and
deficits, which will require a contractionary correction, followed by yet another expansion, etc. This pattern is shown in Panel (b) of Figure 16.12. The opposite
political shift, from left to right, is expected to generate policies that focus more on price stability and are less activist, as shown in Panel (c). The interesting
innovation of the partisan business cycle hypothesis is that by the very nature of the election process, the economy must be subject to a shock after every
election! Rational workers and firms will prepare for an election by expecting some ‘average’ of the parties contending for public office. By construction, they will
be wrong ex post. This shock, combined with the propagation mechanism already discussed, can give rise to business cycles.
The existence of political business cycles is not firmly established, and there are good theoretical and empirical reasons to be sceptical. The general view
that governments can manipulate their macroeconomic instruments to achieve political gain on election day runs sharply against the forward-looking view of
rational expectations adopted in this book. Citizens who care about the future should see through government attempts to manipulate voters’ sympathies. They
should understand, for instance, that a boom designed to arrive on election day will be followed by restrictive policies soon afterwards. Not only should they not
be impressed, but also they could be expected to vote against a government that actually destabilizes the macroeconomy.
Similarly, the idea that partisan governments use macroeconomic policy to favour interest groups is not as straightforward as it may seem. Presumably, voters
span the whole spectrum of opinions, as shown symbolically in Figure 16.13. To be elected, a party must gather support from all the voters who share most of its
views plus the ‘median voter’ in the centre. This median voter, who holds the key to electoral success, does not hold right- or left-wing views, so why should
there be a shift in policies? The only reason is that the median voter is first seduced, only to be cheated upon later. This would explain both partisan policy
swings and the rejection of governments which have served the interests of its constituencies at the cost of incurring the median voter’s wrath. The other
possibility is to form coalitions of left- and right-wing parties that shun the centrist electorate. In that case, the equilibrium of the extremes could still deliver policy
stability.
Fig. 16.12 Political Business Cycles
Panel (a) shows an electoral business cycle. The political business cycle theory predicts that policies are employed to align the peak of the cycle with election dates. Elections
are then followed by austerity-oriented policies. Panel (b) shows the effect of a change from a right-wing to a left-wing government; the opposite change is shown in Panel (c).
In practice, it is probably an oversimplification to presume that left-leaning governments are always Keynesian while right-leaning governments are
neoclassical. When supply-side shocks were dominant, the pendulum shifted towards the neoclassical view. The late 1990s, a period of rising popularity of the
neoclassical views, witnessed the emergence of the ‘third way’ among left-leaning parties in a number of countries (New Labour in the UK, Schröder in Germany,
Clinton in the USA, etc.), while right-leaning parties have adopted the ‘compassionate Conservative’ image championed by politicians like David Cameron and
Angela Merkel. Since 2008, all major political parties in the OECD economies have endorsed some form of fiscal and monetary stimulus for the economy. Given the
magnitude of the global economic demand shock that was associated with the Great Recession, this is not surprising.
Fig. 16.13 Partisan Politics

Voters’ views are spread over the range of ideologies and political preferences. To be elected, a party needs to win a majority, which means its own supporters plus the median
voter.
16.4 Beyond Controversies: The Synthesis

Section 16.2 presented a long litany of ongoing controversies between Keynesians and neoclassical economists. These polar views lead to sharply diverging
conclusions about the role of macroeconomic policies. Is it possible that these disagreements survive 80 years after the publication of Keynes’ seminal work?
There will always be vocal supporters of one of these unreconstructed views, but the weight of the evidence does not support either of the extreme positions. As
a result, common practice is generally more pragmatic. This section briefly reviews the evidence and presents the synthesis that guide actual policies.
16.4.1 The Evidence is in the Middle

The neoclassical view asserts that prices move fast enough to ensure that the economy never strays far from its growth trend and the vertical LAScurve. This is
why they see demand policies as useless at best, possibly even counterproductive. The direct evidence on prices shows that, in normal conditions when inflation
is low or even moderate, prices do not adjust fast. An example is provided in Table 16.2, which is taken from a large international study. Persistence is defined as
the impact of past inflation rates on the current rate, the backward-looking component of underlying inflation in Chapter 13. This coefficient can be between zero
(no persistence at all) and one (total persistence). The table indicates that persistence varies from one country to another but usually falls between 0 and 1. The
exception is Japan that has undergone zero inflation for more than two decades, which makes the question irrelevant.
Table 16.2 Persistence of Inflation
Persistence
Australia 0.53
Canada 0.34
Euro Area 0.60
Japan 0.04
New Zealand 0.57
Sweden 0.35
Switzerland 0.74
UK 0.53
USA 0.39
Sources : Altissimo, et al. (2006).
The same study, and many others, also indicate that persistence varies across industries, which explains differences in persistence across countries by
different patterns of sectoral specialization. For instance, it is generally the case that prices are adjusted more frequently in the industrial sector than in the service
sector.
Neoclassical macroeconomics is rooted in microeconomic principles that suggest that it is irrational for a firm not to adjust its prices quickly in response to
market signals. Looking at the price-setting process in Chapter 13, we saw that prices tend to follow production costs and market signals. How can these views be
reconciled? Various reasons have been advanced and found empirically valid. One reason is related to the battle of the markups presented in Chapter 13. Perfect
competition, as assumed in simple microeconomics, is simply quite rare. Firms are constantly striving to accumulate market power, which allows them to change
prices strategically, balancing many considerations, in order to maximize their profits. Another reason is the existence of ‘menu costs’. These are costs associated
with prices changes, liking printing new menus in a restaurant, or new catalogues in most cases. They also include the relationship with customers who dislike ‐
frequent changes. The study used in Table 16.2 reports that, in general, only 15% of firms change prices in any given quarter. It does appear rational for
producers to react slowly, and possibly deliberately, to market signals and this is indeed what the evidence indicates.
16.4.2 Macroeconomic Policies Face Limits

Price stickiness does not imply that the Keynesian case is solid. It says that there may be room for macroeconomic policies to be effective, but effectiveness is
not guaranteed. We have already seen in Chapter 12 that the exchange rate regime undermines fiscal or monetary policy. Regarding fiscal policy, we have seen in
Chapter 6 that the Ricardian equivalence principle predicts that consumers will alter their saving decisions in a way that offsets government actions. In Chapter
17, we will see that the budget constraint limits the freedom of governments. Regarding monetary policy, we understand that expectations may move faster when
the public at large anticipates interventions. We also observed in Section 16.3 that many lags might undermine what the authorities are trying to achieve.
For all these reasons, governments and central banks have retreated from the policy activism that characterized the heyday of Keynesianism. They certainly do
not attempt to respond to the slightest perturbations of the economy. Both monetary and fiscal policies tend to move slowly and persistently, responding to
major shocks only and maintaining their stances for years in a row. They also recognize that supply-side shocks present them with unpalatable trade-offs, as
described in Chapter 14.
16.4.3 The Synthesis

The many lessons learned over decades leave us with a view that is neither purely neoclassical nor purely Keynesian. The synthesis, which a wide majority of
economists accepts, borrows from both traditions. Neoclassical macroeconomists accept that prices are sticky but insist that macroeconomics cannot be divorced
from microeconomics. The consumption and investment functions must be compatible with optimizing behaviour, and this is indeed the reasoning adopted in
Chapter 8. The Keynesians recognize that this also applies to governments and central banks; for example it underpins the Taylor rule.
This synthesis is now in use around the world. An increasing number of central banks and more governments use large-scale models that incorporate the
principles developed in all the preceding chapters. These models, called Dynamic Stochastic General Equilibrium (DSGE) models, are somewhat
complex and generally require the help of a computer to solve. We now review the characteristics of these models by considering their name in detail. Each word
matters.
General equilibrium means that they incorporate many markets which are simultaneously in some form of equilibrium. This is what we have done, starting with
the goods market and the IS curve. The DSGE models have an IS curve, sometimes broken down in specific markets like industrial goods, services, housing,
equipment, etc., describing optimal decisions by producers and consumers. They also include a TR curve, representing optimal behaviour of central banks as
summarized by the Taylor rule. Finally, they include a Phillips (or AS) curve. Quite often, the Phillips curve is based on the rational decision of firms to change
prices slowly. Firms are described as having some local monopoly power, meaning that they have the choice to set prices because they do not have to blindly
follow ‘the market’.
Dynamic means that the economy does not jump immediately from one (general) equilibrium to another whenever it is hit by a shock. This is captured in several
ways, which should look familiar to readers of this text. The Phillips curve moves slowly over time, responding to rapidly changing forward-looking rational
expectations as well as to slow-moving backward-looking evolution. Demand responds with a lag to exogenous changes, as described earlier in this chapter.
Finally, these models are stochastic. They describe economies constantly subject to random shocks of the types described in Section 16.3. This implies that
policies do not control the economy tightly and that there is considerable uncertainty, which implies that forecasts can never be exactly right, and go badly wrong
when large shocks occur. These random disturbances are introduced on the basis of past experience, which allows the models to mimic reality. However, rare and
large shocks stand to invalidate the model occasionally. This is what happened at the time of the Great Financial Crisis that started in 2007, as shown in Table 1.3.
While the DSGE models represent a Keynesian approach similar to the AS–AD model, they were the outgrowth of a research agenda that strove to explain
cyclical fluctuations solely as the result of real shocks to an economy—an economy in which money was neutral or was completely absent. These shocks were
essentially shocks to total factor productivity. This agenda of real business cycle (RBC) theory was not just intellectual curiosity. The goal was to ask
whether cycles generated in a competitive economy of purposeful, maximizing agents could resemble those observed in the data. In the end, the RBC approach
failed to deliver the silver bullet for understanding the business cycle. But just as with the Adelmans in Box 16.4, this research served as the shoulders on which
many researchers using the DSGE approach could stand. The current synthesis of views is that real, monetary, and financial shocks all drive the business cycles
of modern economies.
Many people have blamed economists for not predicting the crisis or for failing to produce remotely reliable forecasts at the time, despite their sophisticated
DSGE models. Economists have two answers. First, they explicitly acknowledge that the world is stochastic. Surprises, good and bad, always happen. It is true,
however, that the shocks that DSGE models allow for were based on the experience of the previous three or four decades, and often less because of data
availability. The Great Financial Crisis is a very rare event, the last similar one occurred in 1929, almost 80 years earlier. Surely, models should not allow such
shocks to occur more frequently!
The second answer is that the DSGE models were mostly built in the 2000s. They were new and streamlined and did not pay much attention to the financial and
banking sectors described in Chapters 7 and 10. Since the crisis, most DSGE models have been extended to include both financial markets and banks. Time will tell
whether they behave better, but our study of these markets warns us that these markets are volatile because they are driven by expectations that can shift
dramatically in very little time. The next s ection provides a bird’s eye view of the crisis and the challenges that it has posed to macroeconomics.
16.5 The Great Recession and Demand Management: New Challenges or

Old Dilemmas?
16.5.1 The Policy Challenge

The global financial crisis and the Great Recession that engulfed the world beginning in 2008 was by far the most severe economic event of the past 75 years.
Major stock markets around the world plummeted, losing about half of their value within the space of a year. US GDP fell in 2009 by 2.8%, and Eurozone GDP fell
by even more (4.5%). Unemployment in the US shot up in one year from 5.8% to 9.3%, but also rose throughout the advanced economies of the world, and
especially in Europe.
There is little serious disagreement about the proximate source of the Great Recession: an implosion of banks and a collapse of financial markets in the US,
following a bubble period based on persistent overvaluation of real estate and the derivative securities based on those real assets. These assets were sold
throughout the world and, when the bubble burst, provided a ready conduit for a demand contraction in Europe, Asia, and the rest of the world. While the root
cause of the ‘mother of all bubbles’ is still disputed, it is agreed that a combination of rampant financial innovation, low interest rates set by the Federal Reserve,
and lax regulation of financial markets were essential ingredients. US households lost an estimated trillion dollars of wealth (houses and financial assets), which
triggered a precipitous decline in consumption expenditures and, by extension, GDP. Box 11.5 in Chapter 11 and especially Box 11.2 tell a powerful story.
The collapse of aggregate demand in the US had adverse effects throughout the world. A sharp drop in US imports led to retrenchments of aggregate demand
in other countries. The diagnosis of collapsing demand is easy to make through the lens of the AS–AD model as an inward shift of the AD curve. We know that
when inflation and output move in the same direction, the demand side must have been active; Table 16.3 documents that in the Great Recession, sharp declines
in GDP growth throughout the world were accompanied by equally steep declines in inflation rates.
16.5.2 The Policy Response

Given the clear signal in the data of Table 16.3, it seems like a cut-and-dried case for aggressive demand management policies as discussed in Section 16.2. Yet the
global financial crisis and the Great Recession which ensued represented perhaps the greatest challenge to macroeconomics since Keynes wrote The General
Theory. The overall policy response was not a uniform success.
Table 16.3 Changes in Inflation and Growth in the Great Recession, 2008–2009
Change in Change in
GDP growth, inflation rate,
2008–9 (in %) 2008–9 (in %)
World –3.0 –3.6
Eurozone –4.5 –3.3
USA –2.8 –3.8
Germany –5.6 –2.8
France –2.9 –3.2
UK –4.2 –1.4
Sources: IMF World Economic Outlook Database, April 2016.
The initial reaction to the financial crisis came from a tight, well-developed international network of central banks to provide liquidity to banks, to maintain
orderly trading conditions, and to assist healthier banks take over those which had suffered large losses on securitized mortgage investments. In 2009, the US
government nationalized two mortgage finance firms, Fannie Mae and Freddie Mac, as well as American International Group (AIG), an insurance company that
had issued excessive amounts of credit default swaps (CDS) described in Chapter 7 and had not set aside sufficient reserves to cover the growing number of
defaults. In doing so, the US effectively bailed out many European banks that held the CDS and would have been wiped out by a failure of AIG. The UK moved to
nationalize some of its largest banks and its central bank has been prompt to bring interest rates virtually to zero.
But the train wreck was already in progress, and spending was collapsing in the US. There, policy-makers reacted decisively and in a remarkably non-partisan
manner to the severity of the slowdown—a tax rebate was enacted in 2008 to stimulate demand, and the new Obama administration passed the American
Recovery and Reinvestment Act of 2009, a stimulus package of almost $800 billion of tax cuts, infrastructure spending, and aid to US states. In contrast,
aggregate demand management in Europe faced fundamental limitations. Every European country has its own policy-making apparatus and trans-European
coordination has never been Europe’s strong suit, so the fiscal policy response was a national affair by default. The executive arm of the EU, the European
Commission, tried to play a coordinating role but was unable to decide and implement fiscal policy moves directly. Generally, there was a perception that fiscal
stimulus was ‘undersupplied’—each nation stimulated their economies a little less than would have been best, betting that they would benefit from the stimulus
of their neighbours. If all countries do this, too little fiscal stimulus is taken. Second, even though fiscal policy is effective under fixed exchange rates, most
individual European countries are small and open, meaning that their fiscal multipliers are relatively low and national fiscal policies are not very effective. With
many countries burdened by large public debts even before the crisis, many governments did not find expansionary fiscal policies attractive.
Since the Eurozone countries had no independent monetary policy, it was thus up to the only truly powerful organization with teeth, the European Central
Bank, to take up the task of demand management on the monetary front. Yet while the ECB has come under increasing pressure to follow the Fed’s lead, it
remained committed to its mandate of fighting inflation. More crucially, the individual Eurozone countries do not have their own central bank. As their fortunes
differed widely after the crisis, the ECB could not deal with the needs of any particular country. Its policies were too contractionary for some countries, too
expansionary for others.
Figure 14.16 showed that the ECB was slow to cut its interest rate in 2008 and then kept it above the rates set by the Federal Reserve Board and by the Bank of
England. As a result, for reasons explained in Chapter 15, the euro appreciated and the Eurozone failed to resume growth. It took four years for the ECB to move,
and forcefully so. Meanwhile, the economy languished. The truth is that commodity prices increased considerably starting in 2010. As we know, this can trigger
inflation, especially if ‘second-round effects’ are allowed to unfold, meaning that the underlying rate is revised upward and pushes actual inflation up in a vicious
cycle. The ECB chose to focus on the prevention of second-round effects, presumably accepting a delayed recovery.
16.5.3 Diagnosis and Lessons for the Future: Fiscal Policy

It was not only monetary policy that failed to revive the Eurozone economy. Fiscal policies were simply too feeble. Many countries had allowed their public debts
to grow, ever so slowly, over the previous decades. Facing nervous financial markets, they feared that they would find the door shut and be unable to borrow any
more, and experience the fate of Greece and Ireland in 2010, Portugal in 2011, and Cyprus in 2013. In addition, the Eurozone’s Stability and Growth Pact formally
prohibits budget deficits in excess of 3% of GDP. Like the ECB, European governmentswere torn between fighting the recession and letting deficits and
government debt grow. The European Commission, in charge of monitoring the Stability and Growth Pact, adopted a strict interpretation and argued that the fiscal
policy multipliers were too low for governments to be tempted by expansionary policies.
Monetary and fiscal policies were driven by prudence. Fear of inflation and of financial market reactions trumped the fight against recession and
unemployment. These fears eerily echo the events of the 1930s following the crash on Wall Street in 1929. Policy-makers of this pre-Keynes era, both
governments and central banks, refrained staunchly from intervention, arguing that a steady hand was needed to maintain confidence in the government. Was
the synthesis between neoclassical and Keynesian macroeconomics lost on Europe? It certainly failed to convince many governments, faced with the same fears
as their predecessors of the 1930s, to adopt more active policies. Yet the real lesson to be learned in Europe is a lack of concerted and coordinated fiscal policy.
16.5.4 Diagnosis and Lessons for the Future: Monetary Policy

The most distinctive scar in the wake of the Great Recession has been the ineffectiveness of monetary policy. The instability of the money market multiplier
starting in 2009 was already noted in Figure 10.7 and had not seen in such extreme form since the Great Depression. It meant that central banks around the world
were ‘pushing on a string’—increasing the availability of money to banks with little or no effect on their lending behaviour—even while offering very low interest
rates for funding the loans. Why weren’t banks interested in lending? To answer this question we need to return to the lessons learned about banks and financial
stability in Chapter 10.
To start with, banks had taken on too much debt before the financial crisis and funded risky investments. Rather than extending more loans, banks wanted
desperately to deleverage—to reduce their borrowing and lending activities, given their net worth.10 Part of this might reflect the lack of profitable investment
projects—possibly a slowdown in the rate of total factor productivity growth, a phenomenon discussed in Chapter 18. Second, the recession had increased the
number of non-performing loans and reduced profitability. In addition, banks faced new restrictions on their ability to lend. Leverage limits were the logical results
of a decade of bad if not downright illegal banking practices. Third, mistrust among banks had risen to unprecedented levels. This was also described in Chapter
10. Financial markets knew that many among them had taken risky bets over the previous years and that many of these bets could turn sour. The US bank Lehman
Brothers collapsed in September 2008 because it couldn’t raise funds in the interbank market. This spectacular event led financial markets to freeze completely, as
no financial institution wanted to lend to another; rather, they were more interested in holding excess reserves to safeguard against a sudden panic. Finally, the
real sector—firms, households, and governments—were also reducing their borrowing. They suffered large losses on their assets and wanted to reduce debt, not
to borrow for more unsustainable consumption and investment spending.
Box 16.7 This Time It’s Different? Unconventional Policies for Unconventional Times
When central banks reached the end of the Taylor rule rope, they did not stop there. They invented new, non-standard monetary policies already discussed
briefly in Chapter 10. The first innovation is known as forward guidance. Central banks began to commit publicly to keep interest rates low for a long period
of time. As Chapter 10 stressed, monetary policy largely operates through interest rates of longer maturity. Since the shape of the yield curve is driven by
market expectations of future monetary policy decisions, promising that the policy rate would stay low for a very long time should flatten the yield curve,
despite being stuck at zero with the present policy rate. This strategy seemed to work, with longer-term interest rates declining significantly after 2011, as
shown in Figure 10.10.
Still, because consumers and firms were focused on deleveraging, and financial institutions were unwilling to lend, low long-term rates were not
enough to stimulate spending. This led to a second innovation, known as Quantitative Easing (QE). The idea here is to provide financial institutions with
massive amounts of bank reserves, which they should ultimately lend to the non-bank sector and jump-start demand. As described in Chapter 10, the
central bank buys assets and provides money in return. As Figure 16.14 shows, the scale of QE has been unprecedented. The intent is that financial
institutions will prefer to lend the cash, which bears a zero interest rate, and thus receive interest on their loans. Yet, QE can only be effective if the financial
institutions are ready to resume lending. Early evidence suggests that QE has had only a moderate impact on bank lending.
This has led to a third innovation. In 2015, some central banks (Denmark, Eurozone, and Sweden) began imposing a negative interest rate on reserve
deposits of commercial banks at their central banks. For central banks to demand interest or a fee from banks for watching over their money is certainly
unconventional. The idea is to further encourage banks to lend rather than to hold cash. By moving into negative territory, central banks are cautious not to
trigger unintended consequences. In the summer of 2016, the lowest rate to date, at –0.75%, was offered by the central banks of Denmark and
Switzerland. It is too early to know whether this non-standard policy will be successful.
Fig. 16.14 Size of the Balance Sheet of Central Banks
As central banks have pursued quantitative easing policies, they have bought financial assets and created high-powered money (their liabilities). As a result, their balance
sheets have grown at unprecedented rates.
Sources : ECB, Federal Reserve Board, Bank of England.
As a result of all these factors, aggregate demand expanded only slowly after the recession was over. Looking at it in terms of the IS–TR–IFM framework: the
IS curve had shifted sharply to the left, and interest rates set by central banks—following the Taylor rule—hit rock-bottom. As stressed in Chapter 9, it is difficult
or impossible to set negative market interest rates when holding cash is an option, so reducing the neutral rate has its limits, as does the Taylor rule. Even near-
zero interest rates couldn’t convince consumers and firms to borrow more, nor would financial institutions be ready to resume lending, which is always risky for
them. This is why the Fed (in late 2008) and the Bank of England (in 2009) embarked on a series of unconventional or non-standard monetary policies. The ECB
followed, starting in 2013, then in early 2015 with a vengeance. Box 16.7 provides details on these brave new policies.
The lessons from this chapter are mixed. To counteract a once-in-a-lifetime demand shock like the Great Recession, aggregate demand management policy must
be swift and decisive. Fiscal policy is more likely to work if it is done quickly and in a coordinated fashion; otherwise it will fall victim to the Friedman critique.
This lesson is all the more important for Europe, now that monetary policy appears to have been effectively compromised by the zero lower bound and through
widespread deleveraging by both financial and non-financial agents. At the minimum, we need to remember the old banker’s adage, ‘You can lead a horse to
water, but you can’t make him drink.’ Any number of non-standard measures can be implemented, but all of them—save the proverbial helicopter money dropped
by the ECB onto the cities of Europe—require the active participation of a healthy banking sector.
It would be wrong to blame banks or the financial industry for all the macroeconomic malaise which followed the financial crisis. Real interest rates have been
on a downward trend since the 1990s, and are as low as they were in the days of the gold standard. Interest rates might well be low for reasons that have less to
do with ultra-easy monetary policy and low inflation than the natural real rate of interest, if for example, the marginal product of physical capital has declined in a
persistent way. We will revisit this point in Chapter 18 when we address the subject of long-term growth, growth policies, and the real interest rate.
Summary
1 While it is theoretically possible for demand management policies to smooth out business cycle fluctuations, recent experience and economic principles suggest a more
measured approach.
2 A justification for demand management arises when markets do not clear rapidly, leading to low levels of output and employment over a longer period of time. Neoclassicals
and Keynesians mainly disagree on the degree to which prices and markets can achieve continuously efficient allocation of resources and optimal satisfaction of individual
needs.
3 Neoclassicals argue that market-clearing is a good first-order approximation, and that markets are closer to perfection than governments. Keynesians generally believe that
markets suffer from a host of imperfections and that economies can suffer from persistent underutilization of resources.
4 One reason that output may remain below equilibrium output for some time is the speed at which inflationary expectations can be validated and adjusted. Even if economic
agents have rational expectations, they may have taken decisions at an earlier date and locked in outdated expectations. This will affect macroeconomic variables today.
5 Uncertainty plays an important role in the debate. For Keynesians, private decisions are taken with imperfect knowledge of future conditions. This results in wrong pricing
and resource allocation decisions. For neoclassicals, uncertainty means that policy mistakes are more likely to make matters worse than to improve them.
6 The modern conception of the business cycle is the culmination of many small and large random influences. These shocks are difficult, if not impossible, to forecast. It is thus
hard to imagine a government that would be able to offset all these shocks, especially given the recognition, decision, implementation, and effectiveness lags associated with
policy.
7 One way around this problem is to make judicious use of leading indicators. Another is to make extensive use of automatic stabilizers which reduce recognition and decision
lags.
8 Real-life policy-makers do not always have the best intentions, caring more about being reelected instead. This may lead some governments to employ demand management
policies in a politically opportunistic way.
9 Political business cycles may take the form of expansionary policies being introduced just before elections, to be followed by corrective, contractionary policies after
elections. Partisan business cycles may result from the alternation of governments that defend the interests of their constituencies.
10 The sharp disagreements between hard-core neoclassical economists and Keynesians are fascinating but neither camp has much of claim on facts. Prices are sluggish, as
Keynesians argue, but not as much as they claim. Macroeconomic policies are less effective than Keynesians describe but not entirely impotent either. The evidence points
toward a synthesis of both views.
11 This synthesis is now enshrined on large-scale models in use in central banks, government agencies, and international organizations. Dynamic Stochastic General Equilibrium
(DSGE) models are based on the principles presented in earlier chapters: a general equilibrium that includes goods markets (the IS curve), a Taylor rule (the TR curve), and a
Phillips (or AS) curve. They also allow for stochastic shocks as presented in this chapter.
Key Concepts
neoclassicals, neoclassical economists

Keynesians, Keynesian economists
activist demand policies
rational expectations
deflation
inflation-targeting strategy
deterministic, stochastic view of business cycles
impulse-propagation mechanism
demand shocks, supply shocks (or disturbances)
persistence
recognition, decision, implementation, and effectiveness lags
Friedman critique
automatic stabilizers
electoral business cycles
partisan business cycles
political business cycles
median voter
Dynamic Stochastic General Equilibrium (DSGE) models
real business cycle (RBC) theory
Exercises
1 Explain the motivation for Keynesian activist demand policy using a diagram. Explain the neoclassical rejection of that type of policy.
2 Why is the behaviour of underlying inflation so critical to evaluating the effectiveness of policy? What determines this behaviour? How does the exchange rate regime (fixed or
flexible) affect your answer?
3 How important is the slope of the AS curve for evaluating the effectiveness of demand policy? What underlying characteristics of the economy might contribute to a steeper
AS curve?
4 What are the costs of inflation? Are there any benefits of inflation? If so, what are they?
5 Flip a coin 50 times, writing down +1 for heads,−1 for tails (you may use a spreadsheet program if you like). Call this variable εt. Now create a new variable, defined as yt =
1/4εt −1+ 1/2εt + 1/4εt+1, so the new variable is a weighted average of present, past, and future lagged values. (You will not be able to compute the first and the last values of
y .) Plot both time series and comment on them. How do your results change if you use yt = 3/4εt+ 1/4εt−1 instead?
6 Now repeat your experiment, using a normally distributed random variable for εt instead of the coin flip (you will definitely need a spreadsheet for this). Plot your results.
How do they compare to those of the previous problem?
7 Define the Friedman critique of activist macroeconomic policy. What is an essential element for it to be true? What could be done by governments to avoid it?
8 Name some leading, some lagging, and some coincident indicators. Explain the behaviour of these variables, using the IS–TR–IFM and AS–AD frameworks.
9 Why might the slope of the yield curve (long-term rate minus short rate) be a good predictor of the business cycle? How might it depend on the Taylor rule in force? How
would it depend on the exchange rate regime? Draw what you think a Burns-Mitchell diagram might look like and explain why.
10 Describe the non-standard monetary policies adopted in the aftermath of the financial crisis and explain why they were justified.
Essay Questions
1 ‘Aggregate demand policy is based on exploiting misperceptions of the private sector in order to increase GDP.’ Discuss.
2 Can activist demand management policies affect equilibrium unemployment?
3 State and critique the Friedman rule for optimal inflation. How might the accuracy of inflation measures affect your support for this rule?
4 Recent bailout packages for the Greek economy have imposed fiscal austerity measures (spending cuts and tax increases) on that country, which has traditionally run large
budget
deficits. Explain why the IS–TR model as well as the AS–AD predict that these austerity measures will lead to larger reductions in income and inflation than under a system
of fixed exchange rates. Do you think underlying inflation will react more or less strongly to the measure? How should this affect the adjustment process?
5 Some central banks have adopted negative interest rates. Why is it surprising and even counterintuitive? Imagine what could happen if they were aiming at sharply negative
rates such as–5%.
1 Ben Bernanke served as Chairman of the Board of Governors of the Federal Reserve System, the central bank of the United States, from 2006 to 2014. He steered US
monetary policy through the most difficult financial crisis since the Great Depression. His best-known academic research was on the transmission of monetary policy during
that period, knowledge which served him well during the many policy challenges during the Great Recession.
2 The link between fiscal and external deficits is clear from the accounting identity (2.7): NX = (S − I ) + (T) − ( G). The link between deficits and debt is discussed in Chapter 6.
3 Ragnar Frisch received the first Nobel Prize in Economics in 1969. Eugen Slutsky, who perhaps deserves more credit than he received, was a researcher at the Conjuncture Institute
in Moscow during Stalin’s dictatorship and was unable to publish this discovery until eight years later, when his work was finally translated into English.
4 The real interest rate should equal the marginal product of capital less depreciation, as was presented in Box 8.6 in Chapter 8. The real interest rate r is the nominal interest rate i
less the inflation rate i, i.e. r = i−π. When i = 0, it follows that r = −π.
5 Friedman did not advocate handing out money on every street corner! To the contrary, he argued that as monopoly provider of money, the central bank should simply hold down
the opportunity costs of holding money—and the way to do this is to reduce inflation below zero, and drive the nominal interest rate to zero. In the long run, this actually means being
quite stingy with monetary growth!
6 Such a random variable which is identically and independently distributed is often called white noise. White noise has the property that current and past values contain no
information helpful in forecasting future values.
7 A formal analysis is provided in the WebAppendix to this chapter. While such a shock is not commonplace, it is useful to help focus ideas. A similar shock worth discussing is a
permanent increase in the foreign inflation rate under fixed exchange rates.
8 The example was chosen because a single impulse induces a cyclical reaction which resembles a cycle. More generally, this property is not a necessary condition for stochastic shocks
to generate cyclical behaviour.
9 Burns–Mitchell diagrams are used as a descriptive device to summarize the average behaviour of variables over a ‘representative cycle’. They are described in detail in Chapter 1. The
eight economies considered are the UK, Spain, Canada, France, Germany, Italy, Japan, and the USA.
10 The process of deleveraging implies a ‘shortening’ of individual bank’s balance sheets, usually accomplished by refusing to roll over loans that are repaid, and using the proceeds to
retire own borrowings , just as can be seen in the Eurozone during the years 2012–2014 in Figure 10.11.
Fiscal Policy, Debt,
and Seigniorage 17
17.1 Overview
17.2.1 Provision of Public Goods and Services
17.2.2 Redistributive Goals: Equity versus Efficiency
17.3.1 Consumption and Tax Smoothing
17.3.2 Output and Employment Stabilization
17.3.3 Automatic Stabilizers
17.3.4 How to Interpret Budget Figures
17.4.1 The Public Debt with No Growth and No Inflation
17.4.2 The Public Debt with Growth and No Inflation
17.4.3 The General Case: Inflationary Finance When the Economy is Growing
17.5.1 Cutting the Deficit
17.5.2 Seigniorage and the Inflation Tax
17.5.3 Default
17.5.4 Interest Rate Relief
17.5.5 Long-run Economic Growth
Summary
Chancellor: In my old age I have been freed from pain.

Listen and look at this portentous bill
Which has made welfare out of all our ill:
‘Be it known to all men who may so require;
This note is worth a thousand crowns entire.
Which has its guarantee and counterfoil
In untold wealth beneath imperial soil.
And this is hereby a substitute approved
Until such time as the treasure can be moved.
Emperor: And do my people think it negotiable?
Do army and court take it for pay in full?
Strange though I think it, I must ratify it.
Steward: To collect those fluttering notes, one couldn’t try it;
Once issued, they are scattered in a flash.
The Exchanges stand wide open for the queue
Where every bill is honoured and changed for cash—
Silver and gold—at a discount it is true.
And then to butcher, baker, pub it goes,
Half the world only seems to think of stuffing;
While the other half in brand new clothes goes puffing.
The clothier cuts the cloth, the tailor sews.
Long live the Emperor! Makes the cellars gush
In a cooking, roasting, platter-clattering crush.
Goethe’s Faust
17.1 Overview
Governments play an important role in our economic lives. Especially in Europe, they are big, and have been growing bigger for most of the post-war period. Table 17.1 shows that governments in the European Union spend close to half of GDP, leaving the other
half to the private sector. What do they spend their money on? A large part, roughly a quarter to one-third of GDP, consists of transfers and subsidies to individuals (health and unemployment insurance, poverty alleviation). Transfers redistribute income from the
haves to the have-nots, from the lucky ones to the needy ones, expressing a sense of solidarity among citizens and a rejection of excessive inequalities. Governments are also big consumers, constructing roads, purchasing buildings and the services of public
employees, and much more. Of course, to pay for all that, they collect taxes—again about a half of GDP.
Table 17.1 General Government Spending and Finances: Eurozone, USA, UK, and Japan, 2016
Eurozone USA UK Japan

Total spending 47.4 37.9 42.6 41.5
(% of GDP)
Public consumption
as % of GDP 20.8 14.3 19.2 20.6
as % of private consumption 37.1 20.8 29.6 35.5
Budget surplus –2.1 –4.4 –3.4 –4.5
(% of GDP)
Gross debt (% of GDP) 89.5 107.5 89.7 247.5
Source: AMECO on line,European Commission.
Governments are not known to be particularly strict in managing their budgets. In fact they have turned out to be big borrowers. Budget deficits are frequent, and most governments are heavily indebted to the private sector and foreigners. While European
countries have consolidated their budgets in recent years, on average public debt still represents nearly one year’s GDP, and more than that in several countries, including Belgium, Greece, Ireland, Italy, and Japan, to name but a few. But is public debt
fundamentally bad? Or is there a more optimistic view that debt can and should be used to smooth government tax policies when expenditures are temporarily high?
We begin by looking at the economic functions of governments and how they fulfil their tasks. Do governments have an economic role to play at all? This question has been debated since time immemorial between right, left, and centre, between partisans of
laissez-faire and interventionists. In this chapter we focus on two economic functions of governments: the microeconomic function, which includes the provision of public goods and services and income redistribution, and the macroeconomic function, which
aims at stabilizing aggregate activity.
Like households, if they spend more than they earn, governments must borrow and accumulate debt. This process is inherently explosive, since more debt means more debt service, hence the need to borrow even more. To get this process under control,
governments must run a primary surplus in the future, just as we learned from Crusoe in Chapter 6. But governments also have a unique feature: they can also borrow from their central banks, at least in principle, and sometimes only indirectly. Indeed, we saw in
Chapter 10 that central banks conduct open market operations mostly dealing in public debt instruments. We also saw in Chapter 16 that quantitative easing has led central banks to absorb massive amounts of public debt. When central banks lend to
governments, even indirectly, in effect they create money that is used to pay for public spending, just as Mephistopheles did in Goethe’s Faust
In addition, central banking is highly profitable. They print money at very little cost and then sell it to the markets at face value. They pocket the difference. This unique privilege, based on their monopoly right to create legal tender, is called seigniorage. Since
central banks are usually a branch of government, their profits are passed on to the Treasury. Seigniorage thus provides resources to the government, but if abused it ultimately means fast money growth and inflation.
Finally, governments can simply refuse to pay back their debt as originally arranged or default, partially or totally, on their debts. In contrast to private agents, governments can default with impunity, in effect confiscating the wealth of their creditors. This
radical solution is only made use of in times of national emergency.
17.2.1 Provision of Public Goods and Services

Governments ‘produce’ goods and services, mostly for collective consumption.1 Why are they involved in such an activity? The two main reasons are that some goods are fundamentally collective while others are produced under increasing returns.
The defining attribute of public goods is that are non-excludable and non-rival. They cannot be appropriated for individual consumption, and once they are available for one, they are available for all. This applies, for example, to law and order, defence,
public gardens, or foreign affairs. In addition, some public goods exhibit a special characteristic: their use by one person actually benefits others.
This is called an (positive) externality. A good example is education. All of society benefits from mass literacy. For example literate workers, who are in short supply in some poor countries, interact more effectively, which raises their productivity. Indeed, there
is considerable evidence that economic growth is powerfully enhanced by the population’s education level, which we called human capital in Chapter 3. Because public goods cannot be appropriated by anyone, they cannot be sold. For this reason, private
producers would not produce them, and if they are needed, governments must step in.
Furthermore, the existence of externalities implies that individuals would only buy what they need personally, overlooking the benefits to others. In general, they would buy less than is desirable for society as a whole. For example, if no one else knows how to
write and read, my own incentive to learn how do to so is close to nil. This is why basic education is not only public and free, but also compulsory. Police protection is another example. Security could be privately organized, but safe streets are good for everyone,
whether they pay for them or not. In practice, only better-off, homogeneous neighbourhoods could afford to set up public safety on their own. Society has an incentive to combat lawlessness on a wider scale than do individuals interested only in protecting their
own safety.
A similar reasoning applies to the case of goods produced under increasing returns. One example is the usage of streets: the cost of paving and maintaining a road is roughly the same whether there are few or many users. Streets could be private and their use
charged to users. This would undoubtedly relieve traffic congestion in cities, but perhaps too much. Each of us could react to street tolls by reducing our movements to the point where the cost per remaining user would lead to the curtailment of street provision in
some areas, in the end hurting everyone.
While there is a strong justification for some government consumption, there are no clear-cut borders between goods and services that can be provided only publicly and those that could be provided privately. One key role of democracy is to choose those
borders, and adapt them from time to time as economic and social factors dictate. The absence of unequivocal criteria for deciding what should be publicly provided explains why there is much national soul-searching on the issue. In many countries, debates
resurface constantly concerning such cases as education (private schools and universities exist alongside public education), social security (health and retirement insurance are increasingly provided privately), and utilities (highways are built and run by private
companies; electricity and telephone networks are privately owned and operated).
17.2.2 Redistributive Goals: Equity versus Efficiency

A fundamental result from microeconomics is that productive efficiency—the optimal use of available productive resources—is achieved when each factor of production is paid its marginal productivity. This may result in a very unequal distribution of income
and wealth. Indeed, we observe the coexistence of much individual wealth alongside grinding poverty. While this outcome may be efficient from a productive point of view, an altogether different logic emphasizes that human beings have similar basic needs that
should be met under all circumstances. Equity or fairness is often seen as a requirement for society to be cohesive and stable. Yet equity and efficiency often work against each other. There is a fundamental equity–efficiency trade-off.
Governments can and do reduce inequalities. Progressive income taxes reduce the differences in post-tax incomes. Taxes levied on the better-off pay for transfers to the worse-off. In fact, a significant part of public spending is dedicated to income redistribution.
Table 17.2 shows the size of transfers, both as a share of GDP and as a proportion of total government outlays (spending). In some countries, this is the single largest item in the government budget. Not surprisingly, different countries deal differently with the
equity–efficiency trade-off. Austria and Sweden seem to place more weight on equity than Japan or the USA, for example.
Table 17.2 Government Transfers, Various Countries, 1960 and 2010
%of GDP %of government outlays

%1960 2010 1960 2010
Austria 14.8 31.0 51.8 59.1
Belgium 12.7 32.7 44.8 61.7
Denmark 7.6 38.2 35.1 65.4
Finland 9.0 34.3 41.6 62.3
France 16.3 35.7 53.5 63.0
Germany 14.1 29.9 50.2 62.6
Greece 5.3 28.1 30.6 56.0
Ireland 9.6 28.7 38.7 43.0
Italy 11.2 31.6 45.4 62.9
Japan 4.5 24.8 34.5 60.5
Netherlands 8.6 28.8 n.a. 56.3
Portugal 3.7 29.2 24.5 56.8
Spain 2.9 27.8 23.1 60.8
Sweden 8.6 34.9 32.2 65.9
UK 9.0 29.7 30.7 58.7
USA 6.0 30.9 24.4 52.2
Source: European Economy; OECD,Economic Outlook.
There is a trade-off, however. Income redistribution for the sake of equity has disincentive effects. Highly paid—and presumably highly productive—people may reduce their work effort in response to heavy taxation, or may even move abroad. On the other
hand, those who receive transfers from the state may find it pointless to work hard for little net reward. Once again, how to deal with the trade-off is a thorny political issue.

Providing public services and redistributing income does not imply at all that total government spending must systematically exceed revenues. The government could perform its microeconomic functions without running budget imbalances; it only needs to raise
enough tax revenue to meet its expenses. As Table 17.3 shows, however, significant surpluses and deficits are the rule rather than the exception. Do the deficits and surpluses of governments indicate bad planning or irresponsibility? Not necessarily. The second
function of public budgets is to stabilize aggregate income and spending, which stabilizes employment at the same time, as we have in previous chapters. To achieve these goals, governments should dissave in bad years and save in good years, just as many
people do. Public imbalances cannot always be explained by appealing to this motive, however, as we will see.
Table 17.3 Government Budget Balances, Various Countries,1975–2015 (% of GDP)
1975 1980 1985 1990 1995 2000 2005 2010 2015 Average,1975–2015
Austria –2.4 –2.0 –3.0 –2.5 –5.9 –1.9 –1.8 –4.6 –1.2 –2.8
Belgium –6.4 –10.2 –9.9 –6.7 –4.5 –0.1 –2.8 –4.2 –2.6 –5.3
Denmark –2.4 –3.5 –2.1 –1.3 –2.9 2.2 5.0 –2.9 –2.1 –1.1
Finland 5.1 3.8 3.5 5.4 –6.2 6.8 2.5 –2.8 –2.7 1.7
France –1.9 –0.1 –3.0 –2.4 –5.5 –1.5 –3.0 –7.0 –3.5 –3.1
Germany –5.6 –2.9 –1.1 –1.9 –9.7 1.3 –3.3 –3.3 0.7 –2.9
Greece –2.6 –2.3 –10.4 –14.0 –9.1 –3.7 –5.3 –10.4 –7.3 –7.2
Ireland –11.2 –11.2 –10.8 –2.8 –2.1 4.8 1.6 –32.4 –2.3 –7.4
Italy –10.3 –7.0 –12.4 –11.4 –7.4 –0.9 –4.4 –4.5 –2.6 –6.8
Japan –2.0 –3.2 –0.6 2.1 –4.7 –7.6 –6.7 –8.1 –5.4 –4.0
Netherlands –3.4 –4.2 –3.7 –5.3 –9.2 2.0 –0.3 –5.3 –1.8 –3.5
Norway 3.0 5.4 9.7 2.2 3.2 15.4 15.1 10.5 5.7 7.8
Spain –0.2 –3.0 –7.3 –4.1 –6.5 –1.0 1.0 –9.2 –5.1 –3.9
Sweden 5.1 –5.8 –3.7 3.4 –7.3 3.6 1.9 –0.3 –0.0 –0.3
UK –5.2 –3.7 –3.3 –1.8 –5.8 3.7 –3.3 –10.3 –4.4 –3.8
USA –5.2 –2.6 –5.0 –4.2 –3.3 1.5 –3.3 –10.6 –4.4 –4.1
Source: OECD,Economic Outlook, June 2016.
17.3.1 Consumption and Tax Smoothing

In Chapter 8 we learned that people generally dislike fluctuations in their consumption levels. As a result, they tend to borrow in bad years to sustain a level of consumption they are accustomed to, and in good years they pay down on their debts and, if possible,
even save for the rainy day. The citizens’ desire for consumption smoothing applies to their consumption of public goods and services as well, and it is the responsibility of governments—and a determinant of their electoral success—to provide a steady flow
of public goods and services. To do so, like individuals, governments have to borrow and save, depending on the economic situation. This is important because government income is very sensitive to cyclical fluctuations. Tax revenues fall when peoples’
incomes decline, simply because taxes are usually set as a percentage of income. Because taxes reduce individuals’ incomes and, therefore, their private consumption possibilities, it makes little sense to increase them in bad times. Tax smoothing, on the
contrary, is the natural fiscal counterpart to consumption smoothing of households.
This principle has a central implication for the conduct of fiscal policy. If a series of bad years reduces the country’s income, which is also the tax base, the government’s best course of action is not to try to balance the budget. In the interests of its citizens, it
should maintain a steady flow of public goods and services, and finance the revenue shortfalls by borrowing.2 Conversely, a few particularly good years, during which taxable income rises, should not be used to raise government consumption, but to increase
savings. Acting on behalf of the public at large, a government should behave like any economic agent, meeting temporary income disturbances by saving or borrowing, within the limits of its budget constraint. Box 17.1 applies the principle of tax smoothing to the
controversial case of Germany’s unification.
Budget deficits can only be financed by public borrowing, which increases the stock of public debt already in existence. Figure 17.1 shows how public debt (as a percentage of output) has evolved in selected countries. Debt rises, sometimes spectacularly,
during wars and declines afterwards. Wars are periods of unusually high public expenditure, yet they are rarely expected to last very long. The tax-smoothing principle seems to have been applied here (even if some countries eventually defaulted on part of their
debt). Similarly, the oil shocks of the 1970s and the Great Recession were accompanied in many countries by increases in debt accumulation.
Box 17.1 Tax Smoothing after German Reunification
When he endorsed swift reunification early in 1990, Chancellor Helmut Kohl promised that there would be no new taxes for West Germans. Yet the former East Germany entered the Federal Republic with precious little dowry, a hugely inefficient
productive sector, a large external debt (mostly to West Germany), poor infrastructure, and considerable environmental liabilities. In the eastern states, output fell by about 50%, unemployment—official and unofficial—rose to about 30%, and state-owned
enterprises needed cash to stay afloat until they could be sold. From a budgetary viewpoint, the new eastern states required massive public spending but could not contribute much to financing. The pressure on the federal budget is apparent in Table
17.4 and continued well into the next decade. At the same time, and for many years, Germany’s initial current account surplus of 5% of GDP melted to a deficit of 2%, showing that the world also helped finance German unification. This is as it should be in
the face of a temporary shock.
Table 17.4 Fiscal Implication of German Reunification (% of GDP)
1989 1990 1991 1992 1993 1994 1995 2000 2005 2008 2010
Government expenditures 43.1 43.6 46.1 47.3 48.3 47.9 48.3 45.1 47.0 43.7 48.0
Budget balance 0.1 −1.9 −3.1 −2.5 −3.0 −2.3 −3.2 1.3 −3.4 0.1 –4.3
Gross public debt 39.8 40.4 41.1 40.9 46.3 46.6 55.7 60.4 71.1 64.6 83.4
Current account 4.6 3.0 −1.3 −1.1 −1.0 −1.4 −1.2 −1.8 4.6 6.0 5.6
Source: OECD,Economic Outlook.
Fig. 17.1 Public Debt in Four Countries (%of GDP)

The UK had a debt/GDP ratio of over 100% in 1855, a legacy of the Napoleonic wars. During the First World War the British debt rose to about 200%, and during the Second World War to 300%. The effects of war on public finance are also visible for the USA (including the Civil War).
Sources: OECD, Economic Outlook; IMF, http://pages.stern.nyu.edu/~rsylla/.
17.3.2 Output and Employment Stabilization

A cyclical downturn means that personal incomes decline temporarily. The laissez-faire view is that, facing a temporary income fluctuation, individuals should borrow and/or save to smooth their consumption pattern, with government playing no particular role.
This prescription would be correct if all individuals could indeed borrow during a recession. Credit rationing, however, changes the situation. 3 Individuals who cannot borrow, or cannot borrow as much as they need, are unable to smooth out their consumption.
Not only are they hurt, but their declining demand deepens the slowdown through the demand multiplier effect. This effect is illustrated in Figure 17.2 where, starting from long-run equilibrium at point A, a recession occurs, which is captured by the leftward shift
of the aggregate demand curve from ADto AD'. Under fixed exchange rates, the government can use fiscal policy to stop this process. To keep the curve in its AD position and prevent the move from point A to point B, it either increases its own spending or
provides tax relief. In effect, the government borrows on behalf of its citizens who cannot get credit or can only do so at prohibitively high interest rates. Conversely, a demand boom provides the government with the opportunity to run a budget surplus and pay
back the debt accumulated during previous downturns. Both examples of the government leaning against the wind of the business cycle are called countercyclical fiscal policies.
Fig. 17.2 Stabilization Policies
When demand exogenously falls, the economy moves from point A to point B. In the absence of stabilization policy, the economy will eventually move to point C. Fiscal policy can be used to speed up a return to trend output at point A, or even to prevent the AD schedule from shifting.
17.3.3 Automatic Stabilizers

We have argued that there are good reasons for governments to use their budget to smooth income and consumption. Interestingly, when a government refrains from using discretion to vary public spending and tax revenues, its budget automatically leans
against the wind. Public budgets tend to go into surplus during upturns and into deficits during recessions. To see why, it is necessary to separate spending from revenues. While public consumption is less sensitive to cyclical fluctuations, transfers and tax
revenues are fundamentally determined by economic conditions. Many transfers, such as unemployment benefits and welfare payment, are tightly linked to the state of the economy. Similarly, when incomes and spending rise, tax collection automatically
increases. Conversely, during economic slowdowns, tax collection decreases. The reason is that nearly all taxes are set as rates applied to incomes or spending. In the end, given the budgetary process described in Box 17.2, public spending is little—if at all—
affected by business cycles. This means that the budget balance is therefore automatically procyclical—moving from surplus in booms to deficits in recessions. This is confirmed in the case of the Netherlands in Figure 17.3.4 It is possible to avoid cyclical budget
fluctuations only if the government raises taxes in recessions and cuts them in booms—which would probably add to the volatility of the business cycle.
Box 17.2 The Budgetary Process
All democracies follow roughly similar budgetary processes. Each year, the government presents a budget to its parliament, which then debates on—and sometimes amends—each item before voting on it. One part of the budget concerns spending by
the various ministries or departments; the other part concerns tax revenue. The parliament approves tax rates, literally hundreds of them, from VAT to income, from petrol to corporate profits or property. While spending authorizations are set in amounts
(say, euros) and are therefore immune to changes unless the law is amended, tax receipts in euros are uncertain, depending upon how much is to be taxed at the set rates. This is why parliament cannot decide exactly what the deficit or surplus will be.
Instead, it is presented with a forecast of GDP, which underlies a forecast of tax receipts and the associated deficit or surplus. It is well understood that economic conditions will settle the matter as the fiscal year goes on. Many governments have a
tendency to forecast high growth, large tax receipts, and small deficits, since such forecasts are not binding and make the authorities look good, at least for a while. It takes an unusually good year to have a better budgetary outcome than announced,
while a moderate slowdown may easily result in large ‘unexpected’ slippages.
The fact that unemployment rises during recessions provides another rationale for countercyclical fiscal policies. The mere increase in unemployment is not a justification for active fiscal policies. If unemployment rises because its equilibrium level
has permanently increased, the economy will eventually settle along its long-run aggregate supply schedule and demand management is bound to fail. Attempts to use fiscal expansions to keep unemployment below equilibrium will only lead to more
public debt and higher inflation. However, short-run fluctuations in unemployment around its equilibrium rate occur because price and wage rigidities prevent an optimal utilization of available resources. Countercyclical fiscal policy may be a corrective
device to keep unemployment close to its equilibrium level, and output near its trend growth path. Sustaining aggregate demand with public spending when private demand weakens, or directly boosting private demand with tax relief, has the potential to
limit the size of business cycles.
Fig. 17.3 Cyclical Behaviour of the Primary Budget in the Netherlands, 1980–2010
When the economy enters a slowdown (the output gap declines), tax receipts fall and transfers rise, usually with a lag of a year or two. The primary budget—the government budget balance excluding service of the public debt (which is largely acyclical)—moves in the same direction as
the output gap. It is procyclical.
Table 17.5 Expected and Realized Government Budgets in 2010 (% of GDP)
Forecast time Dec June Dec Dec

2008 2009 2009 2010 Actual
France −3.9 −7.9 −8.6 −7.4 −7.0
Germany −1.0 −6.2 −5.3 −4.0 −3.3
Italy −3.1 −5.8 −5.4 −5.0 −4.5
The Netherlands −0.9 −7.0 −5.9 −5.8 −5.3
Spain −3.8 −9.6 −8.5 −9.2 −9.2
Source:: World Bank, OECD.
To summarize, we have seen that, when an economy slows down, the budget deficit will normally increase, or its surplus shrinks or even moves into deficit. This automatic lowering of taxes acts like an implicit fiscal expansion. Conversely, better-than-expected
economic growth reduces the budget deficit or increases the surplus because of enhanced tax income for the government, a contractionary fiscal policy of sorts. In the end, we see that exogenous shifts in private demand are automatically cushioned—but not
completely offset—by budgetary shifts. These are the so-called automatic stabilizers. They work in the absence of any policy action. Simply by enacting the budget, as approved by the parliament, the government finds itself conducting a countercyclical
fiscal policy, dampening both recessions and expansions.
17.3.4 How to Interpret Budget Figures

With automatic stabilizers in place, the budget is partly endogenous. Policy choices plan a surplus or a deficit, but economic conditions determine the outcome. This is why raw budget figures do not always fully reveal the government’s intentions and can be
misleading. Table 17.3 shows that most countries underwent budget deficits in 1975 and 1980. These years coincided with the post-oil-shock recessions and the associated deterioration of public finances. Norway, an oil-exporting country, provides a neat counter-
example: it underwent a boom and tax revenues were further boosted by oil revenues. Similarly, during the post-financial crisis, it has proved difficult for governments in southern Europe to demonstrate credibly that they were trying to reduce their structural
budget deficits. Indeed, cuts in spending and increases in taxes further depressed the economy, which in turn depressed taxes and led to larger, not smaller deficits.
The endogeneity of government budgets means that it is not straightforward to determine whether the stance of fiscal policy is tight or easy. A growing deficit may signal an explicit government decision to respond to a cyclical downturn, but it may also reflect
the automatic worsening of the budget as the economy slows down. This is bothersome, not just for observers and analysts, but for the government as well. How can it know that its response is appropriate if it does not even know what its effect on the budget is?
As explained in Box 17.2, any budget law—which authorizes a given level of spending and sets all tax rates—includes an estimate of the expected balance, explicitly based on a forecast of GDP for the corresponding year. In Figure 17.4 the schedule FP (for fiscal
policy) corresponds to such a budget law, i.e. spending and tax rates are taken as exogenous. The positive slope represents the working of the automatic stabilizer. Given the amount of spending and the tax rates approved by the parliament in the budget law, an
increase in the output gap improves the budget balance because tax revenues increase. Conversely, the budget surplus deteriorates when the output gap declines. For instance, if the budget law assumes that the economy will be on trend during the
corresponding fiscal year, it predicts the budget surplus indicated by point A. If the actual growth outcome is not as good as expected, point A' shows that the budget will end up in deficit.
What if the government decides to react to the economic slowdown by increasing spending or cutting taxes? This discretionary action is captured by a shift of the budget schedule, which moves to FP'. This schedule lies everywhere below FP, simply because
the higher spending level or lower tax rates imply that, for any output gap, the new budget balance will be lower than the previous one. What, then, does this fiscal expansion mean for the budget balance? If the economy does not respond to the fiscal boost
during the fiscal year and the output gap remains the same as the one corresponding to point A', the budget deficit will widen, as shown at point B'. If, on the contrary, the fiscal expansion manages to limit the slowdown, the budget will be described by a point on
the FP'schedule, somewhere up and to the right of point B'. For example, if the fiscal expansion succeeds in keeping GDP on trend, the new situation is described by point B.
The figure illustrates why it is difficult to interpret budget figures. Suppose, for instance, that last year’s situation is described by point Aand that, in the present year, the outcome corresponds to point B'. We observe a sizeable worsening of the budget
balance. Is it due to a fiscal expansion or to an economic slowdown? Both, in fact, but how do we know the role of each contributing factor? This is the question that we need to deal with.
The procedure is to ask what the budget balance would be if real GDP were on its trend path. The corresponding budget balance, which is constructed to be free of cyclical effects, is called the cyclically adjusted(or sometimes structural) budget. Figure 17.4
considers two budget laws, represented by the schedules FP and FP′. The corresponding cyclically adjusted budgets are represented by points A and B, respectively. The distance AB is a measure of the fiscal relaxation corresponding to the shift from FP to FP′.
In general, actual budget outcomes are not equal to the cyclically adjusted budget because the GDP is rarely at its trend level. Looking at the budget law represented by FP, it is easy to see that a positive output gap implies an actual budget surplus larger than the
cyclically adjusted surplus: compare points A and A′′.
Fig. 17.4 Endogenous and Exogenous Components of Budgets

The line FP describes how the actual budget responds to cyclical fluctuations of output about its trend for a given fiscal policy stance. The move from line FP to line FP' describes a more expansionary policy stance. The cyclically adjusted budget is measured assuming a zero output gap.
For fiscal policy stance FP it is given by point A, and for FP' by point B.
Cyclically adjusted budgets are routinely calculated and discussed. For example, the original European Stability and Growth Pact, designed to impose fiscal discipline within the monetary union, considered only actual (i.e. realized) budget outcomes. This led to
significant confusion and debates as countries with excess deficits claimed that it was the consequence of less than expected growth. This is why, following a revision, the pact now emphasizes cyclically adjusted balances. The stark difference between the two
budget measures can be seen in Figure 17.5, which displays changes in fiscal balance between 2008 and 2010 for 20 OECD countries. All countries saw their budgets deteriorate due to bad economic conditions. The blue diagonal divides outcomes into those for
which the official government budget surplus deteriorated more than the cyclically adjusted balance (to the left) or less (to the right). Of all 20 OECD countries represented in the figure, only Greece and Italy tightened their fiscal policies on a cyclically adjusted
basis over the period 2008–2010. The figure shows clearly that despite the sharp increase in budget deficits during crisis, these numbers overstate the expansionary policy of most OECD governments.
According to Table 17.6, public indebtedness in almost every OECD country is large and growing.5 Government debt grew especially during the 1970s, with this growth slowing down for some during the 1980s but accelerating for others. Since the financial crisis,
government debt is once again on the rise. Yet there is good news. The intertemporal budget constraint does prevent the permanent accumulation of public debt, just as Chapter 6 foretold. At some point, budget deficits must be closed and turned into primary
surpluses. Figure 6.8 shows how this rule had worked for several ‘problem cases’. After sharp increases during the recession that followed the Great Financial Crisis, strategies for pursuing debt stabilization are again a central concern, especially in the context
of the Stability and Growth Pact of the European Monetary Union.
Fig. 17.5 Changes from 2008 to 2010 in Actual and Cyclically Adjusted Budget Balances, 20 OECD Countries (%of GDP)
The horizontal axis shows the change of the actually measured budget balance, and the vertical axis displays the change in the cyclically adjusted budget balance over the period of the global recession 2008–2010. By construction, these two measures are equal when the output gap has
not changed—the blue diagonal line. Points located above the diagonal correspond to a recession—a negative change in the output gap—where the cyclically adjusted balance deteriorates less than the actual balance. Conversely points below the diagonal correspond to a positive change
in the output gap. The figure confirms that all countries went into recession. Points along the horizontal line capture a neutral fiscal policy stance, when the cyclically adjusted balance is unchanged and fiscal policy is reduced to the effects of the automatic stabilizers. Above the horizontal
line, fiscal policy is contractionary, and it is expansionary when the country is shown under the line. We see that most countries adopted countercyclical fiscal policies, as pledged at the second G20 Summit in London in April 2009. Two countries with serious debt problems, Italy and
Greece, undertook discretionary fiscal contractions. In fact, a fiscal contraction was requested from Greece as a condition for the loan received from the other Euro Area countries and the IMF. Fiscal policy was most expansionary in Portugal, which had also to apply for support by early
2011.
Table 17.6 shows that despite the debt crisis brewing since 2008, considerable progress has been made in many countries over the longer haul. Spectacular successes—in the sense of stopping exploding debt paths and even reducing high levels of debt—can
be observed in Belgium, Denmark, Ireland, the Netherlands, Spain, and Sweden. Despite the troubles endemic to southern Europe at the moment, these countries made significant progress, in order to fulfil the conditions for adopting the euro.6 Since the crisis,
much of this progress has been overturned, a result of the recession and countercyclical fiscal policies as much as structural deficits and most importantly, rising rates of interest for the issuance of new debt.
Table 17.6 Gross Public Debt, Various Countries, 1970–2015 (% of GDP)
1970 1980 1990 2000 2010 2015

Austria 18.5 35.3 56.0 65.9 82.4 86.2
Belgium 60.1 74.2 125.9 108.8 99.7 106.0
Denmark 7.6 39.2 62.4 52.4 42.9 40.2
Finland 11.4 11.2 13.8 42.5 47.1 63.1
France 21.0 21.0 35.4 58.7 81.7 95.8
Germany 17.8 30.3 41.3 58.8 81.0 71.2
Greece 18.0 22.7 72.4 104.9 146.2 176.9
Ireland 47.6 66.9 90.2 36.1 86.8 93.8
Italy 35.7 54.0 91.7 105.1 115.4 132.7
Japan 11.5 52.5 69.4 143.8 215.8 245.4
Netherlands 49.4 43.5 73.9 51.4 59.0 65.1
Norway 23.6 36.9 32.2 28.0 41.7 31.6
Spain 14.3 16.0 41.6 58.0 60.1 99.2
Sweden 26.2 37.6 39.4 50.6 37.6 43.4
UK 75.8 51.4 31.3 38.9 76.6 89.2
USA 44.3 41.2 62.0 53.1 94.7 105.9
Source: AMECO, European Commission and IMF Debt Database.
We now examine the debt stabilization issue in detail. It will allow us to understand the forces that lie behind the debt accumulation process, the challenge of stopping this unsustainable evolution, and what happens if it is not done.
17.4.1 The Public Debt with No Growth and No Inflation

Let us start with the easiest case, where the central bank does not finance public deficits and we can ignore the role of seigniorage. We also assume that inflation is nil because then we do not need to worry about the distinction between nominal and real variables.
The government budget deficit is the sum of the primary deficit—the excess of public spending G over net tax receipts T—and of debt service—here a single (real) rate of interest r times the existing debt stock B. In the absence of asset sales (privatizations),
foreign aid grants, or default, this deficit will equal the net borrowing requirement of the country—the net issue of new debt. In the absence of monetary financing (borrowing from the central bank), the government issues new debt ∆B held by the public. This can
be formalized as:
(17.1)
If the overall budget is in deficit—technically, ‘the public sector borrowing requirement’ is positive— the government must borrow, which means an increase in the public debt B (∆B > 0). If the budget it is in surplus, the government can retire some of its existing
debt or accumulate assets (∆B < 0).7
Looking at public budgets in this way shows that the debt accumulation process can easily become explosive. The reason is that the debt feeds and grows on itself; the higher the current stock of debt B, the higher is debt service rB, and therefore the larger the
deficit and the need to accumulate more debt. Even when the primary budget is balanced (G − T = 0), the overall budget is in deficit and the debt continues to grow. In this case, the government keeps borrowing to pay interest on the existing debt, which means
more indebtedness. Without any further effort to control it, debt will accumulate at rate r.8 This feature of indebtedness is general and applies to any debt, be it public or private, domestic (i.e. held by domestic residents), or external.
To halt the accumulation of its debt, the government must run a permanent primary surplus large enough to service the existing debt without having to incur more borrowing. Formally, we ask what must happen for ∆ B = 0 in (17.1). It is easy to verify that this
implies:
(17.2)
Stabilizing the absolute level of debt can be a formidable task. The longer the government puts off the day of reckoning, the larger the debt becomes, and the larger is the surplus ultimately required to stabilize it. This observation is illustrated in Table 17.7, which
presents net debt levels and primary budget balances (as a ratio to GDP) for a sample of countries. At the moment, our attention is focused on the third column of this table, which shows the primary surplus required if the government wanted to stabilize the net
debt at its 2015 level, assuming a real interest rate of 5%. Of the countries shown, none of the countries were close to fulfilling this objective. Clearly, the public finances of these countries were hard-hit by lost tax revenues and policy response associated with the
weak recovery (if any) from a deep recession.
In any event, moving the structural (cyclically neutral) fiscal balance by 3–5% is painful, and often politically impossible. Fortunately, the situation need not be so dire. The following sections show that the stabilization of the absolute level of debt is excessively
stringent because it ignores both growth in GDP and the possibility of monetary finance and inflation.
17.4.2 The Public Debt with Growth and No Inflation

It is misleading to compare the absolute size of the government debt of the USA with that of France and Sweden, or Belgium. A country’s ability to service debt (to repay it with interest) is fundamentally related to its economic size, since a given tax rate applied to
a larger economy produces more revenues. This is precisely why all the data in Table 17.7 were presented as ratios to GDP, and why the stabilization of the debt–GDP ratio is a more reasonable objective than the stabilization of the debt level itself.
This distinction matters even more because GDPs tend to grow secularly over time. Box 17.3 shows formally how the debt accounts change when we look at debt–GDP ratios. Beyond the arithmetic details, the logic is easy to understand. When we look at the
evolution of the debt–GDP ratio, we can think of a ‘race’ between the numerator—the level of debt—and the denominator—GDP. The faster the real GDP grows, the less the ratio increases. The level of debt itself may increase while declining as a ratio to GDP; all it
takes is for the GDP to rise faster than the debt stock.9 By the same token, in a growing economy, were the absolute stock of the debt to remain stable, as in the case studied in the previous section (see equation (17.2)), the debt-to-GDP ratio would simply vanish
over time. Since the real debt grows at the rate r, when the GDP grows at the rate g, the debt-to-GDP ratio rises at the rate (r − g).
This means that when the real GDP growth rate g exceeds the real interest rate r, a balanced primary budget is sufficient for the debt–GDP ratio to shrink. When the real interest rate exceeds the economy’s growth rate, however, the debt process is explosive. Yet,
the primary surplus required to stabilize the ratio of the debt to GDP is smaller than the primary surplus required to stabilize the level of the debt because GDP growth helps contain the increase in the debt–GDP ratio. The last column in Table 17.7 drives this point
home. Assuming a 5% real interest rate and a 2.5% real GDP growth rate, we see that in 2015 all countries listed in the table need much less of an adjustment to stabilize their debt–GDP ratios.
These results help us understand why debt increased so dramatically in the 1970s and 1980s. Over the 1960s and early 1970s, real GDP growth exceeded the real interest rate in most countries. For example, real interest rates in the UK during the period 1960–1980
averaged 0.9% while real growth was 2.4%. Under such conditions, budget deficits need not imply a growing debt–GDP ratio. The debt accumulation process was not explosive, at least relative to GDP, a fact that probably encouraged complacency about deficits
and debt. In contrast, over the period 1980–1995, UK real growth hardly changed (2.4%), while real interest rates averaged 4.7%! The debt process became explosive and required prompt and vigorous action. Several countries failed to adjust quickly enough.
It is now easier to understand how problems faced by the southern European countries in servicing government debt have been aggravated by a real growth slowdown over the decade before the crisis. Table 17.8 displays average growth rates for these
countries. Clearly, long-term growth ( g) has declined in the past 10–20 years. This is a cause of concern not only for social and economic reasons, but also because it can endanger the stability of already-high levels of government indebtedness. For this reason,
stabilization programs of the IMF and the European Commission also insist on measures which enhance long-term growth, even while fiscal austerity slows down or even throttles short-term prospects.
Table 17.7 Net Government Indebtedness and Primary Budget Balances, 2015 (% of GDP)
Primary surplus*
Net debt Actual

in 2015 primary budget Required to stabilize the size of debt Required to stabilize the debt/GDP ratio
surplus in 2015
Belgium 106.0 –2.6 5.3 2.7
Germany 71.2 0.7 3.6 1.8
Ireland 92.8 –2.3 4.6 2.3
Italy 132.7 –2.6 6.6 3.3
Netherlands 65.1 –1.8 3.3 1.6
*Required surplus assuming a 5% real interest rate and a 2.5% real GDP growth rate.
Source:Eurostat.
17.4.3 The General Case: Inflationary Finance When the Economy is Growing
Inflationary finance represents an alternative way of relaxing the budgetary stringency required for debt stabilization. Simply put, the government exploits its monopoly right to create money in the narrow sense (monetary base) and uses that money to pay its bills.
Technically, the government borrows directly from the central bank, or forces it to purchase existing debt and to hold that debt on its own balance sheet (see Chapter 10). When the borrowed money is spent, it becomes an addition to monetary base, the
foundation for money creation by the banking system, and, in the long run for a higher price level.
Table 17.8 Economic Growth in Southern Europe, 1980–2015 (% per annum, average)
1981–1985 1986–1990 1991–1995 1996–2000 2001–2005 2006–2010 2011–2015

Greece 0.2 1.3 1.3 3.6 3.9 –0.2 –3.8
Italy 1.7 3.1 1.2 2.0 0.9 –0.3 –0.7
Portugal 1.5 6.2 1.9 4.1 0.9 0.6 –0.9
Spain 1.3 4.7 1.7 4.1 3.4 1.1 –0.1
Euro Area 1.6 3.4 1.6 2.9 1.8 0.8 0.8
EU 1.5 3.4 1.6 2.8 1.5 0.8 0.5
Source: International Monetary Fund,World Economic Outlook Database, April 2016 and AMECO on line
Box 17.3 Debt–Deficit Arithmetic
Growth, no inflation
In order to study the evolution of the debt–GDP ratio (B/Y), recall that it grows at a rate equal to the difference in the growth rates of numerator and denominator. Using the equation for budget accounting (17.1), we see that the numerator grows by (G − T)/B
+ r, while the denominator grows at rate g of GDP (since we are considering longer periods of time, here we use the trend GDP growth rate). With a little bit of calculus we can show that:
(17.3)
As long as the real interest rate exceeds the growth rate, the debt process remains explosive. Yet the debt–GDP ratio grows at the rate r −g, which is lower than r, the growth rate of debt as shown by (17.1). For a given ratio of the primary deficit to GDP,
more debt means more deficit and greater borrowing requirements. The primary budget surplus required to stabilize the debt–GDP ratio sets the left-hand side of (17.3) equal to zero:
(17.4)
If the rate of interest r is below the GDP growth rate g, the debt–GDP ratio can be stabilized while running a budget deficit. Fast-growing countries can outgrow their debt.
Growth, inflation, and seigniorage finance
When inflation is positive, the role of money and money creation in government finance has to be accounted for. The WebAppendix shows how budgetary accounts are modified to recognize that the deficit can be financed by both new debt issue and
seigniorage finance—loans made by the central bank to the government to cover the deficit. These loans are equivalent to the creation of additional monetary base, M0:
(17.5)
Stabilizing the debt–GDP ratio now requires an even smaller primary budget surplus, or can even be achieved with a primary deficit if enough monetary base is created:
(17.6)
Remember that in Chapter 13 we derived the long-run relationship between inflation, monetary growth, and output as π = μ − g. Combining this with the fact that ∆M0 is equal to μM0, the stabilization condition for debt becomes:
Seigniorage relaxes the government budget constraint as long as inflation and growth generate continuously rising demand for money.10
Producing monetary base is virtually costless: a stroke of a pen—more precisely keying in a few zeros in the computer—or activating the printing press. Since the national government uses the money it creates at face value to acquire real resources, it (or the
central bank, which is owned by the government) makes a comfortable profit. The revenue from this lucrative activity is called seigniorage.11 Seigniorage is the main source of profit for central banks. As public institutions, central banks are usually required to turn
over most of their profits to their governments. Although seigniorage can represent a substantial source of government revenue in times of high inflation, it has rarely done so in recent years.
Naturally, the greater use of seigniorage results in faster money growth (via the money multiplier effect) and, eventually, higher inflation. Formally, the role of seigniorage can be seen when the budget account (17.1) is modified to recognize that any increase in
the nominal monetary base ∆M0 provides real resources to the government. The last part of Box 17.3 provides the mathematical details. Since the real value of seigniorage is ∆(M0/P), the account is rewritten as:
(17.8)
Remember that B, G, T, and r are defined in real terms.

Seigniorage is a cheap source of financing for the government. It severs the link that makes the debt process explosive because little or no interest is paid on the monetary base. But the explosiveness is simply transferred elsewhere—into inflation. All
hyperinflationary episodes can be linked to a government’s attempt to break away from its budget constraint. For the same reason, hyperinflations ultimately end when governments close their deficits, or when central banks stop financing them.
Yet inflation, especially when it is unanticipated, can help the government solve its financing problems, via the second mechanism of inflationary finance, the inflation tax. The inflation tax simply means that inflation, which arises from monetary growth,
erodes the real value of obligations the government has issued in its own currency. Most important in this regard is the public debt, which primarily takes the form of non-indexed, nominal bonds, but also includes money issued by the central bank. Because the
nominal face value of the debt remains unchanged, its real value is eroded, and debt-holders suffer a real capital loss. The inflation tax is just the mirror image of this loss: the reduction of the real value of the debt is a gain for the government.
Seigniorage and the inflation tax go hand in hand. Seigniorage leads to money growth and therefore to inflation and debt relief via the inflation tax. Naturally, the inflation tax applies only to debt issued in local currencies. In addition, the inflation tax works only
if inflation is unexpected. The reason is that, when debt-holders anticipate inflation, they demand a nominal interest rate which compensates them for the expected erosion of the principal. This is the Fisher principle of Chapter 14. As the nominal interest rate rises
in line with expected inflation, the real interest rate remains unchanged. In that case, the bondholders do not lose, nor do governments gain from inflation—except for the inflation tax on money balances, which do not bear interest.

What are the options open to a government that wants to stabilize an exploding debt–GDP ratio? This is the central question which dogs not only Europe in the second decade of the twenty-first century but also the USA and Japan. Three ways involve short-term
approaches, while two additional ones are long-run and require a fair amount of patience. In the short run, three and only three ways are available to stabilize the debt–GDP ratio:
(1) fiscal stringency (‘austerity’)—reducing the deficit and achieving the required primary surplus by reducing public spending, raising taxes, or both;
(2) inflationary finance—the monetization of deficits and use of the inflation tax;
(3) outright default on some or all of the existing debt.
All three amount to different forms of taxation: standard taxation in the first case, taxing those who happen to hold nominal assets (money, and nominal bonds) in the second case, and taxing those who own government debt in the last case.
In the medium to long run there are two additional options:
(1) reducing the rate of interest;
(2) raising the long-run rate of growth, or at least achieving the attainable rate of growth determined by factors spelled out in Chapter 3.
In what follows we examine these options.
17.5.1 Cutting the Deficit

Deficit reduction is the virtuous road to debt stabilization. Politically, though, it is also the hardest to implement. Cuts in public spending elicit significant resistance from those directly affected, e.g. government employees who will fight for their jobs, or public
construction contractors that want to keep their businesses going. Raising taxes is notoriously unpopular with everyone. Another possibility is the privatization of state enterprises, the proceeds of which can be used to pay down debt and reduce interest
burdens.
As difficult as it is, deficit reduction as a medium-term policy had been successfully implemented in a number of European countries. As Table 17.9 shows, at the beginning of the millennium, almost all countries were models of fiscal discipline with zero or even
positive primary surpluses; countries with some of the most serious debt problems of previous decades—Belgium, Ireland, and Italy—had actually turned their primary budgets around. This consolidation is attributable largely to the adoption of the monetary
union, as explained in Box 17.4. The problem arises when budget cuts and privatizations are implemented in bad times, i.e. the middle of a deep recession. Such cuts can be highly counterproductive and make the situation worse.
Table 17.9 Primary Budget Balances for Selected European Countries, 2000–2015
2000 2005 2010 2011 2012 2013 2014 2015

Belgium 6.1 1.5 –0.7 –0.9 –1.0 –0.1 –0.3 –0.1
France 1.2 –0.7 –4.5 –2.6 –2.4 –1.9 –1.9 –1.6
Germany 3.5 –1.0 –2.1 1.0 1.7 1.4 1.6 1.9
Greece 2.5 –1.6 –5.7 –3.4 –4.3 –9.5 0.1 –3.9
Ireland 6.5 2.5 –29.9 –10.0 –4.8 –2.2 –0.4 0.5
Italy 4.5 0.2 –0.1 1.0 2.1 1.7 1.4 1.4
Portugal –0.7 –3.9 –8.5 –3.6 –1.4 –0.6 –2.8 –0.2
Spain 1.8 2.7 –7.8 –7.6 –7.9 –4.0 –2.9 –2.4
17.5.2 Seigniorage and the Inflation Tax

In the end, monetary finance of fiscal deficits is just another form of taxation, like excise, income, or consumption taxes. It operates by reducing the value of the money base (the central bank’s liability) and of the public debt (the Treasury’s liability). Inflationary
finance is simply a tax on money and bondholders.12
This result can be seen in a different way. In the budget accounts, (17.1) for example, it is the real interest rate that matters. For the inflation tax to work, the real interest rate must fall, otherwise the real cost of servicing the debt remains unchanged. When
inflation rises unexpectedly and quickly enough, nominal interest rates on long-maturity assets cannot be modified as they are contractually fixed for the whole life of the asset. This explains why ex post real interest rates are just the mirror image of inflation.
To a fiscally undisciplined government, the inflation tax may seem like an easy way to escape its debt obligations. It is not as painless as it looks, though. When inflation rises, buyers of new bond issues will demand higher nominal interest rates to avoid getting
‘burned’. In addition, suspicious lenders will be less and less willing to agree to long-term loans. As the maturity of the debt shortens, the government must constantly issue new bonds to pay back its maturing debt, and it must then pay the new higher nominal
interest rate. If the debt accumulation process remains unstable, the government will be forced to create more money, leading to more inflation, which will soon lead to higher nominal interest rates, calling for yet another increase in inflation, and so on.
This is precisely how many hyperinflations get started. A surprise inflation tax wipes out the value of nominal assets and eliminates the real public debt, but leaves a hyperinflation in its wake. As the recent episodes of Bolivia, Serbia, Ukraine, or Zimbabwe
show, hyperinflations take a terrible toll on the economic and social structure of a country, and stopping a hyperinflation can be very costly too. This is why it is an option only used under extreme political situations. It is no surprise that most hyperinflation
episodes occur during wars (Congo and Serbia in the 1990s) or in their aftermath (Germany, Greece, Hungary after the First World War), or in troubled times (several Eastern European countries at the time of the collapse of the communist regimes) when the
government is too weak to enforce fiscal discipline.
Box 17.4 Euro Area and the Stability and Growth Pact
When a country has its own currency, it can use seigniorage to finance at least part of its deficits. Once it has fixed its exchange rate or joined a monetary union, it loses this source of financing, just as it loses its ability to determine the domestic inflation
rate.13 The consequences for national fiscal policies are profound. First, the debt stabilization requirement becomes more stringent, as a comparison of (17.1) and (17.7) shows. Second, default is the only option for a government unwilling to face the
facts of the intertemporal budget constraint and run primary surpluses. The spectre of default was a great concern for the founding fathers of the European Monetary Union. Europe continues to consist of sovereign countries with their own individual
histories and traditions.
For this reason, admission to the monetary union was restricted to countries that fulfilled criteria of prudent budgetary behaviour.14Also, once a member, each country is subjected to the Excessive Deficit Procedure. This procedure is defined by the
Stability and Growth Pact. In brief, the pact sets upper limits on annual consolidated budget deficits of 3% of GDP. Should a country exceed this limit, it is given an initial ‘early warning’ and must then take prompt corrective action. Failure to correct the
situation exposes the country to a fine. For the reasons presented in Section 17.3.3, during the 2000–2003 slowdown many countries exceeded the 3% limit. In late 2003, the fine procedure should have been triggered against France and Germany, but
politics took over and the pact was ‘suspended’, a decision that subsequently was ruled illegal by the European Court of Justice. This led to a first reform of the Stability and Growth Pact, designed to allow the automatic stabilizers to function in
recessions—thus allowing the deficit to increase—but requiring member countries to run budget surpluses in good times.
The Euro-crisis has renewed discussion of the usefulness of the Stability and Growth Pact. Facing deep recessions, most member governments simply let their deficits grow. Some governments even had to borrow massive amounts to bail out their
banking systems. Simply put, they were not able politically to contain deficits in the face of rapidly rising unemployment. A reform has been introduced to integrate national budgetary processes—including parliamentary deliberations—into a ‘European
Spring’ in which the Commission vets proposals. The effectiveness of the reform has yet to be determined.
17.5.3 Default
The most brutal way for a government to stabilize the debt in the short run is simply to default, i.e. to repudiate it. Except for post-war or post-revolution periods—when the blame can be put on exceptional circumstances or previous regimes—governments rarely
resort to this approach, and do so only under severe stress. Outright default is always perceived as a major breach of confidence and leaves long-lasting scars on the reputation of governments. As Box 17.5 shows, default was used in Italy’s Fascist era to reduce
high interest rates, which reflected lenders’ doubts that stringent budgetary policies would last. Defaults can also be seen as a form of specific taxation that affects bondholders, much as inflation does. Indeed, partial default is exactly equivalent to a tax on bond
income. If, for example, a government reduces the value of its debt by half, this is the same as imposing a 50% tax on interest and repayment of the principal.
Box 17.5 Mussolini and the Public Debt
Italy emerged from the First World War with a large debt and a sizeable budget deficit. Between 1923 and 1926, having eliminated all political opposition, Mussolini re-established near budget balance and brought the debt–GDP ratio down by reducing
spending and raising taxes (Table 17.10). Yet, concerned that the debt was too short in maturity, and therefore vulnerable to market conditions, in November 1926 the government imposed a mandatory conversion of debt of less than seven-year maturity
into fixed-rate (5%) longer-term bonds. In 1934, these bonds were again forcibly converted into 25-year loans bearing a 3.5% interest. The first conversion is estimated to have resulted in a partial default of 20%, the second one in a loss of 30%. After
these moves, the government found it very hard to undertake new borrowing. It was forced to cease issuing short-term debt in 1927 and paid a premium estimated at 2–3% on borrowing from banks.15
Table 17.10 Public Finances in Italy, 1918–1928
1918 1922 1924 1926 1928

Tax revenues as % of public spending 23 46 90 100 90
Debt–GNP (%) 70.3 74.8 65.1 49.7 53.8
Source::Alesina (1988).
The issue of default has different implications when the debt is owned by foreigners. Thus far, it was implicitly assumed that the public debt was held by residents. In that case, debt accumulation or stabilization amounts to a redistribution of income across
generations, between those who are taxed now and those who will be taxed in the future. When debt is owned by foreigners, the situation is different. As Chapters 6 and 15 have shown, honouring external debt implies transferring resources to the rest of the
world. This requires running a primary current account surplus, i.e. spending less than is earned after interest payments. The temptation to default and stop the outflow of resources can be strong, especially when the stock of debt is already large. On the other
hand, a country which has defaulted cannot borrow abroad for a long time, often 10 years or more. During that period of ‘international pariah status’, the country will find it difficult or impossible to run a current account deficit. Comparing the two situations,
honouring the debt requires a current account surplus, possibly for decades, while defaulting only requires a balanced current account, usually for a few years. This explains why sovereign nations are sometimes more willing to default on the external than on the
domestic debt when the going gets rough.
17.5.4 Interest Rate Relief

One mechanism often overlooked for stabilizing an explosive debt situation is a reduction of the interest rate at which countries borrow, either at home or abroad. This is not a trivial exercise, since interest rates are determined in capital markets in which many
players are active, all of which have a view about the debt and whether it will be serviced on time. As was already shown in Box 17.5, sometimes the rates at which the public is willing to lend to government are dominated by a risk premium (Chapter 15) or even the
expectation of outright default. If this expectation is stubborn, it may be impossible for a country to borrow at an interest rate at which it can be reasonably expected to service its debt.
To see this, suppose that a country had a debt–GDP ratio of 100%, long-term growth was 3%, and the real market interest rate for debt was 4%. According to our formula (17.4) a primary budget surplus of 1% of GDP is necessary to keep debt–GDP constant at
100%. But now suppose the markets are risk neutral, but expect the government to default with a high probability in the coming years. Suppose that investors in the bond markets now demand a higher rate, say 6% per annum, from the government in question.
This higher rate is meant to compensate bondholders (lenders) for outcomes in which no payments are made at all (default). At that higher rate, of course, stabilization of the debt–GDP ratio is only possible if the primary current account is 3% of GDP, much higher
than before and rendering the probability of default even higher, and so on. There is a distinct possibility that the refinancing of government debt ‘becomes unhinged’—a situation in which interest rates are high because lenders expect with high probability that
default will occur, and, because the expected default will occur, interest rates are high. This is an example of self-fulfilling prophecy, similar to non-fundamental asset pricing discussed in Chapter 7. Box 17.6 describes the present situation for many countries of the
Eurozone; a glance back at the examples of bubbles in Chapter 7 suggests that the possibility of self-fulfilling prophecies should be taken seriously.
Box 17.6 Interest Rates Unhinged: The Great European Sovereign Debt Crisis
The Great Recession of 2008–2009 inflicted many forms of collateral damage on Europe’s economies—in the first instance, a collapse in export demand, a freezing-up of financial markets, a sharp recession, and a rise in unemployment. One of the
more surprising casualties of the worldwide turndown was the credibility of many European countries as sovereign borrowers. The drop in GDP led to sharp declines in revenues and opened gaping holes in budgets. Pessimism regarding future growth,
given the rise of emerging economies in the rest of the world, gave the feeling that the ‘Emperor had no clothes’—that debt service, believed by all for decades to be a sure thing, was in fact highly contingent. Credit rating agencies, which should know
more about the risk of default of borrowers than anyone else, had given all Eurozone countries very high to highest ratings in the years when the monetary union began. Many countries were now perceived to be in danger of default.
It all began when large financial institutions began to sell off their Greek government bonds in the late fall of 2009. They did this because the new Greek government had admitted a few weeks earlier that the budget deficit was in fact twice as large as
reported in the official statistics. PIMCO, one of the largest bond funds in the world with more than $1.3 trillion of bonds under management, began to sell their holdings of Greek debt. Greece’s budget problems had been known for decades (see Table
17.3), yet investors were convinced when the euro was introduced that no euro-using country would ever default—in a clinch, the Eurozone would take care of it. Other investors, seeing PIMCO selling, began to sell too.
Like a contagious disease, investors fled to the door, selling not only Greek bonds, but those of other troubled Euro countries as well. Because PIMCO also sold their debt holdings of other problem countries, prices of Portuguese, Spanish, and even
Italian bonds fell, raising their interest rates. Figure 17.6 shows how it all began in December 2009 and spread like wildfire. Credit rating agencies, long on the sidelines, saw rising bond yields as a bad sign and proceeded to downgrade the
governments involved, in a series of small steps. As the governments initiated painful budget cuts, the economic situation deteriorated even further, and rating downgrades continued.
Under these conditions it is difficult if not impossible for countries to borrow fresh money—even if they are on a clear and credible trajectory of improvement. Along with the International Monetary Fund, the Eurozone countries undertook to provide large
loans to the countries that had lost access to financial markets: Greece and Ireland in 2010, and Portugal in 2011. Figure 17.6 shows that these loans did not stop the risk premia from rising. The markets looked at the five solutions listed earlier and did
not like what they saw. They considered that the conditions attached to these loans, chiefly the requirement that the primary deficits be quickly curtailed, would cut growth and inflation, making it even harder to stabilize the debt–GDP ratio, for the reasons
already explained. Inflationary finance was ruled out by the ECB, always keen to guarantee price stability. Raising long-term growth was far too far away to deal with an urgent crisis situation.
That left two options: debt restructuring (a polite way of describing partial default) and interest rate reduction. In 2011, Greece was asked to default—the default occurred in early 2012—on its bonds held by the private sector, which by then had sold
much of its holdings to the public sector. Interest rates charged by other governments were reduced. Yet, the crisis countries did not recover access to financial markets. In the end, in the Summer of 2012, the ECB stepped it by promising to do ‘whatever
it takes’ to preserve the Eurozone. Figure 17.6 shows that this statement triggered a lasting decline in interest rates. Greek rates flared up again in 2015 as a newly elected government clashed with the other Eurozone governments, refusing fiscal
austerity and other conditions imposed for previous loans. The Greek government finally gave in, in exchange for further reductions of the interest rate on loans by the other governments, but the country has not recovered from the quarrel.
Fig. 17.6 Yields on Greek, Irish, Portuguese, Spanish, and Italian 10-year Bonds, Relative to Germany, 2009–2011 (%per annum)
The curves depict the difference between market yield on a 10-year bond in each of the five countries and that of Germany. In December 2009 yields of Greek bonds began to rise, followed by Portugal and Ireland, then Spain and Italy. In spite of a series of very large loans by the
IMF and other Eurozone countries, the yields kept rising until the ECB’s pledge in mid-2012 to ‘do whatever it takes’ to preserve the monetary union.
Sources:OECD.
17.5.5 Long-run Economic Growth

The last possible way to get the debt–GDP ratio down is to raise the denominator, GDP, in a sustainable way—this means raising the long-term economic growth rate ( g). This solution is difficult to conjure up in the short run, but the experience of several
European countries in the years 1980–2000—Denmark, Germany, Ireland, the Netherlands, and Sweden—suggests that it is not out of the question. It does require perseverance and sacrifice. In the next chapter we will explore some of these so-called supply-side
policies. They do not offer instant cures, but a prospect of longer-term sustainability.
Summary
1 One fundamental purpose of fiscal policy is to provide public goods and services. The boundary between what has to be produced publicly and what can be produced privately is not clear-cut.
2 A second function of fiscal policy is the redistribution of income and the alleviation of inequities that may be generated by the market mechanism. Doing so, however, may lead to inefficiencies.
3 A third function is to use the budget to offset temporary or cyclical fluctuations. This is done by running deficits in bad years (financed by borrowing) and surpluses in good years (to repay previous borrowing).
4 Countercyclical fiscal policy has three main benefits: (1) tax smoothing, (2) private consumption smoothing, and (3) private income maintenance.
5 The fact that some households sometimes cannot borrow provides a justification for fiscal policy to step in and support private consumption smoothing.
6 If prices and wages are not fully flexible, fiscal policy can be used to stabilize demand, either directly through government spending, or indirectly through taxation by reducing fluctuations in private sector incomes.
7 When they vote on the budgets, parliaments set public spending levels and tax rates. During a recession (respectively, an expansion) tax receipts decline (respectively, an increase) leading to a deficit (respectively, a surplus). As a result, the budget acts as an automatic stabilizer.
8 The operation of automatic stabilizers makes it difficult to interpret changes in the budget. The cyclically adjusted budget balance provides a way of disentangling the endogenous response to cyclical fluctuations from exogenous discretionary government actions.
9 Indebtedness is an inherently explosive process. When the primary budget is balanced and the debt is positive, borrowing is necessary just to service existing debt. The real debt accumulates at a rate given by the real interest rate.
10 To stabilize the level of the real debt, in the absence of real economic growth or monetary finance, the government must run a primary budget surplus equal to the interest charge. The longer it waits, the larger will be the debt and the interest burden that it faces, and the larger the
required primary budget surplus.
11 In a growing economy, stabilizing the ratio of debt to GDP is a less stringent condition than stabilizing the absolute debt level. The required primary surplus is proportional to the difference between the real interest rate and the real GDP growth rate. Not only is this smaller than the
real interest rate, it may well be negative, thus allowing permanent primary deficits.
12 Inflationary finance reduces the debt burden for two reasons. First, seigniorage provides resources directly to the government, virtually free of charge. As money growth eventually leads to inflation, the real value of nominal debt declines. Second, the inflation tax reduces the real value
of nominal government debt which is not indexed. The inflation tax can be collected, however, only if inflation is unexpected. Otherwise the nominal interest rate rises, which protects lenders.
13 In addition to lowering the deficit—through spending cuts or tax increases—or resorting to money finance, debt can be stabilized by default. This drastic form of taxation has severe consequences for a government’s reputation.
14 As long as the public debt is held by residents, debt stabilization or reduction implies income redistribution within the country. When part of the public debt is held by non-residents, stabilization requires a net transfer by residents to the rest of the world. A default redistributes wealth
in the opposite direction.
15 Debt stabilization is also possible in the medium to long term by achieving a reduction in interest rates at which government debt is refinanced. This requires investing in credibility and removing fears of default. Increasing economic growth, or at least reaching the potential level of
output, can also help stabilize the debt–GDP ratio, but requires a great deal of patience.
Key Concepts
public goods
seigniorage
non-rival
non-excludable
externalities
human capital
productive efficiency
equity–efficiency trade-off
tax smoothing
countercyclical fiscal policy
automatic stabilizers
cyclically adjusted budgets
debt stabilization
inflation tax
Exercises
1 Are bank notes and coins part of the public debt? Should they be? Why or why not?
2 A country’s budget law sets public spending at €30,000 million and the tax rate applied to income is 25%. Last year’s GDP was €100,000 million and the forecast is that it will grow by 3%. What is this year’s budget surplus (or deficit) set by the law? What will happen to the balance
if the economy does not grow at all? If there is a deep recession, a decline of output of 3%?
3 A country growing at a rate of 2% has a debt–GDP ratio of 100%. What is the primary budget surplus that keeps this ratio constant when the real interest rate is 2%? When it is 6%?
4 Suppose the debt–GDP ratio is 150%, growth is 3% per annum, and the real interest rate is 5%.
5 What is the primary government budget surplus (as a percentage of GDP) that can stabilize the debt–GDP ratio?
(a) How does your answer change if the interest rate falls to 4%? If growth falls to 2%?
(b) The demand for (real) central bank money, the source of seigniorage, declines with the rate of inflation. Suppose, as an example, that this demand (in billions of euros) is given as follows:
Seigniorage is a tax applied to this demand, whose rate is just the rate of inflation. Compute seigniorage as a function of the inflation rate. (Hint: an inflation rate of 5% corresponds to a tax rate equal to 0.05.) Which inflation rate maximizes seigniorage?
Inflation rate 0% 1% 2% 5% 10% 20% 25% 50%

M0/P 1,000 905 819 607 368 135 82 7
6 Figure 17.5 determines four quadrants (ignoring the diagonal). Characterize and comment on the situation in each quadrant.
7 The budget balance is –300 and GDP is 20,000. Potential GDP, however, is 20,600. Compute the structurally adjusted budget using the following information:
• public consumption is not cyclical;
• transfers are as follows: TR = 13,000 − 0.5Y;
• taxes are: T = 2,000 + 0.3Y.
8 Consider a country where the public debt stands at 60% of GDP. The real interest rate is 3% and the trend growth rate is 2%. In the absence of seigniorage, what is the primary budget required to stabilize the debt level? The debt–GDP ratio?
9 Why does a primary surplus in excess of what is found in (17.2) imply that the public debt will eventually disappear? Why does a primary surplus in excess of what is found in (17.4) imply that the public debt will eventually disappear as a fraction of GDP?
Essay Questions
1 Make a list of all public services that you can think of. In each case, examine whether they can be privately provided.
2 ‘The national debt is a great scam, because it will never be repaid.’ ‘The national debt is irrelevant because we owe it to ourselves.’ Comment on these two popular characterizations of government indebtedness.
3 Critics of the Stability and Growth Pact argue that it prevents the operation of automatic stabilizers during cyclical downturns. The proponents respond that the way to deal with the pact is to run structurally adjusted surpluses. Explain and comment.
4 Between 2006 and 2010 the Irish public gross debt rose from 37.8 to 96.2% of GDP. Collect data on Irish budgets and growth during this period to explain this remarkable evolution (possible sources: OECD Economic Outlook database, Ireland’s Finance Ministry’s website:
< www.finance.gov.ie>).
5 Debt stabilization is controversial. One view is that it requires running balanced budgets, irrespective of the possible cyclical implications. Another view is that it is far better to use fiscal policy to support rapid growth. Evaluate these arguments.
1 Chapter 18 discusses the characteristics of public goods and their role in economic growth in more detail.
2 Public sector borrowing can occur domestically or abroad. A country with a temporary fall in income must either borrow or reduce its absorption. As a first approximation, it does not matter whether it is the private or the public sector that borrows abroad. Chapter 15 has already discussed the
international ramifications of intertemporal imbalances for the real exchange rate.
3 We explain the mechanism of credit rationing and its importance for consumption in Chapters 6 and 8.
4 The budget balance is T − G − TR, where G is public spending on goods and services, TR are transfers, and TX are tax revenues (total spending is G + TR). G is not cyclical, TR is rises in recession, and TX is procyclical (households and businesses pay more taxes when the economy is doing well).
Net taxes are TX − TR; they must be procyclical, i.e. move with the cycle.
5 The table presents gross debt. Net debt figures, which take assets of the government into account, are considered unreliable because of the difficulty involved in valuation of state assets.
6 Chapter 19 provides more details on requirements imposed by the EU and the European Central Bank for countries adopting the euro, of which the most important involve fiscal policy.
7 Note that B represents the net public debt, which is the government’s gross debt less its assets. The reason is that the government should normally receive interest or income from its asset holdings, which tends to reduce the overall budget deficit, exactly in the way debt service tends to increase the
deficit. If L and A are, respectively, the government’s liabilities and assets, net debt is B = L − A, and rB = rL− rA.
8 The tendency for the debt to grow at the rate of interest r can be seen by looking at (17.1) when the primary budget is balanced: then, ∆B = rBor ∆B/B = r, which is the growth rate of the public debt.
9 The rate of growth of the debt–GDP ratio B/Yis a formula that we have already seen in many contexts, beginning with Chapter 5 (Box 5.3). The debt–GDP ratio declines when the real debt grows at a lower rate than the real GDP.
10 It should be remembered that the ‘base’ of the inflation tax is the demand for real high-powered money (M0/P), which is also an endogenous variable. It depends negatively on the opportunity costs of holding it—the inflation rate. If inflation is too high in the long run, the decline in the ‘tax base’
M0/P might ultimately dominate the gain from higher inflation!
11 The origins of this expression can be traced back to the Middle Ages, when local lords had the monopoly over coinage. Often, they abused this power to reduce—debase—the gold content of coins minted in their domains. In a similar fashion, governments exploit the monopoly power of the
central bank in creating the medium of exchange to acquire valuable resources from the private sector. Modern seigniorage is just another form of taxation. The authorities exchange money, which is costless to produce, against goods and services. It is as if these goods and services were simply
confiscated.
12 More generally, unanticipated inflation redistributes wealth from lenders to borrowers when the assets are nominal, i.e. set in money terms and not indexed to a price level.
13 Seigniorage still takes place at the level of the European Monetary Union. Indeed, the European Central Bank (ECB) is required to turn over its profits to participating countries. But the decision on how fast to let the money base grow, and how much seigniorage to collect, no longer lies with
national governments, but solely with the ECB. By virtue of its founding statutes, the ECB is formally independent and required to aim at price stability. It is legally prohibited from financing the government budgets. For a detailed account of the fiscal policy restraints facing countries wishing to join
the European Monetary Union, see Baldwin and Wyplosz (2015).
14 More details on the so-called ‘Maastricht Criteria’—which resulted from the Maastricht Treaty (1992)—can be found in Chapter 19.
15 This box draws on Alesina (1988).
Policies for the Long Run
18
18.1 Overview
18.2.1 The Benchmark of Efficiency: Perfect Competition
18.2.2 Imperfect Competition and Economic Rents
18.2.3 Market Failures and Overall Market Efficiency
18.3.1 Dealing with Externalities
18.3.2 Dealing with Malfunctioning Markets
18.4.1 Taxation as a Necessary Evil
18.4.2 The Effect of Taxation on Efficiency
18.4.3 Adverse Effects of Taxation on the Tax Base
18.5.1 Heterogeneity and Imperfect Information
18.5.2 Imperfect Contracts and Labour Market Regulations
18.5.3 Social Policies Incentives and Taxation
18.6.1 Waiting for Godot? The Long and Variable Lags of Supply-Side Policy
18.6.2 The Political Economy of Supply-Side Reforms
18.6.3 Feasible Supply-Side Policy: Peanuts or Big Payoffs?
Summary
Economic growth—meaning a rising standard of living for a clear majority of citizens—more often than not fosters greater opportunity, tolerance of diversity, social mobility,
commitment to fairness and dedication to democracy . . . But when living standards stagnate or decline, most societies make little if any progress toward any of these goals, and in
all too many instances they plainly retrogress.
Benjamin Friedman
18.1 Overview
In Chapter 16, we explored the role of demand management in macroeconomic policy, and Chapter 17 highlighted the usefulness and limitations of fiscal policy in
the toolbox of demand management. Demand-induced recessions are costly in terms of unemployment and lost output; recovery can be accelerated by well-
chosen activist policies. Yet, since the economy returns to its long-run path anyway, demand-side policies can only have temporary effects. Permanent effects
would require raising the potential GDP path or, even better, the slope of that path—potential growth.
Policy measures designed to raise the long-run or potential GDP are known as supply-side policies. This is shown in Figure 18.1, which plots the level of
output Y itself on the horizontal axis rather than its deviation from potential output as in previous chapters. Successful supply-
side policies raise potential GDP from to faster than if it were left to the normal process of economic growth. The attractiveness of such policies is that
they bypass the uncomfortable trade-off between output and inflation. In the short run, more output and lower inflation are possible as the economy moves from
A to A′.1 At the same time, bad supply-side policies can reduce Y, causing higher inflation in the short run and higher unemployment and lower output in the long
run.
Fig. 18.1 The Macroeconomics of Supply-Side Policies
Supply-side policies aim at mobilizing productive resources to increase equilibrium output. Sustainable shifts in the aggregate supply curve yield both more output and less inflation
in the short run (point A’). In the longer run, inflation returns to its previous level (as indicated by the long-run aggregate demand schedule) but output is permanently higher.
This chapter surveys a broad spectrum of supply-side policies and assesses their effectiveness. These policies can target labour, product, or financial markets.
They may also involve fine-tuning existing government interventions such as taxation or subsidies, or involve changing regulations of product and labour
markets. Some reforms merely require the streamlining of bureaucratic rules, the uniform application of law, or the elimination of corruption. Others work directly on
improving incentives to save, work, and produce. They can increase general efficiency, possibly by requiring less competitive, inefficient industries to shrink,
freeing resources for other, more valuable uses. An important warning regarding supply-side policies is that they do not produce immediate results, often
requiring five to ten years—sometimes even longer.
We examine three broad approaches to increasing the economy’s long-run productive potential. First, good supply-side policy should aim to make product,
labour, and financial markets as efficient as possible; and when markets fail that test, government intervention can improve matters. This was also the message of
Chapters 3 and 17. Second, given that governments are already interfering in the marketplace for both good and bad reasons, they should strive to minimize the
negative impact of their interventions. One example is regulation; another is taxation; yet another is subsidy policy. Third, unemployment remains a deep concern
for Europe, where on average 10% of the labour force is out of work. Yet, in the last 15 years, many countries have made significant inroads against high long-term
unemployment problem, including the UK, the Netherlands, Denmark, Finland, Germany, Ireland, and Sweden. Even in the course of the recent cyclical downturn,
these countries have proved that unemployment need not be a curse. Even if the political resistance to such reforms is strong, all of Europe can share in these
successes. As the quote at the beginning of the chapter suggests, economic growth helps societies sustain their dynamism, maintain social cohesion, and
improve standards of living for all of their members.
18.2.1 The Benchmark of Efficiency: Perfect Competition

Adam Smith (1723–1790), perhaps the most famous economist of all time, claimed in his book The Wealth of Nations that efficient resource allocation was best left
to the market system. One of the greatest achievements of modern economics has been to confirm that, under ideal conditions, market economies can achieve
optimal employment of resources.2 What are those ideal conditions? Most importantly, markets must ‘work’, meaning that prices must be free to adjust. There
must be sufficient competition so that new firms can offer similar or identical products and eliminate excessive profits due to monopoly position. Externalities
discussed in Chapter 17 must be absent. Information should be equally available to all agents. Not surprisingly, these ideal conditions are unlikely to be met in
practice. If not, government intervention can improve the allocation of resources and possibly achieve a greater level of output, and ultimately well-being, in the
economy.
An important example is the labour market, studied in Chapter 4. ConsiderFigure 4.7, which shows equilibrium in an ideal labour market. Firms’ demand for
labour is given by the downward-sloping marginal product of labour curve (MPL). Firms hire labour to the point at which its marginal productivity equals the real
wage. Similarly, workers supply hours of labour up to the point at which the utility loss of forgone leisure time is greater than the gain from goods earned from
working—the real wage. The market-clearing equilibrium point A can be seen as an ideal state. Increasing employment beyond point A in Figure 4.7 reduces the
MPL below what workers are willing to work for. Reducing employment below A would raise the MPL above the real wage which would be sufficient to motivate
labourers to give up their free time. Put another way, if wages are above the market-clearing level, firms would restrict employment and workers would increase
their supply, resulting in involuntary unemployment. If real wages are too low, firms demand more man-hours than workers are willing to supply.
What pushes the real wage towards its equilibrium, market-clearing level w at point A in Figure 4.7? The answer is competition. In an idealized, perfectly
flexible market for labour, a wage lower than w will not attract as many workers as firms want to employ. Competition among firms will lead some to improve their
wage offer. As they succeed in luring workers from their previous jobs, other firms will have to respond with a higher wage offer. Competition will bid up wages
until they reach the market-clearing level. Conversely, imagine that the wage exceeds w. Some will not find a job, or will not be able to work as many hours as they
wish. Forces of competition will trigger a decline in the wage as unemployed workers underbid those who have jobs, eventually driving wages to point A.
That point A represents an ideal state is a general principle: for an economy to achieve the optimum allocation of resources, the market must clear. In fact, all
markets must clear. Either all prices are right and all productive resources are fully employed, or none of them is, and under-utilization and inefficiency spread
throughout the economy. Economists tend to be split into two camps on this issue. Some are convinced that markets are naturally efficient, so there is no need for
governments to interfere. This is the laissez-faire view. Others believe that few markets meet the high standards set forth by Adam Smith. The existence of
market failures, when and where they can be identified, provides a justification for government intervention. Note, however, how cautiously this statement is
worded. Interventions are only justified if two conditions are met. First, they should be limited to clearly identified market failures. Second, they should be targeted
directly at the market failure to avoid creating additional distortions of their own. Misguided policies can do more harm than good. Policy failures, too, must be
eliminated. In the s ections that follow, we identify several generic market failures and discuss the associated range of possible interventions.
18.2.2 Imperfect Competition and Economic Rents

When several firms compete with each other in the same market, they are under constant pressure to adapt the price and design of their product to the desires of
the consumer—otherwise they will lose sales and be forced to exit the market. They will try to operate as efficiently as possible, and in doing so perform the
function society wants them to: allocating available resources to their most efficient uses. And that is exactly what Adam Smith predicted.
Competition, however, can be painful. Every economic agent wants protection from competition. Firms strive to acquire a leading position in the market at the
expense of their rivals, thereby earning above-average profits, that are appropriately called economic rents. How do they do this? Firms normally strive to
establish market power, or even market dominance or monopoly status, by exploiting increasing returns to scale or by differentiating their products. When
production exhibits increasing returns to scale, larger firms can squeeze out smaller firms by producing more, and thus reducing their costs. Eventually only a few
firms survive and then dominate the market (think of the market for automobiles). Firms sometimes restrict competition by preventing others from entering their
markets. Most means of maintaining dominance in a market are prohibited; threats against competitors and extortion are illegal. Yet it is standard practice for
producers to spend significant amounts of money creating brand names and convincing consumers that their products are fundamentally different or unique.
Customers who enjoy consuming products that are different—or merely believe that they are—will be willing to pay more for them. Firms use clever marketing to
obtain and defend this monopoly power, and can justify the large sums of money spent on product advertising by the resulting profits. Product differentiation
thrives on individuals’ desire for variety in the marketplace. It is indisputable that the consumer would prefer variety to monotony, and the quest by firms to
innovate and produce variety is spurred by competition.
Box 18.1 EU Competition Policy and National Preferences
Economists love competition because it generally leads to more efficient outcomes. This enthusiasm is rarely shared by firms operating in non-
competitive markets, where large profits may be at stake, and by their employees who share in the rent. Policy-makers and even consumers may be
lukewarm to competition-enhancing measures. Especially in the European Union, where cultures and traditions are defined by national boundaries,
reservations abound that unbridled competition might weaken or destroy national identities. It is sometimes very difficult to distinguish between respect for
consumer preferences for variety and restraint of trade. The latter means less competition, less efficiency, lower output, and fewer jobs without any real
gain to consumers.
Examples of this dilemma facing the EU abound. In a highly publicized case involving beer quality in the late 1980s, Germany was taken to the European
Court of Justice, which has the final say in matters of EU competition policy. Since 1516, German brewers had produced the golden beverage according to
the Reinheitsgebot (‘beer purity law’), which limited the production of beer to four ingredients (water, yeast, malted barley, and hops). Comparably strict
regulation of beer quality did not exist elsewhere in Europe.3 In effect, the law excluded ‘impure’ foreign beer that did not meet the exacting standards of the
Reinheitsgebot from the German market. The European Court of Justice found that beer brewed using rice and other ingredients should be allowed, while
allowing German brewers to market their own brew as compliant with the previous regime. Many years later and despite the new rules, national tastes
prevailed; most German beer drinkers continue to drink German beer. Similarly, Italians now must have a choice between granoduro pasta and those
made with other wheat varieties. French cheese consumers can try Danish blø if they wish. In theory, the consumer is better off.
The Reinheitsgebot case raises a number of interesting questions. Should the EU forbid the import of beef from the USA that is treated with hormones or
antibiotics? Should genetically modified plants be allowed? Should imports of foreign ‘culture’—in particular, audio and video recordings—be limited? Can
governments set a minimum price for books in bookstores, as in France, Germany, or Austria? Such regulations are often justified using non-economic
arguments. The question remains whether these truly outweigh their economic costs.
Measures which increase competition and thereby economic output are generally referred to as competition policy. Competition policy can take a number of
forms. In many countries, monopolies are regulated or supervised closely. Firms are seldom allowed to acquire excessive shares of their markets. Collusion in the
form of cartels and price-fixing schemes is illegal. More recently, governments have also limited the ability of firms to dominate their input and output markets. One
famous case in Europe involved the Microsoft Corporation and the market power created by its operating system and internet browser programs. Similarly, the UK
government separated electricity generation from electricity distribution as well as train service from rail ownership with the intent of creating more competition.
The EU Commission has moved in this direction at the Community level with respect to energy transmission networks. More generally, as European integration
proceeds and the relevant market expands to the level of the EU, anti-monopoly powers are being transferred to Brussels. Some interesting cases of EU
competition policy are discussed in Box 18.1.
Firms are not the only economic agents which try to avoid competition. Labour markets are frequently characterized by non-competitive behaviour. First, labour
supply is intimately linked with the human condition, and competition among workers for jobs is often considered inappropriate, in bad taste, or even unethical.
Second, wages are often set in bilateral negotiations, and trade unions are seen as protecting interests of employed workers in a vulnerable situation. An
alternative perspective sees trade unions as monopolists with high real wages in mind.4 With high real wages, labour demand is low and equilibrium
unemployment high. If a majority of trade union members accept this trade-off, they will press for an outcome that is not efficient. In a similar way, firms like to
control or even reduce competition on labour costs, and are frequently organized in employers’ associations. In Europe, these associations frequently take
decisions that are also binding for members, and may be followed by non-members (usually smaller firms). If labour and management set wages in collective
bargaining without considering the interest of non-member workers, typically the unemployed, and of non-member firms, the level of competition is reduced, and
the economy is less efficient. This subject will be taken up again in Section 18.4.
Competition in labour markets is not limited to labour unions and management. A number of practices by professional associations—of lawyers, doctors, and
architects, for example—also restrict the supply of labour and raise wages. Limitations on international immigration can be interpreted as the ultimate restriction of
competition for labour. Because so many political and social issues are associated with immigration, it would be premature to advocate open borders for everyone
on purely economic grounds. Yet it remains true that immigration is efficient, usually increasing GDP enough to compensate those in the receiving country who
are made worse off, which is rarely the case. Box 18.2 gives more details.
18.2.3 Market Failures and Overall Market Efficiency

The goal of supply-side policy is to improve the functioning of markets and thereby create efficiency gains. Greater efficiency is like a free lunch—in theory, it can
be redistributed to other members of society without making anyone worse off. This is the Pareto principle presented in Box 18.2. In general, this means
intervening in markets which do not function properly. While Chapters 16 and 17 dealt with temporary interventions to speed up the elimination of demand
shortfalls or to cool off a booming economy, the problem here concerns failures of markets to achieve efficient outcomes. Solutions to these market failures call for
permanent new policies or reforms which ultimately raise equilibrium output. But many of these failures simply cannot be eliminated; they are in the nature of
things. Supply-side policies may help by reducing the extent and the impact of these failures. We start by looking at the origins of ‘natural’ market failures, which
we group in four general categories:
externalities
public goods
increasing returns
information asymmetries.
Externalities
As discussed in the last chapter, an externality occurs when someone’s action inevitably affects others. Pollution is a good example of a negative externality:
when a firm dumps its waste in a river, those who live or work downstream suffer and have to expend resources to purify the water that they drink. Market
outcomes involve too much production of negative externalities, because the firms and individuals do not recognize the costs they impose on the rest of society.
In contrast, positive externalities spill over to others and provide society with a better outcome. A general conclusion is that not enough positive externalities
will be produced. Putting flowers in your windows costs money, while passers-by enjoy looking at them at no cost. Since you bear the costs of putting flowers on
your balcony, you balance the price and the benefits (your pleasure). While putting out even more flowers might further raise society’s pleasure, it is
unreasonable to expect individuals to behave in the public interest (although they sometimes do). The market alone apparently cannot redress the externality
problem. If, however, property rights are well-defined, and it is not too costly to enforce these rights, it can.6 If, for example, the law specifies that everyone has
the right to clean water, the polluting firm will be obligated to expend resources to clean up the water. In facing this situation, the firm will adopt cleaner production
processes. Alternatively, it will offer to indemnify those who live downstream to prevent them from taking the case to court in a costly lawsuit. Conversely, the law
could grant firms the right to pollute. Then the costs would be shifted to the people who live downstream, who would have to pay to clean up the water, or offer
money to the firm for it to reduce its disposal or adopt other production processes. While the second solution may seem unfair (and it probably is), it is no less
efficient than the first. The general point is that, once property rights have been assigned, the market is more likely to deliver the best possible outcome, given the
presumption that nature makes it impossible to produce without creating waste.
Box 18.2 The Supply-Side Economics of Immigration
Because labour is so abundant relative to capital in developing countries, many, many people there are very, very poor. In India, for example, GDP amounts
to less than €4.50 per day. Yet many Indians are highly educated and would readily migrate to Europe where the hourly wage might average €20–25 per
hour or even more. Even those without exceptional skills but who are able-bodied and willing to work could compete with the less skilled. In addition, many
people are seeking refuge from tyranny and war, not just economic betterment. One can hardly blame them for wanting to move. Is it economically efficient
to keep these people out?
The analysis of Chapter 4 suggests that it is not, and Figure 18.2 shows why. Two labour supply curves are drawn, one corresponding to that of the
native population only, without the immigrants. The second curve to the right is the sum of natives plus newly arrived migrants. For the moment, the labour
demand curve is assumed unchanged by migration. It is equal to the marginal product of labour. For a given wage level, the area under the curve and
above that wage represents total profits earned by firms in the economy. In Figure 18.2, profits are thus AWB before the migrants arrive. After they arrive, the
aggregate labour supply curve shifts outwards, wages decline from W to W ′, and employment rises from L to L′. Total GDP rises by the amount BLL′D, the
sum of the additional marginal products of the newly employed. Without migration, this increase in GDP would not have occurred.
So if immigration is economically efficient, why all the fuss? First, the distribution of the increase in GDP is not shared by all: profits rise by WBDW’ and
the migrants earn CDL′L′′, but wages of native workers have in fact declined from WBLO to W′CL′O. Only if the demand for labour shifts out—which may
occur if there is new investment by natives, or if the immigrants themselves bring capital with them—would the outcome be different. Second, immigrants
place extra demands on infrastructure and the social safety net which they have not helped to pay for. Third, the Pareto principle 5 establishes that a change
is socially desirable under two conditions: (1) it increases overall income; (2) those who gain compensate those who lose out. In other words, part of the
surplus BLL′D ought to be redistributed to compensate workers for their loss from WBLO to W′CL′O. This is rarely the case. Finally, although economists
like to emphasize the efficiency of immigration, other sociological aspects may be more important: humankind’s complex tribal, if sometimes crude and
primordial, instincts.
Fig. 18.2 The Supply-Side Economics of Immigration

The consequence of immigration is to shift total labour supply outward. In the new equilibrium, wages decline from W to W′ and employment increases from L to L′. While
GDP rises unambiguously by BLL′D, this gain is not evenly spread: the income of capital increases from AWB to AW′D, migrants now earn CDL′L′′, while the income of
natives declines from WBLO to W′CL′′O.
Externalities can also be transmitted via market prices. For example, the efforts of countries to develop biofuels may lead to higher prices around the world for
grain and other agricultural products, which may in turn lead to malnutrition in poor countries. These pecuniary externalities are properly addressed by
changing the distribution of wealth and income. But other, inherently non-pecuniary externalities are more challenging. Assigning property rights and
creating a market is often not enough to solve the problem. An example is training and education. Better trained people share their knowledge with co-workers and
their own children. Thus their knowledge has a value for society at large and, if they are not compensated, they will not accumulate as much human capital as is
optimal from the standpoint of the economy, even if the firm invests more in workers out of motives of pure profit maximization. For this reason, society has an
interest in stepping in, and subsidizing education. In Chapter 3 we saw how externalities involving education, law and order, and health can influence long-run
growth. We look at the policy implications in Section 18.5.
Public goods
Public goods, such as parks, clean air, or information, are special. They are non-rival, meaning that consumption by one does not make them less available to
others, and they are non-excludable, meaning that, when they are available, everyone can use them freely. As a consequence, no one can be charged for using
them, and they must be provided by the state. Law and order is an externality, but it is also a public good that can and must be provided by a functioning
government. In societies which have become ‘failed states’, law and order are absent, private justice and local warlords take over, economic activity collapses, and
poverty spreads. This has been the case in countries torn by civil wars, such as the Democratic Republic of Congo, the Ivory Coast, Liberia, Libya, Syria, or
Somalia.
Increasing returns
Some industries are characterized by increasing returns to scale. A good example is railways. A railway that just serves two cities will only be used by people
living in or going to these two cities. If it includes a vast network, the link between the same two cities will be used by many more people, many of whom will only
pass through these two cities. The larger the network, the more valuable is each of its sections, and the larger the profits to be earned from each section. Left to
the market, competition can easily lead to a natural monopoly, with just one company owning the entire network. But we know that, once it has established
itself, a monopoly will tend to charge high prices, resulting in the socially inefficient under-utilization of its product. A good case can be made for the state to
intervene, in this case by granting monopoly rights to a company—which can be public—in exchange for a regulation which will prevent monopolistic pricing.
Recent discussions concerning the divestiture of European power transmission networks or the opening of national railways systems to foreign competition
reflect these concerns.
Information asymmetries
It is a simple observation that we know more about ourselves than others. This information asymmetry plays a pervasive role in all sorts of markets. Traders often
hide information that may reduce the price they can get for their products. Two examples: used car dealers and traders in financial markets. Both have little
incentive to tell the truth in their businesses and customers know it, but cannot do much more than be suspicious or avoid their businesses. In the first case, the
market can easily dry up, even if the salesmen really do tell all they know about the cars they sell. In the second, traders make profit by reacting faster to new
information, so they will naturally tend to hide what they know and what they are up to. Moreover, one trader’s move, if interpreted by the others in the wrong
way, can trigger a stampede, as explained in Chapter 7. In both cases, the state can play a constructive role by prescribing what information must be provided by
sellers of used cars or by forcing financial firms to disclose much information that they would prefer to keep to themselves.
Even die-hard free market advocates recognize that markets are subject to failures, that these failures may have serious implications for economic performance,
and that only the state can solve most of the associated difficulties. Should the state intervene and solve all market failures? Here economists often disagree. The
public choice school, founded by Nobel laureates Friedrich Hayek and James Buchanan, maintains that policy failures are more important than market failures; the
government, when it intervenes, does more harm than good. Others view state intervention as unavoidable, especially in modern societies where economic
interactions can be quite complex and thus serve as conduits for the diffusion of adverse effects of market failures. It is fair to say that no country completely
refrains from intervening in the marketplace, whether for the right reasons or the wrong ones.
Fig. 18.3 The Impact of Supply-Side Policies
Beneficial supply-side policies are depicted as an increase in the long-run level of output in Panel (b). As shown in Panel (a), this increase can result from 1) a shift in the
production function due to an increase in total factor productivity or an increase in equilibrium inputs of other production factors while labour input is held constant, or 2) as an
increase in labour input, holding the production function and other inputs constant. The outcome 3) is a shift from Y to Y’ in Panel (b).
Many market failures not only affect the level of output, but also its growth rate. In Chapter 3 we considered the potential for those policies mainly directed at
enhancing growth. Similarly, Chapter 7 addressed policies that deal with financial market failures. In the following sections, policies are discussed which have the
objective of rooting out market failures, improving the functioning of markets by increasing competition or the availability of useful information to market
participants, or simply making government intervention less costly from the economy’s perspective. All are examples of good supply-side policy. They have the
common feature of shifting out the vertical LAS curve in Figure 18.3 to the right, from to . This can occur either as (1) a one-off improvement in total factor
productivity which increases output, holding productive inputs constant, or (2) by increasing the supply of productive factors, holding the level of technology
and the position of the production function constant. Bad supply-side policy has the opposite effect, reducing output produced by the same volume of factor
inputs, or by inducing households and firms to withdraw productive factors from market uses.
18.3.1 Dealing with Externalities

The basic principle is to ‘internalize externalities’. This means making those that create an externality bear its costs or the benefits. In the pollution example, this is
achieved by making clean water a right or an entitlement. We have illustrated this principle with a few important examples which have recently received prominent
attention.
Human capital accumulation

The first example, also mentioned in Section 18.2.3, is investment in human capital. It is one of the most crucial factors of economic development. Human capital
is acquired at school, of course, but also at work. It may be best to spread training over a lifetime to allow people to update their depreciating capital and learn
about newly acquired knowledge. The knowledge of an individual person benefits society in a myriad of ways. It increases the productivity of co-workers, of
course, but better-educated people can make better decisions in everyday life—including taking better care of themselves and their children, or voting as informed
citizens. An individual’s return from human capital is believed to be significantly smaller for the individual than for society. If they invest with only their individual
return in mind, they will underinvest regarding what is justified by the broader social return. For this reason, societies have an interest in encouraging people to
invest in more human capital than they would on their own. Free and compulsory education was the first response to this externality. Universities typically cannot
fully cover the cost of educating students, even if students pay tuition fees. The rest is publicly funded. Subsidizing traineeships in firms or continuing education
have the same goal in mind. In addition, unemployed people tend to see their human capital depreciate. This provides an additional justification for active labour
market policies described in Section 18.5.1.
Law and order

Law and order has all the characteristics of a market failure: it is an externality, it is subject to increasing returns to scale, and it is a public good. It is an externality
since everyone benefits from others being honest. Investment in physical and human capital is threatened when crime robs people and firms of their assets, and
will be less attractive in the absence of law and order. It is subject to increasing returns to scale, since commonly accepted and enforced law works better than
when everyone sets their own rules and practises self-enforcement. Private protection, the spontaneous market response to the absence of law and order, is
inefficient. It also accentuates the effect of inequality, as rich people can better defend themselves than poor people. Law and order is also a public good since it is
non-excludable and non-rival. Thus, on all three grounds, leaving law and order to the market results in a massive failure. Indeed, since time immemorial, one of the
key attributes of any political power has always been the provision of law and order.
Table 18.1 Ease of Doing Business Rankings, 2015
Country Ranking Country Ranking Country Ranking

Singapore 1 Germany 15 Bulgaria 38
New Zealand 2 Estonia 16 Croatia 40
Denmark 3 Ireland 17 Hungary 42
Korea, Rep. 4 Iceland 19 Belgium 43
Hong Kong SAR 5 Lithuania 20 Belarus 44
United Kingdom 6 Portugal 22 Italy 45
United States * 7 Poland 26 Cyprus 47
Sweden 8 France 27 Serbia 59
Norway 9 Spain 33 Greece 60
Finland 10 Czech Republic 36 Luxembourg 61
Romania 37 Albania 98
Source: The World Bank (http://www.doingbusiness.org/rankings).
More insidious is the impact of law and order—or lack thereof—on the performance of the product market. In a dynamic economy, the ability to start a business
is a fundamental prerequisite for experimenting, developing, and bringing innovative new processes and products to the market. Frictions which impede
competition and consumers’ ability to benefit from it is harmful for the supply side of the economy. Some of these frictions may take the form of overzealous
regulation which protects insiders from competition; others may take the form of corruption—ranging from organized crime which shakes down new businesses
for the privilege to exist, as well as government officials who also ask for bribes or payment for permits, or apply regulation in capricious and unpredictable ways.
Since 2004, the World Bank has published an annual evaluation of the ease of ‘Doing Business around the World’, which ranks countries on the ease of
starting a business, obtaining construction permits, getting electricity, registering property, getting credit, protecting investors, paying taxes, trading across
borders, enforcing contracts and bankruptcy resolution. The ranking of the top 10 countries as well as a large selection of European economies appears in Table
18.1. It is not coincidental that those countries who have grown the fastest in the past decades are those that have relatively unbureaucratic, well-regulated, and
low-corruption business environments.
Health
Health is obviously a private ‘good’. Everyone wants to be healthy. It is also a source of externality. People with poor health do not work well and are often
absent; this reduces the productivity of their colleagues. Chronically ill people can lose their jobs and become destitute, relying on society’s generosity, or
resorting to criminal activity to survive. Sicknesses can also be contagious, presenting a vivid example of an externality. Health can be provided privately, as it is
in many countries. Private health programmes generate inequalities, and most people think that it is unacceptable that the rich receive better treatment than the
poor. But, in addition, because of the externalities that it generates, privately provided health is likely to be inefficient, as everyone will only spend up to the point
where their (marginal) benefit equals marginal cost and not consider the positive effects their health have on others. This is why most countries have established
systems of social security and enforce compulsory medical examinations and vaccination. A healthy society is a rich society. Causality runs from wealth to health,
but it can also go in the opposite direction.
18.3.2 Dealing with Malfunctioning Markets

The solution to market failures is not always to produce public goods and services. In fact, it is increasingly being recognized that private producers tend to be
more cost-efficient than publicly owned ones, provided that they are adequately regulated. In this section, we look at monopolies and review some prominent
examples of the privatization process and regulation in practice.
Monopolies
Markets characterized by increasing returns tend to evolve to a situation where a very few firms buy out or eliminate the others. When only one firm survives, it is
a monopoly. Once it has achieved that position, its incentive is to charge high prices, and possibly to stop innovating, and let the quality of its goods decline.
When just two or three dominate, they have incentives to agree among themselves to raise prices and lower competition in terms of quality and innovation. This is
called collusion. Thus, in the presence of increasing returns, markets evolve spontaneously to a situation where competition is insufficient to match the principles
of market efficiency. Examples of industries prone to increasing returns include automobile and aeroplane production, transport, electricity production and
distribution, and telecommunications.
One response is to regulate such industries. Governments step in and either break down the monopolies (e.g. the ongoing deregulation in the electricity,
telecommunications, and airline industries), or prevent mergers, and fight actively against collusion. Because competition increasingly takes place globally, such
anti-trust and anti-collusion policies are often conducted at the supranational level. The European Commission has been granted wide authority in the area of
antitrust law enforcement. For instance, it has blocked several mergers between international giants and imposed pro-competition conditions on others.
Privatization
In many countries, the other response to market failures, especially the existence of natural monopolies, has been to set up state-owned companies that would not
seek to exploit their monopolistic power. This has been the case of railways, electricity generation and distribution, water distribution, telecommunications, etc.
Since the early 1980s, European policy-makers have looked more critically at the performance of state monopolies. The prices that they charged were regulated,
and based on their costs. But who was minding the costs? Suspicion grew that, in the absence of competition, these firms were not producing at the lowest
possible cost. Monopolistic rents were not captured in the form of private profits, but in the form of slack production, poor service quality, technological
backwardness, and sometimes comfortable salaries, and other advantages. The contrast between Europe and the USA, which relied to a much lesser extent on
publicly owned companies, had become glaring.
The response was a wave of privatization, which is still underway. Once a company had become private, it was interested in expanding across borders. The
presence of state-owned monopolies in some countries prevented the entry of foreign competitors, while allowing the national company to expand abroad. Clearly,
this was an unfair situation. The creation of the Single Market in 1992 gave the European Commission the right to force national governments to privatize most of
their state-owned monopolies and to open their markets to foreign competition. One example of successful supply-side reform in Europe is a sequence of
privatizations and deregulation in the telecommunications industry described in detail in Box 18.3. This process has freed a dynamic force for economic growth
and job creation. Twenty years down the road, the process is still far from complete, a testimony of how entrenched the interests are.
Regulation
Regulation is often the best response when markets produce unacceptable outcomes. For instance, Chapter 7 showed that financial markets are essential for
economic growth and, yet, they are prone to undergo violent crises, a manifestation of the asymmetric information problem. As a result, they are regulated. In
principle, banks are not allowed to take excessive risk with the deposits of their customers, who cannot effectively monitor their banks and stand to lose part of
their savings if the bank collapses. They are also required to keep enough funds to absorb likely losses without threatening their depositors. Financial traders are
restricted in the amount of their business which they can fund with borrowed money (leverage), putting a legal ceiling on the risk that they can take, in an effort to
limit the occurrence of financial crises.
That regulation affects nearly every aspect of economic life does not only show how widespread market failures are, but also highlights the tendency of the
state to do too much. Asymmetric information explains why food labels, airline services, or driving licences are regulated. Yet the regulation of store closing times
in many parts of Europe, taxis in Greece, pharmacies in Germany, or chimney-sweeping in Switzerland might really be a case of favouring private interests at the
expense of the public good. The recent controversial discussion of Uber and other new forms of internet commerce in France highlight the difficult questions
involved in regulation and deregulation.
Box 18.3 Telecommunications: Successful Supply-Side Policy
In the early 1990s a number of European countries, led by the UK and later by Germany, began to privatize and deregulate their telecommunications
industries. These had been the domain of the national postal systems, which had been slow to introduce new technologies and continued to charge high
prices for poor services which cost little at the margin to produce. In effect, the Europeans were doing little more than recognizing the successes of the USA
in the early 1980s which resulted from the break-up of the monopoly ATT (American Telephone and Telegraph) and the resulting deregulation of long-
distance telephone services. The Europeans went further than the Americans, however. By agreeing on a pan-European standard for wireless
technologies and wireless application protocols, they created a common market for an activity which involves massive external effects. They also
intensified direct competition with local telecom service provision—which remains a local monopoly in most countries.
The deregulation of telecoms in Europe has led to visible positive supply-side economic effects. The telecom industry has not only been a source of
value added and income growth for Europeans, but has also generated hundreds of thousands of new jobs. Most importantly, consumers have benefited.
In Germany, the overall price index fell by 27% for telecom services, by 65% for long-distance, and by 63% for mobile phone use between 1995 and 2007.
Naturally, much of these price declines was due to technological improvements and associated cost reductions. Yet it is hard to imagine price declines
like these offered voluntarily by a monopolist, be they public or private. While prices were falling all around Europe, the relative price declines in long-
distance service and portable telephone services are closely associated with deregulation in the UK in the late 1990s. More recently, the public auction of
third-generation mobile communication radio frequencies heralds the combination of these technologies with the internet, and promises the introduction
of new products and services that the old national postal services could never have offered.
Subsidies and industrial policy

For a variety of non-economic reasons, many countries operate elaborate systems of subsidies which shield certain firms and industries from the discipline of the
market. Table 18.2 displays the evidence for some OECD countries. Subsidized firms can sustain losses for a long time and avoid adjusting to changing economic
conditions. In doing so, they keep resources, capital, and labour employed, but do so inefficiently and prevent those resources from being used elsewhere. They
do not face the full cost of their operations (part of the costs are charged to taxpayers), or else the factors of production are paid more than their marginal
productivity.7 In fact, they may even keep factor prices artificially high and hurt other businesses that are not subsidized.
Table 18.2 Subsidies in Various Countries (% of GDP)
1975 1990 2009

Belgium 3.0 1.7 1.8
France 2.0 1.8 1.4
Germany 1.7 1.7 1.0
Italy 2.6 1.8 0.8
Spain 0.6 1.0 1.0
Sweden 2.4 3.6 1.6
USA 0.3 0.5 0.4
Public ownership of firms is another, more subtle form of subsidization. Unlike private firms, state-owned enterprises (SOEs) rarely face demanding shareholders
and are almost never shut down. When they lose money, they generally receive public resources. This may come either as an explicit subsidy from the
government to cover the loss, or as a loan to the company at interest rates unavailable to other firms. SOEs operate in virtually every major industry in Western
Europe.
Most countries regard certain economic activities as indispensable for strategic or political reasons. These include defence-related industries, such as steel,
energy, high technology, aircraft, and shipbuilding. As many of these activities exhibit increasing returns to scale, governments often try to guarantee that the
firms are large enough. This is often the underlying logic behind industrial policies. Industrial policies amount to official backing of national corporations or
whole industries. This takes the form of subsidies, public orders, and trade policies. Trade policies include tariffs on foreign goods, quotas on imports, export
credits financed at concessionary rates, and procurement policies, not to mention ‘buy domestic goods’ campaigns. These policies aim to encourage people to
choose domestically produced goods over cheaper foreign ones.
The ultimate effect of these policies is a distortion of prices from levels consistent with competition. Either they are too low, when the government subsidizes an
industry, or too high when the import of comparable foreign goods is impeded. In either case, consumers or taxpayers have to pay. Once again, the principle that
prices reflect efficient production costs is violated. Supply-side considerations have led to a reassessment of strategic requirements. For instance, the European
Single Act of 1992 bans most restrictions to intra-European trade. Yet industrial and trade policies survive, often conducted at the EU rather than the national
level, with such celebrated examples as Airbus, Volkswagen, and Galileo. More recently, private European banks have complained that state savings banks and
their parent entities enjoy hidden subsidies in the form of government guarantees. EU competition authorities have demanded that these be privatized or at least
set on an equal footing, singling out the German Sparkassen and the Landesbanken. These institutions respond that they provide banking services to remote,
rural areas, and provide financing for projects there that private banks generally shun. As the issue of state aids becomes increasingly political, it assumes the
aspects of national preferences discussed in Box 18.1.
18.4.1 Taxation as a Necessary Evil

Because they are non-rival and non-excludable to some degree, public goods are by their very nature difficult to finance. The consumption of non-rival goods by
one person does not affect, broadly speaking, the ability of others to do the same. Since they are non-excludable, beneficiaries of public goods cannot be charged
for their use. Think of a bridge across a river. A toll booth can be installed at the entrance, but what price should its owner charge? Non-rivalry means that the
marginal cost of usage is very low; from the perspective of social welfare the price ought to be low, too. But the costs of providing public goods can be high (a
bridge is very expensive to build!), so how can the producer be compensated? In addition, a bridge is a natural monopoly if it is the only one in the vicinity. If the
owner charges a high price and makes large profits, market competition will lead to a proliferation of bridges, all probably next to each other, which also seems like
an inefficient outcome. It is difficult for markets to cope with such failures because they require some sort of grand plan, which decentralized markets, by their very
nature, cannot (and should not!) have. Most often, public goods are provided collectively (bridges are free to use), or their provision is regulated (privately built
bridges are generally subject to strict regulations, including pricing and quality of service).
Public goods are pervasive: transportation and amenities, but also justice and police, passports, national defence and security, and diplomacy, etc. In each case,
there is a market solution, but it is generally inefficient, as not enough of the good in question—and possible none at all—would be privately provided. And in
each case, the insufficient provision of the public goods would greatly impair economic activity, leading to the breakdown of other, properly functioning markets.
This is why the provision of public goods is a fundamental supply-side policy. The more efficient the provision, the more productive the economy will be. At the
same time, efficiency means that public goods are produced at the lowest possible cost—which also involves issues of public procurement policy. Building
bridges, roads, and other forms of infrastructure involves large sums of money, and corruption or the bias towards local or national producers can represent a
significant efficiency loss.
18.4.2 The Effect of Taxation on Efficiency

Once a society has decided that government should provide certain public goods, public resources need to be raised in order to pay for them. This is done
through taxation of final goods and services, factors of production, and other activities. As a rule, taxation distorts markets by driving a wedge between the cost
of producing goods and services and the price paid by the consumers. This violates a fundamental principle of economic efficiency, which states that the marginal
benefit of the good or service involved should be set equal to the marginal cost of supplying it. When this principle is violated, there is a loss of welfare to both
consumers and producers, which can be quantified. This deadweight loss of taxation is described in more detail in Box 18.4.
In order to provide public goods, the government needs to raise taxes and accept the associated distortions. Still, it should do so in the least inefficient way.
The Ramsey principle of public finance8 states that this is best achieved, first, by spreading taxes as widely as possible over different goods and services
(and to limit the rate of taxation on any single good) and, second, by taxing most heavily those goods with the most inelastic (i.e. steepest) demands and supplies
(because the reduction of spending on those goods would be the lowest). While the Ramsey principle sounds like a good idea, implementing it is politically
contentious. Each producer stands to be hurt when a tax is applied to its production—because the producer surplus is reduced, as Box 18.3 shows—and therefore
lobbies hard to obtain the lowest taxes possible. Consumers, on the other hand, tend to be spread out and difficult to organize politically. As a result, tax reforms
bring a diverse coalition of private interests that often succeed as the silent majority that stands to benefit from them remains inactive. Finally, the principle
contains some distasteful implications—the highest rates should apply to necessities such as lodging, food, fuel, and clothing, while luxury goods with high
demand elasticities should face low taxes. All the same, the Ramsey principle is often applied in practice—most of all because the tax base declines the least when
the elasticities of demand and supply are the lowest.
Box 18.4 The Deadweight Loss from Taxation
Why are taxes a source of market distortion? Figure 18.4 illustrates the general case of an ad valorem tax, which is a percentage of the value of the activity.
Under perfect competition, the demand curve summarizes the marginal willingness of the market to pay for the good, while the supply curve describes the
marginal cost of producing it (both measured in euros). At point E, where the two curves intersect, the marginal consumer is willing to pay exactly what the
marginal producer requires to produce it. Perfect competition achieves the social optimum. Taxes alter this situation; the price paid by the buyer now differs
from the after-tax price received by the seller. The new supply curve S’ shows that the producer receives only a fraction of the market price. At the new
market equilibrium point D, the price is higher and the amount consumed and produced is lower. Both consumers and producers are worse off.
What are the losses to an economy from distortionary taxation? Figure 18.4 gives us a graphical answer. Because of the tax, consumers cannot enjoy
the lower price OG and pay higher price OA instead, purchasing only AD instead of GE. The area traced out by ADEG represents the consumer surplus, or
consumers’ willingness to pay above the market-clearing price, that is lost due to the tax. At the same time, the lower price (net of tax) to producing firms
means that firms will lose profits on goods they would have sold, produced at a cost lower than the no-tax price. The existence of these profits is due to the
fact that the supply curve is upward-sloping. This second area BCEG is the producer surplus lost due to the tax. This consumer and producer surplus are
not lost entirely. Despite the price rise, purchases of AD will still occur. The tax income from an ad valorem tax is given by the rectangle ABCD. This leaves
the two triangles, DEF and CFE, which represent lost consumer and producer surplus, or deadweight loss to society. These losses can be quantified.
Economic consultants frequently estimate deadweight losses as the cost of taxation when considering different policy options.
Fig. 18.4 The Effect of Taxation
Taxes drive a wedge between the price faced by the buyer and the price charged by the seller. Here a sales tax (e.g. VAT) shifts the supply curve upward. The new
equilibrium occurs at point D, with less output, a higher buying price (distance OA), and a lower selling price (OB) than at the tax-free equilibrium point E. Tax revenue for the
government is measured by area ABCD, the quantity sold times the tax rate (the difference between the two prices). Consumers, who could buy quantity GE at price OG,
suffer a ‘welfare loss’ measured by the area ADEG. Similarly, producers suffer a loss represented by the area GECB. Of these two losses—area ADECB—the government
receives ABCD. What is left, the triangles DEF and EFC, represent, respectively, lost consumer and producer surpluses—deadweight losses—arising from the tax.
18.4.3 Adverse Effects of Taxation on the Tax Base

By definition, non-distortionary taxes do not affect economic behaviour, because the tax is independent of economic decisions. An example would be lump-sum
taxes levied on individuals without any reference to incomes, wealth, or spending, or taxes levied unexpectedly on past incomes and wealth so that it is too late to
react. For this reason, non-distortionary taxes are appealing to governments. In practice, retroactive taxation is considered unfair precisely because it takes people
by surprise. Lump-sum taxes are also unpopular, as UK Prime Minister Thatcher’s fateful experience with the poll tax in 1990 showed. She tried to implement it and
was promptly voted out of office. As a result, nearly all taxes are distortionary.
Because distortionary taxes move the economy away from its first-best equilibrium, it is likely to reduce the tax base on which the tax is paid. In fact, it is
conceivable that raising tax rates actually results in lower tax revenues, if the tax base shrinks faster than the tax rate increases.9 This effect is sometimes called the
Laffer curve and is depicted in Figure 18.5.10 More precisely, the Laffer curve describes a theoretical relationship between total government tax revenues on the
vertical axis and the average tax rate (the ratio of tax receipts to GDP) on the horizontal axis. The tax rate ranges from 0 to 100%. At a 0% rate, tax revenue is nil
(point A). When the tax rate reaches 100%, no one is likely to work or produce at all so tax receipts are also nil (point B). At intermediate tax rates, tax receipts are
positive, as at point C. The hump-shape of the curve indicates that the tax rate distorts the economy so much that beyond a certain tax rate, taxable income falls
faster than the tax rate increases. The threshold point D corresponds to the average tax rate for which tax receipts are at a maximum. Any rate of taxation to the
right of point D is inefficient because the same tax income can be raised with a lower tax rate, i.e. less distortion.
Fig. 18.5 The Laffer Curve in Theory

When the tax rate is 0% there is no tax revenue at all (point A). As the average tax rate approaches 100%, tax income is also likely to be zero (point B). Intermediate tax rates
generate positive tax income, e.g. at point C. By continuity, the Laffer curve implies a hump-shaped relationship between tax income and average tax rates. Tax revenues are the
highest at point D.
The Laffer curve was not taken seriously for policy purposes when it was first proposed. This is because its most important detail is unknown: the location of
point D. In the early 1980s, Laffer claimed that the USA had passed this point. When the USA did cut tax rates, tax revenue actually declined. In recent years,
however, economists have begun to pay more attention to the Laffer curve. Figure 18.6 plots Laffer curves that result from inserting labour taxation in a model of
the long run which marries economic growth (Chapter 3) with labour markets (Chapter 4). In this model, taxes do affect equilibrium labour supply and output. For
an individual, the incentives to work and pay taxes at very high tax rates are very low. As a result, rational individuals will work fewer hours and take more
(untaxed) leisure.
Fig. 18.6 Laffer Curves in Reality?
Raising labour taxation beyond a critical rate reduces the return to work and total labour tax revenues. Economists cannot vary taxes in a country as a controlled experiment to
discover this critical rate—but they can isolate it in models which have been ‘calibrated’ to match features of real economies. The figures plot total long-run tax revenues
(measured as an index of average tax revenues in the period 1995–2010) as a function of the labour tax rate in one such model. While the critical tax rate is estimated to be high
—around 55–65% in the countries shown—it is not very far above tax rates displayed in Table 18.5. The model which generated these curves does not include more elaborate
forms of tax evasion, such as tax shifting, capital flight, and the underground economy.
Sources: Trabandt and Uhlig (2012).
In reality, well-paid individuals often expend significant resources to avoid or evade taxes in countries where they are high. This has led to global efforts to
stamp out tax tourism. Several international agreements, adopted in the 2010s, have focused on the banking industry and aim at extensive—possibly automatic—
exchange of information amongst tax authorities. At the same time, high income people are often successful entrepreneurs whose efforts create innovations that
generate jobs and raise productivity, hence GDP potential. These aspects raise disturbing questions. Should tax rates for the most productive individuals be the
lowest? While this may be efficient, it would seem to be distasteful for most—it seems inconsistent with a widely held ethical principle that resources should be
redistributed to limit inequality. Yet the logic of efficiency must be addressed, and will continue to pose a major puzzle for social policy.
Section 18.2 established that markets can fail and require government action. This is doubly true for labour markets. Besides the obvious human dimension—that
individuals and their well-being are associated with labour supply, wages, and unemployment—it turns out that a number of aspects of labour markets make them
fundamentally different from markets for other goods and services, even under ideal conditions. In this section, we will explore a number of these differences. To
the extent that policy-makers recognize and want to correct failures of labour markets or improve badly-conceived regulation, they can design policies which can
increase the employment rate—the fraction of working age individuals in work—or reduce the equilibrium rate of unemployment—the fraction of the labour force
in work in the long run. Both measures lead to a higher level of equilibrium output.
18.5.1 Heterogeneity and Imperfect Information

One of the most striking facts about labour markets is the degree of turnover, or the rate of flow between the states of the labour force shown in Figure 4.14.
Chapter 4 noted that annual inflow into and outflows out of unemployment are frequently larger than the stocks of unemployment at any given point in time. The
high rate of in- and outflow points to two important aspects of labour markets: (1) heterogeneity of labour, and (2) incomplete information about its quality.
First, neither workers and the labour hours they supply nor their workplaces are identical. We say that workers and jobs are heterogeneous. A trivial example of
heterogeneity lies behind the turnover noted in Table 4.7, which is concentrated among young, female, and unskilled workers. High turnover is frequently a
transient phenomenon in the life cycle. It is more characteristic of newcomers to the labour market, and reflects a complicated process of ‘picking and choosing’.
Even in high labour turnover countries like the UK or USA, it turns out that most workers stay at a given job for a very long time.
Second, because workers and jobs can be so heterogeneous, information is likely to be incomplete. Even with the help of modern technology, such as the
internet, it is still impossible to know about all jobs on offer at any point in time. It may be necessary to search harder to find the most attractive jobs. Some
employers do not actively advertise job openings and may open a position only after meeting an acceptable candidate. Workers must decide either to accept a job
or reject it, searching further without knowing what may lie ahead. Most likely, they will have had some employment experience in the past, which guides their
expectations concerning pay and work conditions, yet offers they can expect in the future may not always correspond to patterns in the past. For example, it is a
fact that computer operators—workers who once programmed and serviced large mainframe computers—were highly trained and well-paid professionals in the
1970s and 1980s. As personal computer and local networking technologies became dominant, the demand for computer operators shrank dramatically. Those who
lost their jobs may have had difficulties finding new ones at a similar pay level to that in the past. Similarly, a job and the employer who is deciding how to fill it
have particular needs and expectations which may not be met by every applicant. In times of tight labour markets it may not be very easy for employers to find
their dream candidate.
It is for this reason that unemployment must be interpreted carefully. To the extent that the unemployed are really ‘unemployed resources’, they represent a lost
opportunity and a reason that trend GDP is lower than it could be—a fundamental inefficiency. Indeed, Chapter 4 gave a number of reasons why workers ready to
work at current wages are unable to do so. Yet unemployment also contains an element of efficiency. Workers who lose their jobs due to lay-off or plant closure
frequently possess industry- or firm-specific human capital which would be wasted or under-utilized if they were to accept the first new job that came along. A
crude indicator of the ‘efficiency’ of job matching in a given labour market is the extent to which job openings—called vacancies—and job-seeking,
unemployed workers coexist at the same time in a labour market. Because job requirements and worker qualifications are often specific, one can speak of a job-
matching process. Forcing jobs and workers together indiscriminately would certainly not be efficient.
The extent of mismatch in the labour market is summarized by the Beveridge curves of Figure 18.7. Named to honour the work of British economist William
Beveridge in the 1940s, Beveridge curves display unemployment and vacancy rates in an economy over time. Movements in the north-west and south-east
directions are associated with the business cycle: firms offer fewer jobs and unemployment is higher in recessions, while the vacancy rate rises and the
unemployment rate declines in economic expansions. The position of the Beveridge curve, on the other hand, indicates the longer-run efficiency of job-matching
in labour markets. The coincidence of a large number of vacancies with high unemployment—a Beveridge curve far away from the origin—indicates either that
workers are badly informed, unable to take up the offers for lack of mobility or adequate skills, or unwilling to change and adapt.
In the mid-1970s, the Beveridge curve seemed to shift away from the origin, suggesting increasingly inefficient or mismatched labour markets—in fact, a rise in
the equilibrium rate of unemployment. In the USA, this shift was of temporary duration. It has been much longer-lasting in the UK, while in Germany, the return to
the earlier position has been even slower. In contrast, the Great Recession has been associated with a worsening of the Beveridge curve in the USA shifting
outwards, while the trade-off in the UK and especially Germany has shifted inwards.
Shifting the Beveridge curve towards the origin—put differently, reducing the unemployment rate at any level of vacancies—can be good supply-side policy.
For policy-makers, it means implementing measures which improve the job-matching process. One is to improve information on job openings by increasing the
number and effectiveness of job agencies. Advances in networking technology have made it possible for the unemployed seeking work to know about job offers
all over a country—as long as they are posted. In the years 2003–2004, Germany embarked on a comprehensive reform of its national employment agency, making
data on jobs and training positions available nationwide. Second, the adaptability of workers’ skills may be enhanced via job retraining programmes. This means
giving workers whose skills are unwanted or obsolete new ones which are in demand. Another possibility is to increase workers’ and firms’ geographical mobility,
e.g. by making the housing market more efficient or providing subsidies to firms in search of new locations. In recent years, countries have begun to make more
use of active labour market policies. This is nothing but the creative and flexible use of all of the measures already mentioned and more—as opposed to the
passive policy of simply paying out unemployment benefits. Especially in the past 20 years, the Nordic countries have combined ‘European’ solidarity with the
unemployed in the form of generous unemployment benefits with activation policies designed to keep the unemployed in touch with the labour market. Despite a
number of adverse shocks in the past decade, the Nordic countries continue to boast the world’s lowest long-term unemployment rates. Table 18.3 shows the
variety of approaches currently employed in the OECD.
Fig. 18.7 Beveridge Curves in the UK, Germany, and the USA, 1960–2010
The Beveridge curve is the inverse relationship between vacancies and unemployment observed in all economies. In recessions, unfilled vacancies tend to decline while
unemployment increases. The more efficient the job-matching process—the way unemployed workers and unfilled job vacancies find each other in the labour market—the closer
to the origin the Beveridge curve is. Over the past 30 years, the Beveridge curve seems to have shifted outward in several European economies, and much less so, if at all, in the
USA. In both Germany and the UK, the curve is now shifting back, the result of numerous labour market reforms. The recent Great Recession in the USA seems to be associated
with an outward shift of the Beveridge curve, although that interpretation is disputed.
Sources: OECD, Economic Outlook; Eurostat.
18.5.2 Imperfect Contracts and Labour Market Regulations

Because of their social and political aspects, labour markets are often heavily regulated. Regulations cover a wide range of aspects: paid holidays, the length of
working days and weeks, safety standards, works councils, union representation, and other facets of the employment relationship. These regulations often have
efficiency costs and sometimes seem only marginally motivated by good economic arguments. The columns of Table 18.4 report the relative severity of the most
common types of labour market regulation in OECD Europe.
Table 18.3 Expenditures on Labour Market Programmes, 2013 (% of GDP)
Denmark Finland Norway Sweden Francea Germany UKa

Active Measures:
Training 0.53 0.15 0.11 0.13 0.33 0.24 0.01
Employment incentives 0.40 0.49 0.09 0.64 0.03 0.02 0.01
Supported employment and rehabilitation 0.57 0.16 0.17 0.28 0.09 0.03 0.00
Direct job creation 0.00 0.10 0.00 0.00 0.13 0.02 0.01
Start-up incentives 0.00 0.10 0.00 0.01 0.04 0.01 0.00
Total active measures 1.50 1.00 0.37 1.06 0.62 0.32 0.03
Passive Measures:
Out-of-work income maintenance and support 1.41 1.61 0.33 0.68 1.41 0.96 0.32
Early retirement 0.25 0.01 0.00 0.00 0.01 0.05 0.00
Total passive measures 1.66 1.62 0.33 0.68 1.42 1.01 0.32
Administration 0.32 0.15 0.13 0.29 0.25 0.35 0.20
Total expenditures 6.64 5.39 1.53 3.77 4.33 3.01 0.90
Unemployment rates (% of working age population)
Overall unemployment rate 6.9 8.0 3.3 7.2 9.5 5.2 7.5
Long-term unemployment rateb 1.7 1.7 0.3 1.2 3.8 2.3 2.7
aData for 2012.
bPercentage of labour force which is unemployed with current duration of at least twelve months.
Source:OECD, Employment Outlook.
Imperfect mobility and regulation

One situation of possible exploitation relates to worker mobility. Mobility of demanders and suppliers is a central mechanism for correcting imbalances in a
perfectly competitive market. If they feel they are getting a bad deal, they simply move elsewhere. In labour markets, things may not be so simple. The suppliers of
labour are human beings. Mobility means changing jobs, an industry, occupation, or residential location, and people by their very nature tend to be immobile.
Furthermore, they may even value their ‘immobility’, preferring to take a pay cut or even risk unemployment to stay put rather than to move to a faraway place or
accept a radically new occupation. This is one reason that, even in the UK with its highly dynamic labour markets, differences in unemployment rates between
North and South, between Scotland and the Midlands, and between London and the entire country have persisted for decades.
An unscrupulous firm might try to exploit the immobility of its workers by cutting wages after the employment relationship has begun. In perfect market
situations, the worker would simply leave the firm, and look for work elsewhere. In doing so, mobility costs must be incurred. In advanced economies these costs
may be rather significant, especially for those low-wage workers. Minimum wages can be seen as a response to such local market power by employers. Another
response is to extend union contracts to uncovered workers and firms in the sector. Similar arguments are often invoked to justify work-week and work-time
regulations, job protection, and rules concerning job safety. Naturally, minimum wages have a number of disadvantages as well, mostly related to the closing
down of labour markets for young workers and those with little skill or training; these were discussed in Chapter 4.
Table 18.4 Measures of Employment Protection, 2013
Protection of permanent workers against Regulation of temporary forms of Specific requirements for Aggregate
individual dismissal employment collective dismissal index
Eurozone
Austria 2.1 2.2 3.3 2.4
Belgium 2.1 2.4 5.1 3.0
Germany 2.5 1.8 3.6 2.8
Greece 2.1 2.9 3.3 2.4
Ireland 1.5 1.2 3.5 2.1
Italy 2.5 2.7 3.8 2.2
Netherlands 2.8 1.2 3.2 2.9
Portugal 3.0 2.3 1.9 2.7
Spain 1.9 3.2 3.4 2.4
Average 2.3 2.2 3.4 2.6
Eurozone
Other EU
Denmark 2.1 1.8 2.9 2.3
Sweden 2.5 1.2 2.5 2.5
United 1.2 0.5 2.6 1.6
Kingdom
Average 1.9 1.2 2.7 2.1
Other EU
Other OECD
Norway 2.2 3.4 2.5 2.3
Switzerland 1.5 1.4 3.6 2.1
United 0.5 0.3 2.9 1.2
States
Average 1.4 1.7 3.0 1.9
Other EU
Source: OECD (www.oecd.org, OECD Indicators on Employment Protection).
Regulation of dismissal
A related argument has been invoked for regulations of dismissal ‘without cause’ or for ‘economic reasons’, i.e. not due to malfeasance or misbehaviour on the
part of the worker. Firing workers for reasons related to the business cycle are associated with uncertainty and economic dislocation. In many European countries,
prior notice of termination of employment must be given to workers—during which time the employee remains on the payroll and represents a cost to the firm.
‘Social plans’ are often required for large-scale redundancies, in which the exact list of employees is decided using criteria like age, family status, and re-
employability. Severance payments are often legally mandated for workers dismissed for economic reasons. While common in EU countries, job protection is by
no means the norm in the OECD. The UK, the USA, and Denmark are examples of countries where ‘employment at will’ contracts prevail.
While severance regulations make it difficult for firms to reduce employment in the short run, they also increase the effective cost of labour to firms. They make
firms more reluctant to hire in good times, precisely because they worry about consequences in bad times. Firms will tend to use the existing workforce through
overtime or conversion of part-time into full-time before hiring new workers. Table 18.4 reports a ranking of the degree of ‘job protection’ provisions offered in a
number of industrial countries. It is noteworthy that countries in which youth unemployment rates are high tend to have the most restrictive dismissal laws.
Surveys consistently reveal that workers in those countries where dismissal laws are the most liberal—the USA, the UK, and Denmark—feel the least insecure
about their job. This perception, which is hardly intuitive, must have to do with the perception workers have about the ease of finding a new job, should their firm
go out of business or move abroad. Because firms are less concerned about shedding labour in bad times, they are less reluctant to hire workers in good times.
As the economy becomes increasingly subject to global influences and technological advances, the nature of employment relationships has changed
fundamentally. Employment is less likely to involve a lifetime relationship, and is increasingly likely at some point to be ‘restructured’ or reoriented towards
completely new areas of activity. Severance rules of the type described are likely to inhibit the growth of new firms as well as the expansion of existing ones. As
the pressures for more flexible employment grow, firms have become increasingly creative in finding ways to undo the intended effects of the legislation. For
example, they may chain together several short-term contracts for the same employee, or hire workers from temporary help agencies, shifting the burden to others.
In the end, it may be worth considering market solutions in which workers accept to work under ‘employment at will’ for a wage premium. This would allow firms to
pay for additional flexibility while preserving employment protection.
18.5.3 Social Policies Incentives and Taxation

The social safety net
The social safety net refers to the system of transfers and benefits designed to help the disadvantaged and vulnerable in society. These include unemployment
benefits, social welfare, old-age pensions, early retirement, health insurance, and disability payments. A large gap divides European countries, which transfer 20–
30% of their national income to individuals or firms, from the USA, Japan, and Switzerland, which transfer only 10–15%. This might lead a casual observer to
conclude that high European unemployment is a product of the ‘social welfare state’, which puts too much weight on solidarity at the cost of productivity and
economic efficiency. Yet it is overly hasty to claim that Europeans have erred too far in the direction of social protection, in comparison to the rest of the OECD.
To some extent, the high level of transfers observed in Europe are a response to high unemployment, which may have other underlying causes. At the same time,
these transfers—in the form of unemployment benefits, welfare, and premature retirement and disability pensions—take the pressure off workers and firms to
adjust to a changing world economy. The greatest danger is that the safety net becomes a trap, leading to long-term unemployment.
It is useful to utilize the tools developed in Chapter 4 to help think about the adverse effects of the safety net on incentives. The social systems of most
countries share two institutional features. First, poor or unemployed people receive transfers—income maintenance programmes or unemployment benefits—from
the state. Second, income taxes are progressive: the rate of taxation increases as income rises. Taking up a job not only means receiving a salary, but also paying
taxes if the salary is high enough, and losing eligibility for income maintenance programmes. It is conceivable, then, that people can be financially worse off by
taking a job, not to mention incurring a loss of leisure, and possibly some activity in the underground economy. Implicitly, these people face an effective marginal
tax rate—considering the overall effect of work on their income—in excess of 100%. Box 18.5 shows how safety net programmes may lead to a welfare trap,
inducing people to remain unemployed or stay out of the labour force, thereby reducing the productive potential of the economy. Recent experience of ‘work-to-
welfare’ in the UK and the USA indicates that the incentive aspect is important for bringing workers on social assistance back to work.
Box 18.5 Crusoe Caught in the Safety Net
In Figure 18.8, Crusoe is faced with a budget constraint which is no longer strictly linear. When not working at all in the market he receives an
unconditional transfer of T from the government—unemployment benefits, welfare, or other payments associated with the social safety net. In this particular
social security system, Crusoe must give up his benefits, coconut for coconut, by exactly the amount of his additional income—an effective marginal tax rate
of 100%. This is representative of the current situation in many European countries. Since Crusoe loses a euro of benefit for every euro he earns in work,
the budget line is flat up to the point at which Crusoe receives T, whether he works or not. Crusoe values leisure, so it is hard to see why he would accept
part-time work in this regime. Unless the after-tax wage is high enough, Crusoe is unlikely to work.
Figure 18.8 also shows the impact of reducing the effective marginal tax rate from 100% to some lower rate which is nevertheless higher than that for
someone already in work and not receiving welfare payments, holding T constant. As the tax rate declines, the budget line for the household becomes
steeper. At some point Crusoe can make himself even better off than he was at A, by choosing point B. In doing so his income is added to the tax base and
contributes to higher GDP. The effect can be intensified if the carrot (the lower effective marginal tax rate) is combined with a stick (lowering T ), although
this type of reform is politically difficult to implement.
In Germany, the Hartz labour market reforms (implemented after 2004) adopted several of the principles discussed in this chapter. One measure
reduced the effective marginal tax rate on work by exempting some earnings from counting against social benefit. Effectively, this becomes a top-up
provision. Another measure implemented benefit cuts (‘sanctions’) for those who consistently turn down job offers. Job centres were reorganized and on
the basis of a modernized database were better able to ‘activate’ the unemployed with credible job offers. This is one key reason why the unemployment
rate was cut by half over the following decade.
Fig. 18.8 Incentives and the Social Safety Net

In a social welfare system with benefit T for those out of work but with 100% marginal tax rate, the individual depicted in this figure chooses not to work (point A), because
starting at zero, small increases in labour supply yield no gain in net income. The budget constraint is If the individual is allowed to keep some additional income
(changing the budget constraint to they can improve their well-being by working (point B). The most important problem with lowering the effective marginal tax rate to
unemployment benefit and welfare recipients is the cost to the state, which can be considerable.
Labour taxation
Because labour is so essential to economic activity, it is natural to expect governments to tax it. Perhaps because the Ramsey principle of public finance (Section
18.4.2) is so compelling, labour is one of the most highly taxed ‘commodities’. As Box 18.6 explains, labour is subject not only to income taxes paid by households,
but also to a number of social security contributions by both employees and employers. It is also natural to expect that labour taxation might influence the demand
for labour. Higher taxes raise the real cost of labour faced by firms, leading to lower employment in the sector that pays the tax. 11 Table 18.5 shows the rate of
taxation on labour in various countries in its various guises. Despite high unemployment, efforts have been rather slow in Europe to reduce the tax burden placed
on the labour market. France, for example, has exempted or reduced social security contributions by low-wage workers.
Box 18.6 Taxes and the Labour Market in Europe
Labour taxes can be grouped into three classes: (1) income taxes, (2) social insurance charges paid by employers, and (3) social security contributions
paid by the employees. All three can be added up, as they either reduce the workers’ net receipts and thus affect the supply of labour, or increase the cost
of labour and reduce demand (governments prefer to call the last two ‘contributions’—dedicated to government funds for unemployment benefits, national
health insurance, retirement and pension benefits, disability insurance, solidarity with various causes, including low-cost housing and special retraining
programmes, etc.). Let τF be the employer’s wage tax rate, let W be the wage of the employee before income taxes and employee social security
contributions, let τW be the tax rate of contribution of the employee, and let τP be the personal income tax rate. The worker’s take-home pay after taxes is
Wtake-home = (1 − τW)(1 − τP)W, while the cost of labour to the firm is given by Wlabour costs = (1+τF)W. As a result, the effective labour cost is equal to the
take-home pay multiplied by the factor which is often called the labour tax wedge. Holding Wtake-home constant, increases in the tax
wedge raise labour costs and reduce the demand for labour. In addition to all this, personal income tax paid by workers on their salary may be progressive,
meaning that it increases at the margin as taxable income itself rises.
Whether high labour taxation leads to high unemployment or not depends on how the supply of labour or the collective bargaining system reacts.
Equilibrium unemployment rate remains unchanged only if wages after taxes fall by exactly the amount of tax—i.e. if workers shoulder the entire burden.
But the collective labour supply curve will not be vertical in general. Households may find it attractive at high taxes to work less, work in the underground
economy, or take overtime pay in the form of a holiday (leisure), as is often done in Germany and Scandinavia. It is thus likely that such high taxes make
hiring labour unattractive, and that a reduction of such taxes could increase employment. The problem for governments is how to replace the lost revenue.
Table 18.5 Labour Taxation in 2015 (in %)
Country Total tax wedge of which: Top marginal personal income tax rate
Personal Social security contributions
income tax
Employee Employer
Euro area
Austria 49.5 13.1 14.0 22.4 50.0
Belgium 55.3 21.6 10.8 22.9 45.3
Finland 43.9 18.4 6.7 18.7 49.1
France 48.5 10.7 10.3 27.5 47.5
Germany 49.4 16.1 17.2 16.2 50.0
Greece 39.3 7.1 12.4 19.7 47.0
Ireland 27.5 14.2 3.6 9.7 47.0
Italy 49.0 17.5 7.2 24.3 48.8
Luxembourg 38.3 16.0 11.4 10.9 43.6
Netherlands 36.2 15.2 12.1 8.9 49.2
Portugal 42.1 14.0 8.9 19.2 50.3
Spain 39.6 11.6 4.9 23.0 46.0
Other EU
Denmark 36.4 35.8 0.0 0.8 55.8
Sweden 42.7 13.5 5.3 23.9 57.0
UK 30.8 12.8 8.4 9.7 45.0
Switzerland 22.2 10.5 5.9 5.9 36.1
Czech Rep. 42.8 9.2 8.2 25.4 20.1
Hungary 49.0 12.5 14.4 22.2 16.0
Poland 34.7 5.0 15.3 14.4 20.9
Other OECD
Australia 28.4 22.7 0.0 5.6 49.0
Canada 31.6 14.1 6.8 10.8 49.5
Japan 32.2 6.7 12.4 13.1 55.7
New Zealand 17.6 17.6 0.0 0.0 33.0
Norway 36.6 17.9 7.3 11.5 39.0
United States 31.7 16.5 7.0 8.1 46.3
Source: OECD Centre for Tax Policy and Administration. Columns may not add due to rounding.
Yet, the net effect of high taxes on employment depends on the elasticity of labour demand and supply. If collective labour supply is relatively inelastic—
perhaps reflecting the inelastic labour supply of households—the burden of the tax will fall primarily on wages, and employment will be relatively unaffected. In
Europe, however, wages are set in collective bargaining, leading, in all likelihood, to a flatter collective labour supply curve. Under these conditions, high labour
taxes reduce employment significantly, and are likely to be associated with a loss of consumer and producer surplus.12 In the very long run, the elasticity of
labour demand is likely to be high, since firms may simply move to other locations where labour costs are lower. Not surprisingly, the integration of European
economies is driving a de facto harmonization of taxation across national boundaries, with or without explicit government coordination. This pattern is readily
recognized in Table 18.5.
18.6.1 Waiting for Godot? The Long and Variable Lags of Supply-Side Policy
Supply-side policy is like an investment. It involves sacrifices today, with gains only at some future date, often at a future date which is highly uncertain and
dependent on decisions taken by future policy-makers. For this reason, supply-side policies are a hard sell—politically speaking. As with demand-management
policies, supply-side reforms require time to show their beneficial effects—usually much more time than it takes for a tax cut to stimulate demand. Even the
most.pngted politicians will find this difficult to endorse, since the likelihood that they will be around to enjoy the credit for implementing those policies is small
indeed. While Wim Kok, Ronald Reagan, and Margaret Thatcher were able to reap some of the benefits of their supply-side policies, the same cannot be said of
Jimmy Carter, Leszek Balcerowicz, or Gerhard Schröder, all of whom implemented bold reforms and were either punished or ignored for them.13
Consider the types of policies discussed in the last three sections. The failure of labour markets to clear due to excessive real wages cannot be corrected
overnight, but rather only in small concerted steps over many, many years, especially if a currency depreciation is impossible (as it is within the euro area),
whereas the overvaluation may have taken a decade or two to evolve. Similarly, deregulation and privatization take a long time to bear fruit. Deregulation must be
well-planned to prevent exploitation of temporary market dominance, and must also allow entry of new firms, including foreign ones. It takes several years before a
wave of new firms arrives, increasing competition and forcing relative price reductions and quality improvements. Privatization is the most difficult of supply-side
policies to implement, since simply transferring the assets from public to private hands is politically unacceptable and will always yield sour grapes in the future,
either on the part of the government vis-à-vis the winners who made fortunes from the assets, or the losers who ‘paid too much’ for them.
If this weren’t enough to discourage policy-makers, the time dimension is compounded by the overall credibility of announced policies. The sacrifices
demanded of households at the outset of reform are often significant. The policy change itself may not be irrevocable, so economic agents may not expect reforms
to be long-lasting or permanent. Most obviously, a new government could be elected at a future date which has different preferences and priorities and simply
reverses the policy. It is even possible that the same government may change its mind after the private sector has acted on its announcement! This so-called time
consistency problem was first identified by Nobel Prize laureates Finn Kydland and Edward Prescott, and is concerned with the incentives of those who
announce a policy to actually follow through with that policy after others have reacted to the announcement. Time inconsistency is a very general phenomenon,
which applies to many fields besides economics. A son will promise to drive carefully if lent his parents’ car; a person in prison always pledges not to engage in
unlawful activity if released early; politicians promise the moon if elected. All of these promises are time-inconsistent. The risk is that only time-consistent
solutions, which are less desirable for all involved, are adopted: no son is lent a car; no inmate is released early; and politicians are never trusted. Under such
conditions it is hardly surprising that the private sector needs convincing of the permanence of supply-side reforms.
18.6.2 The Political Economy of Supply-Side Reforms

Supply-side reforms are difficult to enact because there are always losers. These losers are often few in number and have lots to lose. The winners are more
diffuse, numerous, and thus harder to organize. The political difficulty of implementing supply-side policies is best illustrated by the example of labour market
reforms. While unemployment is generally regarded as a curse of modern market economies, joblessness in Europe has been high for almost three decades. In
addition, it was seen in Figure 1.1 that, in contrast to the USA, European unemployment has risen steadily over successive business cycles. Unemployment
represents under-utilized labour resources. If this under-utilization is not the conscious choice of households, it is involuntary and represents a supply-side
problem. As the leading quote at the beginning of this chapter indicated, an important challenge for the nations of Europe is to address this issue, perhaps using
the ideas and concepts developed here.
Suppose there is agreement—among policy-makers, at least—that labour markets should be reformed. How should a country go about it? Several problems
arise with simply ‘deregulating’. Consider the example of abolishing job protection. First, many companies that had wanted to reduce employment under the old
regime will certainly take advantage of deregulation, and lay-offs are likely to increase in the short run. In most European countries, the political consequences of
such a move are severe enough to make reform unthinkable. Second, the issue of time consistency already discussed in Section 18.6.1 arises. Suppose a firm that
would like to hire more employees expands its employment in response to deregulation. At this point, a government that is fishing for political support might re-
impose the dismissal regulations, ‘trapping’ the firms at higher employment levels. Firms will anticipate this, with the sad result that no new workers will be hired,
despite deregulation! Governments must pre-commit to a ‘no-reregulate’ regime before employers are convinced.
It is also true that an overwhelming majority (90–95%) of workers are employed, so that their domination over the unemployed might even be regarded as
democratic, and is generally mitigated by unemployment benefits. Governments often cannot reform labour market institutions because their interventions are
often regarded as interference in the collective bargaining process. One noteworthy approach followed by Spain was the expansion of the use of temporary
contracts for workers, with some restrictions.14 Such contracts allow firms to hire workers without restrictions on a limited-time basis. By some accounts, as much
as four-fifths of all new hires in Spain since the late 1980s has been under such temporary contracts. This response suggests that firms are willing to hire more
workers when job security obligations are absent. Second, young people were the primary beneficiary of such reforms, which has helped relieve an acute youth
unemployment problem. Finally, increasing the relative number of workers who have jobs with less protection increased the consensus for broader reforms which
followed. By increasing the power of the outsiders, labour market reform became more acceptable.
The political economy of labour markets may be the most important element determining the success or failure of reforms. Efforts to increase the attractiveness
of part-time work are often blocked by unions, which see their bargaining power diluted by new employees who are unlikely to be members. Work-sharing
proposals which increase the cost of labour are resisted fiercely by employer associations. Successful reform efforts must involve management and labour putting
many different elements on the negotiating table, trading reduced working time and relief on labour charges for flexibility, wage moderation, and reform of the
unemployment benefits system.
18.6.3 Feasible Supply-Side Policy: Peanuts or Big Payoffs?

Let us assume that the supply-side reforms are credible, actually believed by economic agents, and acted upon as intended. Five to 10 years after the reform was
enacted the obvious question arises: was it at all worth the effort? Another difficult aspect of supply-side policies is that the gains are uncertain. Because the sum
of millions of individuals’ behaviour will ultimately determine the outcome, it is difficult to know how trend output will react.
It is natural to suspect that the gains from supply-side policies are often of a secondary order. After all, we have talked a lot about deregulating individual
sectors of the economy, reforming the tax system, or altering the equilibrium of labour relations. Rather than presenting a complex model used by economists to
answer these questions, we present two examples of clearly identified and successful policies in Europe in the past 25 years: the Netherlands and Ireland. As our
examples will show, the gains can be substantial.
The Netherlands
Plagued by the ‘good fortune’ of natural gas discoveries in the 1970s, the Netherlands was the mother of the Dutch disease: increasingly valuable natural
resources led to overvaluation of the currency, the guilder, which put much manufacturing industry out of business. Most important was the unemployment rate,
which almost reached 12% in 1983. A generous social security system was proving unsustainable at those rates of unemployment. A stagnant employment rate in
the light of increasing demand by women for access to market work put politicians under pressure to find solutions. Female labour force participation rate in the
Netherlands was among the lowest in the EU.
The solution to the Netherlands’ problems began with the Wassenaar Accord of 1982, a long-term agreement between labour, management, and the government
which moderated real wage growth in the public and private sectors in exchange for better treatment of part-time employment and government tax reductions.
While not officially part of the agreement, the government followed up with tax breaks for part-time jobs as well as cuts in public sector employment and wages.
Most importantly, tax rates on labour were reduced, blunting the net impact of real wage moderation on household incomes. Real labour costs in the Netherlands
grew much more slowly than in neighbouring EU economies in the 1990s, more closely resembling the USA than neighbouring Germany! In the years that
followed, other reforms were enacted: the system of unemployment benefits was reformed and the employment agencies also reformed to provide more carrot-and-
stick incentives for the unemployed to return to work. The benefit was strong growth in employment, especially part-time employment of women, and a drop in
unemployment.
The Wassenaar Accord was supply-side policy primarily targeted at increasing potential employment and reducing the unemployment rate. But as Figure 18.9
shows, this policy also had beneficial effects for GDP per capita in this country in the longer run, as the production function would predict from an increase in
labour input per capita.
Ireland
With its anaemic growth rates, high budget deficits, and unemployment reaching a staggering 18% in the late 1980s, Ireland was called the ‘sick man of Europe’ by
the late German-American economist Rudi Dornbusch.15
Emigration continued to bleed the economy of its most talented young people to the USA, the UK, and continental Europe. Starting in late 1987, a programme of
fiscal consolidation and wage restraint was implemented with the agreement of the collective bargaining parties. More important, the social partners (labour
unions and employers) agreed to a series of centrally bargained ‘Social Partnership’ agreements stipulating modest increases in money wages. The climate
between labour and management improved. In a development similar to that in the Netherlands, the government responded by reducing taxes in the early phase of
the reform programme.
Fig. 18.9 Two Successful Supply-Side Reforms: the Netherlands and Ireland
The three panels show the economic performance of the Netherlands and Ireland relative to France along three dimensions: GDP per capita, unemployment rate, and employment-
to-population ratio. The crucial years are 1983 for the Netherlands and 1987 for Ireland. As the charts show, benefits come, but with considerable delay, and often due to a
cumulation of additional actions taken after the initial reforms were accepted. These benefits are more concentrated on the employment front in the Netherlands, as opposed to
economic growth in Ireland.
Sources: AMECO on line, European Commission.
In the early 1990s, Ireland cut corporate taxes and introduced incentives for foreign investors to locate in the country. It increased spending on education and
infrastructure. Active labour market policies were implemented to help people into gainful employment again. The economy’s product markets were deregulated
and bureaucracy streamlined to increase attractiveness to foreign businesses considering investing in Ireland.
Figure 18.9 shows that it is hardly an exaggeration to speak of an Irish economic miracle. In the period from 1992–2007, the Irish economy grew at an almost
Asian real rate of 6.3% per annum. After 15 years of extraordinary growth, Ireland managed to raise its GDP per capita above that of the UK and France—with the
reminder, of course that GDP is not GNI, and that some of Ireland’s value added represented income to many UK, US, French, and German capital owners who had
built factories and office buildings and contributed to its economic success. In addition—the most convincing sign of all—the population of Ireland stopped
shrinking for the first time in decades, rising from 3.55 to 4.34 million, or 23%, during the same period. Hundreds of thousands of expatriate Irish in the USA, the
UK, and elsewhere returned home to make this miracle happen.
Since 2007 Ireland has been the focus of the world’s attention for another, less favourable reason. It was an epicentre of the financial crisis, suffering a collapse
and near-complete nationalization of its banking system. After 2001, the financial sector began to swell out of proportion, not only due to the enormous growth in
the property market but also because of dubious investment projects. The collapse of the Irish property market and world demand for Irish exports took their toll
on GDP, which plummeted by 10.1% in the years 2007–2010. Yet the bust which followed the boom will hardly reverse one of the great economic success stories of
the last century: over the years 1989–2007, the Irish economy grew by more than 190%! The lion’s share of this increase represents sustainable potential output
and permanent increases in Irish living standards.
Box 18.7 Supply-Side Reforms: A Solution to Greece’s Problems?
Europe’s sovereign debt crisis began in Greece, so it is natural to focus attention on this country’s efforts to solve it. At the end of 2011, the Greek public
debt–GDP ratio exceeded 165%. Chapter 17 tells us that any sustainable solution to Greece’s debt crisis must involve a combination of large and painful
primary government budget surpluses for some time, lower borrowing interest rates (r), higher economic growth (g), debt relief, or default. The ‘Troika’—
the IMF, the ECB, and the European Commission—has stressed all of these elements, but also called for supply-side measures to raise the long-run
growth path. Given that Greek GDP per capita is less than two-thirds of either France or Germany, the potential for further growth is significant.
The Troika has produced a lengthy list of supply-side problems in the Greek economy. The list includes low tax collection, overly generous pensions,
uncompetitive markets due to collusion among producers, a large number of protected professions (from pharmacies to notaries and taxi drivers),
unproductive state-owned firms, an antiquated health care system, an inefficient and sometimes corrupt public administration, and more. The Greek
authorities were requested to conduct wide-ranging reforms as a condition for debt relief.
Since these conditions have been put forward, successive governments have made some progress, but not enough in the eyes of the Troika. The
government blamed fierce opposition from a myriad of interest groups and the lack of cooperation from the administration. It has faced violent street
protests and a succession of general strikes. According to the World Bank’s annual ‘Doing Business around the World’ survey Greece ranked 100th in
2011 but by 2015 had risen to 60th (see Table 18.1). This represents real progress, but Greece still ranks behind Costa Rica, Turkey, and Mongolia and
most of Europe. This illustrates the dilemma of supply-side policy: how to conduct reforms and yet remain in power. Indeed, in 2015 a far-left government
came to power in a coalition with the far right, as the two moderate parties that had succeeded each other since the start of the crisis massively lost
support. The new government was elected on a promise to resist ‘foreign orders’ but it had to give in and grudgingly resume supply-side reforms when
foreign support was cut off.
The Irish episode also showed that financial market excesses can ruin a party—as the boom comes to an end, financial markets often play on in a game of
double-or-nothing. Lax supervision by Irish banking authorities was the real culprit which laid the foundations for the economic crash after 2007. Good banking
regulation is also good supply-side policy.
Supply-side reforms: The silver bullet?

The cases of the Netherlands and Ireland demonstrate that the supply-side policy can have long-run effects on an economy’s long-run potential productive
capacity. Performance measures in Figure 18.9, which are based on a benchmark EU country, France, also show how long it can take for the reforms to have a
noticeable effect. The political willpower to stay the course in such situations may be difficult to sustain.
Ten years after the introduction of the euro, it has become evident that the member countries’ economies were not developing in a synchronized way—some
were lagging behind others, leading to current account imbalances and exchange rate misalignments which can cause problems in a monetary union.
16 Supply-side policies represent one important dimension of present and future attempts to bring the slow-growing parts of the Eurozone back into the growth
club. The fact that they are painful does not make them undesirable per se, but can make them more difficult to enact. Box 18.7 presents Greece as an example of
the difficulty with enacting supply-side policies.
Summary
1 Supply-side policies are appealing because, in contrast to demand-side policies, they do not imply a short-run trade-off between unemployment and inflation. They increase
output permanently at any given level of inflation and economic growth, and may even increase the rate of growth itself. At the same time, they often take considerable time
to take effect and require a longer-term perspective that goes beyond the planning horizons of most politicians.
2 One principle underlying supply-side policies is that markets do not function perfectly. By removing market imperfections, the economy’s overall output and productivity
can be enhanced.
3 There are three main sources of market failures: (1) externalities, (2) increasing returns to scale, and (3) asymmetric information.
4 Externalities occur when someone’s economic activity has an effect, positive or negative, on others. Positive externalities imply that one does not recognize the benefits to
society of one’s actions, and will undertake less than is socially desirable. Negative externalities imply that one does not recognize the costs to society of one’s economic
actions, and will undertake too much of them. Pecuniary externalities are solved once property rights are ascertained. Non-pecuniary externalities call for government
interventions.
5 Increasing returns lead to monopolies. Some monopolies are natural, inherent to the task itself. The solution used to be state ownership. Increasingly, state monopolies are
privatized and government interventions take the form of regulations.
6 Asymmetric information occurs when one’s actions are not known to others. It may lead to inefficient outcomes. Regulation can be designed to alleviate this problem.
7 Public goods that are non-rival and non-excludable tend not to be privately provided. Governments can and should step in, and provide these goods which can be highly
productive.
8 Taxation is necessary to pay for the operation of government. It is also a source of inefficiency because it drives a wedge between the price paid by the consumer and the price
received by the producer, reducing demand and supply.
9 Because the operation of governments requires resources with alternative uses in the private sector, supply-side considerations call for limiting public spending to the
production of goods and services that cannot be produced by the private sector. There is much debate on the correct size of government.
10 Governments often subsidize firms and industries. Although the objective of subsidies is to protect firms, they remove the incentive to compete, and ultimately cost jobs.
State ownership has similar effects. In some cases, good supply-side policy may imply reductions in subsidies to firms and the privatization of state enterprises.
11 High structural unemployment is a supply-side problem. It arises as a result of labour market distortions, some of which are due to private agents and others to interventions
of government. Eliminating structural unemployment is possible by better management of labour taxation, severance regulations, labour relations, and the social safety net.
Active labour market policies can help prevent the emergence of long-term unemployment.
12 Supply-side reforms are like social investments which involve high costs up-front for gains which are not realized for many years. The immediate gains to policy-makers are
not obvious and for that reason these reforms are often postponed or avoided. Evidence suggests that the payoff of good supply-side policies can be substantial.
13 If households and firms understand the political economy of supply-side reforms, it is less likely that they will believe that announced policies will actually be adopted. In
addition, the incentive for policy-makers to change their plans after the fact further reduces the credibility of announced reforms and thus reduces the effectiveness of such
reforms.
14 Labour market reforms are highly politicized and controversial. Because beneficiaries of reform are often in the minority—for example the unemployed—their interests may
be difficult to protect. Broad-based reforms almost always require give and take of the involved parties. Whether conflictual or consensual, reforms require time—as long as a
decade—to have a measurable effect.
Key Concepts
supply-side policies
market-clearing
laissez-faire
market failures
policy failures
economic rents
competition policy
Pareto principle
positive and negative externalities
property rights
pecuniary and non-pecuniary externalities
non-excludable
natural monopoly
human capital
privatization
industrial policies
trade policies
non-rivalry
deadweight loss of taxation
distortionary taxation
Ramsey principle of public finance
Laffer curve
job matching
Beveridge curve
active labour market policies
welfare trap
labour tax wedge
consumer and producer surplus
time consistency problem
Exercises
1 Why are supply-side policies both more promising and more difficult to implement than demand-side policies?
2 Why is cutting taxes on capital a supply-side policy? Why is it so controversial, especially in Europe?
3 According to the Ramsey principle of public finance (see Box 18.4), on which would you levy higher taxes, jewellery or petrol? Labour income or capital income?
4 It is often alleged that the Laffer curve is more likely to be relevant in countries with a large underground economy. Explain. How might the underground economy contribute
to the ‘unemployment trap’ described in Box 18.5?
5 Chapter 15 discussed the ‘Dutch disease’, the reaction of the real exchange rate to an increase in domestic wealth associated with a resource discovery. In the case of Britain
and Norway, both countries enjoyed the benefits of the North Sea oil discoveries of the 1970s. Norway subsidized its exporting industries as a response, while Britain used
the resources to help balance the budget and therefore pay for transfers and government spending. Why might a subsidy be good supply-side economics in this case?
6 It is sometimes claimed that overtime working contributes to the unemployment problem. In particular, hiring and firing costs make it more attractive to pay current workers
to work more (the intensive margin) than to hire more workers (the extensive margin). Furthermore, tax provisions may shield overtime income from normal labour taxation.
How might reform of overtime working be difficult under these conditions? Can you think of reform measures that might encourage fewer overtime hours?
7 What is an unemployment trap? How does it occur? Explain why workers at lower pay levels are more likely to be caught in the unemployment trap.
8 Show diagrammatically how a leftward shift of the collective labour supply curve can lead to a long-run decline in output.
9 What is the difference between active and passive labour market policies?
10 Trace through the macroeconomic effects, in the short and long run, of a policy that reduces equilibrium output (for example, an increase in labour taxation). Under what
conditions could this lead to a permanent increase in the rate of inflation? (Hint: use the AD–AS framework developed in Chapter 14.)
Essay Questions
1 ‘A cut in income taxes in Europe would have significant supply-side effects.’ Comment.
2 Why do some countries find it easier to carry out supply-side reforms than others?
3 How would you try to convince workers that they would globally benefit from lower employment protection?
4 Taxes are distortionary but they are used to finance public goods. How would you appraise a proposal for severely cutting public services in order to reduce tax distortions?
5 Define economic rent. Identify an economic rent that you find particularly objectionable and propose ways to eliminate it.
1 Note that the short-run effect of the shift is a reduction in the rate of inflation, as both short-run AS and long-run LAS curves shift rightwards. After the adjustment is
complete, inflation will return to its previous level, the target inflation rate set by the central bank (point A”) (see Chapter 14).
2 In the 1950s, Nobel laureates Kenneth Arrow of Stanford and Gerard Debreu of Berkeley showed how Adam Smith’s intuition could be rigorously established. They identified the
conditions—which are quite numerous!—under which a market economy delivers the socially optimal allocation of resources.
3 Beer can be made with any number of ingredients, and some forms of home-grown competition may have posed an existential threat at various times in German history. Ironically,
the Bavarian Reinheitsgebot , which celebrated 500 years of existence in 2016, was passed to protect beer drinkers from the health effects of bad brew, but has also functioned to protect
state monopolies and tax revenue involved in the production of hops and malt, and safeguard the supply of wheat and rye needed for bread production.
4 The analysis of trade unions is developed in Chapter 4.
5 Named after Vilfredo Pareto (1848–1923), an Italian economist who taught in Lausanne.
6 This result is known as Coase Theorem, named after Ronald Coase, a US economist who received the Nobel Prize for his work on the importance of property rights in a market
economy.
7 For example, after the oil shocks of the 1970s, reduced world demand for tankers and the emergence of Asian competition (Japan, Korea) created major difficulties for European
shipyards. The UK, Germany, France, and other countries reacted by subsidizing their shipbuilding industry. In the end, the costs became too high and the situation too hopeless for the
subsidies to be maintained. While the subsidies did save jobs for a few years, they did so in a very inefficient way.
8 Named after Frank Ramsey (1903–1930) a philosopher, mathematician, and economist at the University of Cambridge.
9 A recent European example was an increase in the cigarette tax in Germany, which was followed by a decrease in cigarette tax revenues. Taxes make up 80–85% of the retail price of
cigarettes.
10 Economist Arthur Laffer, then from Chicago, is reported to have been influential in persuading US President Ronald Reagan to cut taxes in the early 1980s.
11 This qualification is important. If some type of labour is untaxed, demand for that type of labour will rise when labour taxes increase, increasing employment and wages. The obvious
example is the underground economy, which always thrives when taxes are high. Another is self-employment, a status in which workers must provide for their own health and pension
outside the state social security system.
12 Chapter 4 provides more details on the collective labour supply curve. Producer and consumer surplus are defined and discussed in Box 18.3.
13 Wim Kok was a Dutch union leader and prime minister, who helped engineer the Wassenaar Accord in 1982 between labour, management, and the government in the Netherlands,
described in more detail in Section 18.6.2. US President Ronald Reagan permanently cut and simplified personal income taxes in the USA in the 1980s. UK Prime Minister Margaret
Thatcher deregulated and reformed UK labour markets, significantly restricting the influence of labour unions. US President Jimmy Carter deregulated the airlines, railroad, trucking, and
brewing industries. Polish Finance and Deputy Prime Minister Lezek Balcerowicz imposed ‘shock therapy’ on Poland after the fall of communism and the dismantling of central
economic planning. German Premier Gerhard Schröder introduced the Agenda 2010 and the Hartz reforms of German labour market law, which tightened eligibility and duration
requirements for unemployment benefits.
14 For example, firms were not allowed to ‘roll over’ the same worker in a series of short-term contracts. Employers have devised numerous loopholes to deal with this restriction.
15 Rudiger Dornbusch (1942–2002) was professor at Rochester, Chicago, and MIT, advisor to Latin American countries, and the author of a best-selling macroeconomics textbook (one
that sold considerably more copies than the one you are reading!). He is best-known as the father of the overshooting model of nominal exchange rate adjustment under sticky prices
which was sketched out in Chapter 15, the model which married the long run of purchasing power parity with uncovered interest rate parity.
16 In the next chapter we will see that relatively synchronized GDP growth is a key feature of well-functioning monetary unions.
The Architecture
of the International
Monetary System
19
19.1 Overview
19.2.1 The Gold Standard and How It Worked
19.2.2 The Inter-War Period
19.2.3 The Bretton Woods System of Fixed Exchange Rates
19.3.1 IMF Assistance and Conditionality
19.3.2 Special Drawing Rights
19.3.3 Surveillance
19.4.1 The Impossible Trinity
19.4.2 Booms and Busts
19.4.3 Crises Driven by Fundamentals
19.4.4 Non-fundamental Crises
19.5.1 An Old Debate: Fixed versus Flexible Exchange Rates
19.5.2 The New Debate: Financial Liberalization
19.5.3 Currency Boards and Dollarization
19.5.4 Monetary Unions
19.5.5 The European Monetary Union
Summary
When we understand that Lombard Street is subject to severe alternations of opposite causes, we should cease to be surprised at its seeming cycles. We should cease too, to be surprised at the sudden panics. During the period of reaction and adversity, just even at the last instant of
prosperity, the whole structure is delicate. The peculiar essence of our banking system is an unprecedented trust between man and man: and when that trust is much weakened by hidden causes, a small accident may greatly hurt it, and a great accident for a moment may almost destroy it.
Walter Bagehot (1873)
19.1 Overview
If history has taught us anything, it is that currency and banking crises keep recurring, despite the best efforts of some politicians to ban them. In the years 2007–2012, the financial crisis started as a mortgage banking crisis in in the USA, spread to Wall Street and
other money centres, then caused a collapse of global credit and the greatest economic slowdown since the 1930s. It spread to Europe and struck at the heart of the common currency project in ways that economists did not predict. It made clear the dominant role
of international financial markets.
This is not the first global crisis. Indeed, the international community had long been ready for emergency action. The International Monetary Fund (IMF), in particular, was set up after the Second World War to provide emergency funds to lessen the
burden of hard-hit countries and to try and prevent contagion—but in a time of fixed exchange rates. After the end of the Bretton Woods System, it has served as an international watchdog, providing objective surveillance of countries for international private and
public lenders. With the European sovereign crisis, the IMF has assumed a central role alongside European authorities. Yet, its actions, and those of other international institutions, are often criticized. Could the crisis not have been prevented? Was the
international response adequate? Over the last decade, there have been frequent calls to overhaul the international financial system, but very little has happened.
It may well be that financial crises are inherent in financial markets; this certainly seems to be the opinion of William Bagehot, writing in the late nineteenth century. It was once said that international finance has progressed in the same way international air travel
has: there is more and more of it, it is terribly important but often superfluous, and if there is an accident, it is nasty and a lot of people get hurt. Similarly, the international monetary world has become a dangerous place. It is also constantly changing. Emerging
economies that were once dirt-poor are catching up with the developed world. They have joined the bandwagon of globalization and expect to be treated like equals. Many of the questions on the table are age-old. What exchange rate regime to adopt? Should
capital movements be restrained? How to balance the relationship between the developed North and the poorer South? What role for international financial institutions like the IMF and the World Bank? In this chapter, we revisit these old questions, as well as the
European question: how much sovereignty must a country give up in order to participate in this brave new world?
Section 19.2 provides a brief overview of the history of the international monetary system that will help us to understand the current situation. Section 19.3 describes the role and structure of the IMF, the linchpin of the present system, in some detail. The crises
of the 1990s are presented and interpreted in Section 19.4, paving the way for the enduring question of the appropriate exchange rate regime. The last section provides a cautious assessment of a situation in Europe and elsewhere, which is still evolving, but lies at
the centre of the European question: what does it mean to be a sovereign nation in a world of globalized trade and, most importantly, highly integrated financial arrangements?
19.2.1 The Gold Standard and How It Worked

For centuries, both domestic and international trade was conducted using gold and silver. Metallic monies were used for thousands of years because, as explained in Chapter 9, they were easily recognizable and acceptable by others. Because they were scarce,
precious metals were a reasonably stable store of value: they were not easily subject to manipulation, and thus were a reliable medium of exchange. National currencies as we know them today did not exist.
Progressively throughout the nineteenth century, banknotes began circulating alongside gold and silver. These notes were a promise to pay the bearer in precious metal. They were as good as the name of the issuer, mostly his honesty in not issuing more
notes than he had precious metal. Initially at least, the notes were issued by private bankers, who often failed to exercise adequate self-discipline. Even if they did, banknotes were not always fully backed by metallic reserves, which led to occasional banking crises
when their owners tried to redeem their notes all at the same time. This is one reason why central banks were created, and why they displaced private banks as issuers of paper money. Central banks were formally required to hold close to 100% gold or silver to
back their issues of banknotes. These notes were convertible into gold, coins, or bullion at the holder’s request, and conversion was indeed routine. With close to 100% backing, banknotes simply represented a more convenient way of holding gold or silver.
This international arrangement was known as the gold standard. It was ultimately formalized and lasted from 1879 to 1914, less than 40 years (see Box 19.1), collapsing one month before the outbreak of the First World War. The gold standard era is sometimes
nostalgically associated with the fast growth and rapid industrialization of the time, and is regarded as a great economic success story. This ‘success’, in turn, is often attributed to the gold standard’s automatic adjustment mechanism. In fact, the gold standard
had many problems, and adjustment mechanisms were by no means as automatic as is often believed. But it remains a benchmark and, in many respects, new developments, such as monetary unions and currency boards (studied in Section 19.5.3), attempt to re-
create some of its most desirable features. So it is well worth a long, hard look.
Domestic operation of the gold standard

In principle, the gold standard was a simple affair. Money was gold, or paper that was ‘backed’ by gold. This meant that the gold was literally redeemable on demand. The demand for that money was stable, driven by the need to carry out everyday transactions.
Supply too was quite stable. Because the stocks of coinage were so large, even large gold discoveries and the resulting flows of newly coined gold or silver amounted to small monetary shocks. The role of the monetary authorities was merely to establish and
guarantee the gold content of their own currencies, the gold exchange rate.
International operation of the gold standard: Fixed exchange rates

By triangular arbitrage, pegging a currency’s value to gold was sufficient to determine all exchange rates vis-à-vis other currencies pegged to gold. For instance, if the Dutch guilder was set at some fixed price, or parity, of 50 per ounce of pure gold (the usual
weight reference) and sterling was set at £25 per ounce, the guilder was worth £0.5. If the market exchange rate were to decline to £0.4 (a depreciation of the guilder relative to sterling), it would make sense to purchase gold in the Netherlands with guilders, ship it
to the UK, sell it to the Bank of England in exchange for sterling, and convert sterling into guilders. For every 100 guilders tendered in the Netherlands for gold, the transaction would yield 2 ounces of gold, sold in the UK to acquire £50. Selling these sterling
balances against guilders at the 0.4 exchange rate would yield 125 guilders, a 25% profit! Such a prospect was sure to trigger large sales of sterling and large purchases of guilders in the exchange markets, promptly appreciating the guilder’s value back to its only
sustainable sterling value, £0.5. To be sure, the example ignores transaction costs, especially the cost of transporting the gold across the English Channel and insuring it in transit. Once these costs are accounted for, the market exchange rate can deviate
somewhat from the bilateral rate implied by the gold parity, leaving a band of fluctuation within which it is not worth undertaking the buying, selling, and shipping. These bands were known as ‘gold points’, and are similar to exchange rate intervention bands
employed in modern fixed exchange rate systems. They implied margins of fluctuation of about 1%. This is another application of the no-profit condition presented in Chapter 7.
Box 19.1 Bimetallism and Gresham’s Law
It was only at the end of the nineteenth century that gold became the premier international medium of exchange. For centuries, silver and gold had competed against each other. Bimetallism, as the system was called, established a fixed parity between
gold and silver, and coins in both metals were usually accepted for all transactions, both nationally and internationally. The relative value of gold and silver was set by international agreements, which were occasionally called into question as new
discoveries of either metal threatened to upset the parity. Troubled times then followed with the operation of Gresham’s law. This principle states that the currency (metal) that is more valuable (in non-monetary markets) than its official rate stops
circulating: ‘bad money chases out good’.1
Partly because silver became more plentiful, bimetallism ceased to exist in Europe in the 1870s. The last major countries to defend bimetallism formed the Latin Monetary Union in 1865, setting a parity of 15.5 ounces of silver for 1 ounce of gold. This
union consisted of Belgium, France, Italy, and Switzerland (for this reason all, with the exception of Italy, adopted the ‘franc’ as the name of their national currencies). In the USA, where a central monetary authority was absent, bimetallism survived for a
longer time. The final blow occurred when the newly created German state switched to gold and unloaded large amounts of silver on the free market. The risk of complete gold loss in a world under a gold standard forced the remaining countries to
abandon bimetallism entirely.
The Hume Mechanism

In the absence of advanced capital markets and international financial institutions, flows of gold between nations served as the primary means of financing payment imbalances. This adjustment was identified more than two and a half centuries ago by David
Hume, the Scottish economist and philosopher who first described it. The Hume mechanism (sometimes called ‘gold-specie mechanism’) is still highly relevant for understanding the origins, evolution, and resolution of balance of payments imbalances.
The Hume mechanism works like this. A country which ran a trade deficit would automatically lose gold to its trading partners. The metal was often physically shipped abroad to pay for the excess of imports over exports. 2 The exported gold coins were then
minted and coined in the currency of the surplus country. Thus, a trade deficit implied a shrinking money supply, a surplus meant an expanding money supply. This had three consequences. First, the reduction of the money supply led to tighter monetary
conditions and higher interest rates in the deficit country, which encouraged capital inflows. In this way, capital account surpluses could finance trade deficits. Second, higher interest rates tended to slow down economic activity, which reduced imports at given
exchange rates (think of the net export (NX) function introduced in Chapter 11).3 Third, the contraction of the money supply associated with gold outflow would reduce the domestic price level directly. Hume saw this effect through simple monetarist lenses (lower
gold supply meant lower prices in the long run) but an economic slowdown will also lead to deflation, as rising unemployment weakens workers’ bargaining power and firms’ mark-ups decline. These price declines (or lower inflation) would improve the deficit
country’s competitiveness and restore trade balance.
In the export surplus country, the process works in the opposite direction: balance of payment surpluses lead to gold inflows, which expands the money supply and depresses interest rates. An economic expansion results, leading to higher prices, which
reduces the surplus directly (growing economies will import more) and indirectly (as the real exchange rate appreciates and competitiveness is eroded).
The Hume mechanism is a symmetric adjustment process, that worked silently and automatically under the gold standard. It could only be interrupted or interfered with if governments intervened. We will see later that within fixed exchange rate regimes, the
Hume mechanism is weakened if imbalances are not allowed to give rise to corrective mechanisms. One assessment of the current problems of monetary union is the failure of the Hume mechanism to function properly.
The main benefits of the gold standard

The world has never seen, and probably never will see, a true pure gold standard. Still, some observers regret the passing of the gold standard that existed. Why? First, the Hume mechanism had the virtue of credibility. Under a gold standard, monetary policy is
entirely determined by the stock of gold. In principle, it is out of the politicians’ hands. Second, with the money supply naturally constrained by the availability of a rare resource, inflation is not likely to emerge on any significant scale, and it didn’t. This is
documented in Table 19.1. Third, there is no need for a particular country to be at the centre of the world monetary system, avoiding competition and conflict regarding which country that should be. As we will see later, the last century of monetary relations has
been dogged by this issue, even long after gold had left centre stage.
The gold standard certainly produced a stable price level. People were able to base their expectations on a fairly predictable stable evolution of the value of goods in terms of money—in Chapter 16 we saw the advantages of low inflation rates. In the theory of
economic policy, the Hume mechanism and the gold standard illustrate the advantages of rules over discretion. When expectations are being formed, private agents try hard to guess the intentions of policy-makers. In Chapter 13 we saw that workers and firms
need to take into account the future path of inflation, which involves guessing future central bank policy. Often they base their expectations on past behaviour. If the central bank has misbehaved in the past, these expectations will be pessimistic. Thus, even if the
central bank has good intentions, a bad outcome is possible: agents expect high inflation, which the central bank may then feel compelled to validate with an expansionary monetary policy, usually to avoid a recession. The gold standard, by implementing
automatic rules, made it easy for economic agents to reckon with the economic policy. By removing policy-makers’ discretion to adjust policy, rules can lead to a more credible environment for economic agents.
Table 19.1 Inflation Rates in Five Countries, 1900–1913 (annual average rate of increase in GDP deflator, %)
France Japan USA Germany UK

0.9 2.8 1.3 1.3 0.9
Source: Maddison (1995).
Limits of automatism
Despite these appealing aspects, the gold standard had several limitations. Sterling was the main currency, backed by the most developed financial centre at the time, London. Britain had been on the gold standard since 1819 when the Bank of England received its
key statute (the Act for the Resumption of Cash Payments, known as Peel’s Bill), and longer than other countries. Furthermore, as the largest creditor country, Britain provided the rest of the world with sterling balances which often ended up as reserve currency
held by other central banks.4 Three consequences followed from Britain’s dominant, or hegemonic, position. First, the Bank of England was able to set the interest rate for the rest of the world, but naturally concerned with British economic conditions. Second, the
demand for sterling as a reserve currency allowed Britain to finance long-running balance of payments deficits, paid with sterling-denominated liabilities of the Bank of England. In doing so, Britain could evade the automaticity of the Hume mechanism: its money
supply did not decline when it ran deficits with the rest of the world, because gold was not shipped to cover its deficits. Third, the widespread acceptability of sterling balances allowed the Bank of England to maintain a ratio of reserves to deposits—known as
‘the Proportion’—well below 100%. At the height of the gold standard late in the nineteenth century, the Proportion fluctuated between 30% and 50%. To the extent that its gold stock was shielded from the vagaries of the balance of payments and since the
Proportion could vary, the Bank of England possessed considerable freedom to set its interest rate.
Britain was not the only country that tinkered with Hume’s mechanism. Some countries actually imposed limits on gold exports and imports, as well as on minting and coinage. Many central banks accumulated sizeable reserves of foreign currencies, first and
foremost in pounds sterling. Thus, the assumed link between metal and money supply was less than fully automatic. Furthermore, the gold reserves of the Bank of England eventually fell below the value of the Bank’s liabilities towards other central banks, leading
to an ‘overhang’ of unbacked British debt. Many scholars consider it a miracle that the overhang never threatened the credibility of external sterling liabilities; others see it as testimony to the power of credibility and trust in monetary systems.
The limits of a metallic standard

The automatic mechanism that is often considered to be the central advantage of the gold standard does not come for free. It can have significant cost in terms of economic instability. While the gold standard years were characterized by low inflation, they also
exhibited greater output variability. As already described, there is a trade-off between rules and discretion. If respected, the rules of the gold standard are very strict: the money supply is determined solely by the balance of payments, so macroeconomic
adjustments must be dealt with entirely through wage and price changes. If wages and prices adjust slowly in a recession—that is a central message of Chapters 13 and 14—this adjustment process may take a long time. In the meantime, the economy ‘goes
through the wringer’ of unemployment and recession. Despite the credibility of the Hume mechanism, the gold standard has severe costs in some situations.
Another problem is that the overall gold stock—and thus the money supply—depends on natural discoveries. Economic growth, on the other side, implies a continuously expanding demand for real balances. If gold discoveries do not match the demand,
increases in the real money supply can occur only if the price level declines—i.e. the price of gold must rise.5 Figure 19.1 shows that the gold supply has increased steadily, with periods of scarcity following great discoveries in California, Alaska, and South
Africa, as well as technological advances in mining and mineral processing. As these shocks were largely random, the money supply and the price level were hardly stable, even if the average inflation rate was low.
Fig. 19.1 Monetary Gold Stock and Cumulative Gold Production, 1840–1980
Although gold production has continued over time, its rate of increase has tapered off. At the same time, the demand for gold for industrial purposes has increased over the past 100 years. Perhaps not coincidentally, gold stocks held by central banks have remained flat for several
decades.
Source: Cooper (1982).
19.2.2 The Inter-War Period

Inter-war monetary arrangements can be conveniently grouped into three sub-periods. The first sub-period ranges from the end of the First World War to the return to the gold standard in 1926. The gold standard was then maintained until 1931, and was then
followed by a period of managed float marked by competitive devaluations and a collapse of world trade as a consequence of the Great Depression.
The free float period (1919–1926)

In 1919 two countries, the UK and the USA, dominated the world monetary scene. During the First World War, the USA remained on the gold standard, but the UK had suspended the mechanism and allowed the pound to depreciate by about 30%. Under the
leadership of a young Chancellor of the Exchequer, Winston Churchill, Britain made the fateful decision to return to the gold standard at the pre-war parity. With strongly deflationary policies in place, the pound was brought back to its pre-war parity in 1925, but
remained overvalued due to persistent inflation since 1914. Germany and a number of other Central European countries returned to the gold standard only after experiencing and vanquishing their celebrated hyperinflations. France devalued the franc immediately
after the war but underwent rapid inflation in the years 1922–1926. Its return to the gold standard in 1926 after the Poincaré stabilization marked the return to the pre-war situation, albeit at a devalued parity. In doing so, France was able to avoid much of the
macroeconomic distress suffered by the UK, just as the Mundell–Fleming framework would predict.
Ephemeral gold standard (1927–1931)

The newly restored gold standard was a mere shadow of its predecessor. It had two competing centres, London and New York. Many currencies were badly misaligned: some were overvalued, like sterling; others were undervalued, like the French franc. Gold
holdings became an ever smaller part of foreign exchange reserves as most central banks were accumulating dollar and pound balances. Free convertibility between banknotes and gold was suspended, and most central banks actively discouraged or prohibited
the circulation of gold coins.6 The Hume mechanism was often circumv ented by sterilization operations. The only remaining principles of the system were the principles of currency convertibility 7 and fixed exchange rates.
When the Great Depression hit after 1929, the gold standard was already weak. With its overvalued currency, Britain was particularly vulnerable. Its gold reserves shrank quickly while France, with an undervalued currency, was accumulating gold and selling off
its sterling balances. Soon Britain’s official liabilities exceeded its gold reserves and it had to suspend convertibility in September 1931 and let sterling float. The gold standard was over.
The managed float (1931–1939)

Britain allowed the pound to depreciate sharply to about $3.3/£, and a number of countries holding large sterling balances followed suit, as can be read from Table 19.2. Formerly overvalued currencies became undervalued. At a time when all countries were
struggling against the Great Depression, these devaluations were a tempting means of gaining a competitive trade advantage at the expense of other countries. A gold bloc, including France, Belgium, the Netherlands, Italy, Switzerland, and Poland, was
established to resist the temptation of retaliatory depreciations. The situation worsened seriously in 1933 when the USA, the remaining centre of the gold standard, imposed an embargo on gold exports, introduced exchange controls, and depreciated the dollar
from $20.67 to $35 per ounce of pure gold. The coup de grâce was the dissolution of the gold bloc following the devaluation of the Belgian franc in 1935. ‘Beggar-thy-neighbour’ policies (competitive devaluations) followed, but were self-defeating since each
country attempted to devalue vis-à-vis all the others. The next step was a ‘tariff war’: each country raised its tariffs to restrict imports, thus encouraging the substitution of domestically produced products. Imports declined but so did exports, as other nations
followed suit. While the aim of boosting output failed, international trade collapsed, as shown in Figure 19.2.8
19.2.3 The Bretton Woods System of Fixed Exchange Rates

The principles
Preparations for a new monetary order were laid at the Bretton Woods conference of July 1944, convened by the Allied nations before the end of the Second World War. 9 The conference led to the creation of the International Monetary Fund (IMF), as well as
the International Bank for Reconstruction and Development (known today as the World Bank) and the General Agreement on Tariffs and Trade, the precursor to the World Trade Organization. The new world monetary order was conceived to prevent a repeat of
the competitive devaluations and unilateral trade restrictions that could threaten a post-war economic recovery:
Table 19.2 Beggar-thy-Neighbour Depreciations, Various Countries,
1931–1938 (value of currencies as percentage of 1929 gold parity)
1931 1932 1933 1934 1935 1936 1937 1938

Belgium 100.1 100.2 100.1 99.9 78.6 72.0 71.7 71.8
Denmark 93.5 70.3 55.8 50.0 48.5 49.0 48.6 48.1
France 100.1 100.3 100.0 100.0 100.0 92.4 61.0 43.4
Germany 99.2 99.7 99.6 98.6 100.3 100.1 99.7 99.6
Italy 98.9 97.4 99.0 97.0 93.0 82.0 59.0 59.0
Norway 93.5 67.2 62.7 56.3 54.5 55.2 54.7 54.1
The Netherlands 100.1 100.3 100.1 100.0 100.0 94.9 80.9 88.8
Switzerland 100.6 100.6 100.2 100.1 100.0 92.6 70.2 70.0
UK 93.2 72.0 68.1 61.8 59.8 60.5 60.0 59.3
USA 100.0 100.0 80.7 59.6 59.4 59.2 59.1 59.1
Source: League of Nations, Statistical Bulletins.
Exchange rates were to be fixed; realignments required prior IMF approval.

The IMF could provide loans as an alternative to devaluation for countries facing balance of payments difficulties.
The dollar was the centre of the system. All countries officially declared a fixed parity, called a par or central value, vis-à-vis the US dollar, which itself pegged to gold directly. Currencies were allowed to deviate by no more than 1% from the par value.
Exchange controls and tariffs were allowed only as temporary measures for the immediate post-war period. In the event, full currency convertibility was only achieved in Europe in 1958, and a number of developing countries still have non-convertible currencies and capital controls.
Fig. 19.2 The Decline of World Trade During the Great Depression
During the period 1929–1933, the enormous increases in world trade that had been accomplished in the previous three decades were wiped out by a spiral of protectionist measures. This famous illustration by Professor Charles Kindleberger of MIT, taken from a 1934 publication of the
League of Nations, shows just how quickly trade wars can get out of hand.
Source: Kindleberger (1973).
Fig. 19.3 The Three Layers of Bretton Woods
The three layers of the Bretton Woods system consisted of gold, the US dollar, and the other participating currencies. The USA declared a gold parity for the dollar, thereby pegging to gold. Intervention took place mostly in dollars, but there was an implicit understanding that gold stood
behind the dollars.
Gold and the dollar

Officially, all currencies were defined in terms of gold. Yet, at the end of the Second World War, the USA held about 70% of all gold reserves and was the only country credible enough to set a gold parity. With the US Marshall Plan providing European countries
with dollar balances, the most obvious approach for them was to declare a parity vis-à-vis the US currency.10 The outcome was a de facto three-tier system, represented in Figure 19.3. Gold remained the fundamental standard of value, but for all currencies this was
mediated by the dollar, hence the name gold exchange standard given to the Bretton Woods system. The system thus relied on the ability of the USA to maintain the declared parity of $35 per ounce of gold.
The International Monetary Fund reengineered

For a long while, the Bretton Woods system worked rather well. After a rash of post-war parity adjustments—including an unauthorized devaluation of the French franc in 1948—exchange rate stability prevailed. Trade expanded quickly and was easily financed by
dollar balances, provided initially by the Marshall Plan, then by US trade deficits and the resulting capital flows.11 The IMF became the respected watchdog of the fixed exchange rate system. It developed an elaborate system of loans to countries suffering
balance of payments difficulties. Its resources are provided by member-country deposits. The size of a country’s deposit, based on its size, determines its quota. Quotas determine each metmber country’s voting weight and its borrowing rights, and are updated
every five years. Details on its current role are provided in Section 19.3.
Devaluations were in principle restricted to cases of ‘fundamental disequilibria’, i.e. balance of payments deficits not of a temporary (cyclical) nature. In order to help member nations avoid devaluation, the IMF made, and still makes, resources available for
immediate lending—called ‘purchase agreements’ when effected and ‘repurchase’ when reimbursed. Each member country is eligible for immediate lending up to its quota. Beyond that, lending becomes conditional. The IMF requests a formal agreement on
specific policy steps and results designed to solve the ‘non-fundamental’ part of external disequilibria. This conditionality became the central source of power of the IMF that would survive the collapse of the Bretton Woods system.
The Triffin paradox and the collapse of the Bretton Woods system
As economies grew and international trade developed, more ‘international money’ was needed. Since the US dollar was the international money, more dollars would have to be made available to the world economy. For internationally held dollar balances to grow,
it was necessary for the USA to run balance of payments deficits, just as Britain did during the days of the gold standard. Inevitably, US official liabilities abroad outgrew the country’s gold reserves. Figure 19.4 shows that this happened in 1964. Yet at this point,
the USA could no longer guarantee the gold value of the US dollar. This is the Triffin paradox,12 a fatal weakness of the gold exchange standard.
A conjunction of economic and political events brought the situation to a climax. The Vietnam War and ambitious domestic social programmes (President Johnson’s ‘Great Society’) led to increased public spending in the USA, which accelerated growth and
inflation, and deepened the current account deficit. At the same time, countries critical of the Bretton Woods arrangement began to protest loudly. French President de Gaulle publicly complained about the ‘privilège exorbitant’, which allowed the USA to use
seigniorage to finance its political activities (the Vietnam War) and economic power (the acquisition of European corporations by US companies at the time). In a dramatic gesture, France began to swap dollars for gold in the mid-1960s, increasing its precious metal
stock from $3.7 to $5.2 billion between 1964 and 1966.
The markets took notice. Anticipating an increase in the price of gold, they sought to buy it while it was still cheap. The response of the monetary authorities was to form the Gold Pool (Belgium, Italy, the Netherlands, Switzerland, West Germany, the UK, and
the USA, with France inactive after 1967), an agreement to sell gold to maintain the $35/ounce parity. As the drain on official gold holdings accelerated, the Pool pulled out of the gold market and declared that they would henceforth trade gold only among
themselves—they would neither sell to nor buy from private parties—at the official price. The market price of gold rose substantially higher than the official parities. Tensions within the Gold Pool grew until President Nixon’s historic decision to suspend the gold
parity of the US dollar on 15 August 1971.
Fig. 19.4 US Official Liabilities and Gold Reserves, 1950–1970

As long as foreigners were willing to hold dollars, the USA could finance its large balance of payments deficits by increases in foreign holdings of official assets (dollars held by central banks). Yet the gold reserve of the USA declined over the entire period shown, as foreign central
banks occasionally tendered their dollars for gold. Sometime in 1964, the stock of external official claims against the US gold exceeded the dollar’s gold backing. At that moment, the credibility of the gold exchange standard was called into question.
Sources: Dam (1982); IMF.
Fig. 19.5 Inflation Rates: USA, UK, Germany, and Italy, 1960–1976
Inflation rates during the Bretton Woods era moved closely together, which is characteristic of a fixed exchange rate regime. As soon as the system collapsed in 1971, inflation rates diverged sharply.
Source: IMF.
From the Smithsonian Agreement to Jamaica

The severing of the gold–dollar link destroyed a key component of the Bretton Woods arrangement, but the gold crisis was not the sole factor in its demise. Inflation had been rising in most countries in the late 1960s, but at increasingly different rates (Figure
19.5). This led to misalignments which challenged exchange rate parities, which had been virtually unchanged since the late 1940s.13 Speculative capital movements followed. Britain and Italy came under IMF conditionality in 1969. The pound was devalued in
1967, followed by the French franc in 1969, while the deutschmark was revalued. The delinking of the dollar from gold opened the Pandora’s box for further realignments. The credibility of the exchange rate system was severely damaged.
The last major effort to save the sinking ship was an agreement reached in December 1971 during a conference at the Smithsonian Institution in Washington, DC. The dollar was devalued vis-à-vis gold to $38 per ounce, yet remained inconvertible into gold,
even among central banks. Some currencies were revalued, others devalued. The margins of fluctuations were enlarged from 1% to 2.25% around par value, while the European countries maintained a reduced (half) margin. This was the ‘snake’ arrangement. By the
end of 1972, the pound was floating, soon to be followed by the Swiss franc, the Italian lira, and the Japanese yen. In March 1973, the remaining ‘snake’ members decided to float jointly vis-à-vis all other currencies, including the dollar, within a wider 2.25% margin.
France left, then re-entered, as did Italy; then both left again. Sweden and Norway joined informally. By 1975 the principle of fixed exchange rates was more or less dead, at least for the convertible currencies of the industrialized countries. In January 1976, the
Jamaica Agreement made official the new role of the IMF. From then on, it would be in charge of overseeing a world monetary system of increasingly flexible exchange rates.
The influence of the IMF is probably stronger today than in the heyday of the Bretton Woods system, if only because it is now in charge of a system that is less internally inconsistent than before. With the demise of the gold exchange standard
and a floating dollar, the Triffin paradox was no longer relevant. Each country is free to choose its own exchange rate regime, and there is no agreed-upon international currency; the US dollar plays this role de facto, a feature that can be
challenged every day in the marketplace. Gold has long been ‘demonetized’, it is no longer a reference, and many countries have since sold large parts of their gold stocks.
19.3.1 IMF Assistance and Conditionality

Countries continue occasionally to face balance of payment problems. In principle, they are now free to depreciate their currencies in response to adverse shocks, and over the longer haul such policies can bring the current account back to levels consistent with
an intertemporal budget constraint. Frequently, though, the authorities prevent their exchange rate from adjusting to shocks by intervening in the foreign exchange markets, sometimes with borrowed foreign exchange. Eventually they may exhaust their foreign
exchange reserves—reserves built up during the past by running surpluses on the sums of current, financial, and capital accounts. What can they do then? An obvious answer is to treat the causes, and not the symptoms of the crisis. Taking remedial action,
usually moving away from undisciplined monetary and fiscal policies, requires time, however, and the house is burning. The solution is to call the IMF, the international firefighter.
When called upon in an emergency, the IMF proceeds in three steps:
(1) It assesses the situation and makes recommendations to the authorities.
(2) Conditionality follows: an emergency loan is extended to the troubled country, which in return commits to a number of policy actions.
(3) As the loan is disbursed, typically in several instalments, the IMF monitors the implementation of the agreement and may suspend further disbursements if the country in question is violating its commitments.
In an average year, some 10–20 loans are arranged.14 The average volume of loans outstanding represents about 12% of world exports. In 2016, 189 countries were members of the Fund.
It is important to stress that the Fund’s assistance is not a.pngt, but a loan, generally with a maturity of one to four years. The interest rate charged is slightly above the market rate to discourage using the Fund as a cheap source of money. Most private
financial institutions deal only with sovereign borrowers in good standing with the IMF, so the Fund usually has a great deal of leverage over borrowing countries. This explains, of course, why countries in trouble turn to the IMF and are also willing to accept the
conditions that it requests. Despite popular beliefs to the contrary, most countries actually repay their loans. The reason is that the IMF has priority over all other creditors, and not repaying the IMF effectively means being excluded from all other sources of
international financing. Only a handful of ‘pariah’ countries have defaulted on IMF loans. They will have to pay back if and when they want to re-enter the international financial arena. Most eventually do.
The Fund’s resources—the amounts it can lend—are limited to three sources. Own resources are the deposits of member countries—equal to their quotas. One quarter of a country’s deposit must be in ‘strong currencies’, with the remainder in its own currency.
The deposited strong currencies thus constitute the first usable resource. Second, as explained in the next section, the IMF has created the Special Drawing Rights (SDR), which are rights to borrow strong currencies from issuing countries. Finally, the Fund has
the right to borrow funds, up to a limit, from the international financial markets. In early 2016, available resources amounted to about $600 billion.
This is a small amount. Individual countries may need a bailout on the order of $100 billion, sometimes more. This is why the IMF is often bringing together a group of countries willing to help out, by lending bilaterally to a crisis country alongside the Fund. The
usual practice is that the IMF manages the loans and sets the associated conditions. During the Eurozone crisis, however, it changed its approach. It lent to Greece, Ireland, Portugal, and Cyprus one third of their estimated needs and agreed to co-manage the
interventions with the European Commission and the European Central Bank. For the first time, it accepted a junior status in a major rescue operation.
19.3.2 Special Drawing Rights

Despite having lost the official status it enjoyed in the Bretton Woods system, the US dollar remains the de facto means of payment for international trade and the foreign exchange reserve currency of choice at central banks. Potential competitors arise from time
to time, such the euro or the Chinese renminbi, but the more interesting question is: why should the international means of payment be any particular country’s currency? In the late 1960s, many countries lobbied for the creation of a new world currency, reviving
an old idea which had been defeated at the Bretton Woods conference in 1946. 15 If the IMF could be transformed into the world central bank, it was thought, it could issue its own currency for all central banks to settle their payments, borrow from each other, and
intervene on exchange markets. Unsurprisingly, the USA, the Fund’s largest shareholder—with a veto right—objected.
With the objective of increasing and stabilizing the supply of international liquidity, it was decided at the Rio Conference in 1967 to create the special drawing right (SDR). SDRs can be seen as a line of credit allocated by the IMF to each country in
proportion to its quota. Each member country can draw on its line of credit to obtain convertible currencies from the Fund. At the time of its conception, the SDR was valued at the rate of SDR 35 per ounce of gold, thus making it worth exactly US$1. The SDRs
were not, however, backed by gold or dollars. Just like money created by banks, SDRs are valued simply because they are accepted.16
After gold convertibility was suspended, the SDR’s value was redefined as a basket of four currencies. In 2016, the renminbi was included in the basket. 17 SDRs yield an interest rate—the weighted average of interest available on the four underlying currencies.
Symbolically, the SDR is the IMF’s unit of account. As a basket, it is less volatile than any of its components, which has made it convenient for other purposes. Some countries peg their exchange rate to the SDR. Private debt issues have been denominated in
SDRs, although technically they are just a basket of the constituent currencies. Some 9 billion SDRs were initially created in 1970, and more were subsequently added in 1981 and in 2009 to a cumulative total of 204.1 billion.
19.3.3 Surveillance
The IMF is not just a firefighter. It is also a safety inspector. It continuously monitors the macroeconomic scene in each member country in order to detect possible risks to its currency and to keep abreast of the local situation, should an emergency arise.
Surveillance takes several forms:
Annual visits and evaluations (called Article IV consultations). The IMF’s economic assessments and recommendations, once highly confidential, are now posted on the Fund’s website (each country has the right to refuse the release of this information, but that is considered a bad signal).
Twice a year, the IMF publishes the World Economic Outlook, which outlines its views of the situation in the world and in member countries.
For countries in difficult situations, the IMF conducts ‘enhanced surveillance’, which means more frequent evaluations and recommenda-tions.
For countries that have borrowed from the Fund, the so-called programme countries, the monitoring is more or less permanent, based on agreed-upon targets for policies and outcomes.
Each government knows that its policies are monitored and that the conclusions are presented to the Executive Board of the Fund. Box 19.2 explains how decisions are made. Surveillance is justified as a preventive means of avoiding disruptive, ‘beggar-thy-neighbour’ policies that wrecked
the world economy during the inter-war period. When countries pursue economic policies that are criticized by the IMF, this disapproval is noticed by the outside world and usually results in internal and external political and financial pressure, which can go far in correcting aberrant
policies.
Box 19.2 How the IMF is Managed
The ultimate authority is exercised by the IMF’s Board of Governors which meets, in principle, once a year. The governors are the finance ministers or central bank governors of all member countries. Quotas are distributed according to a formula that
involves GDP, current account transactions, foreign exchange reserves, and a measure of volatility of international transactions. Voting rights are proportional to quotas plus a fixed number of ‘basic votes’ designed to increase the rights of the smaller
countries. Table 19.3 displays the quotas and voting rights of the top 20 countries. The most important decisions require 85% of the votes, which gives the USA de facto veto power.
In practice, the Board of Governors delegates managing authority to the Executive Board, which consists of 24 Executive Directors. They are located at the IMF’s headquarters in Washington and meet nearly every day. The largest-quota countries (the
USA, the UK, Germany, France, Japan, China, Russia, and Saudi Arabia) have one executive director each, while the other executive directors represent several countries, grouped along regional lines. The Executive Directors select the Managing Director
to run the professional staff. 18 All decisions are taken by the Board of Executive Directors, which cast votes in the name of each country according to its quota. In this way the Executive Directors represent the interests of their countries, while the staff and
the Managing Director represent the institution.
Table 19.3 IMF Quotas and Voting Rights (January 2016)
Quotas Votes
% of total % of total
Japan 6.51 6.18
China 6.44 6.12
Germany 5.63 5.35
France 4.26 4.05
Italy 3.19 3.04
India 2.77 2.65
Russian Federation 2.73 2.61
Brazil 2.33 2.23
Canada 2.33 2.23
Saudi Arabia 2.11 2.02
Spain 2.02 1.93
Mexico 1.88 1.81
Korea 1.81 1.74
Australia 1.39 1.34
Belgium 1.35 1.31
Switzerland 1.22 1.18
Indonesia 0.98 0.96
Source: IMF.
19.4.1 The Impossible Trinity

An important implication of the IS–TR–IFM model is the impossible trinity. We know from Chapter 12 that monetary policy is non-existent when the exchange rate is fixed and capital is fully mobile. In that case, as a consequence of the interest rate parity
condition, the domestic interest rate i is fully driven by the foreign interest rate i* and expected exchange rate change, which is nil if the fixed exchange rate is fixed. Put simply, the impossible trinity states that the following three aspects of a monetary system are
jointly incompatible:
(1) full capital mobility;
(2) fixed exchange rates;
(3) monetary policy independence.
Taken in pairs, however, these features are compatible and feasible. During the Bretton Woods era, capital controls were the rule rather than the exception. The impossible trinity was satisfied and (2) and (3) above could coexist. The coexistence of fixed exchange
regimes and capital controls is more than coincidental. Capital controls enable countries participating in fixed exchange rate arrangements to preserve some monetary autonomy. Over time, however, the globalization of international financial markets brought near-
perfect capital mobility along with it, allowing investors and speculators to swap billions of short-term assets at the push of a button. This was a game changer, which many policymakers did not recognize for some time.
19.4.2 Booms and Busts

Boom-and-bust cycles are a classic feature of currency crises, and deserve special attention here because they represent an excellent application of the principles we learned in Chapter 14. They share a number of common features. The boom starts with
liberalization in a country long accustomed to operating a fixed exchange rate system under the umbrella of capital controls. The liberalization attracts foreign investment. The central bank lets its foreign exchange rate reserves grow and fails to offset this source of
monetary expansion. On the contrary, it usually marvels at the effect of foreign investment and lower interest rates on growth. Growth becomes too rapid to be sustainable as aggregate demand increases (the AD schedule shifts to the right) much faster than
aggregate supply.
Soon inflationary pressures arise. The authorities, on the other hand, are so pleased that they overlook the growing overvaluation of the exchange rate—who wants to spoil the party with a devaluation? These easy years are usually characterized by
unbounded optimism and laxity, and considerable risk-taking. Domestic banks have little trouble obtaining funds from abroad in foreign currencies and lend them on freely in domestic currency. Governments too often borrow abroad in foreign currency, getting
better interest rates and betting on the continuation of the miracle. The IMF is either as optimistic as the local authorities, possibly because the rising current account deficit appears to finance fixed investment (including housing construction) and looks
sustainable—or is a voice in the wilderness as its danger warnings go unheeded. International investors marvel at the success story, but keep a critical eye on the situation.
Then something happens. The current account deficit deepens too much, or some miracle country elsewhere in the world stumbles. Then the investors move out faster than they came in, accompanied, if not preceded, by local investors. The exit is vaguely
reminiscent of the bank runs encountered in Chapters 9 and 10. The crisis erupts and output falls, often deeply.
Currency crises are like automobile accidents: they seem to happen randomly and unexpectedly, but this is generally not the case. A first class of explanations maintains that dangerous drivers are more likely to have more accidents. Not only can we see the
accidents coming, but we can do something to prevent them, either by sending bad drivers to school or by increasing the penalties for hazardous practices. Not all crises can be easily explained by such fundamental factors. Many occur for no apparent reason.
Self-fulfilling crises occur merely because they are expected to occur. While countries with impeccable drivers may be reasonably immune to such accidents, most others are not and can be hit, sometimes as part of a contagious process. We look at these two major
interpretations in the context of currency crises.19 This type of crisis occurs most visibly in countries that operate a fixed exchange rate system, although brutal depreciations also occur in countries that allow their exchange rates to float.
19.4.3 Crises Driven by Fundamentals

Very often currency crises are the unavoidable outcome of policies that are not sustainable. Currency crises occur when policies are inconsistent with a fixed exchange rate regime, typically in defiance or ignorance of the impossible trinity. Governments may let the
domestic money supply grow too fast, often because this looks like an easy way to finance a budget deficit. We know what happens next: interest rates decline and capital flows out. If the flow is not too large, because of limitations to capital mobility or because
real-life markets are a bit more hesitant than theory claims, the authorities believe that they can have their cake and eat it too. They expand the domestic money supply, selling foreign exchange to maintain the peg, possibly even sterilizing to maintain the
semblance of monetary control.
Recall from Box 12.2 in Chapter 12 that the monetary base M0 can be linked to the asset side of the consolidated banking system as the sum of the central bank’s foreign exchange reserves R and total domestic credit DC, that is, M0 = R + DC. For a given M0,
this policy implies that as DC increases, R must decline. Foreign exchange reserves are limited, so the policy is clearly unsustainable. Something must give. The rational solution is to stop expanding money or, even better, to reverse the policy. If that does not
happen, foreign exchange reserves will continue to decline inexorably and the time will come when the central bank cannot guarantee the exchange rate. Recall the interest rate parity:
(19.1)
As long as markets believe that the fixed exchange rate will be upheld, if only for just another day, the expected change in the exchange rate (∆ S/S) is nil, and the domestic interest rate i is equal to the foreign rate i*. This implies that the demand for money stays
constant. But once it emerges that the exchange rate parity cannot be upheld, the expected change in exchange rate (∆S/S) will become negative. At this moment, the domestic interest rate rises and money demand declines. With domestic credit DC still growing,
by assumption, R must decline even faster. Market traders will not wait idly by until the exchange rate collapses; capital outflows swell, rapidly exhausting foreign exchange reserves. The central bank must give up the peg in the midst of a currency crisis, which
takes the form of a cascade of sales of the domestic currency and steadily worsening expectations.
This simple story captures the essence of an exchange crisis. Two aspects are quite striking. First, while it might appear that it is the attack that causes the collapse of the exchange rate regime, nothing could be further from the truth. The exchange rate regime
was doomed long before the crisis, its day of reckoning only being delayed by the existence of a large enough stock of foreign exchange reserves. The attack merely determines the timing of the collapse. Second, no one is surprised. All was quiet before the storm,
but it was deceptive. Everyone saw it coming and was just waiting for the right time to act. The crisis was fully anticipated. In real life, of course, there is some uncertainty and things are not quite so clean. Box 19.3 recounts the story of Iceland.
The reasoning so far has emphasized financial causes of crises, such as a monetary policy in violation of the impossible trinity principle or, in Box 19.3, misguided actions by commercial banks. The crisis occurs when exchange reserves are rapidly drawn down
because markets anticipate a depreciation, which can happen for other reasons as well. There are many other, non-financial reasons that can lead to this final meltdown. Greece, for instance, had allowed wages to increase much faster than labour productivity. In
this case, as we know from Chapter 4, labour costs increase and firms become uncompetitive. The ensuing current account deficit acts as a drain on foreign exchange reserves, which will ultimately trigger similar capital outflows. In 2015, Russia suffered large
current account deficits when oil prices abruptly declined; eventually its currency lost half of its value. Similarly, continuing budget deficits lead to current account deficits according to the identity (2.7), a frequently observed source of currency crises such as
those seen in Portugal (2012) or in Ukraine (2015) where, in addition, output fell and inflation soared in the wake of the Maidan revolution and war with Russian separatists. More generally, political instability and wars are often accompanied by currency crises, e.g.
Egypt in 2012–2013 or Venezuela in 2015–2016.
The common feature of all these crises is that they are the predictable result of fundamental flaws, which are there for all to see. For a while, a slow erosion of foreign exchange reserves gives the false impression that the situation is sustainable. If the
fundamental flaws persist, it is not and, at some point, the financial markets will come into play. They will conclude that the exchange rate is under threat. The result will be large-scale sales of the domestic currency and an accelerated depletion of reserves, leading
to a full-scale crisis.
Box 19.3 Icelandic Aftershocks following the Financial Earthquake

In the early 2000s, commercial banks in Iceland expanded their activities at amazing speed, both domestically and internationally. Credit boomed and the central bank was as enthusiastic as commercial bank managers about the newly found Eldorado.
For a while, as is usual with booms, growth accelerated and foreign exchange reserves grew as the banks repatriated profits earned abroad and foreign investors rushed to receive a share of the pie. But the US financial crisis brutally reminded investors
that banks are highly fragile and that prudence is an essential virtue of safe banking. Capital exited as quickly as it arrived. Even though the Icelandic krona was not pegged, the central bank rushed to slow its free-fall, just as it was pouring cash into
banks that were facing massive withdrawals (see Figure 19.6). Eventually, the central bank realized that its quickly depleting foreign exchange reserves were insufficient and it had to accept a deep depreciation of the exchange rate. Iceland asked the IMF
for emergency support and quickly nationalized and restructured its banks. Domestic credit growth was curtailed and reserves were once again built up over time.
Fig. 19.6 Fundamental Crisis in Iceland in 2008

The evolution of domestic credit and reserves in the months preceding the 2008 currency and financial crisis conforms well with the theory of crises. Easy money (a rising volume of credit) made a crisis inevitable.
Source: IMF.
19.4.4 Non-fundamental Crises

In contrast with crises that are the outcome of fundamentally unsustainable conditions—one might say, crises that are well-deserved—some currency crises can be self-fulfilling. This term suggests that they occur because they are expected to occur. The
reasoning may sound circular, and it is. A simple variant of the previous example illustrates starkly how crises can be self-fulfilling. The central bank is now assumed to keep domestic credit DC constant. The exchange rate peg is credible, as it should be—no one
expects the exchange rate to change and the interest rate parity condition indicates that the interest rate is constant. The demand for money is therefore also constant, as is the stock of foreign exchange reserves. The situation appears perfectly stable and could
have gone on forever, were it not for a sudden loss of confidence in the domestic currency, for reasons that are soon to be discussed. If such an exogenous loss of confidence occurs, domestic money is sold and the central bank is forced to spend its foreign
exchange reserves to uphold the exchange rate peg. The more it sells its reserves, the weaker it looks and the faster capital flows out as expected depreciation rises, and so does the interest rate, which further reduces the demand for money.
A natural reaction to this line of reasoning would be to think that any country can be subject to a non-fundamental crisis. Indeed, countries seem to play the same role as commercial banks do in a modern fractional-reserve banking system—as long as there is
confidence a particular bank is sound, there won’t be a run on it. There are two questions that need to be answered. The first one is: why have the markets lost confidence in the first place? We will return to this question shortly.
The second question is: does the central bank have the will to fight, or is it rather likely to give up? This is a matter of determination and resources. Central banks typically react to a self-fulfilling attack by declaring their firm intention to dig in and fight to the
death. But are they willing and able to do so? The central bank must be willing to accept a high and rising interest rate, high enough to reverse the tide of capital outflow. In the meantime, high interest rates depress consumption and investment spending, which
can lead to a recession. The central bank must be able to afford the cost of an exchange rate defence, which can be high in both economic and political terms. It must have sufficient reserves to mount the defence. A large stockpile, alongside unflinching central
bank determination to resist, may turn sentiment around. If the stockpile is small, it can be enlarged by borrowing from the IMF or other central banks. Box 19.4 explains how East Asian countries have attempted to organize themselves to resist speculative
attacks on their exchange rates. Finally, capital outflows can be restricted with controls. The objective is to slow the speed at which reserves are depleted. Box 19.4 also touches upon this issue.
Not all countries, however, need fear a self-fulfilling attack. A currency crisis occurs when ‘the market’ fears that a currency could be depreciated—similarly banking and debt crises reflect a fear that banks will fail or debts will be defaulted upon. However, as
explained in Chapter 14, ‘the market’ brings together a large number of traders who compete fiercely against each other, mainly by taking opposite bets. A speculative attack requires that a large number of traders take a similar view. For that to be the case, there
must be some pre-existing vulnerability that makes a depreciation plausible. Vulnerabilities are not fundamental weaknesses that make a crisis unavoidable. They are conditions that will make a central bank defence either impossible or too costly.
There are two broad categories of vulnerabilities. The first involves macroeconomic factors. A good example is high unemployment. A restrictive monetary policy response will worsen an already bad situation, to the point where the markets calculate that the
central bank will prefer to let the exchange rate go. Another example might be a stock market perceived to be a speculative bubble at risk of crashing. A sharp increase in the interest rate would prick the bubble (see Chapter 7) and could tip the whole market into an
uncontrolled free-fall. Examples of crises associated with macroeconomic vulnerabilities include Sweden in 1992, Mexico in 1995, and Argentina in 2001.
The second category concerns ‘balance sheet weaknesses’, i.e. a lack of net worth in household, business, or government sectors. This often occurs when the banks, firms, or the government have borrowed money in foreign currency. A depreciation would
increase the domestic value of these liabilities and make the debtors insolvent. In that sense, the crisis is self-fulfilling: they are not insolvent in the current situation but a depreciation can bankrupt them. The East Asian crises of 1997–1998, described in Box 19.4,
provide examples of this type of crisis. Another example is the case of a highly indebted government. With debt service already high, a sharp increase in the interest rate raises debt service, increases the overall budget deficit, and sends the debt even higher,
possibly to levels deemed unsustainable.
The upshot is that self-fulfilling attacks can occur, but only if some underlying vulnerabilities already exist. The vulnerability is not lethal, as in fundamental crises, but when combined with an attack, it makes the cost of a defence of the exchange regime
unacceptably high. The existence of a vulnerability is not a guarantee that an attack will occur; it may or it may not, and if it does, it will succeed. An important aspect of non-fundamental crises is the contagion phenomenon. Contagion occurs when a crisis in one
country starts a chain reaction, spreading to others which would have been considered safe under the status quo. A good illustration is offered by the Asian crisis of 1997–1998, which is described in more detail in Box 19.4. It is frequently argued that the
Eurozone crisis, which started in 2010, had elements of a self-fulfilling crisis.
Box 19.4 Two South-East Asian Crises and the Aftermath: 1997–1998 and 2007–2008
In June 1997, pressure started to build on the baht, the Thai currency. On 2 July, the Bank of Thailand abandoned its peg, which was followed by an immediate 20% depreciation. Speculation moved promptly to the Philippines peso and the Malaysian
ringgit. The peso floated (within bands) on 11 July, and the ringgit followed on 14 July. Next in the eye of the storm, the Indonesian rupiah too was left to float on 14 August. By mid-October, bowing to months of pressure, Vietnam widened the band of
fluctuation of the dong, and the Taiwan dollar was devalued. Brazil and Argentina started to feel the pressure at the end of October, while the Bank of Korea started to intervene heavily in defence of the won. When it had to give up on 17 November, the won
promptly fell by 10%, which triggered a new wave of attacks against the other currencies in the region (Figure 19.7). Stability finally returned when Korea reached an agreement with the IMF on 3 December, involving the largest loan ever. A conspicuous
exception was Indonesia, where a political crisis was underway. The rupiah continued to collapse until President Suharto resigned in late May 1998. By then the rupiah had shed 75% of its initial value.
In 2007–2008, another crisis hit, and many countries from East Asia (Thailand, Indonesia, the Philippines, Malaysia, Korea) faced intense speculative pressure. The crisis started in Thailand, which faced a fundamental crisis. It then spread to other
countries in the region, a contagion process that had a strong self-fulfilling flavour. Thailand, Indonesia, the Philippines, and Korea asked the IMF for help. The conditions imposed by the IMF were seen as humiliating and all countries of the region were
determined never again to have to borrow from the IMF. To that effect, the Chiang Mai Initiative, named after the city in Thailand where it was signed, organized a system of mutual conditional loans among a dozen of countries. They also undertook to
accumulate large stocks of reserves. Under attack in 1998, Malaysia decided not to ask the IMF for help. Instead, it imposed strict capital controls. Since then, the East Asian countries have further developed their cooperation in exchange reserve pooling.
Within the G20 Korea is also actively promoting agreements among the world’s large central banks for automatic loans.
Fig. 19.7 Currency in Crisis (index 100 ∙ January 1997)
The Thai currency was the first to fall in June 1997. The other Asian currencies soon followed, a spectacular case of contagion. When the dust settled, most currencies had lost one third of their initial value, and the Indonesian rupiah had depreciated by 70%.
Sources: IMF.

Ever since the end of the gold standard, policy-makers and economists have debated the merits of fixed versus flexible exchange rates. The choice of exchange rate regime was always perceived to be between fixed-but-adjustable exchange rates of the Bretton
Woods system type, or more or less freely floating rates. Fashions have come and gone, from Bretton Woods to Jamaica. Nowadays the pendulum seems to have swung back to a revival of pseudo-gold standard arrangements (monetary unions, currency boards,
dollarization) which rely largely on the Hume mechanism. Prompted by the globalization phenomenon and recent crises, the debate has also raised the old question of the desirability of capital liberalization. In this section we review the old debate and move on to
the more recent ideas.
19.5.1 An Old Debate: Fixed versus Flexible Exchange Rates

The case for flexible exchange rates
Arguments for flexible exchange rates take two forms: either they praise the virtues of flexible rates, or criticize the limitations of fixed rates. Overall, they can be boiled down to the view that it is better to leave the exchange rate to the markets than to the
authorities. On the pro-flexible side:
(1) Exchange rate changes are frequently needed to compensate for inflation differentials (the PPP principle). Fixed exchange rates can only be adjusted sporadically, which leaves long periods when they are misaligned. In addition, such realignments are easily predictable and lead to
speculative attacks.
(2) Exchange rate changes are also needed to cope with shocks which alter external competitiveness, e.g. changing energy prices, the emergence of new competitors, etc. With a fixed exchange rate regime, either all prices have to adjust, or the exchange rate must be changed. With wage and price
rigidity as a fact of life, the first solution can be protracted and painful, possibly requiring pressure on wages and prices to be brought about by the Phillips curve mechanism, i.e. unemployment.
The most common arguments against fixed ex-change rates go like this:
(1) We do not know with much precision what the equilibrium value of the exchange rate should be. Policy mistakes are likely to arise from ignorance, or from misguided political motivations.
(2) Fixed exchange rates are vulnerable to crises. As noted in Section 19.4.4, only those countries with impeccable credentials (no vulnerability, a highly credible central bank) may consider themselves immune to speculative attacks. All the others may be subject to a crisis, with devastating
consequences.
The list of cases of exchange rate mismanagement is impressive. It starts with Britain’s painful return to an obviously overvalued pre-First World War gold parity in 1925, to the dollar overvaluation that preceded the collapse of the Bretton Woods system, to
numerous cases where thriving black markets indicate that the official parity is off the mark. More recent cases include the decisions of Italy and the UK to leave the European fixed exchange rate mechanism (ERM), the Asian crisis, and several crises in
Latin America.
The case for fixed exchange rates
This is really a case against flexible rates, and against the view that markets do a better job than the authorities. The case is built on the observation that flexible exchange rates tend to fluctuate widely, too widely to be explained by inflation differentials or real
disturbances. Two explanations are usually offered, which are not mutually exclusive:
(1) Overshooting implies that exchange rates tend to move far from their equilibrium level.
(2) Exchange markets deal with considerable uncertainty with large payoffs when betting right, and large losses when wrong. This leads markets to move in fits and starts, imparting additional uncertainty and instability to the economy.
For good or bad reasons, most European countries have demonstrated a keen attachment to exchange rate stability since 1945. The fear has always been that exchange rate volatility would hurt intra-European trade and threaten the Common Market. The decision
to create a monetary union may be seen as the last step in continuous efforts at keeping intra-European exchange rates stable.
19.5.2 The New Debate: Financial Liberalization

Capital controls: The pros and the cons
The process of financial integration, which began in the 1980s and accelerated in the 1990s, has become controversial, if only because it has been linked to the wave of currency crises that followed. One after another, developed and then developing countries
have dismantled capital controls which were put in place at the end of the Second World War, and sometimes long before. The restrictions can take a variety of forms: (1) outright prohibitions of export or import of money and other financial instruments, (2) limits
on such transfers, (3) dual exchange markets (one fixed, for commercial transactions, one flexible for financial transactions), and, more recently, (4) the Tobin tax which is described in Box 19.5. While capital controls are highly controversial, the principles involved
are quite straightforward.
Critics of capital controls argue that restrictions to the free movement of capital prevent savers from getting the best available returns, and prevent firms from borrowing on the best possible terms. Saving and investment both suffer, with adverse effects on
long-term growth. They further observe that, by isolating its domestic financial markets, a government can ‘milk’ them to finance its own budget deficits at costs lower than the international capital market would offer. Furthermore, when capital controls are
effective, national interest rates may cease to reflect the local economic situation. Since high and rising interest rates signal a worsening situation, they tend to discipline imprudent governments. Shutting down the signal offers relief to governments, but at the
cost of a worsening situation in the future, possibly leading to a crisis on the way.
Those in favour of capital controls present three main arguments. First, following on an argument initially spelled out by Keynes, they claim that financial markets are unstable, prone to fads and panics. The result is volatility that is unjustified by underlying
economic fundamentals, and is costly to firms and households.
Second, following on the Mundell–Fleming result that full capital mobility prevents the use of monetary policy under fixed exchange rates, and makes fiscal policy impotent under flexible rates, they argue that restricting capital mobility restores the option of
using demand management instruments. When applied to monetary policy, this principle is the impossible trinity already discussed in Section 19.4.1.
Third, saving is a source of growth if it is invested in productive uses such as plants, machinery, schooling, and training, etc. The proponents of capital controls note that the bulk of international capital movements are of a very short-term nature, even aimed at
intra-day trading opportunities, rather than long-term investment in physical or human capital.
Box 19.5 The Tobin Tax (a.k.a. Financial Transactions Tax)
Back in 1972, James Tobin, a Yale economist and Nobel Prize laureate, proposed to ‘throw sand in the wheels of international finance’. He identified two main objectives, reducing exchange rate volatility and preserving the autonomy of macroeconomic
policy (see the main text for the argument), and concluded that restraining capital movements was a worthy effort. The proposal was not well received. As Tobin recalls, ‘It did not make much of a ripple. In fact, one might say that it sank … I realize that I
am opposed by a powerful tide. A wide-spread orthodoxy holds that financial markets know best, that the discipline they exert on central banks and governments is salubrious’ (Tobin 1996).
Tobin claimed that capital movements ought to be slowed down. He was certainly not in favour of the market-unfriendly, administrative restrictions in use at the time. His starting point was the observation that the vast majority of foreign transactions
involve round trips of seven days or less, speculative money which serves no productive investment or saving purpose, but rather to enforce the ‘no-profit condition’. He argued that a single flat tax on every foreign exchange market transaction of very small
magnitude would deter the unproductive short-term trips without much affecting long-term capital movements which underlie productive foreign investment. This is because the longer the holding period, the less relevant the transaction tax becomes.
Investors compute profits and losses from any transaction in annualized terms. For example, a small tax of 0.1% per transaction implies a total cost of 0.2% for a round trip (invest, earn your profit, and bring it back). If the round trip takes a year, it means
an annualized cost of 0.2%. If the round trip is quicker, say half a year, the annualized value of the tax approximately doubles (the law of compounding applies, so it is a bit more but negligibly so). As the horizon shortens, the annualized tax becomes very
high, e.g. 10.9% on a week-long trip. As Table 19.4 shows, for very long-term investments the tax becomes negligible.
Recently, the Tobin tax has undergone a revival among both economists and opponents to globalization. One of the motives is the tax revenue that its proponents expect, by some calculations, $300 billion annually with a 0.1% tax. Being an international
tax, it could be used for international purposes: the UN and its agencies, relief funds, NGOs, etc. For the same reasons that the Laffer curve exists, there is widespread scepticism that a Tobin tax would yield such revenues, because the markets are likely
to react, first by reorganizing themselves to minimize transaction volumes, and second by migrating to safe havens—at a pinch, a ship in international waters, stuffed with trading terminals and computer servers.
Table 19.4 Impact of a 0.1% Tobin Tax and the Holding Period of Investments
Holding period of investment 1 day 2 days 1 week 1 month 6 months 1 year 5 years
Implicit tax (annualized basis) 55.2% 24.6% 10.9% 2.4% 0.4% 0.2% 0.04%
The link with exchange rate regimes

The macroeconomic policy autonomy argument has a direct bearing on the debate on the choice of an exchange rate regime. The main weakness of fixed exchange rate regimes under full capital mobility is that in most cases they require the abandonment of an
independent monetary policy. It requires some discipline for central bankers to give up any hope of influencing local monetary conditions, and few central banks have lived up to the requirement. As a result, fixed exchange rate regimes are prone to speculative
crises and have shown rather limited survival ability.
The argument can be turned on its head, however. A direct implication of the impossible trinity principle is that fixed exchange rate regimes are more likely to survive when capital controls are in place. Thus the choice is not just between fixed and flexible
exchange rates, but also involves the capital mobility regime. Countries which value exchange rate stability should not rule out restrictions on capital movements altogether. Similarly, countries which favour full capital mobility should not stick to a fixed exchange
rate regime for too long.
However, if full capital mobility is considered an inexorable evolution of economic relations, the fixed exchange rate option may be going the way of the hula hoop and bell-bottomed pants. The impossible trinity means that monetary policy autonomy must be
sacrificed, but few countries are prepared for such a step that may be difficult to defend on the domestic political front. Furthermore, the possibility of self-fulfilling crises suggests that it may take years, possibly decades, before the central bank has achieved a
level of credibility which eliminates vulnerabilities. In the meantime, the threat of currency crises looms large. The hollowing-out hypothesis maintains that the choice of exchange rate regimes is no longer between floating rates and soft pegs—the traditional
fixed-and-adjustable exchange rates—but between floating rates and hard pegs (regimes described in the next section). Critics of the hollowing-out hypothesis argue that soft pegs are still feasible, provided bands of fluctuations are large enough or the peg is
allowed to vary (e.g. crawling pegs) to account for changing economic conditions. Box 19.6 describes how the EU countries have dealt with this choice.
Optimal sequencing
Back in the early 1950s, nearly all countries operated in strictly controlled environments. Many prices were fixed, some goods were even rationed, commercial banks were heavily regulated, and financial markets were limited or non-existent. External controls
regulated exports and imports, tariffs were heavy, and capital flows were essentially forbidden. What a difference a half-century can make! Liberalization seems an inescapable trend, and basic economic principles support this trend. On the other hand, liberalization
has not always been an easy journey. The Asian crisis once more showed that there can be serious setbacks on the way.
Is there a better way of liberalizing? The response is to adopt a proper sequencing of liberalization. McKinnon 21 has proposed an optimal order of sequencing. It is based on two main ideas. First, liberalization should start with the restrictions that are
costliest in terms of economic efficiency. Second, some steps need to be taken before others to avoid inconsistencies. The first step should be the creation of well-functioning domestic goods markets: free prices and abolish rationing. It makes little sense to have
free external trade if domestic trade is heavily constrained. The second step should be the gradual liberalization of international trade, starting with administrative measures (quotas on exports and imports), moving on to eliminating export tariffs, and then finally
reducing import tariffs to avoid shocks to domestic producers.
Soon after the first step, the domestic financial sector should be liberalized, under competitive conditions. Banks should be allowed to set interest rates freely on deposits and loans, to open branches as they see fit, and to choose the range of services that they
offer their customers. Bank regulation and supervision should be developed in parallel to ensure the soundness of the banking system.
Domestic financial markets come next. Bonds and stock markets are allowed to compete with the banking system to both collect savings and finance borrowing by firms and public entities. Here again, regulation and supervision must proceed in parallel to
guarantee a proper functioning of naturally unstable markets. Finally, external financial liberalization follows. The Asian crisis well illustrates the risks of full capital mobility when the domestic financial sector is not functioning properly. Furthermore, the exchange
regime must be adapt-
ed to changing circumstances, as rigidly fixed exchange rates are unlikely to survive capital mobility when full capital mobility is established.
19.5.3 Currency Boards and Dollarization

The menu of exchange rate regimes is not just binary. There are many ways of fixing or of floating. Three varieties of hard pegs have been observed which are worth noting: (1) dollarization and euroization, and (2) currency boards.
Box 19.6 Exchange Rate Regimes in EU Member Countries
With tightly integrated goods markets, the EU countries inevitably care about competitiveness and therefore. their exchange rates. At the same time, they have also forfeited capital controls in any form. They are therefore subjected to the stark choice
predicted by the hollowing-out hypothesis or the impossible trinity: either let the exchange rate float freely or rigidly fix it. The monetary union offers one solution, but it also removes any pretence of monetary policy independence. Exchange rate fixity is the
choice of the 19 Eurozone countries, as well of Denmark, which has pegged its exchange rate to the euro within narrow (± 2.5%) margins of fluctuations and Bulgaria which has adopted an even more rigid link, the currency board arrangement explained
below. The remaining seven countries have chosen monetary policy independence and have left their exchange rates to float freely, although their central banks are known to intervene whenever they observe too much volatility in the exchange rate
markets (see Table 19.5).
Table 19.5 Exchange Rate Regimes in European Union Countries, 2016
Eurozone members Fixed exchange rate Floating exchange rate

Joined EU Adopted euro Joined EU Joined EU
Austria 1995 1999 Bulgaria 2007 Croatia 2013
Belgium 1957 1999 Denmark 1973 Czech Republic 2004
Cyprus 2004 2008 Hungary 2004
Estonia 2004 2011 Poland 2004
Finland 1995 1999 Romania 2007
France 1957 1999 Sweden 1995
Germany 1957 1999 United Kingdom 1973
Greece 1981 2001
Ireland 1973 1999
Italy 1957 1999
Latvia 2004 2014
Lithuania 2004 2015
Luxembourg 1957 1999
Malta 2004 2008
Netherlands 1957 1999
Portugal 1986 1999
Slovakia 2004 2009
Slovenia 2004 2007
Spain 1986 1999
Source: IMF.
Dollarization and euroization

Dollarization is the unilateral adoption by a country of the US dollar as sole legal tender, which can be thought of as a one-sided monetary union with the USA. Similarly, euroization involves adopting the euro. It is as close to the gold standard as a monetary
system can be, without having gold itself circulate. It functions in the same way, including Hume’s mechanism. It can be seen as a variety of hard pegs.
A number of countries never had their own currency. Panama and Liberia have had dollarized economies since their independence. Ecuador and El Salvador adopted the dollar in 2000 and 2001, respectively. Argentina flirted with the idea in 1999. In order to
bring its hyperinflation to a halt, Zimbabwe effectively adopted the dollar in 2009. Kosovo has euroized. One reason for dollarizing is the perception that a foreign central bank will do a better job at enforcing price stability than an indigenous one. Another reason
is a proven inability to come to grips with inflation, as in the case of Ecuador. If trade links with the country whose currency is adopted are intensive, it seems like a good idea. It remains, however, that the domestic interest rate is driven by completely foreign
economic conditions, which may be awkward.
Currency boards
Currency boards used to be the arrangement of choice in the British Empire. They have made a comeback, starting with Hong Kong in 1983, followed by Argentina in 1991, Bulgaria and Bosnia-Herzegovina in 1997, and Estonia in 1992, and Lithuania in 1994 until
they joined the Eurozone. Currency boards resemble dollarization, except that the local currency is maintained. There are three key features of a currency board:
(1) A fixed exchange rate is established vis-à-vis an anchor currency. The local currency is fully convertible in unlimited amounts into the anchor currency at that rate.
(2) The local currency is fully backed by reserves. This is required to ensure full and unlimited convertibility.
(3) Currency boards often hold reserves of 105% or 110% of their liabilities, a precaution since most money is produced by commercial banks which are not restricted to 100% backing. In practice, it means that the high-powered money supply is entirely driven by the balance of payments
via Hume’s mechanism. Monetary authorities are completely passive.
With the exception of Hong Kong, currency boards have usually been adopted by countries that have long suffered high inflation and felt that there was no political will to establish a full-blown independent central bank dedicated to price stability. One transition
country, Estonia, started off with a currency board, and its success at avoiding inflation has inspired Lithuania and Bulgaria.
It might seem strange for independent countries to give up control over their monetary policy. Yet it is the logical consequence of the impossible trinity: with full capital mobility, fixed exchange rates imply the loss of monetary policy autonomy. Even with a
currency board, however, the threat of crises remains. After 10 years of using a dollar-based currency board, Argentina gave up and let its currency float, with painful consequences. The only way to eliminate that threat is to eliminate the currencies themselves.
Since there is no real policy autonomy to lose, the system can only be strengthened. While this speaks for a monetary union, giving up monetary or exchange rate policy completely has its own consequences and costs.
19.5.4 Monetary Unions

A monetary union involves the irrevocable fixing of exchange rates and the abandonment of margins of fluctuation among a number of countries. In fact, it means that individual currencies are no longer distinguishable; a common currency is used. The immediate
implication is that individual central banks lose any remaining autonomy, although one central bank is needed to manage the common currency. This is a special case of the N−1 problem spelled out in Box 19.7. The union’s central bank manages the overall
money supply. Interest rates are the same across the union since money can flow freely. National money supplies are then determined via the Hume mechanism.22
In principle, if a country in the monetary union runs an overall balance of payments surplus, money should flow in, the national money supply rises, and interest rates fall. A deficit results in a loss of money, a rise in interest rates, and an associated contraction.
So goes the theory, at least. In the Eurozone crisis the management of the balance of payments became derailed. Box 19.8 explains how this happened, and that, like currency crises, the situation has two valid interpretations.
Box 19.7 The N–1 Problem

Take N countries with N currencies. There are N–1 independent bilateral exchange rates, as Figure 19.8 illustrates. All the other bilateral rates can be retrieved from these N–1 rates via triangular arbitrage. Now link these currencies together, either in a
monetary union or just a system of fixed exchange rates. As N–1 independent bilateral exchange rates are fixed, it follows that N–1 central banks lose their independence, and the Nth remains free of policy constraints. In the Bretton Woods system, all
central banks were pegged to the US dollar, leaving the Fed with the task of pegging the dollar to gold. In the EMS, the Nth country was not explicitly designated, but the Bundesbank captured the Nth degree of freedom. In a monetary union, such as the
Eurozone, all N central banks lose their policy-making autonomy and a new N+1th central bank is created to manage the new currency.
Fig. 19.8 The N–1 Problem

Two countries which decide to fix their exchange rate lose one degree of freedom. When N countries form a fixed exchange rate system, they commit to N–1 exchange rates, or N–1 degrees of freedom. All bilateral exchange rates can be calculated from just the N–1 rates shown in
Panel (b) by triangular arbitrage, allowing us to draw all the missing arrows.
There is a well-developed theory of optimum currency areas which spells out criteria for creating a monetary union. A region constitutes an optimum currency area when its use of a common currency implies no loss of welfare.23 The best way of thinking
about it is to ask what is lost when the exchange rate instrument is abandoned. Problems arise if the union is hit by asymmetric shocks, i.e. shocks which affect some members but not others. It is precisely under these conditions—and when wages and prices are
sticky—that an adjustable exchange rate can be useful. Without it, adjustment to asymmetric shocks must take place through prices, a process which can be very painful.
The theory of optimal currency areas looks for criteria which make such adjustment unnecessary. Robert Mundell proposed two main criteria:
(1) Factor mobility. Consider the case when an adverse shock, e.g. a loss of competitiveness, hits a country. The country soon goes into a recession, and laments the loss of its exchange rate, which could have been handy to restore its external competitiveness. If, however, its factors of
production could move out to more fortunate members of the monetary union, the pain would be spread out, and the common central bank could use the common external exchange rate to adjust optimally. If capital or labour, or both, are mobile, there would be little cost in unemployment
of factors, and the loss of internal (within the union) exchange rates would be trivial.
(2) Absence of asymmetric shocks. Most of the shocks concern the world demand for and supply (competitiveness) of locally produced goods. If the members of the union produce a similar menu of goods, then the likelihood of asymmetric shocks diminishes. The same applies if the
member countries produce a very diversified menu of goods. In that case a particular shock is likely to be of little import.
Practical examples abound which call for a common currency. No one would doubt that London should use the same money as the rest of England, or that Paris should use the same currency as the rest of France. Large, open cities clearly meet the conditions for a
currency area. First, national capitals are the epicentres of labour and capital in a country, and mobility—especially of young people—into cities is very high. Second, the shocks, both positive and negative, which hit these capitals certainly have strong
correlation with the rest of the country. At the same time, the sceptic might ask: what about Scotland or Northern Ireland in the UK, or Catalonia in Spain, or the new federal states of Germany? Here a single currency has been used, sometimes begrudgingly, for
some time; but this was not always the case. Obviously economic actors can be conditioned to make do with a common currency, with behaviour and economic structures adapting accordingly. The symmetry of shocks between Belgium, the Netherlands, and
Germany has increased significantly since the 1960s, so that these countries are as fit for a monetary union now as the German states were in the 1960s. Sometimes politics are more important than economics. As Box 19.9 shows, this was true for the first,
unsuccessful European monetary union—under Charlemagne in the ninth century—as it is for the present one.
Box 19.8 The TARGET-2 Controversy
As explained in Chapter 10, many interbank markets ceased to function when the financial crisis hit in 2007. Banks simply stopped lending to each other. It happened in the Eurozone as well, and the problem became particularly severe after the
sovereign debt crisis in 2010. This was a problem, because the Hume mechanism requires deficit countries to send more funds abroad through commercial bank transfers than they receive, and this would indeed lead to a reduction of bank deposits,
i.e. broad money, reducing demand in the deficit country. But what happens if banks are not dealing with each other?
Here is a chance to learn about the plumbing of the Eurozone, called TARGET 2.24 In normal times, this automated system records and settles payments between commercial banks in different Eurozone countries, debiting the paying bank’s account
with its own central bank and crediting the payee’s account. This is how money circulates across national borders. Once panic struck and commercial banks stopped lending to each other, paying banks were allowed to borrow from their own national
central banks and payees in the receiving country received a credit with their own central bank (i.e. additional bank reserves). But that meant that a deficit country’s central bank would run up a debit vis-à-vis TARGET 2, with a corresponding credit for the
central bank in the surplus country. In that way, the system of European central banks stepped in to keep money flowing, using central bank credit to fund current accounts imbalances which had been financed by private lending (normal bank credit)
before the crisis. As Figure 19.9 shows, surplus countries like Germany quickly became net creditors to TARGET 2, while central banks of Greece, Spain, Ireland, and Italy became net debtors. As the crisis intensified and capital flight increased, TARGET-
2 balances reflected the cumulation of financial as well as current account deficits.
Just as with currency crises, there are two views of the TARGET-2 imbalances. One is that balance of payments deficits are fundamental and reflect long-term real exchange rate misalignments—imbalances that the IMF would have policed and
corrected in the old Bretton Woods system. But the Eurozone had no such mechanism. The other interpretation is that a crisis of confidence among banks, initiated by financial crisis, exposed vulnerabilities in deficit countries that had long been swept
under the rug, and these led to a self-fulfilling prophecy, as bankers rushed for the exit. That the imbalances in Figure 19.9 erupted so suddenly around the time of the financial crisis despite long-standing chronic deficits in the past is evidence for the
latter hypothesis.
What would happen if the indebted countries left the Eurozone and defaulted? It would not be pretty, but losses would be shared by all the central banks, not just the central banks of the surplus countries. Second, these being debts and credits among
central banks, potential losses could always be paid for by new money creation. This would leave no one harmed and would not be inflationary, since newly created money would just replace money that had been destroyed when the debtor country left
the Eurozone.
Fig. 19.9 TARGET-2 Imbalances; 2001–2016 (€ bn)
Credit and debit positions within TARGET 2 grew as a result of the Eurozone crisis. These balances replaced actual money flows because the interbank market stopped functioning and private capital ceased to flow to deficit countries.
19.5.5 The European Monetary Union

We conclude this chapter with a quick sketch of the European Monetary Union or Eurozone. This unique arrangement, in which sovereign states share a common central bank and a common money, has existed since 1999; euro coins and banknotes have
circulated in the ‘euro area’ since 2002. The Eurozone is an increasingly important component of the global financial architecture. Starting with 11 members, it had grown by 2016 to 19 countries with a population of 340 million—more inhabitants than the USA. The
role of the euro in international payments is well-established and will probably grow over time.
This is because, at the outset, Europe didn’t seem to meet the necessary conditions for a monetary union put forth by Mundell.25 By most accounts, Europe is still far from achieving the status of an integrated single economic region such as the states of the
USA, or the provinces of Canada, or Mexico, or Brazil. The recent Euro-crisis, which is primarily a fiscal crisis with first-order ramifications for monetary and financial stability, has sharpened this critique. The European Monetary Union faces significant challenges
to its existence in the years to come. Box 19.10 summarizes some of the most salient of these challenges, including those that can only be settled in the political arena.
Yet ever since the founding of the Economic Community in 1957, Europeans have believed that increasing integration was the secret to securing lasting peace and preventing future armed conflict. Ironically, Europe has a chronic history of wars over the past
two millennia, many of which coincided with or were followed by monetary unions. In particular, monetary union frequently coincided with the rise of new nation-states. Box 19.9 gave one example of an early attempt to institute a monetary union in Europe. It
failed, yet it shared many of the motives of the euro—to create the fundamental conditions for greater peace and prosperity. It highlights how economic, political, and social progress in Europe was rough sailing at the outset. In times of growing globalization and
international turbulence, the ‘European project’ may finally reach smoother waters as the continent comes to political and social terms with itself.
Box 19.9 Charlemagne’s European Monetary Union
One of the most interesting attempts to use monetary union as an instrument of political union was made by the first Carolingian king, Charlemagne. In his quest to unify Europe, Charlemagne was not loathe to employ the sword, but understood the
importance of a single, harmonized currency for the promotion of commerce and trade. At the Council of Frankfurt in 794—which was ostensibly about religious matters—a supposed minor detail was also on the table. Charlemagne, who was fascinated
by economic affairs, saw the advantage of a common standard for coinage in the conquered territories of Spain and Southern and Central Europe. Even if the conversion of precious metal into coins was decentralized, it would be advantageous to use a
single definition and metal content for those coins. Charlemagne seemed to understand that the gold standard could have deflationary consequences; because it was so rare, trade using gold was impractical for all but the most valuable transactions.
So the gold sou was replaced by the silver livre carolinienne (which eventually became the pound!). It was worth 240 denarii, possibly inspired by the success of the Islamic dinar. The new common currency, with common minting standards, enlarged the
trading area and reduced transaction costs. It was also legal tender:
Everywhere, in every city and every trading place, the new denarii are also to be legal tender and to be accepted by everybody. And if they bear the monogram of our name and are of pure silver and full weight, should anyone reject, in any place, in any
transaction of purchase or sale, he is to pay 15 soldi [roughly the price of a cart load of wheat]. (Article 5, Protocol of the Council of Frankfurt)
After Charlemagne died, his realm was partitioned among his sons and the political union of Europe soon dissolved into the Middle Ages. Obviously, monetary union was not a sufficient condition for the political unification of Europe, but it did sustain
the economy of the Carolingian empire for a number of centuries. Later, the Edict of Pitres in 864 (during the reign of Charles the Bald) would tighten the grip on seigniorage rights, significantly restricting the number of authorized coinage sites. Perhaps
the wisdom of Charlemagne was undermined by the greed of his grandson?
Box 19.10 The Euro-Crisis: Fundamental or Non-Fundamental?
The European Monetary Union crisis started in late 2009 with a sharp rise in yields on government debt of Greece, spreading to other southern European countries, Portugal and Italy, that were seen at the time as fiscally sound. It also spread to Ireland
and Spain, two countries with solid public finances that were hit by the bursting of private real estate bubbles. Facing severe difficulties in the banking sector, the Irish and Spanish governments were led to recapitalize the banks and to guarantee
depositors, which opened up gaping holes in their budgets.
One interpretation of the Euro-crisis—a crisis of banks, government deficits, and public debt—is that it is fundamental. Nations of the Eurozone are sovereign, and the Stability and Growth Pact, designed to impose fiscal discipline, obviously failed.
Furthermore, the Maastricht Treaty did not explicitly provide for intervention at the EU level in case governments find themselves unable to borrow on financial markets at a reasonable interest rate. In addition, since the advent of monetary union in 1999
and until the crisis, some countries (Greece, Portugal, Ireland) have run large persistent current account deficits while others (Germany, Netherlands, Finland) ran surpluses. In principle, these imbalances should be self-correcting, in line with the Hume
mechanism described in Section 19.2. Figure 19.10 shows the extent to which prices and labour costs—especially for non-traded goods—have deviated across countries since the beginning of the monetary union. A crisis seemed unavoidable.
Another plausible interpretation of the crisis is that it is non-fundamental, like the speculative attacks described in Section 19.4.4. Such an attack requires a vulnerability; in this case it was long-lasting budget deficits and quickly mounting public debts. If
the financial markets conclude that several Eurozone countries lack the political will to stabilize their debt–GDP ratios, they come to expect defaults and require higher bond yields. These higher interest rate costs, in turn, require higher primary surpluses
to fulfil the conditions for solvency, which raises the probability of defaults and calls for further increases in bond yields. The markets began to ask whether financially healthier countries of Europe would provide money needed to avoid defaults. This
made the crisis a collective one. According to this view, in order to convince the markets that the non-fundamental expectations are simply wrong, a sufficiently strong backstop is needed—either a greatly expanded European Financial Stability Fund, or a
European Monetary Fund, acting just like the International Monetary Fund in the Bretton Woods era.
Fig. 19.10 Cumulative Price Level and Unit Labour Cost Changes Relative to Eurozone-12 Average, 1999–2010
This figure cumulates changes in the price level (Panel (a)) and unit labour costs, the ratio of wages to nominal productivity (Panel (b)) for 12 Eurozone countries that were in the monetary union from the beginning. For each country, pink bars represent the contribution of tradable goods
sectors to the total change in unit labour costs; orange bars show the contribution of non-tradable sectors. The sum of the two bars represents the total deviation from the average development in the Eurozone-12 group over the period. Evidently, the lion’s share of the misalignment
originates in non-traded output.
Source: Brede (2011).
Summary
1 International monetary arrangements initially arose from the need to provide international trade with an easy means of settling cross-border payments. For centuries, both domestic and international trade was carried out using gold and silver. The gold standard in its pure form lasted less
than 40 years, from 1879 to 1914.
2 Taken literally, the gold standard implied a rigid monetary rule and a fixed exchange rate regime. By the Hume mechanism, a trade deficit caused a shrinking money supply, while a surplus meant an expanding money supply. Both act to equilibrate trade imbalances.
3 The evolution of the monetary system after the First World War can be seen as a series of ad hoc responses to international crises and system inadequacies. In particular, the Bretton Woods system was designed to avoid the competitive devaluations of the inter-war period by
establishing a system of fixed exchange rates based on the US dollar’s link to gold. This ‘gold exchange standard’ was not a gold standard in the strict sense.
4 The collapse of the Bretton Woods system was due to the internal inconsistencies of a system that required increasing amounts of international reserves to be provided by the USA, which were, in theory, convertible into gold. Large US balance of payment deficits in the late 1960s
created an overhang of official external dollar liabilities which far exceeded the USA’s gold assets.
5 The IMF fulfils two main roles: (1) it exercises surveillance over member countries on a routine basis, and (2) it provides emergency assistance to countries which face balance of payments difficulties. Its loans are conditional on the adoption and implementation of programmes designed
to cope with the source of payment imbalances.
6 Currency crises can be divided into two types. (1) First-generation crises occur when domestic policies are incompatible with the exchange rate peg. They are usually anticipated. (2) Second generation crises are self-fulfilling. They afflict central banks which appear vulnerable or
uncommitted to an exchange rate target, or which have not acquired sufficient credibility.
7 Currency crises seem to be contagious for three reasons: (1) first-generation contagion through loss of competitiveness; (2) second-generation contagion when markets discover similar vulnerabilities or lack of central bank commitment; and (3) investors’ contagion when losses in one
country prompt international players to withdraw from other countries which suddenly appear risky.
8 Capital liberalization brings about long-term benefits. But financial markets are prone to bouts of instability which may result in currency crises. In addition, full capital mobility severely restricts the ability to carry out macroeconomic policies (Mundell–Fleming).
9 The choice of an exchange rate regime involves various trade-offs. In the end, small, open economies may favour some degree of exchange rate stability, while larger countries may prefer to integrate themselves in the world economy at fluctuating real exchange rates.
10 The widespread shift to capital account liberalization has had the effect of sharpening the choice between floating and fixed exchange rates. Soft pegs are increasingly seen as dangerous, hence the fashion for hard pegs: monetary unions, currency boards, and dollarization.
11 The European Monetary Union faces challenges which have to do not only with fiscal and current account imbalances, but also with its very architecture. It presumes that the member countries in fact meet the criteria for a viable currency area. Like the Bretton Woods system before it,
imbalances can develop that can be eliminated only via the Hume mechanism, or the intervention of a central authority like the IMF. In Europe, this central authority has yet to be established.
Key Concepts
International Monetary Fund (IMF)

gold standard
parity
band of fluctuation
bimetallism
Gresham’s law
Hume mechanism
balance of payments
Bretton Woods Conference
gold exchange standard
quota
conditionality
Triffin paradox
special drawing right (SDR)
impossible trinity
boom-and-bust cycles
self-fulfilling attack
speculative attacks
vulnerability
exchange rate mechanism (ERM)
hollowing-out hypothesis
sequencing
N−1 problem
optimum currency areas
European monetary union (EMU)
Exercises
1 Why can’t beggar-thy-neighbour policies work? What is the difference with tariff wars?
2 Does the Triffin Paradox still apply to the US dollar?
3 China has been fighting hard to include its currency in the SDR. What difference will that make?
4 Does the Hume mechanism work within a monetary union?
5 ‘Currency crises cannot be foreseen, markets attack as soon as they expect a crisis.’ Comment. In your answer, you may distinguish between first-generation and self-fulfilling crises.
6 Draw the implications of the N−1 problem for the international monetary system and for regional exchange rate arrangements.
7 Can self-fulfilling crises also affect the banking system?
8 ‘The developing countries cannot borrow in their own currencies. This puts them at permanent risk of a currency crisis.’ Explain and comment.
9 The Eurozone countries cannot borrow in their own currencies. This puts them at permanent risk of a financial crisis.’ Explain and comment.
10 What is the difference between a currency board and monetary union membership?
Essay Questions
1 The poorer countries regularly propose that the IMF issue more SDRs and that they be distributed mostly to the poorer countries to support their development. The richer countries refuse, contending that this would be inflationary. Discuss.
2 It has been said that the Bretton Woods system was doomed from its start. Explain why. Could its demise have been avoided? How?
3 Is a return to the gold standard feasible? Desirable?
4 The choice of an exchange rate regime is difficult. The European Monetary Union combines fixed exchange rates among its members with a floating exchange rate vis-à-vis the rest of the world. Explain how this mixture of exchange rates regimes works. Is it a good or a bad idea?
5 Quotas and voting rights at the IMF were changed in April 2008. The decision is presented in http://www.imf.org. How would you evaluate this change?
1 Living in the sixteenth century, Sir Thomas Gresham had been in charge of royal finances, then became a foreign exchange trader in Antwerp until he created the Royal Exchange, better known today as the London Stock Exchange. It is sometimes argued that the gold standard in the UK
was an artefact of Isaac Newton’s decision in 1717 to undervalue silver in terms of gold. Within little time, Sir Isaac had only gold on his hands.
2 Considering the high shipping, insurance, and security costs involved, this was a very expensive way of settling international payments. To save on these costs, some central banks, such as the Federal Reserve Bank of New York, offered (and still offer) custodial services for gold owned by other
central banks. In the late 1990s, the New York Fed was the guardian of about 270 million troy ounces of monetary gold, or about one third the holdings of all central banks. At current market prices this gold is worth between 400 and 500 billion dollars.
3 In the time of the gold standard, there was no monetary authority (central bank) to set interest rates in some regular way like the Taylor rule. The money supply simply interacted with money demand to produce market-clearing interest rates, just as described in Chapter 9.
4 During the 40 years preceding the First World War, some 20% of British savings were invested abroad.
5 Higher gold prices will also discourage its use as jewellery or increase the use of substitutes in industry, thus freeing up some extra gold for monetary purposes. These effects are relatively modest, however.
6 In some cases, e.g. Britain, convertibility was possible only for large denominations, since the Bank of England restricted its conversion to bullion (as opposed to coins). The system is sometimes referred to as the ‘gold bullion standard’.
7 A currency is convertible when holders, both private and official, may exchange it without restriction. Convertibility does not necessarily imply a fixed exchange rate, since a floating rate system also allows participants to freely purchase and sell foreign exchange.
8 Remember: Y = C + I + G + X − Z: ceteris paribus, reducing Z raises Y, but if X falls by the same amount, there is no net gain.
9 Named after a small ski resort in the US state of New Hampshire. The conference considered two plans published in 1943, prepared for the USA by Treasury Secretary Harry White, and for the UK by John Maynard Keynes. The White plan eventually prevailed.
10 The Marshall Plan was a massive aid programme for post-war Western Europe and Japan funded by the USA.
11 This was the first sign that the Hume mechanism might run into trouble: the USA provided the ‘gold’ (dollars) by running balance of payments deficits with the rest of the world, that is, by exporting dollars in return for goods, services, and investment positions.
12 It is named after the Belgian economist Robert Triffin (1911–1993) who identified the ‘fundamental flaw’ of the Bretton Woods system.
13 The French franc was devalued in 1958; the German mark and Dutch guilder were revalued in 1961.
14 The IMF offers a large menu of lending facilities tailored after particular needs. They can be reviewed at the IMF website <www.imf.org>.
15 The ‘Keynes Plan’ envisioned the creation of an international reserve currency, the bancor, which would play the role gold did in the old gold standard era, but which would be supplied by an international agency, i.e. not the USA.
16 Hence the following quote by the economist Fritz Machlup: ‘Now the forward-looking experts of the Fund and the negotiating governments have proved that their reputation for backwardness in economic thinking had been undeserved. All that matters for the acceptability of anything as a
medium of exchange is the expectation that others will accept it . . . Money needs takers, not backers’ (quoted by Dam, 1989: 152).
17 The currencies, and their weights in the basket, are the US dollar (41.73%), the euro (30.93%), the renminbi (10.92%), the yen (8.33%), and the pound (8.09%), reflecting their international use. The weights are revised every five years.
18 A long-standing gentlemen’s agreement that the IMF’s Managing Director is from Europe while the President of the World Bank is from the USA has recently been challenged and criticized by developing countries.
19 We ignore banking and debt crises, although the process is similar.
20 In a famous metaphor, a key proponent of flexible rates, Chicago economist and Nobel Prize laureate Milt on Friedman, noted that the shift to summer time can be achieved by having everyone adapt behaviour and do the same things an hour earlier, or by moving the clock ahead by one hour. The
latter is much easier, he argued, than changing the habits of millions of people. Changing the exchange rate is easier than changing millions of price s.
21 The late Stanford economist Ronald McKinnon (1935–2014) contributed widely to international monetary economics and the role of finance in development. He also contributed to the theory of optimum currency areas.
22 The mechanism described here is somewhat more general than Hume’s in the following sense: an expansion of the local money supply is not predicated on a trade surplus, but could also be achieved with a capital account surplus (an excess of private capital flows).
23 The seminal work on optimum currency areas is by Robert Mundell, the same economist who shaped the Mundell–Fleming framework presented in Chapter 12.
24 TARGET stands for Trans-European Automated Real-time Gross Settlement Express Transfer. The number 2 refers to the second version of the system.
25 The influential economist Martin Feldstein stated in a widely-publicized article in 1997 that European Monetary Union could even lead to civil unrest and war.
Epilogue
20
20.5 New Keynesian Macroeconomics: The Latest Synthesis
20.7 Labour Markets
20.10 Demographics, Low Productivity Growth, and Secular Stagnation
20.11 Conclusions
The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled
by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist.
J. M. Keynes
Throughout this book, we have emphasized the usefulness of macroeconomics as a tool for understanding and improving the way the world works. To this aim, we have presented
a unified treatment of the field, and have downplayed the historical evolution of ideas over the years as well as important controversies, past and present, which have accompanied
these ideas. Focusing on controversies can be fascinating, but it can also cloud the extent of agreement and common understanding, leaving the unsatisfactory impression that
macroeconomics is too conflict-ridden to be of any practical use. Once the basic framework is well understood, it is both interesting and illuminating to track the field’s intellectual
history. In this concluding chapter we present a highly compressed survey of the major steps of the field’s development, the key players, and the policy debates. Its emphasis is
on the European scene.1
John Maynard Keynes,

1883–1946
Source: Copyright Hulton-Deutsch Collection/Corbis.
The birth of macroeconomics is conventionally associated with the publication in 1936 of Keynes’ The General Theory of Employment, Interest and Money. Its
influence has been phenomenal. Some even claim that it changed more lives in the twentieth century than any other single work. Many factors explain this
success.
The book came out towards the end of the Great Depression and can be seen as the response of economic research to challenges of the time. To
contemporaries who had witnessed the rapid rise of mass unemployment, the classical laissez-faire view looked factually wrong and almost immoral. Others had
already moved in the same direction, but Keynes brought together many apparently disparate themes. The idea that prices do not necessarily clear the goods
market at full employment had been put forward by the British economist Robert Malthus (1766–1834). Knut Wicksell (1851–1926) and his successors, who came
to be known as the ‘Stockholm School’, had gone quite a long way towards what was to become the IS curve. Other economists—including Hjalmar Schacht,
who single-handedly vanquished the German hyperinflation in 1923 and later went on to become Hitler’s Finance Minister—had long advocated deficit spending
on public works during the Great Depression. But Keynes’ attack was unique in many ways. The General Theory proposed a wholly new concept of equilibrium,
even though it took decades to decipher fully what it really meant. It was a combination of scholarly analysis, strident criticism, and practical policy
recommendations that appealed to both theoreticians and policy-makers. Keynes did not just aim at fellow researchers. He frequently descended the ivory tower
of academia to promote his ideas in the media, where his reputation as a brilliant and provocative polemicist was well established.
The truth is that Keynes was a man of many talents and, by 1936, of great experience as well. As a young economist during the First World War, he had
worked in the UK Treasury, which he represented at the Versailles Peace Conference, until he resigned in a rather undiplomatic fashion. In The Economic
Consequences of the Peace, he criticized the harsh reparations imposed on Germany, maintaining that it would be destabilized by the economic burden of the
Treaty. His analysis turned out to be prophetic, and established his reputation among policy-makers. The book also strained Keynes’ relations with British
government circles until the Second World War. His strident criticism of Chancellor Winston Churchill’s decision to return the pound to its pre-war parity did not
help in this regard. Maynard, as his friends called him, was also a charismatic intellectual leader. He assembled a group of brilliant economists at Cambridge
University who went on to dominate the profession in Britain and beyond.
One important message of the Keynesian revolution was that fiscal policy can be used to fight recessions, in particular when monetary policy is ineffective—
either because expansionary monetary policy no longer lowers the nominal interest rate or when investment spending is depressed by bad ‘animal spirits’. Deficit
spending, as it was then called, was taken on board in many countries after the war, in effect becoming conventional wisdom. German-speaking countries too
have been influenced by Keynesian ideas, but scepticism there has been present and, to this day, a large segment of the policy-making and academic
establishments see them as dangerous. German reluctance towards Keynesianism is linked to the role that deficit finance had in the hyperinflation of 1922–1923.
This mistrust of Keynesian policies has found its way, in a subdued form, into the monetary union’s Stability and Growth Pact and into the statements of the
European Central Bank (ECB), in which fiscal deficits tend to be seen as a source of concern rather than as a potential means of output stabilization. In most of
Europe, though, Keynesian ideas are still alive, although their limits—described in Chapters 16 and 17—are generally well recognized.
Keynes’ theory was not fully worked out. The General Theory is difficult to read, frequently lacks precision, and can sometimes be downright confusing. It fell
upon Keynes’ disciples to dot the i’s. Most of the effort was conducted in his native Britain and in the USA, with some important contributions from other
countries, including that of the Polish economist Michal Kalecki (1899–1970), who had anticipated many of Keynes’ ideas and went on to try to merge Keynesian
and Marxist schools of thought.
Beyond clarifying Keynes’ views, one task that his disciples had to grapple with was to reconcile the Keynesian construction with generally accepted theories.
Indeed, it soon emerged, that the attack on the ‘classics’, as Keynes labelled established neoclassical economics, was not at all general and rested on the
assumption that the price level is constant. This assumption was acceptable in situations of low employment, like the Great Depression, but was seriously at
odds with post-war economic conditions, characterized by full employment and, later on, rising inflation. The necessary reconciliation effort, the neoclassical
synthesis introduced in Chapter 13 and 14, was carried out mostly in the USA, with Nobel Prize laureate Paul Samuelson (1915–2009) and his MIT colleagues at
the forefront, but also by Keynes’ colleague in Cambridge, John Hicks (1904–1989), and by Don Patinkin from Hebrew University in Jerusalem (1922–1995).
European macroeconomists were not particularly productive during this period, with a few notable exceptions. Some Swedish economists, under the leadership
of Assar Lindbeck (1930–), and their Norwegian colleagues developed a small open economy version of the Keynesian model. Nobel laureates Jan Tinbergen,
from the Netherlands, and James Meade (1907–1995), a student of Keynes’ in Cambridge, also made major contributions in extending Keynes’ framework to the
small open economy case. The most innovative construction in the Keynesian tradition was the Mundell–Fleming model presented in Chapter 12. Its architects
were Marcus Fleming (1911–1976), a British economist working in the International Monetary Fund, and Nobel laureate Robert Mundell (1932–), who was
undoubtedly inspired by the smallness and openness of his native Canada and of Switzerland, where he lived for a time.
An implication of Keynesian economics was that countries as a whole could be a research subject. Today, it is hard to believe that pre-Keynesian economics
was mostly preoccupied with sectors and firms, and had little to say about questions such as growth or employment. In fact, aggregate data, like GDP, the
unemployment rate, or the consumer price index, were sporadically collected and seldom the subject of great research interest. The rise of Keynesian economics
prompted a vigorous effort at developing the relevant concepts and assembling the data. This effort started in the late 1930s at a time when most of the leading
economists were either in the USA or in Great Britain. Unsurprisingly, therefore, the main contributions were developed in these two countries, with early
pioneers such as Simon Kuznets (1901–1985) from Columbia University and Richard Stone (1913–1991), a Keynes student from Cambridge, both of whom were
eventually awarded the Nobel Prize for their work. Once data were available, and with the advent of the first computers, economists undertook to build large-scale
models that were meant to mimic the economy. Following early work by Italian-born Nobel laureate Franco Modigliani, these large models have become standard
fare in most finance ministries, international organizations, and economic forecasting companies, where they are routinely used to produce forecasts and simulate
the effects of policy decisions. Despite the subsequent decline of Keynesian economics, these models continue to exert great influence on day-to-day decisions
made by governments, banks, and businesses.
The neoclassical synthesis shows that the Keynesian equilibrium is a special case, which applies when prices are sticky. Obviously, the next task was to
explain how prices move, when they eventually do. This led to a search for what was known as the ‘missing equation’. This equation was discovered as an
empirical regularity by A. W. Phillips (1914–1975) at the London School of Economics—his contribution is extensively discussed in Chapter 13. This discovery
prompted the next question: what is the theory behind the Phillips curve? Work on this question, mostly in the USA, was well under way just when the curve
started to vanish. The disappearance of the Phillips curve, correctly anticipated by Friedman in the late 1960s, paved the way for the rise of the monetarists, a
rival school of thought committed to exposing fundamental flaws in Keynesian economics.
Milton Friedman,
1912–2006
Source: Copyright Hulton-Deutsch Collection/Corbis.
By the late 1940s, the Keynesian school had established a strong foothold in the USA, where most of macroeconomic research was conducted, but it never
enjoyed total supremacy. The University of Chicago, in particular, remained the bastion of the classical economics that Keynes had sought to upend. Keynesian
ideas certainly attracted attention at Chicago. In the 1940s, Chicago economist Lloyd Metzler published an influential attempt to characterize Keynes’ ideas
formally. But the Chicago academic tradition must have seen a fundamental threat in Keynesian macroeconomics. It is thus not surprising that the ‘Chicago
School’ led an intellectual attack against the Keynesians. Eventually, it was as successful as Keynes’ own attack against the classics. Part of the success of the
Chicago School is due to Milton Friedman, whose many talents matched those of Keynes himself.
Friedman combined extraordinary intellectual vigour, leadership, charisma, government experience, and communication skills. Like Keynes, he spent the war
years at the Treasury, the US Treasury in his case, where he contributed to the war effort. Like Keynes, he assembled a group of young economists, who
regularly met in the ‘Workshop in Money and Banking’ and went on to rewrite macroeconomics. Like Keynes, he devoted much time and effort to popularizing
his ideas, writing a regular column in the US magazine Newsweek and becoming a popular guest on television shows. And, like Keynes, he did not shy away from
contact with politicians, providing advice to unsuccessful presidential candidate Barry Goldwater, as well as to the considerably more successful President
Ronald Reagan and to Prime Minister Margaret Thatcher. Many of his Chicago associates became known as the ‘Chicago boys’. They achieved considerable—
and still controversial—influence in South America and elsewhere.
Friedman pursued several ideas, all of which undermined the key building blocks of Keynesian economics. First, he was an unabashed defender of free
markets, which Keynes saw as chronically prone to failures. This led him to actively promote the view, long advocated by the Austrian-born economist Friedrich
von Hayek (1899–1992), that governments are a threat to freedom, and not just in economic matters.2 Friedman and his colleagues resuscitated the influence of
the laissez-faire school, which had been shattered after the Great Depression.
Second, Friedman confronted Keynes’ view that fiscal policy is a useful tool for macroeconomic stabilization and that monetary policy is useless. The label
‘monetarist’, widely applied to the Chicago School, comes from this aspect of Friedman’s work. 3 His A Monetary History of the United States, 1867–1960, written
in 1963 jointly with Anna J. Schwartz (1915–2012), is generally regarded as a masterpiece that fundamentally changed the way we look at monetary policy. At the
empirical level, this book attributes the Great Depression to bad monetary policy, in contrast with Keynes, who tended to blame ill-timed fiscal policies. At the
theoretical level, the book re-established the classic ‘quantity equation’ MV = PY, where V is the velocity of money. This equation, which was dismissed by the
LM equation, brings home the neutrality of money: if velocity V and Y are taken as exogenous, the price level P is directly driven by money. In the classical view,
Y is at full employment and V is constant, whereas in the Keynesian view Y is highly variable, P is constant, and V depends on the interest rate.4 Monetary
neutrality, an old wisdom of classical economics discarded during the Keynesian heyday, returned and has not left us ever since. Its implications are profound
and lie at the core of the theory and practice of central banking, as explained in Chapter 10.
Third, in a careful study of consumption patterns in the USA, A Theory of the Consumption Function, a book published in 1956, Friedman argued that the
Keynesian function C = C(Y) had little theoretical foundation and questionable empirical validity. Instead, he put forward the permanent income hypothesis,
which relates consumption to permanent income, or wealth. This effectively reinvented the intertemporal analysis presented in Chapters 6 and 8, and previously
explored by US economist Irving Fisher (1867–1947) of Yale Universiy. The important consequence of this work was to weaken the significance of the Keynesian
multiplier and the view that fiscal policy can be a tool for output stabilization. Later on, the Keynesians restored some of the clout of the old consumption
function by arguing that many consumers are credit rationed, as explained in Chapter 8.
Finally, in what may have been his greatest triumph, Friedman explained why the Phillips curve, then still considered as the missing equation linking the short
and long run, would vanish as soon as the authorities attempted to exploit the output–inflation trade-off. Not only did he restore the importance of expectations
—and thus established the expectations-augmented Phillips curve—but he restated the long-run neutrality proposition—and thus invented the long-run vertical
aggregate supply schedule. His work was published in 1968, and the Phillips curve went awry thereafter, in the early 1970s. Not only were the Keynesians proven
wrong, they were once more ‘missing an equation’. Monetarism became the new accepted wisdom, in academic circles first, and then among policy-makers. It is
important to note, however, that Nobel laureate Edmund Phelps (1933–), from New York’s Columbia University, who had reached the same result as Friedman and
at roughly the same time, regarded himself as a Keynesian. Phelps essentially foreshadows the eclectic future of modern macroeconomics, as presented in this
book, which accepts the expectations-augmented Phillips curve as the missing equation, even if it means that there is no lasting trade-off between output and
inflation.
In general, Europe was slow to recognize the power of the monetarists’ attack, and did not contribute much to the research effort. In the UK, the academic
establishment was dominated by Keynesians, most of whom refused to acknowledge that a major battle had been lost. The election of Mrs Thatcher changed all
that. She brought in Milton Friedman as an adviser, proclaiming that her government would follow the master’s precepts, including rolling back government,
pushing for wage stability by destroying the trade unions’ grip on labour markets and, of course, a strict application of the monetary neutrality principle. She had
been converted to monetarism by two close advisers working in a think tank that she had created, the Centre for Policy Studies: Alan Walters (1926–2009), a
British economist then working at Johns Hopkins University in the USA, and Patrick Minford (1943–), who had resisted Keynesian influence in his bastion at
Liverpool University. When, early on during her first term, the scope of Thatcher’s policy intentions became clear, 364 academic economists signed a manifesto
that promised disaster if these policies were implemented. Two decades later, most of the signatories agree that ‘we all are Thatcherites now’.
Elsewhere in Europe, the evolution was gradual, mostly the result of generation changes, as freshly graduated macroeconomists started to popularize either
monetarist ideas or less orthodox versions of Keynesian economics. Still, in some countries, such as France, Keynesian ideas remain to this day the dominant
reference in policy-making circles. In the 1970s, two French economists, Edmond Malinvaud (1923–2015), who served for two decades as Head of the National
Statistical Institute, and Jean-Pascal Benassy (1948–) of the Centre National de la Recherche Scientifique, had already developed a disequilibrium interpretation of
the Keynesian model. The particular feature of this interpretation, which has now been abandoned, is that it assumes that there can be lasting excess demand or
supply in goods and labour markets.
In German-speaking countries, as mentioned earlier, Keynesian ideas never quite displaced the classical view, so there was little need for a monetarist counter-
revolution. Economists and policy-makers saw the movement as a vindication of their own views, even though monetarism is considerably more subtle than
classical economics. During the years of Keynesian domination, the flame of classical economics was carefully maintained at the annual Konstanz seminar, which
was initially created by two early monetarists, the Swiss economist Karl Brunner (1916–1989), who worked at Rochester University, and Allan Meltzer (1928–)
from Carnegie Mellon University in Pittsburgh.5 The Konstanz Seminar still meets every year.
The Chicago school also contributed much to our understanding of the open economies. Much of Mundell’s work was produced when he was in Chicago,
where he also trained a generation of international macroeconomists who developed the ‘monetary approach to the exchange rate’. This approach shapes much
of Chapter 15, including the stylized facts proposed by Michael Mussa (1944–2012) and the overshooting hypothesis of German-born Rudiger Dornbusch (1942–
2002), both students of Mundell. Many other Chicago economists—including Mundell himself—worked at the IMF, where they forged the Fund’s doctrine and
produced important work under the leadership of Dutch economist Jacques J. Polak (1914–2010).
Yet, Friedman’s ideas have not always been widely accepted. One of his other major contributions is the intellectual defence of freely flexible exchange rates,
as noted in Chapter 19. He gathered ammunition for this position when he was in Paris in 1950, working at the US governmental agency which administered the
Marshall Plan. At the time, he concluded that the European Common Market could not work with fixed exchange rates and he considered the European Monetary
Union to be a mistake. This view receives much support in the UK.

Robert Lucas Jr,
1937–
Source: The Nobel Foundation.
Thomas Sargent,
1943–
Source: New York University.
The attack on Keynesian economics was by no means over yet. Another blow came with the rational expectations revolution. The expectations-augmented
Phillips curve of Friedman and Phelps had left an important question unanswered: what drives expectations? Most economists thought that inflation expectations
gradually caught up with actually observed inflation, i.e. they were only taking account of what Chapter 13 defines as the backward-looking component of
underlying inflation. Although Phelps had made some headway in introducing the forward-looking component, the next major step was achieved in Chicago
again where Nobel Prize laureate Robert E. Lucas Jr, a student of Friedman, spearheaded the rational expectations revolution. 6 Rational expectations are
presented in Chapter 6 and this idea permeates much of this textbook. Lucas and his colleagues 7 argued that if the forward-looking component dominates and if
expectations are not systematically biased, the Phillips curve is always vertical, in the short as well as in the long run. As a result, they asserted, systematic
policy cannot work. In particular, monetary policy affects output and employment only to the extent that it creates inflationary surprises. Since creating short-
lived surprises is hardly a basis for macroeconomic policy, the circle was closed. Friedman’s contribution meant that fiscal policy is not helpful but that monetary
policy is a powerful instrument, although one whose effects are eventually dissipated in inflation. The message of the rational expectations revolution was that
macroeconomic policies should not be used on and off with complete discretion. Instead, policy should obey rules and aim at establishing credibility for adhering
to the rules.8 This was not a complete vindication of the classic laissez-faire approach, but an indictment of Keynesian policy activism.
The view that ‘only unanticipated money matters’ was never very popular with policy-makers. One could say that it is hardly surprising that central banks
explicitly reject the view that their role is limited to creating surprises. Yet, empirical evidence failed to support this view, paving the way for the New Keynesian
macroeconomics.

Finn Kydland,
1943–
Source: Carnegie Mellon University.
Edward Prescott,
1940–
Source: Federal Reserve Bank of Minneapolis.
Because of its compelling logic, the rational expectations hypothesis attracted immense interest and opened the way for further innovations in other directions.
Clearly, if it is appropriate to assume that expectations are rational, then why shouldn’t all other economic decisions be rational as well? Researchers at
‘freshwater universities’ in the USA—Chicago, Minnesota, Rochester, Carnegie-Mellon, and University of Pennsylvania—have established the microeconomic
foundations of the consumption, investment, and primary account functions studied in Part II of this book. European economists from all countries—many after a
sojourn at those US universities—are deeply involved in this research programme.
Insisting on the rigorous discipline of microeconomic foundations may be intellectually attractive, yet business cycles remain a fact of life that must be
explained. This led neoclassical economists to the Real Business Cycles (RBC) research programme. The aim of this effort was to show that models with flexible
prices and fully rational agents—in brief, the Robinson Crusoe parable developed in Part II—can reproduce the key features of actual business cycles. The ‘RBC
school’, inspired by the American Ed Prescott from Arizona University and Norwegian-born Finn Kydland from Carnegie Mellon University, has a significant
following in Europe. These researchers received the Nobel Prize for their work in 2004.
20.5 New Keynesian Macroeconomics:

The Latest Synthesis
Michael Woodford,
1955–
Source: Photo courtesy of Michael Woodford.
John Taylor,
1946–
Source: www.stanford.edu.
Despite its intellectual attractiveness, the RBC approach was not a great empirical success. Many of the most important stylized facts of the business cycle
remain unaccounted for. Price stickiness simply appears to be a fact of economic life. 9 This opened up an opportunity for the New Keynesians, who were already
at work on their own response to the rational expectations revolution. Their main aim has been to show that price stickiness is not incompatible with
microeconomic foundations and full rationality. New Keynesians have thus been able to produce a new synthesis, which fully rests on rational behaviour but
delivers the traditional Keynesian results. Much of this work has been carried out at traditionally Keynesian ‘saltwater universities’ in the USA (Harvard, MIT,
Yale, Princeton, Berkeley), 10 with some important contributions from Michael Woodford, now at Columbia University in New York. In Europe, Jordi Gali, from
Pompeu Fabra University in Barcelona, has further developed the microfoundations of the expectations-augmented Phillips curve.
The synthesis starts with RBC microeconomic foundations—complete with rational expectations—and adds price stickiness. The result turns out to be very
similar to the AS–AD presented in Chapter 14. It contains an IS curve, which incorporates the aspect that next period’s demand affects that of the current period
—and reflects the idea that households strive to smooth their consumption. Second, it also includes a Phillips curve that is almost identical to the one initially
proposed by Friedman and Phelps. It allows both for rational expectations of price setters, but admits that some agents do not change prices very often, or do not
have the information or the wherewithal to do so. Most importantly, the new Keynesian Phillips curve implies that ‘anticipated money matters’, so that monetary
policy can systematically affect output. Third, it also includes the Taylor rule, named after John Taylor from Stanford University, seen as the deliberate and
systematic response of monetary policy to fluctuations in inflation and output.
The New Keynesian framework has been wholeheartedly embraced by policy-makers who now read the same books as economists. Importantly, the new IS
curve and the old-new Phillips curve attract attention to the crucial role of expectations, which has led central banks around the world to become more
transparent about their own forecasts and intentions. From the policy perspective, the view that fiscal and monetary policies can play a role as tools for output
and employment stabilization is now generally accepted. So too is the recognition that the role of expectations requires much more prudence and care than the
traditional Keynesians dared to admit, as explained in Chapters 16 and 17.

Friedrich August von Hayek,
1889–1992
Source: Carnegie Mellon University.
James Buchanan,
1919–2013
Source: Photo courtesy of James Buchanan.
Since the rational expectations revolution in the early 1970s, macroeconomics has managed to avoid further paradigmatic earthquakes. A number of innovations
have occurred at the frontier between economics and political science. This ongoing research programme starts from the obvious observation that policy actions
are not taken in a vacuum, but by policy-makers, who are real-life politicians and keenly sensitive to public opinion as they seek re-election and power. In doing
so, they need to calculate what will be the effects of their actions on the economy—the traditional macroeconomic question—and their voters’ reactions.
Once these questions are asked, it becomes clear that political systems matter a lot. We need to look at the respective influences of government and
parliament, at the degree of independence of the central bank, and at the electoral rules. Here, the variety of institutions across Europe offers a unique source of
observation. It comes as no surprise that European economists—some of whom are based in the USA—have often played a leading role in this area of research.
The roots of this new approach are both old and interesting as it has brought together some very different traditions. German-language economists in the
tradition of the Historical and Austrian schools have long explored these issues and their Ordnungspolitik (this term is hard to translate but formally means
‘policy of establishing or maintaining order’ or, more precisely, the institutional framework for economic activity) remains very influential in this part of Europe.
However, with few exceptions, 11 these approaches have not exerted much international influence, partly for language reasons, and partly because they rejected
the formalization of their ideas. The same questions were explored independently by US-based economists of the ‘Public Choice School’, including Nobel Prize
laureates James Buchanan (1919–2013) and Douglass North (1920–2015). Most economists associated with the public choice research programme consider
themselves more aligned with the laissez-faire view. 12 Coming from a radically different perspective, New-Keynesian economists began to study why
governments make policy mistakes, rather than simply criticizing them, as do laissez-faire economists. They stress that sometimes policy-makers pursue bad
policies because they have distorted incentives. Since incentives are determined by institutions, to improve policy-makers’ incentives, institutional reforms may
be needed. Important contributions have been made by Alberto Alesina, an Italian economist at Harvard, Torsten Persson (1954–) from Stockholm University,
and Guido Tabellini (1956–) from Bocconi University, among others. Finally, a few French microeconomists—brought together at Toulouse University by Jean
Tirole (1953–) and Jean-Jacques Laffont (1947–2004)—have explored the question of incentives and decision-making under uncertainty. Their results are
gradually percolating into macroeconomics, where they allow the study of the interaction between policy-makers and the private sector.
This work indicates that the kinds of differences that oppose laissez-faire advocates and interventionists are too blunt when stated as ‘the government should
stay out of economics’ versus ‘governments must take responsibility for economic welfare’. There are some tasks that some governments can usefully perform
and others cannot, and tasks of general interest that are better left to the markets. For example, fiscal policy may play a stabilizing role, but governments tend to
suffer from a deficit bias, especially in political regimes where decisions are made by divided parliaments. Equally important is the realization that some economic
policies cannot be improved unless the political institutions are first reformed. A good example is that central bank independence is generally a precondition for
good monetary policy. Employment and growth are two further examples of the important innovations brought about by this fundamental intuition.
20.7 Labour Markets

Richard Layard,
1934–
Source: British Academy.
Stephen Nickell,
1944–
Source: Bank of England.
One of Europe’s sad distinguishing features is the high rate of unemployment which has prevailed since the mid-1970s. In a number of reassuring cases,
however, some—mostly smaller—European countries have been able to roll back unemployment. As Chapters 4 and 18 emphasize, the problem lies in labour
market structures, institutions, and policies. Here again, comparison and analysis of the diversity of situations in Europe has offered a wealth of lessons about
the nature of unemployment and the ways to deal with it. European economists have made significant progress, if only to develop a genuine understanding of
their own labour markets, which differ profoundly from those in the USA.
Much of the pioneering effort has been conducted at the London School of Economics’ Centre for Labour Economics, led by Richard Layard (1934–) and Steve
Nickell (1944–). Other important early contributors are Edmond Malinvaud (1923–2015) from France, Jacques Drèze (1929–) from the University Louvain-la-Neuve
in Belgium, Herbert Giersch (1921–2010) from the University of Kiel in Germany, and Assar Lindbeck (already mentioned), who, together with Dennis Snower
(1950–) now at the Institut für Weltwirtschaft in Kiel, Germany, developed the insider–outsider theory. According to this theory, labour representatives defend
the interests of the employed workers—the insiders—at the expense of those who are not employed—the outsiders. The overwhelming evidence is that labour
market rigidities lie at the root of Europe’s unemployment problem. This assessment is not much disputed today but was initially rejected by Keynesian
economists, who blamed instead restrictive demand management policies. One way of simplifying the debate is to establish whether the problem lies with a high
equilibrium unemployment rate (the case of rigidities) or whether actual unemployment is simply above its equilibrium rate. Interestingly, monetarists always took
the view that the equilibrium unemployment rate had risen. The diagnosis that structural problems are at the root of the unemployment problem in Europe is
hardly disputed, except by a handful of die-hard Keynesians. In fact, it has been accepted by many governments.
This conclusion has triggered a search for appropriate solutions. Structural problems call for structural reforms, and reforms are always controversial. In the
area of labour markets, the controversies are laden with emotional political and social undertones; in some countries deep ideological battles have resurfaced.
The reason is simple, and is reminiscent of the problem of dynamic efficiency discussed in Chapter 3. To get to where we want to go, up-front sacrifices are
necessary. Real wage moderation, cuts in unemployment benefits, reform of job protection, and the deregulation of product markets may well hurt many
individuals today, even if later ultimately leading to future increases in employment and GDP. Finding out how to convince and compensate the losers in reforms
is the magic formula which smaller economies in Europe seemed to have found. Several European economists are actively investigating conditions under which
politically difficult economic reforms can be adopted. What has to be done is now subject to much agreement, with important detailed work carried out by the
economic staff of the OECD and IMF. Some countries have implemented many of these measures, and unemployment has indeed declined, sometimes
significantly. In other countries, governments have been too sensitive to even acknowledge the need for action.

Peter Diamond,
1940–
Source: ©The Nobel Foundation
Photo: Ulla Montan.
Dale Mortensen,
1939–2014
Photo: Ulla Montan.
Christopher Pissarides,
1948–
Photo: Ulla Montan.
Unemployment is such a big issue in macroeconomics that it has perspective—how you look at it determines how you think about it. Chapters 4 and 18
established that unemployment is a supply-side problem in the long run, and those more interested in the long run see labour market institutions and preferences
as playing the predominant role. Yet throughout our book, unemployment has also been seen as a short-run topic of first-order political and social importance—
determined by the ups and downs of the business cycle. Those who are more interested in the short-run perspective stress that sudden declines in aggregate
demand lead firms to reduce employment, reduce hires, and increase unemployment.
Yet unemployment is so complex that it takes on yet another, different dimension when viewed from the ‘bathtub’ or dynamic perspective of inflows and
outflows, as stressed in Chapter 4 (Section 4.4). The enormous turnover of workers and jobs despite large, stable stocks of unemployment signals that the
uniqueness of each worker and job—and the uniqueness of their coming together, or matching—makes search worthwhile. Reshuffling the personnel deck can
actually be productive for an economy. Research shows that even people who have jobs are often on the lookout, even if they are choosier or even deny it
entirely. In addition to the fundamental insight that workers and employers search purposefully, Peter Diamond of MIT also recognized that searchers who do
‘find one another’ in such markets have something special—an economic rent or surplus—that cannot be costlessly recreated without renewed time-intensive
search. As a result of this rent, wage determination takes on the role of allocator of surplus, and has no obvious value as it does in neoclassical economic theory,
except that it must leave the match at least as attractive to both parties as the option of abandoning the match and resuming search.
Since workers and firms are constantly on the lookout, it is of great importance to understand their behaviour. Offers usually come randomly, but not all at
once. The longer and more intensively we search, the more likely we are to receive offers. Dale Mortensen (along with John McCall, a pioneer in operations
research) stressed that optimal search behaviour of workers and firms under such conditions generally takes the form of a stopping rule or reservation wage—a
critical value of the offered wage (or more generally, a critical level of satisfactory workplace attributes) below which the offer is rejected as unsatisfactory. This
reservation wage will depend on many factors, including unemployment benefits, impatience, and the shape of the distribution of wage offers.
If all that were not complex enough, markets with search frictions are fraught with externalities. Students looking for an apartment in Berlin often have to wait
for hours in line for a chance to inspect the premises and compete with hundreds of other candidates. Residents of Paris who work a little late have great
difficulty finding a parking space, if only because many people have the same work habits. Clubs in London often organize ‘ladies’ nights’ to make their premises
more attractive to certain male visitors (and also to female ones!). Christopher Pissarides of the London School of Economics was interested in the implications of
the fact that the probability of an unemployed person making contact with any job offer depends crucially on the number of unemployed already in the market
relative to available open positions. Ceteris paribus , every newly unemployed person depresses the chances for success of all those already in the unemployed
pool, and increases the probability that a vacancy already posted will be filled. This fact, combined with wage determination as a surplus allocation device, can
explain why wages depend on ‘labour market tightness’, yet may not decline automatically in the presence of high unemployment.
The deep insights of markets with search and matching frictions are not limited to the study of unemployment. Markets for financial intermediation (especially
banking), for lawyers and architects, for marriage or partners, for apartments and houses, and even the ‘market’ for parking spaces can be thought of as matching
markets—even if the price of a parking space is zero! A serious branch of monetary economics considers the use of money as a solution to a matching problem.
Combined with powerful mathematical methods to study optimal dynamic decisions under uncertainty, search and matching theory has contributed to the design
and implementation of labour market reforms in modern developed economies.
Robert Solow,
1924–
Source: The Nobel Foundation.
Paul Romer,
1955–
Source: Hoover Institution.
Robert Barro,
1944–
Source: Hoover Institution.
Much as our understanding of labour markets has greatly benefited from institutional economics, one of the most important—and more vexing—issues involving
the wealth and poverty of nations has also been profoundly rethought. As explained in Chapter 3, research on the neoclassical growth model13 conducted at
MIT by Nobel Prize laureate Robert Solow, had two key implications: (1) capital is more productive where it is scarce, and (2) the key source of sustained growth
is unexplained—or exogenous—technological progress. Both implications were deeply unsatisfactory, and both have motivated important innovations to
conventional growth theory.
As noted in Chapter 3, Robert E. Lucas Jr challenged economists to explain why capital doesn’t flow from rich to poor countries. Poor countries are
characterized by low capital intensity and, in theory, must have a much higher marginal productivity of capital than rich countries. One important solution to his
puzzle is that high productivity may be high in theory, but in fact is significantly reduced by the prevalence of poor institutions that allow corruption, instability,
and war to discourage investment.
However, if technological change is exogenous, it is not possible to explain why institutions matter. In the mid-1980s, Paul Romer (1955–) from Stanford
University showed how technological progress could be treated as endogenous. The crucial step was to recognize that knowledge does not suffer from
decreasing returns. This established a link with Lucas’ question: education is an investment in human capital, and it is deterred by poor institutions in exactly the
same way as investment in physical capital. This discovery allowed many others to explore the process of growth and economic development. Much empirical
evidence has since been produced by Robert Barro (1944–) at Harvard University, in collaboration with Catalan Spaniard Xavier Sala-i-Martin (1963–) from
Columbia
University, with important contributions by many others, including French-born Philippe Aghion (1956–) from Harvard University.
The result of this research has been a thorough reappraisal of underdevelopment and of policies that try to deal with extreme poverty in many parts of the
world. The emphasis has shifted from earlier recommendations by the rich countries to ‘do as we do’ to try to encourage the establishment of better institutions
that would provide political leaders in the poor countries with the incentives to adopt pro-growth policies. This literature has also profoundly affected the
international financial institutions, in particular the World Bank and the regional development banks. The results are slow in coming but some important
successes have been achieved. The same message also concerns the rich countries, especially Europe, which stopped catching up with the USA in the mid-
1980s. The message is that reforms are needed to make the political system interested in supporting agents of change rather than established economic interests.
20.10 Demographics, Low Productivity Growth,

and Secular Stagnation
Alvin Hansen,
1887–1975
Source: The History of Economic Thought
website http://www.hetwebsite.net
Lawrence H. Summers,
1955–
Source: Lawrence. H. Summers, CC-BY-SA 3.0
https://creativecommons.org/licenses/by-sa/3.0/
Robert J. Gordon,
1940–
Source: Photo courtesy of Robert Gordon
Interest rates are currently low, lower than they have been in decades. Some economists and politicians have criticized the Fed, the ECB, and the Bank of England
for their ultra-low interest rate policies. We saw in Box 11.5 how a short-run decline in interest rates could easily be explained by a leftward shift of the IS curve
and the zero lower bound for the TR curve. Yet the persistence of very low rates around the world despite economic recovery has prompted some economists to
wonder if only low inflation and excessive liquidity created by the world’s central banks are to blame. In fact, real interest rates have been falling for almost 25
years, so it is natural to suspect the equilibrium real interest rates as the culprit—the neutral real interest rate, corresponding to the marginal product of capital.
This is an intriguing possibility indeed. Although some of the analysis is beyond the level of this textbook, it is easy to construct the basic argument using the
Solow growth model and the Golden Rule of economic growth from Chapter 3. The Golden Rule states that maximal long-run consumption per capita is attained
when the marginal product of capital per capita is equal to the sum of the rate of technical progress, the rate of population growth, and the depreciation rate of
capital. Even in a world where the golden rule is not fulfilled and savings rates are not optimal, the Solow model robustly predicts a positive influence of those
factors on the equilibrium interest rate. A persistent reduction in population growth, an increase in savings rates, or a slowdown in the rate of technical progress
will exert a depressing influence on the real rate of interest in the long run. Lawrence Summers, University Professor and former president of Harvard University,
chief economist of the World Bank, and US Treasury Secretary, has argued that persistent weakness in investment demand relative to the volume of savings is to
blame. This argument receives further thrust from research finding that the rate of technological progress, the source of long-run economic prosperity, has
slowed. Professor Robert Gordon of Northwestern University has argued in a series of influential publications, as is recounted in Box 3.5, that total factor
productivity growth in the twentieth century was the exception rather than the rule, and points to meagre gains despite the impressive technological progress
that cheap computing, telecommunications, internet, and social media have generated. As Box 11.5 stresses, a glut in global savings has arisen, partly because an
ageing population wants to provide for longer periods of old age, but also because increasingly affluent Indian and Chinese lack a solid public social security
system. The secular stagnation hypothesis is not a business cycle phenomenon. Since the factors that led to a strong rise in world savings are long lasting (TFP
growth slowdown, demography, the emergence of high-saving economies), slow growth may be here to stay.
This is not by any means a new argument. In 1938, Alvin Hansen—also from Harvard and a teacher of legendary Nobel laureates Paul Samuelson and James
Tobin—warned in an influential address to the American Economic Association that structural change could lead to a sustained period of ‘secular stagnation’.
Hansen, who spearheaded Keynes’ ideas in the United States during the 1930s, argued that economic growth was doomed to decline because the most important
technical innovations had already occurred and population growth was slowing. Hansen argued for large-scale rural electrification projects, slum removal, urban
renewal, and natural conservation programmes to sustainably increase aggregate demand. Looking back at the widespread damage the Great Depression had
wrought on the United States, it was understandable that he was pessimistic.
Yet scholars are in complete agreement that Hansen was at the wrong place at the wrong time. World War II would raise US government expenditures tenfold
and double real GDP, wiping out any trace of secular stagnation. The war would also take the lives of millions of soldiers and civilians, while destroying trillions
of dollars in productive capital. The baby boom following World War II would give many countries a labour force dividend and a reason for high investment
rates. Technological progress continued unabated until the 1970s. Despite the contextual irrelevance of secular stagnation at the time, the idea’s ultimate validity
has not been lost. Only time will tell if Alvin Hansen (and Robert Gordon and Larry Summers) were right in the end.
20.11 Conclusions
Because economists are not free from their own prejudices, it is crucial that they focus as much as possible on developing rigorous theories and conduct
dispassionate evaluations of their policy implications. By and large, this is what they have done. Controversies abound, but over time intellectual exchange and
the search for unifying truths have brought economists closer together in an ever-increasing degree of agreement. For example, Patrick Minford was harshly
criticized in the 1980s when he argued that high unemployment benefits discouraged unemployed workers from looking for new jobs. His former Keynesian critics
have now accepted the view that unemployment benefits need to be structured in a way that reduces the potentially adverse effects on work incentives. In
contrast to much of the public debate, professional macroeconomists see their field as an intellectual challenge to solve pressing problems, and not as an
ideological exercise.
Macroeconomics was born in Europe as a revolt against classical economists in the wake of the Great Depression. Much of the ensuing research was carried
out in the USA. These developments clarified the limits of Keynesian economics and eventually allowed a field to emerge. It is not yet fully unified but
controversies are now well understood and circumscribed. Prompted by problems of low growth and high unemployment, European economists have made
important contributions that make macroeconomics a lively field with its own flavour. Although there is no such thing as European macroeconomics, a number of
macroeconomic issues in Europe merit more attention than they receive in the USA. These range from the role of regulations, taxes, and transfers in labour
markets to the functioning of monetary unions. More than anything else, Europe consists of ‘small and open’ independent countries, and this fact sets it apart
from the highly integrated states of the USA. If the last two decades of our textbook provide any evidence, it is these differences which make macroeconomics in
Europe so special.
Essay Question
1 Make your list of future Nobel Prize winners, and explain why you chose them.
1 In preparing this chapter, we have benefited from very useful comments and suggestions from Charles Bean, Irwin Collier, Barry Eichengreen, Hans Genberg, Francesco Giavazzi,
Guido Tabellini, and Jürgen von Hagen.
2 Friedrich von Hayek (1899–1992), another Nobel Prize winner, was a prominent product of the Austrian School, which was disbanded in the late 1930s when the Nazis came to
power. Hayek moved first to the London School of Economics and in 1950 to Chicago, where he was a colleague of Milton Friedman.
3 In a gesture of modesty, Friedman is known to have given credit to Henry Simons (1889–1946) for founding the Chicago School.
4 The debate was really about which assumption one is willing to make. There is no incompatibility between the quantity and TR equations: the demand for money M/P = k(i) Y
implies that V = 1/k(i). Inserting a Taylor rule for i yields an expression for velocity. But this must be considered along with the IS equation to explain the interest rate, and with the
AS curve to explain inflation, while the quantity equation was originally meant to be sufficient for understanding the price level and inflation. The quantity equation is also known as
the ‘Cambridge equation’, in deference to pre-Keynes Cambridge.
5 The tradition at the Konstanz seminar is to display a flag that displays the quantity equation ‘MV = PY’.
6 Lucas was not the first to formulate this view. It was first advanced by a number of American economists from Carnegie Mellon University—they inspired Lucas, who spent several
years there before taking up a chair in Chicago—John Muth (1930–2005), Ed Prescott (1944–), and Finn Kydland (1943–).
7 In particular, Thomas Sargent (1943–) and Neil Wallace (1939–).
8 The preference of rules over discretion was first expressed by Milton Friedman in an essay written in the 1950s.
9 This is hardly news! Milton Friedman and Anna Schwarz wrote extensively on long and variable lags with which monetary changes affect the price level. Even the old Scotsman
David Hume (1711–1776) was fascinated by the fact that gold inflows in seventeenth-century Spain had so little short-run influence on the price level.
10 Saltwater universities are located on the East and West coasts of the USA, while freshwater universities are inland, often close to the Great Lakes.
11 Prominent examples are Bruno Frey (1941–) of Zurich and Roland Vaubel (1948–) of Mannheim University.
12 Many members of the Public Choice School are part of the Mont-Pélerin Society, a group created by Friedrich von Hayek in Mont-Pélerin near Lausanne, Switzerland.
13 This theory is called neoclassical because it relies on standard microeconomic principles, much like neoclassical macroeconomics. But Bob Solow, as he is generally called, has been
a leading and enthusiastic proponent of Keynesian economics.
References
Adelman, I. and Adelman, F. (1959), ‘The Dynamic Properties of the Klein–Goldberger Model’, Econometrica, 27.
Alesina, A. (1988), ‘The End of Large Public Debts’, in F. Giavazzi and L. Spaventa (eds), High Public Debt: The Italian Experience, Cambridge University Press.
Altissimo, F., Ehrmann, M. and Smets, F. (2006) ‘Inflation Persistence and Price-Setting Behaviour in the Euro Area,’ Occasional Paper Series No. 46, European
Central Bank, June 2006.
Baldwin, R. and Wyplosz, C. (2015), The Economics of European Integration, 5th edn, McGraw Hill.
Bank for International Settlements (2007), ‘Foreign Exchange and Derivatives Market Activity in 2007’, Triennial Central Bank Survey, December.
Barro, R. J. and Sala-i-Martin, X. (1991), ‘Convergence across States and Regions’, Brookings Papers on Economic Activity, 1.
Barro, R. J. and Sala-i-Martin, X. (2004), Economic Growth, 2nd edn, MIT Press.
Bloom, D., Canning, D., Graham, B., and Sevilla, J. (2000), ‘Out of Poverty: On the Feasibility of Halving Global Poverty by 2015’, CAERE Discussion Paper No. 52,
Harvard Institute for International Development.
Brede, M. (2011) ‘Tracking the Deterioration of Competitiveness within the Eurozone 1990–2010: Theory and Evidence’, Bachelor Thesis, Humboldt University
Berlin.
Bruyn-Hundt, M. (1996), The Economics of Unpaid Work, Thesis Publishers.
Burda, M. and Gerlach, S. (1993), ‘Exchange Rate Dynamics and Currency Unification: The Ostmark–DM Rate’, Empirical Economics , 18.
Burns, A. and Mitchell, W. (1946), Measuring Business Cycles . New York: National Bureau of Economic Research.
Chinn, M. D. and Ito, H. (2008). ‘A New Measure of Financial Openness’, Journal of Comparative Policy Analysis , 10(3), 309–22.
Collier, P. and Gunning, J. W. (1999), ‘Why Has Africa Grown Slowly?’, Journal of Economic Perspectives , 13(3), 3–22.
Cooper, R. (1982), ‘The Gold Standard: Historical Facts and Future Prospects’, Brookings Papers on Economic Activity, 1.
Dam, K. W. (1982), The Rules of the Game, University of Chicago Press.
DeLong, J. B. (1997), ‘What Do We Really Know about Economic Growth?’, <http://www.j-bradford-delong.net>.
Easterly, W. (1999), ‘On Good Politicians and Bad Policies: Social Cohesion, Institutions and Growth’, The World Bank.
Feenstra, R., Inklaar, R. and Timmer, M. (2015), ‘The Next Generation of the Penn World Table’, American Economic Review 105(10), 3150–82.
Fisher, I. (1930), Theory of Interest, Pickering Masters Series.
Fracasso, A., Genberg, H., and Wyplosz, C. (2003), ‘How Do Central Banks Write?’, Geneva Report on the World Economy, Special Report 2, CEPR, London.
Friedman, M. (1968), ‘The Role of Monetary Policy’, American Economic Review, 58.
Garber, P. (1990), ‘Famous First Bubbles’, Journal of Economic Perspectives , 4(2), 35–54.
Goodhart, C. (1988), The Evolution of Central Banks , MIT Press.
Gordon, R. (2000), ‘Does the “New Economy” Measure up to the Great Inventions of the Past?’, NBER Working Paper 7833, August.
Gordon, R. (2016). The Rise and Fall of American Growth: The U.S. Standard of Living since the Civil War. Princeton University Press 2016.
Hall, R. and Jones, C. (1999), ‘Why do some Countries Produce so Much More Output per Worker than Others?’, The Quarterly Journal of Economics , 114.
Harding, D. and Pagan, A. (2001), ‘Dissecting the Cycle: A Methodological Investigation’, Journal of Monetary Economics , 49(2), 365–81.
Henry, J., Hernández de Cos, P., and Momigliano, S. (2004), ‘The Short-Term Impact of Government Budgets on Prices: Evidence from Macroeconometric
Models’, Working Paper No. 396, ECB.
Heston, A., Summers, R., and Aten, B. (2006), Penn World Table Version 6.2 , Center for International Comparisons of Production, Income and Prices at the
University of Pennsylvania, September.
Heston, A., Summers, R., and Aten, B. (2011), Penn World Table Version 7.0 , Center for International Comparisons of Production, Income and Prices at the
University of Pennsylvania, May.
Hume, D. (1752), Of Money.
Keynes, J. M. (1936), The General Theory of Employment, Interest and Money, Macmillan.
Kindleberger, C. (1973), The World in Depression, University of California Press.
Kose, A. and Prasad, E. (2004), ‘Liberalizing Capital’, Finance & Development, September, p. 51.
Kuhn, T. S. (1982), The Structure of Scientific Revolutions , University of Chicago Press.
Lane, P. and Milesi-Ferretti, G. (2007), ‘The external wealth of nations mark II: Revised and extended estimates of foreign assets and liabilities, 1970–2004’, Journal
of International Economics, 73, 223–50, plus update <http://www.philiplane.org/EWN.html>.
Lucas, R. E., Jr (1980), ‘Methods and Problems in Business Cycle Theory’, Journal of Money, Credit and Banking, 12.
Lucas, R. E., Jr (1990), ‘Why Doesn’t Capital Flow from Rich to Poor Countries?’, American Economic Review, 80(2), 92–6.
Mackay, C. (1841), Extraordinary Popular Delusions and the Madness of Crowds , New York: Harmony Books, republished 1980.
Maddison, A. (1991), Dynamic Forces in Capitalist Development, Oxford University Press.
Maddison, A. (1995), Monitoring the World Economy 1820–1992, OECD Development Centre.
Maddison, A. (2006), The World Economy: Volume 1: A Millennial Perspective, OECD Development Centre Studies, Paris.
Maddison, A. (2007), ‘Historical Statistics for the World Economy: 1–2003 AD’ in Contours of the World Economy 1–2030 AD: Essays in Macro-Economic
History, Oxford University Press.Marx, K. (1867), Capital, Foreword to Volume I.
Mitchell, B. (1978), European Historical Statistics , Columbia University Press.
Mitchell, B. (1983), International Historical Statistics , Macmillan.
Mitchell, B. (1998), International Historical Statistics , Macmillan.
Mitchell, W. (1951), What Happens During Business Cycles , New York: National Bureau of Economic Research.
Ndulu, B. and O’Connell, S. (1999), ‘Governance and Growth in Sub-Saharan Africa’, Journal of Economic Perspectives , 13(3), 41.
Nurkse, R. (1961), Equilibrium and Growth in the World Economy, Harvard University Press.
Riksbank, Monetary Policy Update, December 2007.
Ritschl, A. (2004), ‘News from Tobin’s q: Capital Markets were Efficient in 1929’, Working Paper, HU.
Schneider, F. (2015), ‘Size and Development of the Shadow Economy of 31 European and 5 other OECD Countries from 2003 to 2015: Different Developments’,
mimeo, University of Linz, January.
Schneider, F. and Enste, D. (2000), ‘Shadow Economies: Size, Causes, and Consequences’, Journal of Economic Literature, 38.
Smith, A. (1776), The Wealth of Nations.
Tobin, J. (1996), ‘Prologue’, in M. ul Haq, I. Kaul, and I. Grunberg (eds), The Tobin Tax, Oxford University Press.
Trabandt, M. and Uhlig, H. (2012), ‘How Do Laffer Curves Differ Across Countries?’, in A. Alesina and F. Giavazzi (eds), Fiscal Policy after the Financial Crisis ,
The University of Chicago Press.
Wyplosz, C. (1999), ‘Macroeconomic Lessons from Ten Years of Transition’, in B. Pleskovic and J. E. Stiglitz (eds), Annual World Bank Conference on
Development Economics , The World Bank.
Glossary
The glossary presents brief definitions of the key concepts listed at the end of each chapter. Numbers refer to the corresponding chapter(s).
absolute purchasing power parity (5): theory asserting that price levels are equalized across countries once they are converted into a common currency
absorption (2): total national (private and public) spending on goods and services, irrespective of their provenance
accelerator principle (8): the positive effect of an increase in GDP on the rate of investment
accounting identities (2): relationships linking macroeconomic magnitudes to each other by definition
active labour market policies (18): programmes involving direct job creation, retraining, relocation of families from distressed regions, or special
programmes to get young people started in the job market
activist demand policies (16): government policies that attempt, through demand management, to keep current GDP close to its potential or trend level
aggregate demand (1, 11): the sum of planned consumption, investment, government purchases of goods and services, plus net export of goods and
services (the primary current account)
aggregate demand curve (14): downward-sloping curve expressing a negative relation between aggregate demand for goods and services and the inflation
rate
aggregate production function (3): a function linking aggregate output (GDP) to the input of resources such as capital services, labour, and other factors of
production
aggregate supply (11, 13, 14): total volume of goods and services brought to market by producers at a given inflation rate
aggregate supply curve (13, 14): upward-sloping curve linking inflation and aggregate output of goods and services supplied by firms
animal spirits (8): term referring to entrepreneurs’ optimism and willingness to undertake risky investment projects
appreciation (exchange rate) (5, 12): a market-determined increase in the value of a currency in terms of foreign currencies see also depreciation;
revaluation
arbitrage (5, 7): the simultaneous purchase and sale of assets of identical characteristics to earn a profit without bearing risk
AS–AD model (14, 15): a description of the macroeconomy which accounts for output and inflation as the outcome of interactions of aggregate demand (AD)
and aggregate supply (AS)
asset price channel (10): the channel of monetary policy to affect aggregate demand, by which changes in market liquidity influence the market valuation and
thereby consumption and investment
assets (7): forms of wealth held by households, firms, or governments, such as stocks and bonds, bank deposits, cash, or real estate
autarky (6): the state in which a country operates when it does not trade with the rest of the world
automatic stabilizer (16, 17): the economic mechanism that automatically cushions the impact of exogenous changes in aggregate demand, via the effect of
income on savings, imports, and especially taxation
average or unit costs (13): total production costs per unit of output
bail-in (10): the involvement of bondholders and other senior creditors in the rescue of a financial institution, by which a part or all of lending is converted to
equity
bailout (9): the rescue of a failing financial institution by which the government, central bank, or a private/public consortium injects new money into the
institution, usually by taking an ownership stake or making loans at concessionary interest rates; the term can also be applied to sovereign borrowers
balance of goods and services (2): includes merchandise trade as well as trade in intermediate inputs, goods repair, goods held in ports, and non-monetary
gold. The balance of trade in services incorporates a wide and growing variety of invisibles such as transport and travel, communication, insurance, financial
and other services, and also includes royalties and licence fees
balance on international income (2): the net income of a nation which originates abroad, consisting of the sum of primary international income and
secondary international income
balance on primary international income (2): net flow of international income earned abroad by resident entities in production or other economic
activities, subtracting incomes generated by non-resident entities within the country
balance on secondary international income (2): net flow of international payments representing.pngts, transfers from abroad, and tax receipts from
foreign residents; a surplus means a net inflow of these payments
balance of payments (2, 19): a summary of all real and financial transactions of a country with the rest of the world; also sometimes used to refer to the sum
of a country’s current, financial, capital, and financial accounts (plus errors and omissions), corresponding to the change in official institutions’ holdings of
the country’s foreign exchange and other central bank liabilities
balance sheet (9): a statement of the financial position of a firm or other entity at a particular point in time, indicating its assets, liabilities, and net worth
Balassa–Samuelson effect (15): the observation that price levels in richer nations or those with higher productivity in traded goods are systematically higher
than in poor ones, as a result of higher non-traded goods prices
band of fluctuation (19): the range in which the central bank is ready to allow market price of foreign exchange (usually a single currency) to fluctuate without
intervening directly in the market
bank deposits (9): liabilities of commercial banks which are used as money by households, firms, the government, and other financial institutions and represent
by far the largest component of the money supply
bank reserves (9, 10): the central bank liabilities (deposits at the central bank and vault cash) that commercial banks choose or are required to hold to meet
demands of depositors and/or the requirements of regulators
bank run/bank failure (9): a situation in which a large number of customers attempt to withdraw their deposits at the same time from a bank and cannot be
fully honoured, due to difficulties converting sufficient, less liquid assets on the balance sheets to cash, sometimes even inducing bankruptcy
battle of the mark-ups (13): the interpretation of the wage- and price-setting mechanism whereby firms set prices as a mark-up over costs, including wages,
while employees try to have wages grow faster than the inflation rate
basis points (10): in finance, a measure of interest rate differentials equalling one hundredth of a percentage point (5 basis points = 0.05 per cent)
beggar-thy-neighbour (12): policies, especially exchange rate policies, designed to improve competitiveness and divert demand to domestic goods at the
expense of other trading partners or geographic neighbors
Beveridge curve (18): an empirical, downward-sloping relationship between the unemployment rate and the rate of posted vacancies; the position of the curve
is a crude indicator of the efficiency of job-matching
bid–ask spread (7): in financial and especially foreign exchange markets, the spread—usually quoted as a percentage—between the price at which one can sell
a security (bid price) and the price at which one can buy it on the market (ask price)
bimetallism (19): the use of both gold and silver as a commodity money standard
bond (7): a standardized marketable instrument for borrowing, issued by large firms or governments with clearly stipulated conditions for payment of principal
and interest by the borrower, with prices set in financial markets
boom-and-bust cycles (19): period of expanding/contracting aggregate economic activity, usually associated with excessive expansions and contractions of
domestic credit
Bretton Woods Conference (19): meeting held in 1944 and attended by officials from 45 nations to shape a new international money order after the Second
World War
British terms (5): one of two ways of quoting the exchange rate, here in units of the foreign currency per one unit of domestic currency (e.g. for UK residents
US$1.30 for £1); see also European terms
budget line (6): the line expressing the resource constraint of households in consuming today and tomorrow, the slope is the negative of the gross interest
rate.
Burns–Mitchell diagram (1): a diagram displaying the behaviour of macroeconomic variables over the typical business cycle as a deviation from their values
at the cyclical peak
business cycles (1): succession of periods of rapid growth and slowdown or decline in which output fluctuates around its long-run trend
Cambridge equation (5): the simplest formulation of the demand for money, expressing it as a constant (k) times nominal GDP (PY): M = kPY
capacity utilization (rate) (1): the proportion of installed equipment that is currently employed; higher rates occur during booms, lower rates correspond to
recessions
capital (1): most frequently: refers to one of the major factors of production; usually refers to plant, equipment, inventories, structures, residential housing, as
well as computer software; can also refer to the net worth of a bank
capital account (2): component of the balance of payments accounts that records acquisition and sales of non-produced, non-financial, and intangible assets,
such as the right to use land or the acquisition (or loss) of good will through ownership transfer, debt cancellation, or expropriation
capital accumulation (3): the increase of the physical stock of productive capital, sometimes called net investment or net capital formation. It differs from
gross investment, which also includes the capital put in place to replace depreciated equipment
capital adequacy (10): requirement for banks and other financial institution that their capital (i.e. net worth) be large enough relative to their total assets,
usually expressed as some minimum (for example 9%)
capital controls (12): restrictions on the movement of assets into and out of a country, especially bank accounts
capital depreciation (2, 3, 12): the loss of original value of a physical asset owing to use, age, and economic obsolescence
capital gain/loss (15): refers to the gain or loss that is realized upon sale of an asset relative to its acquisition price
capital–labour ratio (3): the ratio of the stock of capital to employment of labour measured in person-hours or employed persons
capital widening (3): the fact that growth in labour input (population, labour force participation) make new investment necessary, merely to keep the capital–
labour ratio at its current level and to prevent it from being diluted
capital-widening line (3): the straight line in the diagram of the Solow model with population growth, which shows the investment per capita needed to
maintain a constant per capita capital stock per unit per capita (in efficiency units)
circular flow (2): the fact that each final sale of a good or service represents income to factors of production employed to produce it; similarly, income to
factors of production is either spent or saved, while savings are used to finance final purchases of goods by others
collateral (9, 10): assets that are offered, or pledged, by borrowers as a means of securing a loan, and are forfeited in case of default
collective labour supply curve (4): the link between the amount of man-hours that workers supply collectively (via wage negotiations or through their
unions) and
the real wage
commodity money (9): forms of money that have intrinsic value in other uses, or derive their value from the commodity out of which they are made, usually
gold or silver
competition policy (18): policies aimed at decreasing monopoly power and increasing rivalry among sellers in markets
conditional convergence (3): the notion that economies converge to well-defined levels of economic prosperity (steady states) which need not be identical,
but depend on individual characteristics of the nation or region under consideration
conditionality (19): requirements imposed by the IMF on member-countries’ macroeconomic policies for obtaining certain types of loans
consol (6, 7): a bond, usually issued by governments, which pays a certain fixed payment each period forever (sometimes called perpetuity)
constant returns to scale (3): property of a production function (or a function in general) in which simultaneous equiproportional increases in inputs (factors
of production) result in an equiproportional increase in output
consumer price index (2): an index of prices
of a basket of goods representative of the consumption pattern of the ‘average consumer’, using fixed quantity weights in some base year
consumer surplus (18): the difference between the amount that consumers would be willing to pay for a specified quantity of goods and what they must
actually pay for it
consumption (2): goods and services produced and sold to households for the satisfaction
of wants; goods that generate utility for households
consumption function (8, 11): a symbolic way of stating that the aggregate consumption is positively related to aggregate wealth and,
if a significant proportion of households is constrained in credit markets, to disposable income
consumption–leisure trade-off (4): the fundamental determinant of the labour supply decision: in order to consume, we need income and therefore we need
to work, which means giving up leisure time
consumption smoothing (8, 17): optimal choice by households to smooth out the impact of temporary disturbances to income on consumption plans by
either borrowing (in the case of a negative shock) or saving (in the case of a positive shock)
contagion (10, 19): the rapid spread of a bank run crisis to otherwise sound financial institutions due to vulnerabilities or interbank liabilities; can also be
applied to indebted countries
contractionary (11): a characterization of policies or shocks that reduce aggregate demand and lead to lower economic growth
convenience function (9): the benefits of using money to its users, especially the advantage of not having to sell assets or to engage in barter when
purchasing goods on the market
convergence hypothesis (3): the hypothesis that countries achieve the output per capita with the passage of time, of a negative association between per
capita growth and initial per capita GDP
coordination failure (1): situation occurring when agents (households, firms) fail to realize that their actions are interdependent, and that acting jointly might
benefit all
core or underlying inflation rate (13): the inflation rate taken into account during wage bargaining to anticipate future inflation or to recuperate losses from
past inflation
countercyclical (1): term used to describe an economic variable when it is negatively correlated with the state of the economy; that is, it moves in the opposite
direction to aggregate output over the business cycle
countercyclical fiscal policy (17): corrective
device intended to keep the economy near its equilibrium level by increasing or decreasing aggregate demand via public spending or
tax policies
covered interest rate parity (CIRP) (15): a no-arbitrage condition equating the difference between domestic and foreign interest rates to the forward
exchange discount
credit channel (10): channel of monetary policy which operates beyond the interest rate in that credit may be limited in quantity to borrowers irrespective of
interest rates or the volume of lendable funds at banks
credit default swaps (7): a type of derivative security which is based on a particular loan contract, typically a bond issued by a government, a bank, or a
company which pays when interest payments are missed or the loan contract goes into default
credit rationing (8): a condition in loan markets in which there is excess demand for loans at the market interest rate
currency (9): paper cash (banknotes) or coinage in circulation
currency crises (15): episodes of sudden capital outflows from countries—most often which have fixed their exchange rates—leading to sharp losses of
foreign exchange reserves held by the central bank and ultimately to a devaluation of the currency
current account (2): the sum of a country’s trade in goods, services, and unilateral transfers with the rest of the world
cyclically adjusted budgets (17): budgets
adjusted for the effect of the business
cycle on tax revenues
deadweight loss of taxation (18): the loss of consumer and producer surplus which is caused by the introduction of taxes on goods and services, effectively
driving a wedge between consumers’ marginal valuation and the marginal cost of production
debt stabilization (17): the process of arresting growth in the debt–GDP ratio, usually achieved by cutting government expenditures and raising taxes
decision lag (16): time lag in policy effectiveness needed by government to formulate policy
decreasing returns to scale (3): describes a production function for which an equiproportional increase in the factors of production results in a less than
equiproportional increase in output; see also constant returns to scale
deflation (16): a period of sustained decrease in the general price level
demand disturbance/shock (14, 16): a sudden exogenous shift in the level of aggregate demand at given levels of inflation, underlying inflation, and nominal
interest rates
demand for money (5): the level of money, either narrow or broad, which households, firms, and governments choose to hold, given the level of economic
activity, nominal interest rates, and the cost of moving between money and other assets
demand side (1): those factors contributing to and influencing the spending decisions by economic agents
deposit insurance (10): insurance of depositors, up to some upper limit, against loss of their deposit at a bank, usually provided by the government or a
consortium of banks
depreciation (capital) (2, 3, 12): the loss of original value of a physical asset owing to use, age, and economic obsolescence
depreciation (exchange rate) (5, 12): a market-determined decrease in the value of a currency; see appreciation, revaluation
depreciation line (3): the straight line in the diagram of the Solow model without technical progress and population growth, which shows the investment per
capita needed to maintain a constant capital stock per capita
derivatives (7): securities that derive their value from the prices or price behaviour of other underlying securities
derived demand (9, 10): a demand for something which arises as the outcome of demand for something else
desired demand function (11): total planned spending given the interest rate and real GDP, the sum of consumption expenditures, investment spending,
government purchases of goods and services, and net exports
deterministic view of business cycles (16): a theory of the business cycle expressing current output as a mechanical function of the economic past
devaluation (12): decision by the monetary authority to reduce the pegged value of the currency; see revaluation, appreciation, depreciation
dichotomy (11): the situation pertaining when equilibrium values of nominal variables can
be determined independently of real variables; the real side of economic activity (growth, unemployment, etc.) is affected only by technology and tastes
dichotomy principle (5): dichotomy is the delinking of the nominal and real sides of the economy; this follows from the monetary neutrality principle
diminishing marginal productivity (3, 6): the tendency that, as the inputs into production are increased, the increments of output will decline
discounting (6, 7): valuing future goods or money in terms of goods or money today; see also intertemporal price
disinflation (14): a sustained reduction in the inflation rate brought about by contractionary demand policy
distortionary taxation (18): see tax distortions
diversification (7): the practice of holding wealth in a number of different assets, with the objective of averaging out the risk implied by any individual asset in
the portfolio
dynamic inefficiency/efficiency (3): an economy
is dynamically inefficient when a reduction
of current savings can make all generations better off; it is dynamically efficient when future generations can be made better off
only by reducing consumption (i.e. increasing savings) today
dynamic stochastic general equilibrium (DSGE) models (16): mathematical models which stress microeconomic foundations of economic agents as well
as their economic constraints and the internal consistency of these
economic agents (1): term used to denote individual decision-makers, households,
and firms
economic growth (1, 3): secular increases in the output of an economy, usually measured by the annual growth in GDP per capita
economic rents (18): returns to factors of production that exceed the minimum amount necessary to keep those factors of production
in operation
effective exchange rate (5): an index consisting of a weighted average of a country’s exchange rates vis-à-vis its main trading partners
effective labour (3): a measure of labour input which accounts for not only the number of hours worked but also for the effect of technical progress on the
productivity of
those hours
effectiveness lag (16): time lag resulting from a slow or delayed impact of economic policies
on real activity
efficient market hypothesis (7): the hypothesis that when markets are efficient, prices fully reflect all available information and that arbitrage has eliminated
all opportunities for systematic excess profits
efficiency wages (4): wages paid in excess of the marginal productivity of labour in order to induce sufficient effort on the part of the workers
elastic (4): the sensitivity of demand or supply
to price (usually measured in percentage terms)
electoral business cycles (16): business cycles which result from governments’ use of macroeconomic policies to boost their re-election prospects, implying
that the economy should be booming and employment is high on election day and that in the aftermath of its election, the government tightens its policies
employers’ associations (4): organizations of employers which represent their interests, especially in collective bargaining
endogenous and exogenous variables (1): endogenous variables are explained by economic principles; exogenous variables, in contrast, are determined
outside the system under study
endowment (4, 6): the exogenous resources that economic agents expect to have in the present and in the future
equilibrium GDP (11): the GDP level at which the desired demand for goods and services is equal to the supply
equilibrium (long-run) real exchange rate (15): the theoretical level of the real exchange rate (the nominal exchange rate doubly deflated by price indexes
at home and abroad) necessary to enforce the national budget constraint with
the rest of the world
equilibrium unemployment (4): corresponds to the situation where the wage matches the collective preference of workers but leave some individual without
jobs
equilibrium unemployment rate (4): the unemployment rate that occurs when employment and unemployment stabilize,
i.e. when aggregate demand for labour is met by aggregate supply. Because labour supply may not perfectly reflect individuals’ preferences, this
unemployment may in part be involuntary (structural unemployment), but it may also reflect the efficiency of the labour market (frictional unemployment)
equity–efficiency trade-off (17): the fact that improving equity among society’s members has a negative impact on the economy’s efficiency
European Monetary Union (EMU) (19): the group of countries that share the same currency,
the euro
European terms (5): one of two ways of quoting the exchange rate, here in units of domestic currency per one unit of the foreign currency (e.g. 7.8 Swedish
krona or 1.1 Swiss franc for $1); see also British terms
excess demand (11): a market situation in which demand exceeds supply at prevailing prices
excess supply (11): a market situation in which the quantity supplied exceeds desired demand at prevailing prices
exchange market intervention (12): the central bank buys or sells its own currency against foreign currencies on the foreign exchange market in order to
prevent unwanted exchange rate movements
exchange rate (1, 5, 12, 15): the rate at which foreign money is traded for domestic money
exchange rate anchor (12): a strategy for pursuing monetary policy which fixes the nominal exchange rate to another currency in order to lower expectations
of future inflation
exchange rate depreciation (12, 15): a loss of the domestic currency’s value in terms of another currency
Exchange Rate Mechanism (ERM) (19): the fixed exchange rate arrangement of the European Monetary System
exchange rate overshooting (15): a depreciation of the nominal exchange rate in response to an increase in the money supply, which exceeds, or
overshoots, the extent implied by the principle of monetary neutrality
exchange rate regime (12): description of the exchange rate system adopted by a country: the exchange rate may be freely floating, or can be fixed so that
the central bank maintains the value of the domestic money in terms of another currency or group of currencies
expansionary monetary policy shocks (11): a shift downward in monetary policy of the central bank described by the Taylor rule, due to a decrease in the
target or neutral interest rate
expected inflation targeting (10): central bank policy by which the nominal interest rate is set to keep the rate of inflation—as forecast by the central bank
itself—within a certain range of a preannounced target
export function (11): function representing the demand for a country’s exports, linking them positively to foreign spending and negatively to the real exchange
rate
external competitiveness (5): the relative ability
of countries to export goods to others at given levels of nominal wages, nominal prices, and nominal exchange rates
external terms of trade (5): the ratio of export to import prices, or how many units of imports are needed to purchase one unit of exports
externalities (17, 18): activities that affect the welfare of economic agents not undertaking them directly
factors of production (1): inputs in the production process, such as labour, capital, or land, which create value added (in contrast to intermediary inputs)
fiat money (9): money which the state declares to be legal tender although its intrinsic value may be little or nothing
final and intermediate sales (2): final sales refer to sales of goods and services to the consumer or firm that will ultimately use them; contrast with
intermediate sales to producers who use and transform these goods or services as part of their own production of goods and services
financial account (2): net sales of foreign assets by private domestic residents (a purchase by domestic residents worsens the financial account balance, a
sale improves it)
financial intermediaries (7, 9): economic entities that collect funds from depositors and lend them to borrowers
financial intermediation (2): the channeling of savings of households by banks and other financial institutions to those willing to undertake physical
investment
finding rate (4): the rate at which unemployed workers find a job, calculated as a ratio of job finds (per month or per year) to total unemployment
fiscal policy (1): the use of the government budget to affect the volume of national spending, or more generally to provide public goods and services, as well as
to redistribute income
Fisher principle/equation (14): the decomposition of the nominal interest rate (i) into the sum of the real interest rate (r) and the expected rate of inflation (πe)
fixed capital formation (6): see investment
flows and stocks (2): a flow is an economic variable measured between two periods of time; a stock is a magnitude measured at a given time
foreign exchange market interventions (2, 12): purchases and sales of foreign money in exchange for domestic money undertaken by monetary
authorities
foreign exchange reserves (12): foreign currencies held by the monetary authority for the purpose of intervening in the exchange markets
foreign or international rate of return (12): the total rate of return earned in per cent on a short-term foreign asset, including gains or losses due to
movement in the exchange rate, which makes investors indifferent between investing at home and abroad
foreign reserves account balance (2): net lending transactions performed by the monetary authority on the foreign exchange market (net purchases of
foreign exchange)
forward discount (15): when the forward price of foreign exchange is lower than the current spot price, implying an expected exchange rate depreciation
forward exchange rate (15): an exchange rate agreed upon today for a currency transaction
to occur at a future date
forward premium (15): when the forward price of foreign exchange exceeds the current spot price, implying an expected exchange rate appreciation
free banking (9): a system of banking without a central bank or central regulatory authority which relies on competition and depositor due diligence to ensure
the integrity
frictional unemployment (4): unemployment which results from individuals’ changing
jobs, losing and finding jobs, or entering the labour force
Friedman critique (16): the view that inherent lags in decision, design, implementation, and effectiveness of activist monetary and fiscal policies may negate
its influence on the business cycle or even increase its magnitude
fundamentals (15): factors driving the exchange rate; the net external position, and determinants of the primary current account as well as monetary conditions
and the degree of price rigidity: in general, the underlying real factors that determine the value of an asset
GDP deflator (2): a (Paasche) price index for total value added of an economy, given by the nominal GDP divided by the real GDP (GDP valued at price of some
base year)
GDP per capita (2): the ratio of GDP to population size
general equilibrium (11, 12, 13): condition of equilibrium applying simultaneously to several markets at the same time, recognizing the interdependencies
between markets
global imbalances (6): large and lasting current account deficits and surpluses that emerged in the 2000s
globalization (1): the process of increasing internationalization of economic relations, especially with respect to trade in goods, services, and financial assets,
but also the increasing mobility of labour and capital ownership
gold exchange standard (19): the system established at the Bretton Woods conference in 1944 whereby gold was the fundamental standard of value, but for
all currencies the gold parity was mediated by the dollar
gold standard (19): a system whereby a country defines its monetary unit in terms of gold
golden rule (3): proposition that per capita consumption is maximized in a growing economy at the point at which the marginal product of capital is equal to the
growth rate
goods market equilibrium (11): the situation in which the desired demand equals supply
Gresham’s Law (19): the proposition that money which is more valuable than its official exchange rate will disappear from circulation: ‘bad money chases out
the good’
gross domestic product (GDP) (1, 2): a location-based measure of a country’s productive activity, corresponding to the value added generated by factors of
production, both local and foreign-owned, within a country
gross investment (3): an economy’s total expenditure on capital goods—equipment and machines, the construction of housing, and inventory changes—
including those which are intended to replace losses due to wear, loss of value, or obsolescence
gross national income (GNI) (2): total value added attributable to the activities of residents, independently of where this value added
takes place
growth traps (3): phases of economic development when low growth leads to low investment in physical, human, and infrastructural capital, which in turn
reduces growth prospects
hard pegs (14): a fixed exchange arrangement which intentionally makes it costly or even illegal to devalue, thereby increasing the regime’s credibility
hedging (15): techniques used to protect oneself against foreign exchange fluctuations; more generally, any trading strategies or techniques used to eliminate
risk
hollowing-out hypothesis (19): the hypothesis that the menu of exchange rate regimes has been narrowed to either fully floating rates or hard peg, implying
that fixed-but-adjustable exchange rates are no longer feasible
household labour supply curve (4): the number of hours a household is willing to work as a function of the wage per hour
human capital (3, 17, 18): the education, training, and work experience acquired by individuals
human rights (3): the rights of individuals in a society to express views or religious beliefs, to associate with others, to engage in economic activity, and to be
free from political persecution
Hume mechanism (19): the process by which trade imbalances were equilibrated under the gold standard system: a trade deficit (surplus) implies a reduction
(increase) in gold and money supply, which leads to falling (rising) prices, higher (lower) interest rates, thereby slowing (stimulating) demand and improving
(worsening) the country’s competitiveness
hyperinflation (1, 5): term used to describe periods of extremely high inflation, usually when the monthly rate exceeds 50%
implementation lag (16): time lag in policy effectiveness as a result of the time taken by parliaments and ministries to pass and originate legislation
import function (11): function linking imports of goods and services positively to domestic spending and positively to the real exchange rate
impossible trinity (19): the result that it is impossible to simultaneously operate a fixed exchange rate regime, allow full capital mobility, and conduct an
independent monetary policy
impulse-propagation mechanism (16): mechanism that transforms shocks (impulses) into irregular oscillations like the business cycle
income effect (4): the portion of change in quantity demanded which is attributed to the change in effective real income that results from the price change
increasing returns to scale (3): a characteristic of the production function which occurs when a simultaneous equiproportional increase in the factors of
production results in a more than equiproportional increase in output
index (1): a number that has no dimension (i.e. is not expressed in units such as DM, tons, hours, etc.); it is usually set to take a simple value like 1 or 100 at a
specific date
indexation (13, 14): a provision in wage or other contracts by which nominal values are adjusted frequently to reflect changes in some price index and to
maintain the real value of the contract’s provisions
indifference curves (4, 8): a graphic representation of all possible combinations of two items that will yield equivalent utility (satisfaction)
industrial policies (18): these amount to official backing of national corporations or whole industries, taking on the form of subsidies, public orders, or trade
policies
inelastic (4): insensitivity of demand or supply to price, all other things equal
inflation (1): the sustained increase in prices over longer periods of time, as measured by the rate of change in a price index (an average of many or all prices in
an economy)
inflation differential (5): the difference between the domestic and foreign inflation rates
inflation targeting (10): a policy approach taken by some central banks in which an inflation rate or band of inflation rates is explicitly and publicly
announced; see also expected inflation targeting
inflation-targeting strategy (14, 16): the central bank announces a target for the inflation rate that it aims to reach within 2–3 years, publishes its inflation
forecast, and adjusts its policy in reaction to the difference between the target and the forecast
inflation tax (7): real revenue that the government obtains by inflation, by eroding the real value of nominal assets and improving the financial condition of the
government, reducing the real value of its nominal liabilities
information asymmetry (10): a situation in which one party has better information than the other/s about the probability of an outcome
insiders and outsiders (4): a distinction applied to workers who are already employed in long-term employment relationships
installation costs (8): the costs of installing new productive equipment
instruments (10): monetary policy decision variables that are under more direct control of the central bank, for example the bank refinancing rate, the deposit
rate, the marginal lending rate, the volume of open market operations conducted with commercial banks, or the exchange rate, when pegged by the central bank
interbank interest rate (9, 10): the interest rate at which banks lend high-powered money (reserves) to one another
interbank market (9, 10): a wholesale market for money, which brings commercial banks together. Also called the open market
interest rate parity condition/interest rate parity (12, 15): the condition that interest rates are equalized across countries after taking account of expected
exchange rate changes; see also covered and uncovered interest rate parity condition
internal terms of trade (5): ratio between non-traded and traded good prices
international financial markets (IFM) line (12): the line in the open economy IS–TR diagram describing the interest rate at which net capital inflows are
zero, i.e. when the interest rate at home is equal to the required rate of return available abroad, measured in the domestic currency
international Fisher equation (15): uncovered interest parity and purchasing power parity imply that the real interest rates are equal across countries ex ante
International Monetary Fund (IMF) (19): an institution set up at the Bretton Woods conference in 1944 to promote international monetary cooperation and
exchange rate stability, to establish a multilateral system of payments for current account transactions, and to assist members facing balance of payments
difficulties
intertemporal budget constraint (6): the relationship summarizing resources and opportunities available in the present and the future to a household for
consumption; the present value of spending must be less than, or equal to, wealth
intertemporal price (6): the price of goods tomorrow in terms of goods today; how much we would be willing to pay for—or sell for—the good today for
delivery at some future date
intertemporal trade (6): trade conducted by households and firms across time
interventionism (1): policy whereby a government supports, coordinates, and even controls certain aspects of private activity; see laissez-faire
investment (2, 6): the acquisition of productive equipment for later use in production; also called fixed capital formation
investment function (8, 11): relationship between investment and its fundamental determinants: aggregate investment depends positively upon Tobin’s q and
GDP growth, and negatively upon the real interest rate
investment funds (7): a collection of assets, assembled by specialized financial institutions, in which investors can purchase shares
involuntary unemployment (4): includes those who wish to work at the current wage level but cannot find a job
IS curve (11): for given values of exogenous variables, the combinations of nominal interest rate i and real output (GDP) that are consistent with goods market
equilibrium
job-finding rate (4): the average proportion of unemployed people who find a job during a given period, e.g. during a year
job matching (18): the matching of job offers of firms and unemployed workers
Keynesian assumption (11): the assumption that the evolution of the price level is insensitive to aggregate demand in the short run
(Keynesian) demand multiplier (11, 12): a ratio indicating the effect of increases in exogenous components of aggregate demand on total aggregate demand
Keynesianism (16): the view that government demand management policy should play a key role in macroeconomic policy: Keynesians hold that markets
suffer from imperfections—for example slow clearing of labour and product markets—which are responsible for the occasional underutilization of resources
labour (1): factor of production, usually measured in man-hours, i.e. the total number of hours worked in a firm, an industry, or a country
labour demand (4): the relationship linking the number of man-hours that firms wish to hire and the cost of labour
labour force (1, 4): the total number of individuals who are either working or actively looking for a job
labour force participation (4): the proportion of working-age people who are in the labour force
labour market institutions (4): formal and informal arrangements that regulate wage negotiations, working hours, health safety regulation, workers’ influence,
etc.
labour share (1): the fraction of national income or aggregate value added paid to labour as wages or other forms of compensation, including payments of firms
to social insurance schemes on behalf of their employees
labour tax wedge (18): the difference between labour’s cost to firms and wages actually received by workers
LAD line (14): The long-run aggregate demand curve, which states that inflation is ultimately set by demand, more precisely monetary policy, either at home
(under flexible exchange rates) or abroad (under fixed exchange rates)
Laffer curve (18): the relationship between government tax revenues and the average tax rate: beyond some point, increases in tax rates are associated with
decreases in tax revenues, because the distortionary effects outweigh the revenue gained
laissez-faire (1, 18): term used to describe the view that properly functioning markets will deliver the best possible social outcome, and that intervention by the
government in economic affairs should be rejected; see also interventionism
Law of One Price (5): asserts that domestic and foreign prices are equal when converted in the same currency. It can apply good by good or to a basket of
goods
leisure (4): time spent not working which confers utility on the person involved
lender of last resort (10): the central bank, in its implicit commitment to protect bank customers by providing failing banks with sufficient monetary base to
prevent collapse
life-cycle consumption (8): theory that consumption choices are made with a planning horizon equal to the individual’s expected remaining lifetime; that an
individual will build up savings during working years and exhaust them during retirement years
liquidity service (9): the benefit provided by money as a form of wealth that allows transactions, i.e. conversions of that wealth into goods and services or
financial assets without delay or risk of capital loss
LM curve (11): in the old IS–LM model derived by J.R. Hicks which is the predecessor of the IS–TR, this expresses a positive relationship, for given values of
the exogenous variables and the price level, between the combinations of real output (GDP) and interest rate for which the money market is in equilibrium
logarithmic scale (1): on a normal scale, one moves up by the same distance when going through 1, 2, 3, etc. On a logarithmic scale, the same distance takes
us to the squared value of the previous step; for example 10, 100, 10,000 are equally spaced; a variable growing exponentially is plotted as a line
long run (13): what economists mean when they talk about the behaviour of variables over a period of decades, rather than over short time spans of a few
quarters or years
long-run equilibrium (14): the state of the economy (output and inflation) when aggregate demand equals aggregate supply and actual inflation is equal to
underlying or expected inflation
macroeconomics (1): the study of the aggregate or average behaviour of the economy, as opposed to microeconomics, the behaviour of individual
households, firms, and markets
macroprudential (10): policies paying close attention to possible risks posed by systemically relevant financial institutions and their financial dealings
marginal cost of capital (8): the cost of an additional increment to productive capacity
marginal productivity of capital (MPK) (3, 8): additional output produced by employing an additional unit of capital
in the production process
marginal productivity of labour (4): additional output produced by employing an additional unit of labour in the production process
marginal rate of intertemporal substitution (8): describes how much of today’s consumption a consumer is willing to exchange for tomorrow’s
consumption
marginal rate of substitution (4, 8): the rate at which one commodity can be substituted for another without changing the level of utility
market-clearing (18): term describing a market that works perfectly by equalling demand and supply at every instant
market failures (18): when markets are not functioning as in theory, for example because competition is imperfect with dominant players or when all the
relevant information is not available to all market participants
market liquidity (7): a financial market is liquid when it is easy at any particular time to find counterparts when selling or buying assets; the opposite case is
that of thin markets
market maker (7): traders or institutions that stand ready to deal in a particular asset
market power (13): the ability for producers to set a price that differs from those of close competitors. This is trivially the case of monopolists but can also
occur when producers are able to differentiate their products (often using brand names), thus creating some limited monopoly power. The limit is that excessive
prices may lead consumers to choose another brand. In this case, we talk of monopolistic competition
mark-up pricing (13): the percentage by which a firm increases the selling price of goods above the average or unit costs of production
maturity (7): the length of time before an agreed-upon financial transaction will take place
maturity premium (7): the premium borrowers are willing to pay for borrowing over a longer period of time
maturity transformation (9): banks do maturity transformation when they borrow for a short term (e.g. they accept deposits from customers) and invest in
long-term assets
median voter theorem (16): if voters’ preferences are evenly spread along some dimension, then a political party’s maximizing election strategy is to
advocate policies that are most favoured by the median (‘middle’) voter
minimum wages (4): the lower bound set on wage rates that may be paid to workers, usually but not always by law
misalignment (15): a persistent deviation of the real exchange rate from its equilibrium value
monetary aggregates (9): various definitions of the money stock, differing largely by their degree of liquidity
monetary base (9): the sum of currency in the hands of the public and bank reserves
monetary disturbance or shock (11): a shift to monetary policy, such as a change in the target or neutral interest rate, or more generally the target inflation
rate
monetary economy (1): the part of the economy dealing with monetary and financial, nominal phenomena
monetary policy (1, 7, 10): actions taken by central banks to affect monetary and financial conditions in an economy
monetary policy autonomy (12): the ability of a central bank to decide and implement the interest rate without fear of political interference
monetary policy shock (11): a shift upward in monetary policy of the central bank described by the Taylor rule, due to an increase in the target or neutral
interest rate
monetary targeting (11): monetary policy which focuses on setting money aggregate levels such as M1 or M2 with no regard for the resultant level of interest
rates
money growth targeting (10): a central bank strategy that sets the rate of growth of a chosen money stock aggregate (e.g. M1, M2, M3)
money illusion (13): term used to describe the failure to distinguish monetary from real magnitudes
money market (9, 10): this is where banks and other financial institutions borrow from each other, usually for short periods. Central banks may intervene on
the money markets. Also: interbank market, open market
money multiplier (9, 10): the ratio of monetary aggregates (M1, M2) to the monetary base
Mundell–Fleming model (12): the open economy version of the IS–LM model
N–1 problem (19): in a fixed exchange rate system with N countries, the fact that N−1 bilateral rates can be sufficient to determine all, leaving one degree of
(monetary) independence
natural monopoly (18): occurs in industries exhibiting increasing returns (telecommunications, transport, etc.)
NDP (net domestic product) (2): in the national income accounts, GDP less depreciation
neoclassical (16): model claiming that flexible prices clear all markets even in the short run
neoclassical assumption (13): the view that prices adjust even in the short run, so that the economy is always dichotomized
net exports (11): the difference between exports and imports of goods and services, often expressed as a function which is negatively related to output/GDP,
positively related to foreign output/GDP and negatively related to the real exchange rate
net investment (3): the increase in the capital stock taking into account depreciation
net lending/borrowing (2): ignoring the capital account, a country's capital account or, equivalently, the balance of its financial accounts; when negative,
also called net borrowing
net taxes (2): the government’s tax income from households and firms after transfers have been subtracted
neutral interest rate (10, 14): the interest rate at which monetary policy is neither expansionary nor contractionary
neutrality of money/monetary neutrality (5): the principle that the money supply does not affect real variables such as real output or unemployment, but
rather the price level
NNI (2): Net national income, defined as gross national income less depreciation, usually financial depreciation charges against income earned by residents
no-profit condition (7): the requirement that it is not possible to make an obvious profit without taking associated risks on financial markets
noise traders (7): irrational or misinformed traders who cause deviations of stock prices from their fundamental value for a long time
nominal exchange rate (5): the value of foreign currency in terms of domestic money
nominal GDP (2): total value added measured at current prices
nominal interest rate (8, 14): an interest rate measured in terms of percentage per period of time paid on a nominal investment made in the local currency
non-excludable (17, 18): a good or service is non-excludable when making it available to one person makes it available to all. Examples are national security,
public health
non-performing loans (10): loans held on the balance sheets of banks or other financial institutions that are technically in default or in arrears, or have been
rolled over despite poor repayment prospects
non-price competitiveness (15): relative attractiveness of our goods holding the real exchange rate constant
non-rivalrous (17, 18): a good or service is non-rivalrous when its usage by one person does not detract from others’ ability to do so; examples are clean air or
knowledge
non-standard or unconventional policies (10): alternative instruments of monetary policy which are designed to restore effectiveness to the conventional
transmission channels, such as negative interest rates for bank reserve deposits, quantitative easing, or even ‘helicopter money’
normative economics (1): economics that passes judgement or provides advice on policy actions; see positive economics
objectives (10): in central banking the primary goals with which the central bank is entrusted, such as price stability/low inflation, maintaining high employment
and growth, or safeguarding financial stability
oil shock (13): a sharp increase in oil prices
Okun’s Law (13): the observed inverse relationship between fluctuations of real GDP around its trend growth path and fluctuations of the unemployment rate
around its equilibrium level
open market operations (10): purchases or sales by a central bank of securities in exchange for its own liabilities, with the desired effect of supplying bank
reserves to, or draining them from, the banking system
open position (15): a trader has an open position in a given asset when they stand to gain or lose if the asset price changes. The opposite, a cover position,
can be arranged by committing to buy or sell the asset at a pre-arranged price
openness (1): the ratio of exports or imports—or the average of exports and exports—to GDP, which is a measure of the extent to which a country trades with
the rest of the world
opportunity cost (6, 8): the value of a resource in its best alternative use
optimal capital stock (8): the stock of physical capital that maximizes the value of the firm, for which the marginal productivity of capital is equal to the
marginal cost of investment
optimum currency area (19): a region for which no welfare loss is implied by the use of a common currency
output cost of disinflation (14): the sacrifice ratio, which compares the cumulated increase in the rate of unemployment with the reduction in inflation
achieved over some period of time
output gap (1, 10, 13): temporary deviations of GDP from its trend or equilibrium level
output–labour ratio (3): the ratio of output to the labour used to produce that output
overvalued (15): see undervalued
Pareto principle (18): the principle that economic interventions should only be undertaken if they make at least one economic agent better off without making
any other worse off, or if transfers can be arranged to ensure that this condition holds
parity/central parity (19): when the exchange rate is fixed vis-à-vis another currency, the monetary authorities declare the official value of the currency; this is
called the parity. Usually, the exchange rate is allowed to vary around a value called the central parity
partisan business cycles (16): business cycles resulting from the succession in power of parties with different economic priorities and preferred policies; see
political business cycles
payments system (9, 10): the network of financial intermediaries, other economic actors, or data repositories that process and record economic transactions in
an economy, usually the commercial banking system
PCA function (15): an expression describing how the difference between exports and imports is related to domestic spending, foreign spending, and the real
exchange rate
pecuniary/non-pecuniary externalities (18): externalities that are/are not transmitted by the market’s price mechanism
permanent income (8): the flow of income which, if constant, would deliver the same present value as the actual expected income path
perpetuity (7): a loan agreement with an infinite maturity; see consol
persistence (16): long-lasting effect of a shock hitting the economy
person-hours (4): a measure of labour input which is equal to the number of people employed times the average number of hours spent working
Phillips curve (13): an empirical relationship linking the inflation rate negatively to the unemployment rate
Phillips trade-off (13): the notion that lower unemployment can be traded off for higher inflation; or that higher unemployment is the price of a lower rate of
inflation
physical capital (2): a factor of production consisting of durable inputs such as machines, buildings, computer hardware and software, and physical
inventories
policy failures (18): when a government intervention in the marketplace brings about a reduction of overall welfare, loss of efficiency, or wasting of resources
political business cycles (16): business cycles resulting from the use of macroeconomic policies to improve the state of the economy just before elections;
see partisan business cycles
positive economics (1): the description and explanation of economic phenomena;
see normative economics
positive money/Vollgeld (9): a proposal which seeks to abolish fractional reserve banking, requiring 100% backing of the means payment (bank deposits) by
standardized, possibly government-issued securities and forcing the separation of means-of-payment from investment and intermediation functions of banks
positive/negative externalities (18): an externality occurs when one’s action has an impact on others. It is positive when the effect on others goes in the
same direction as on the agent at the source of the externality (e.g. reducing pollution). It is negative in the opposite case (e.g. using a seat in a public park)
potential output (10): the level of output which is consistent with the sustainable employment of capital, labour, and other production functions given the
state of technical progress, summarized by the production function and delivered by long-run models of economic growth like the Solow model
PPP (purchasing power parity) line (12): a horizontal line corresponding to the foreign inflation rate, because at fixed exchange rates purchasing power
parity rules out permanent differences between domestic and foreign inflation
present discounted value (6): the value of a stream of income or spending spread over time and valued at today’s price; see intertemporal price
price flexibility (11, 13): prices are flexible when they respond immediately and completely to market disequilibrium
price level (1): the average level of prices in an economy
primary budget deficit (6): the budget deficit net of debt service (i.e. net of the payment of interest on the public debt)
primary budget surplus (6): the government budget balance (the difference between receipts and expenditures) from which interest receipts or payments have
been removed
primary current account (6): the current account less net interest payments (net investment income); alternatively, the difference between gross domestic
product output and aggregate domestic spending when unilateral transfers are equal to zero
primary current account function (15): the relationship which relates the primary current account surplus to the real exchange rate (positively) to domestic
GDP Y (negatively) and to foreign GDP Y* (positively)
primary income (2): international income earned abroad by resident entities in production or other economic activities
principle of monetary neutrality (5): asserts that nominal variables do not affect real variables
private income (2): income to the private sector which remains after taxes have been removed from, and transfers have been added to national income (more
precisely, GDP plus net factor income earned abroad)
privatization (18): the sale or transfer of part or all of state-owned enterprises to the private sector
procyclical (1): an economic variable that is positively correlated with the state of the economy; that is, it moves in the same direction as aggregate output
producer surplus (18): difference between the price that a producer actually receives for a given quantity of goods and the amount corresponding to the
minimum price at which they would be willing to supply the same quantity
product differentiation (13): a strategy used by firms to make consumers perceive their products as different from those of their competitors
production function (3, 6): theoretical relationship linking aggregate output to inputs of factors of production
productive efficiency (17): the optimal use of available productive resources in the sense that no alternative arrangement yields more output
property rights (3, 18): rights to ownership of private property and the means of production, especially the protection from capricious seizure or expropriation
public goods (10, 17): goods and services that are available free of charge—either paid for by the state or by private individuals—for which the consumption
by one person does not prevent the consumption by another person or be appropriated by another (characterized by some degree of non-excludability and/or
non-rivalry)
public infrastructure (3): physical means collectively provided that raise a country’s productive capacity. Some means are freely available (e.g. streets or
public schools), others can be purchased (e.g. access to electricity, internet, or public transportation)
purchasing power parity (PPP) (5): an economic theory that compares different countries’ currencies through a market ‘basket of goods’ approach.
According to this concept, two currencies are in equilibrium or at par when a market basket of goods (taking into account the exchange rate) is priced the same
in both countries
q-theory of investment (8): theory linking investment expenditures (I) to Tobin’s q, the ratio of firms’ market value to the replacement cost of installed capital
quantitative easing (10): a non-standard form of monetary policy implying aggressive and sustained open market purchases conducted by the central bank
with the intent of increasing the monetary base and lendable funds available to banks or of raising bond prices and depressing market yields
quota (IMF) (19): a country’s voting and borrowing rights in the IMF, based on its initial deposit upon joining
Ramsey principle of public finance (18): principle that, for a given amount of revenue to be raised, goods with the most inelastic demands and supplies
should be taxed most heavily in order to minimize overall loss of consumer and producer surplus in an economy
random walk (8): a variable that changes randomly from period to period, where the only change between its value today and its value tomorrow will be white
noise and can be positive or negative
rate of depreciation (3, 8): the rate at which the capital stock loses economic value, either by becoming obsolete or by wear and tear, usually expressed as per
cent per annum
rational expectations hypothesis (6, 16): hypothesis
asserting that agents evaluate future events using all available information efficiently so that they do not make systematic forecasting errors
real business cycle (RBC) theory (16): theory of the business cycle which explains economic fluctuations primarily as a consequence of technology shocks
assuming price flexibility
real consumption wage (4): the ratio of nominal wages to the consumer price index; a measure of the price of leisure (or the return to work) in terms of
consumption goods
real disturbance or shock (11): exogenous shifts to the real level of economic activity, i.e. at a given level of interest rates or the price level
real economy (1, 5): term referring to the production and consumption of goods and services, and the (inflation corrected) incomes associated with productive
activities; see monetary economy
real exchange rate (5, 15): the cost of foreign goods in terms of domestic goods, defined as the nominal exchange rate adjusted by prices
at home and abroad
real GDP (2): total value added measured at constant prices, i.e. using prices observed in a reference year
real interest rate (5, 6, 14): the difference between the nominal interest rate and the expected rate (ex-ante rate) or realized (ex-post rate) of inflation
real vs. nominal disturbance (11): real disturbances reflect changes in exogenous real variables (demand components, external shocks); nominal
disturbances reflect changes in nominal variables such as the stock of money or a change in the nominal exchange rate
real wage rigidity (4): rigidity arising when unemployment fails to cause real wages to decline
realignment (12): a change in the official exchange rate parity
recapitalization (10): the injection of fresh funds into a bank in distress, usually associated with change in ownership and management structure
corresponding to the size of the new capital injection, possibly with direct government involvement
recessions (11): intervals of declining economic activity or output below trend potential output for an extended period, usually lasting from 4–8 quarters
recognition lag (16): time lag in discovering that policy intervention is called for
refinance (10): practice of borrowing, issuing credit instruments or issuing other commitments in order to finance a bank or financial institution’s own lending
activity, often after the loans have been made
relative price (4): the price of one good in terms of another, usually computed as the ratio of two nominal prices
relative purchasing power parity (5, 14): situation occurring when the cost of the same basket of goods in different countries increases at the same rate
once converted into a common currency
reserve multiplier (9): the ratio of deposits to reserves held by banks; equal to the money multiplier when currency holding by the
non-bank public equals zero
reserves ratio (9, 10): the ratio of a commercial bank’s reserves (vault cash or deposits at the central bank) to the total demand deposits it
has issued
returns to scale (3): the impact on output of an increase in all inputs by the same proportion:
if output increases equiproportionally, the production function is said to exhibit constant returns to scale; if output increases more or less than proportionally,
we have respectively increasing or decreasing returns to scale
revaluation (12): decision by the monetary authority to increase the value of the currency; see devaluation, appreciation, depreciation
Ricardian equivalence proposition (6): hypothesis that the time profile of taxes employed by the government to finance a given stream of government
purchases has no effect on agents’ intertemporal budget constraint and therefore on real spending and saving decisions; in which case the public debt is not
considered private wealth
risk averse (7): behaviour characterized by a preference to avoid risk
risk premium (7): compensation above and beyond the expected rate of return on an asset required by agents to hold it
saving (2): postponement of consumption using some part of disposable personal income
saving schedule (3): the schedule shows how much a nation saves when the capital stock increases
secondary income (2): income from.pngts, transfers from abroad, and tax receipts from foreign residents
secular stagnation hypothesis (3, 11): the hypothesis that growth has slowed for a longer period than the typical business cycle, because technological
progress or population growth has slowed, or because of high savings and limited productive investment opportunities
securitization (7): a financial operation which transforms an asset that is not traded (e.g. a housing loan) is transformed into a tradable asset, primarily through
standardization
seigniorage (17): exploitation by the government of the monopoly power of the central bank to create money as a means of raising real resources
self-fulfilling attacks (19): exchange rate attacks that are not justified by the exchange rate fundamentals, but occur because, if they succeed, the authorities
will relax monetary policy, proving the attack to be rational ex post
self-fulfilling crises (15): a crisis that occurs simply because it is expected to occur
separation rate (4): the rate at which employed workers become unemployed per unit of time; see job-finding rate
sequencing (19): principles that indicate in which order a country that has long prevented the normal operation of markets and has at least partially isolated its
economy can remove existing restraints and integrate itself in the world economy
share (7): a claim on a part of the profits or earnings of a firm after operating costs and interest have been paid
short-run equilibrium (14): the state of the economy when aggregate demand equals aggregate supply
short-run neutral interest rate (11): the interest rate that, if maintained, is consistent with output at its potential and inflation at its
target level
Solow decomposition (3): the three-way decomposition of the sources of economic growth into capital accumulation, increase in labour utilization, and the
Solow residual capturing technological progress
Solow growth model (3): a theory that analyses growth as being driven by exogenous technological change and the accumulation
of factors of production
Solow residual (3): the part of GDP growth unexplained by the increase in factors of production and conventionally ascribed to technological progress
sovereign borrowing (6): borrowing undertaken
by national governments vis-à-vis foreigners, usually in the form of bond issues or loans granted by international banks
spatial arbitrage (7): concerns the same assets traded in different locations
special drawing rights (SDRs) (19): a reserve money created by the IMF in 1967 and allocated on the basis of quotas; used among central banks as an
additional source of liquidity
speculative attacks (19): sudden loss of foreign exchange reserves of central banks, arising when exchange market participants anticipate an imminent
devaluation
speculative bubbles (7): persistent deviations of market prices from their fundamental values
spot exchange rate (15): the exchange rate that applies to an immediate currency exchange
stagflation (13, 14): periods when both inflation and unemployment increase
steady state (3): a hypothetical state in which all variables have responded fully to exogenous changes in the environment
sterilized intervention (12): exchange rate intervention by central banks combined with measures that offset the impact on the domestic money supply,
usually a money market purchase or sale of securities in the same amount as the foreign exchange market intervention
stochastic view of business cycles (16): the view that business cycles do not respond to systematic inherent causes but to events that happen randomly
structural unemployment (4): unemployment arising as the result of a mismatch of demand and supply of labour
stylized facts (3): regularities in macroeconomic data which guide economists in their search for models to account for economic phenomena
substitution effect (4): the component of the total change in quantity demanded that is attributable to the change in relative prices
supply disturbance/shock (13, 16): a sudden exogenous increase in non-labour production costs
supply side (1, 18): the productive potential of an economy and the factors that determine its overall efficiency
swap (15): exchange of sums of money of the same currency but on different terms, for instance selling francs for delivery now while simultaneously buying
them back for delivery in three months’ time
systemic risk (10): the risk of a generalized collapse of the banking system, arising because banks and financial institutions hold large amounts of each other’s
liabilities
target inflation rate (10): the inflation objective of the central bank, usually stated publicly, which guides monetary policy via the Taylor rule
target interest rate (11, 14): the interest rate which the central bank would choose to set when output equals its trend or equilibrium level; is the anchor for
the Taylor rule describing monetary policy; see also neutral interest rate
targets (10): intermediate goals or signposts of monetary policy that, while not under the direct control of the central bank, can be influenced using monetary
policy instruments, with an eye to fulfilling monetary policy objectives, see also instruments and objectives
tax distortions (18): effects on real behaviour arising from the wedge that taxes introduce between the price received by the provider of a good or service and
the price paid by its consumer
tax smoothing (17): the proposition that a government should not change tax rates in response to temporary causes of budget deficits, but should borrow
instead
Taylor rule (10, 11): a simple description of how central banks set the interest rate in response to output fluctuations and to deviations of inflation from its
desired rate
technological progress (3): the contribution to economic growth of technological change, usually captured by the rate of increase of total factor productivity
technological shocks (18): exogenous impulses, such as discoveries, inventions, product innovations, or process improvements, that alter the productivity of
factors of production, change the environment of economic agents, and cause them to change their behaviour
time consistency problem (18): the possibility that a policy or a sequence of actions which is optimal before they are taken is no longer so afterwards,
especially if other agents act in response to that policy or sequence of actions
Tobin’s q (8): the ratio of the present value of the return from new investment to the cost of installed capital; often approximated as the ratio of share prices to
the replacement price of equipment
total factor productivity (3): productivity in the production process that is attributable not to any particular factor of production, but to all; growth in total
factor productivity is often measured as a weighted average of growth in average productivities of all factors of production
TR curve (11): a graphical representation of the Taylor rule, which states that central banks adjust the interest rate to reduce fluctuations in output given the
central bank’s neutral rate and inflation target
trade policies (18): policies designed to support a domestic product’s sales through tariffs on foreign goods, or quotas on imports
trade unions (4): organizations of workers formed for the purpose of taking collective action against their employers to obtain improvements in pay and other
working conditions
transmission channels of monetary policy (10): mechanisms by which monetary policy conducted by the central bank affect aggregate demand, including
the direct effect of interest rates, asset prices, bank credit, and the exchange rate
trend (1): long-term tendency in a time series
triangular arbitrage (7): applies mostly to foreign exchange markets and is possible when the relative prices of three—or more—currencies are not consistent
with each other
Triffin paradox (19): the inconsistency of the US dollar (a national currency) as a world reserve currency with its gold backing: in order for internationally held
dollar balances to grow with the world economy, the USA had to run balance of payment deficits over time which eventually outstripped its gold reserves
uncovered interest rate parity (UIRP) (15): the condition that rates of return on assets of comparable risk are equalized across countries once expected
exchange rate changes are taken into account
underground economy (2): economic activities from which income earned is not reported and therefore is untaxed
underlying inflation rate (13): the inertial rate of inflation in an economy, which is driven by both expectations of future inflation as well as past expectations
of present inflation based on earlier information and embedded in contracts and informal agreements
undervalued/overvalued (15): a currency is undervalued/overvalued when its exchange is below/above its long-run equilibrium value, or the level consistent
with its long-run fundamentals
unemployment (1, 4): individuals without a job who are actively seeking work
unemployment benefit (4): financial assistance to those seeking a job but unable to find suitable employment
unemployment gap (13): the difference between the actual and equilibrium unemployment rate
unemployment rate (1): the ratio of the number of unemployed workers to total labour force
unsterilized intervention (12): exchange rate intervention by central banks in which no accompanying interventions impacting the domestic money supply
are taken; see sterilized intervention
user cost of capital (8): the effective cost to a firm of using the production factor physical capital in a unit period of time, including the opportunity cost of
resources tied up in the capital, depreciation, changes in the value of capital, as well as tax treatment of these factors
utility (4, 8): the satisfaction that a consumer derives from the consumption of goods and services
value added (2): increase in the market value of a product at a particular stage of production; calculated by subtracting the value of all inputs bought from
other firms from the value of the firm’s output
vehicle currencies (15): currencies that are widely used to trade or save outside of the country that issued them
Volcker Rule (10): proposal by former Federal Reserve Chairman Paul Volcker to prevent banks that accept bank deposits from trading in securities markets
voluntary unemployment (4): includes those who do not wish to work at the current wage level
voluntary/involuntary unemployment (4): unemployment resulting from the fact that individual workers or trade unions ask for higher real wages than if the
market were perfectly competitive, which may be involuntary from the perspective of individuals
vulnerabilities (19): economic or financial conditions that make a self-fulfilling crisis possible
wage bill (13): the sum of all gross wages and salary payments paid by firms to workers plus fringe benefits as well as contributions to social insurance funds
(health care, pension, unemployment, etc.)
wage inflation (13): the annual rate of growth of nominal wages
welfare trap (18): situation in which payments from the welfare state discourage work, keeping recipients dependent on welfare payments
yield (7): the rate of return on an asset (measured in per cent per annum) based on the purchase price of that asset and the payments generated by the
ownership of that asset
yield arbitrage (7): concerns two assets which happen to offer different returns
yield curve (7): a curve which represents how the interest rate changes according to the maturity of assets
zero lower bound (10): a situation where a central bank has reduced its policy interest rate to 0%, and therefore cannot go lower
Index
A
absolute purchasing power parity
meaning of 561
type of PPP 136–7, 138
absorption
meaning of 39, 561
accelerator principle
investment and 208–9
meaning of 561
accounting identities
central role of 29
economics and 43–4
gross domestic product and 29–37
incomes/expenditures, flows of 37–44
meaning of 561
active labour market policies
meaning of 561
activist demand policies 423, 561
actual unemployment
equilibrium unemployment and 118
AD curve See aggregate demand curve
aggregate demand
goods market and 277–83
long run, in the 359–60, 371–2
meaning of 561
aggregate demand and supply, fixed exchange rate
AD curve, movements along/shifts of 361–2
aggregate demand curve/short run 360–1
aggregate demand/long run 359–60
complete system 362–3
demand disturbances/fiscal policy in long run 364
demand disturbances/fiscal policy in medium run 364–7
demand disturbances/fiscal policy in short run 363–4
fiscal policy under 365
LAD line 359, 362
monetary policy and realignments 367–9
relative purchasing power parity 359
short-run versus long-run equilibrium 362–4
stagflation 364–5
aggregate demand and supply, flexible exchange rate
AD curve, movements along/shifts of 373–4
aggregate demand/long run 371–2
complete system 374
Fisher equation 370
monetary policy/long run 374
monetary policy/medium run 375–6
monetary policy/short run 375
short-run aggregate demand curve 372–3
target/neutral interest rate 372
aggregate demand curve
meaning of 561
short run, in the 360–1, 372–3
aggregate demand management 15
aggregate production function
meaning of 561
technological progress and 76
aggregate spending
consumption/investment elements 194
real interest rate and 370
aggregate supply
meaning of 561
aggregate supply curve
factors that shift 351–2
inflation and 330
meaning of 561
Phillip’s curve/Okun’s law combined 338
animal spirits
investment and 210
meaning of 561
arbitrage
meaning of 561
no-profit condition and 182, 191
purchasing power parity and 136
spatial arbitrage 182–3
triangular 183, 391, 579
yield arbitrage 182
AS–AD model
demand disturbances 379–80, 431–2
disinflation policy 380–4
exchange rate regime 378
impulse-propagation mechanism in the 431–3
indexation effects 376
lags and time horizons 376
meaning of 561
persistence 432
propagation framework 432
supply shocks 376–9, 431–2
using the model 376–84
asset
bonds 178–9
credit default swap 180
derivatives 180
financial markets and 169
investment funds 180
meaning of 561
prices and yields 178–82
securitization 180
shares See stocks
stocks 179–80
asset markets
bonds 173
characteristics of 170–1
derivatives 180
durability of 170, 172
economic functions of 172–8
financial intermediaries 172–3, 190
forward-looking 172
goods and services distinguished
170, 171
implications of 171–2
investment funds 180
macroeconomy and 189–90
market efficiency See market efficiency
no-profit condition 172, 190, 191
overview 169
pricing the future 173–4
profitability 172
risk allocation 174–5
asset markets (continued)
risk diversification 176
risk return curve 177
risk, price of 174–5, 177
risk-return trade-off 176–8
securitization 180
shares See stocks
stocks 170
trading hours of 171
types of 170
volatility of 169–172, 190
well-organized trading platforms 171
yield curves 174
asset price channel
meaning of 561
monetary policy 261
asset prices
exchange rate as 395
volatility 169, 172
yields and 178–82
autarky
endowment and 146
meaning of 561
automatic stabilizer
budget as an 470
demand management policy 434, 445
fiscal policy 454–5
meaning of 561
recessions and 434, 466
average/unit costs 341, 344, 561
B
Bagehot Rules 269
bail-in 264
bailout 228, 562
balance of goods and services
meaning of 44, 562
balance of payments
capital accounts 44–5
current accounts 44–5
errors and omissions 47
examples of accounts 46–7
financial accounts 45–6
imbalances 511
meaning of 562
meaning of the accounts 49
National Income Accounts links 47
net borrowing/lending 45
various countries 49
balance on international income
meaning of 562
balance on primary international income
meaning of 41, 562
balance on secondary international income
meaning of 41, 562
balance sheet
meaning of 562
Balassa–Samuelson effect
meaning of 407–8, 562
bancor 521
band of fluctuation
‘gold points’ 511
meaning of 562
Vietnam 527
bank accounts 223
bank deposits
commercial banks and 224
meaning of 562
bank failure See bank run
bank notes
legal tender, as 226
bank regulation/supervision
confidence-building measures as 264–7
bank reserves
forms of 229
meaning of 562
monetary policy and 245
bank run 227, 562
banks
central See central banks
commercial See commercial banks
credibility/trust 226, 239
custodian of payments system 245
financial intermediation and 173
financially healthy 266–7
liquidity service 226–7
Barro, Robert 555
basis points 259, 562
Basel Regulation
capital adequacy and 266
battle of the mark-ups 343–4,
345, 562
beggar-thy-neighbour
meaning of 562
monetary expansion and 319–20
Beveridge curve
labour market, mismatch in 492
meaning of 562
bid–ask spread
foreign exchange markets and 183
meaning of 562–3
bimetallism
fixed parity between gold and silver 511
meaning of 563
bonds
consol 178
definition of 173
discounting 148, 178
financial market for 169
meaning of 563
perpetuity bond 178
prices and yield 179
selling of 178
yield 178
yield arbitrage 182
boom-and-bust cycles
currency crises 523–4
meaning of 563
borrowing
constraints 10, 203
consumption smoothing and 198, 201, 215
cost of 151
economics and sociology of 147, 149
European Debt Crisis 155
governments and 154
intermediation and 172
investment and 211
net borrowing See net borrowing
overview 144
prosperity and 169
public 158
rational expectations hypothesis 145–6
real cost of 132
sovereign 162–3
Bretton Woods Conference
definition of 563
International Monetary Fund and 570
new monetary order, preparation for 515
Bretton Woods International Monetary System
collapse of 518–19
fixed exchange rate system 515–19
gold exchange standard 517
International Monetary Fund reengineered 517–19
layers of 517
principles of 515–17
Triffin paradox 518
British terms
exchange rates and 132
meaning of 563
See also European terms
bubbles
generally 185–8
rational 191
speculative 185, 578
Buchanan, James 551
budget constraint
endowment and 146–8
intertemporal 146–9
nation’s 161–4
public See public budget constraint
public/private consolidated 156–7
budget line
meaning of 148, 563
budgets
budgetary process 455
budget figures, interpreting 456–7
cyclically adjusted 457
endogenous/exogenous components 457
Burns-Mitchell diagrams
business cycles, common features of 13–17
meaning of 563
variables and 434–5
business cycles
Burns-Mitchell diagrams 13–17
definition of 563
deterministic view of 566
electoral 436–7, 566
naming 18
partisan 436, 445, 574
political, 436–7, 575
real, theory of 576–7
stochastic view of 428–30, 578
trends, distinguished 4
C
Cambridge equation
definition of 563
demand for money 126, 137
capacity utilization
inflation rate changes and 10
meaning of 563
capital
depreciation, definition of 565
factor of production, as 8
meaning of 563
capital account
balance of payments and 44, 45
meaning of 563
capital accumulation
depreciation 66–7
gross investment 66
growth accounting 79
growth miracles 68
meaning of 563
net investment 66
saving schedule 66
savings for growth 68–70
savings/investment and 65–6
steady state 67–8
capital adequacy
meaning of 563
minimal 267
capital controls
capital mobility 307–9, 313, 324
definition of 563
free trade without 304–5
pros and cons of 529
capital depreciation
capital stock and 88, 215
definition of 563
rate of 81
technical progress 77
capital gain/loss
bubbles and 186
meaning of 563
uncovered interest rate parity and 391, 393
capital-labour ratio
definition of 563
economic growth and 62
capital loss See capital gain/loss
capital widening 73
meaning of 563
capital widening line 73
meaning of 563
catching up
convergence hypothesis 81, 407
emerging economies 509
Europe 556
when does it occur 67
central banks
Bagehot Rules 269
bank reserves 225, 239
banking, influence on 245–6
collateral 230, 248
commercial banks, direct lending to 249
currency production 225
derived demand 248
generally 225
hyperinflation and 252
inflation and 329
influencing monetary conditions, objectives for 250–1
interbank market 248
interbank interest rate 248
lender of last resort 267–8
monetary base 225
monetary policy 245–50
money creation, influence on 245–7
money market influence 247
money supply control 229–31, 239
non-performing loans 248
open market operations 246–9
payments system 245
refinance 249, 269
regulation/supervision by 263
reserve multiplier 231
reserve ratio 230, 239, 245, 252
reserve requirement, legal minimum 250
reserves 245
reserves-money stock link 231
setting interest rate 248
Taylor rule 257–9
central parity See parity
child care
female labour force, effect on 101
China
fixed exchange rates 316–17
foreign exchange reserves 50
permanent income hypothesis 201
circular flow
diagram 37–41
meaning of 563
Cobb-Douglas Production function 61
collateral
central banks and 248–9
control of money supply 230
definition of 564
collective bargaining
economics of labour unions 105–6
employment effects of 106–8
rationale for labour unions 104–5
collective labour supply curve
definition of 564
commercial banks
bailout 228
balance sheet of 227, 249
central bank borrowing 249
collateral and 248–9
financial intermediaries 225
generally 225–8
payment system custodians 225
sight-deposit accounts 229
commodity
high demands, effect on price 329
commodity curve
definition of 564
commodity money
bimetallism 563
edging out of 221
intrinsic value of 224, 564
competition policy
definition of 564
forms of 477
national preferences and 477
concepts of macroeconomics
factors of production 8
financial markets 10–11
income and output 5
income distribution 8
inflation 8–10
openness 11
unemployment 5–8
conditional convergence
definition of 564
missing inputs and 83–7
conditionality
definition of 564
IMF assistance and 520–1
confidence building measures
263–7
consol
definition 564
promise of fixed amount payment 148
‘ultimate bond’ 178–9
constant returns to scale
Cobb-Douglas function 61
definition 564
production function and 62, 63, 85
See also decreasing returns to scale
consumer price index
definition of 8, 33, 564
Italy 34
price levels and 19, 20
Paasche index 34
Zimbabwe and 128
consumer surplus
definition 564
loss of 498
consumption
definition of 564
function 194, 203–4, 215
households and 39
consumption (continued)
indifference curves, use of 195
intertemporal trade and 146–7
life-cycle consumption 199
optimal 195–7
real interest rate and 199–201
smoothing 198
temporary/permanent income changes 197–8
utility 195
wealth or income driven 201–3
consumption function
definition of 203–4, 215, 564
determinant of demand 278
macroeconomic fundamental tool 194, 204
consumption-leisure trade-off
definition of 564
indifference curves 93
labour supply and 93–7
utility 93
definition of 198, 564
fiscal policy 452
Greece/Germany 1995-2016, 200
rational consumers and 215
contagion
bank failure and 265, 267
definition of 564
contractionary
aggregate demand, effect on 384
central bank 297
definition of 564
fiscal shock 363
monetary policy and 294–5, 372, 378, 442
revaluation 314
convenience function
definition of 564
money and 220, 239
convergence hypothesis
2% convergence rule 83
conditional convergence 83–4
economic growth and 81–3
explained 564
human capital inputs, effects of 84–7
coordination failures
decentralized markets 20
definition of 564
core inflation rate
definition of 564
countercyclical
definition of 564
unemployment rate and 10, 15, 16
wage share, nature of 345
countercyclical fiscal policy
benefits of 470
budgets 456
corrective device 454
definition of 565
Keynesianism and 367
covered interest rate parity (CIRP)
capital loss 392
definition of 565
forward discount 392
riskless yield arbitrage and 391–3
crawling pegs 531
credit
financial transaction, as a 44
sociology/economics of 147
credit channel
definition of 565
monetary policy channel, as 261
credit default swap
definition of 565
important functions of 181
type of derivative 180
credit rationing
consumers and 202
definition of 565
output/employment stabilization 453–4
restrictions on borrowing 160
currency
appreciation of 132
bid-ask spread 183
boards 531–3
circulation, in 221–2, 245
common 378, 533
control of 239
convertibility 515
crises See currency crises
definition of 565
depreciation 50, 124, 137
domestic 132–3, 312
exchange rate regimes 309–10
foreign 132–3, 312
GDP measures 37
international 134
nominal exchange rate 137
paper cash/metal coins 221, 223
parity change 314–15
pegging 316, 380
monetary base 225, 230
money 223
seigniorage and 466
vehicle 389
currency boards 531–3, 541
currency crises
boom-and-bust cycles 523–4
contagious feature of 539
definition of 565
exchange rate volatility and 411–12
fundamentals, driven by 524–5
Iceland fundamental crisis 525
impossible trinity principle 523, 530
non-fundamentals 526
self-fulfilling attack 526
South-East Asian Crises 1997-1998, 2007-2008, 527
speculative attacks 526
types of 539
vulnerability, pre-existing 526–8
currency depreciation
avoiding 50, 124, 137
current account
components of 44–5
definition of 565
primary current account 401–3
primary current account function 402
current account openness See trade openness
current disposable income
aggregate consumption and 215
cyclically adjusted budgets 457, 565
D
deadweight loss of taxation
definition of 565
market efficiency and 487–8
debt stabilization
deficit reduction 457, 463–4
definition of 565
medium to long term 471
public debt 459–63
strategies for pursuing 457–8
decision lag 433–4, 565
deflation
definition 565
economic slowdown 512
optimal rate of inflation 429
demand disturbance/shock See demand shock/disturbance
demand for money
Cambridge equation and 126
definition of 565
demand function
desired 565
demand management policies
activist demand policies 423
automatic stabilizers 434
business cycles, stochastic view of 428–30
deterministic economy 428
dynamic stochastic general equilibrium models 439–40
electoral business cycles 436–7
equilibrium/disequilibrium 422–3
exchange rate anchor 425
Friedman critique 433–6
Great Recession 440–4
impulse-propagation mechanism 430–1
income and wealth distribution 424–6
inflation, cost of 424–6
inflation, underlying 423–4
Keynesian view 422–4
lags 433
median voter 436–7
neoclassical case 422–4
optimal inflation 429
partisan business cycles 436, 445
partisan politics 437
persistence of expectations 423–4
political business cycles 436–7
political constraints on 436–7
price signals 427–8
rational expectations hypothesis 423–4
real business cycle theory 440
uncertainty, effect of 433
demand shock/disturbance
definition of 565
fixed exchange rate, under 311, 381
flexible exchange rate, under 318–19, 381
Great Recession 437–8, 444
shifts of the IS curve 318–19
demand side
definition of 565
deposit insurance
confident-building measure 263–4
definition of 565
depreciation
capital accumulation and 66–7
nominal exchange rates 133, 136
depreciation line 66–7, 70, 565
depreciation rate
capital accumulation 66
investment and 207, 208
low inflation and 137
derivatives
asset, as 180–1
definition of 565
derived demand
definition of 565
money market equilibrium 235, 239, 248
reserves and 252
desired demand function
definition of 565
goods market equilibrium and 280
devaluation
definition 566
exchange rate changes 314
Diamond, Peter 553–4
dichotomy principle
definition of 130–1, 566
Keynesian assumption and 276
money neutrality principle and 137
Cobb-Douglas Production Function 61
economic growth and 62, 63,
75, 77
intertemporal budget constraint 150
principle of 566
direct investment
financial accounts 44, 47
discounting
bond prices and 147–8, 178
definition of 566
disinflation
definition of 566
macroeconomic equilibrium 380–4
output cost of 381
distortionary taxation
definition of 566
losses to the economy 488
diversification
definition of 566
risk reduction and 175–7, 180
dollarization
currency boards and 531–3
dollarization/euroization 532–3
Dutch disease 405, 502
dynamic inefficiency/efficiency
definition of 566
golden rule and 71–2
dynamic stochastic general equilibrium (DSGE) models 439–40, 445, 566
E
East Germany
current income and spending 204
economic agents
definition of 566
demand side and 20, 565
firms 20, 477
governments as 436
households 20
rationality principle and 145
Economic Freedom Index 86
economic growth
aggregate production function 60–3
capital accumulation/depreciation 65–7
diminishing marginal productivity 62
dynamic inefficiency 71
dynamically efficient 71
golden rule 70–2
growth accounting 78–81
Kaldor’s stylized facts 63–5
logarithmic scale and 12
long run macroeconomics 12–13
meaning of 566
overview 58–9
phenomenon of 59
population growth and 72–5
returns to scale 62
rich/poor countries, reasons for 81–6
savings/investment 65–6
savings, role of 68–70
Solow growth model 78–81
sources of 60–3
steady state 65, 67–8
stylized facts of 63–5
technological progress and 75–7, 80
economic rents
imperfect competition 476–8
definition of 566
education/training
human capital and 84–5
effective exchange rate
computing and comparing 135
definition of 566
nominal and real 134
effective labour
definition of 566
technological progress and 76–7
effectiveness lag 433, 566
efficient market hypothesis
asset prices and 182
definition of 566
implication of 183–4
efficiency wages
definition of 566
real wage rigidity and 110, 119
elastic
definition of 566
labour supply 119
money supply 237
electoral business cycles 436–7, 566
employers’ associations
definition of 566
unions and 103
employment growth
monetary policy objective 251–2
endogenous variables
exogenous distinguished 21
meaning of 21, 566
interest rate 283
endowment
budget constraint and 146–8, 152–3
definition of 566
exogenous 146–9
output and 92
Solow model and 81, 94, 100
equilibrium GDP
definition of 567
goods market equilibrium and 280–1
equilibrium real exchange rate
anchor, as an 408
Balassa-Samuelson effect 407–8
misalignment 405
natural resources 404
net external position 404
non-price competitiveness 403–4
primary current account and 401–3
primary current account function 402
stable or constant view of 407–8
undervalued/overvalued 403
equilibrium unemployment
actual and 118
concept of 113–15
definition of 567
rate of 567
stocks/flows and 111–112
equilibrium unemployment rate
actual and equilibrium unemployment 118
concept of 113–15
European experience 115–18
equity-efficiency trade-off
definition of 567
redistributive goals and 450
‘equity stacks’ 147
EU Competition Policy 477
Euro
role of banks in creating 226
monetary neutrality, evidence for
127
Euronext 144, 170
Europe
competition policy 477–8
market hours at work 101
privatization and 484
unemployment and 475, 496
European Central Bank
commercial banks, financing 249
interest rates 250
monetary policy under duress 298
European Debt Crisis 155, 162
European economists 25, 549, 552, 558
European labour market policy 117–18
European Monetary Union 536–7, 567
European terms
definition of 567
European unemployment
oil shocks and 115–16
supply-side problems and 501
Eurozone
fixed exchange rate regime 304
excess demand/supply
definition of 567
economy off IS curve 284–5
higher prices 278
IS-TR model 293
labour market equilibrium 99
lasting 547
production and 281
excess supply See excess demand/supply
exchange market intervention
definition 567
foreign See foreign exchange market intervention
sterilized 312
unsterilized 312
exchange rate 11
anchor 408, 567
asset price, as an 391
capital loss 392
currency crisis 411–12
definition of 567
depreciation 565, 567
dollar 137
effective 566
equilibrium real exchange rate See equilibrium real exchange rate
fixed See fixed exchange rate regime
flexible See flexible exchange rate regime
floating exchange rates 400
foreign exchange rate See foreign exchange rate
forward 390, 568
forward premium 393
fundamentals 399–400
interest parity conditions See interest rate parity conditions
long run, determination in the 401–8
long run, in the 137–8
monetary authorities and 50, 53
monetary policy 21
Mussa’s stylized facts 397
nominal See nominal exchange rate
nominal, fundamental determinants of 399–401
overshooting 399
overview 388
primary current account 401
purchasing power parity See purchasing power parity
real interest rate arbitrage 394–5
real See real exchange rate
risk premia 394
short run, determination in the 395–401
spot 578
targeting 257
volatility and predictability 30, 411–12
exchange rate anchor 320, 408, 567
Argentina and Bulgaria 425
exchange rate appreciation 132, 318–19, 561
exchange rate depreciation 567
exchange rate overshooting 399, 567
exchange rate parity
central banks 314, 325
realignment and 577
exchange rate regime
choices See exchange rate regime choices
definition of 567
fixed See fixed exchange rate regime
flexible See flexible exchange rate regime
Mundell-Fleming 304–5, 320
exchange rate regime choices
capital controls, pros and cons 529
capital mobility/exchange rates linkage 530–1
currency boards 533
fixed exchange rate See fixed exchange rate regime
fixed or flexible 528–30
flexible exchange rate See flexible exchange rate regime
hollowing-hypothesis 531
monetary unions 533–7
optimal sequencing 531
Tobin Tax 530
exchange rate targeting
exogenous variables
definition of 566
endogenous distinguished 21
goods market and 283
expansionary monetary policy
fixed exchange rate regime, under 368
gold standard and 512
monetary disturbance 295, 317–18
shocks 567
short run/long run effects 375
expected inflation targeting 21, 254, 567
export function
definition of 567
determinant of demand 279–80
external competitiveness
definition of 567
exchange rates and 146, 315,
400, 528
inflation and 365
external terms of trade
definition of 567
real exchange rate 135
externalities
definition of 567
fiscal policy and 449
health 483–4
human capital, investment in 482
law and order 482–3
pecuniary/non-pecuniary 480
positive/negative 449, 478, 575
product market policies and 482–4
property rights 479
F
factors of production
definition of 567
households 37, 39
income distribution and 8
valued added 31
female labour force
child care, effects of 101
participation rates 102
fiat money
definition of 568
legal tender, as 221
modern systems 229
final sales
definition 568
intermediate sales contrasted 30
financial account
categories of transactions affecting 44, 47
definition of 568
financial intermediation
asset markets and 171–2
commercial banks 225
meaning of 39, 568
financial markets
concept of 10
financial openness/financial account openness 305
financial pyramid schemes 163–4
financial stability
Bagehot rules 269
bail-in 264
bank regulation/supervision 264–6
bankers’ mistrust 264
Basle Regulation 266–7
capital adequacy 267
confidence-building measures 263–7
contagion 254
deposit insurance 263–4
financial healthy banks 266–7
fractional reserve banking 263
information asymmetry 265
lender of last resort 267–8
macroprudential measures 265–6
monetary policy prerequisite 262–9
public good and 263
recapitalization 267
systemic risk 265
technological innovation and
268–9
Volcker rule 269
finding rate
definition of 568
job 111
fiscal policy
automatic stabilizers 454–5
budgetary process 455–6
consumption smoothing 452
countercyclical 454, 456,
459, 470
definition of 568
demand management instrument, as 20
disincentive effect 451
economic welfare and 449–51
equity-efficiency trade-off 450
externalities and 449
inequalities reduced by governments 450
lessons for the future 442
output/employment stabilization 453–4
positive externalities 449
productive efficiency 450
productivity efficiency 450
public goods/services, provision of 449–50
redistribution goals 450–1
tax smoothing 452
Fisher principle/equation
aggregate demand 370–1
definition of 568
fixed capital formation 568
firms and 150
See investment
fixed exchange rate regime
aggregate demand/supply See aggregate demand/supply, fixed exchange rate
China and 316–17
demand disturbances 311, 313
demand shocks 313
devaluations 314–15
Eurozone 304
exchange market interventions 310
exchange rate anchor 320
flexible or 320–4, 528–30
foreign exchange reserves 312
IFM line shifts 313–14
international financial disturbances 313
international financial flows and 309
international financial shock 313
IS curve shifts 311–12
monetary policy autonomy, loss of 310–11
money markets under 311
output and interest rate determination under 310–15
parity change 314–15
realignment of policy 315
revaluations 314–15
sterilized foreign exchange intervention 312
TR Curve 310–11
unsterilized foreign exchange intervention 312
what is the regime 310
flexible exchange rate regime
aggregate demand/supply See aggregate demand/supply, flexible exchange rate
beggar-thy-neighbour effect 319–20
definition of 528–9
demand disturbances under 318–19
depreciation 317–18
exchange rate appreciation 318–19
fixed or 320–4, 528–30
IFM line shifts 319–20
international financial disturbances 319–20
IS curve shifts 318–19
monetary policy disturbances 317–18
output and interest rate determination under 315–20
TR curve shifts 317–18
floating exchange rate
flow variable
stock variables contrasted 29
flows and stocks
definition 568
forecasting 23–4
foreign exchange market interventions
definition of 568
monetary authorities and 48
monetary side effects of 310–12
sterilized 272, 312
unsterilized 312
foreign exchange rate
forward exchange rate 390–1
instruments of 390–1
main characteristics of 389–90
triangular arbitrage 391, 412
spot exchange rate 390
vehicle currencies 389
foreign exchange reserves 312, 568
foreign/international rate of return
definition of 568
foreign reserves account balance
definition of 568
forward discount
covered interest rate parity and 393
definition of 568
forward exchange rate
definition of 568
foreign exchange rate and 390–2
forward exchange rate premium
covered interest rate parity and 393
definition of 568
internationally mobile capital and 412
fractional reserve banking
inherent instability of 263
France
gross domestic product 1870-2015, 12
free banking
definition of 568
Scottish banks and 237
freshwater economists
saltwater economists contrasted 25
frictional unemployment
concept of 115
definition of 568
Friedman critique 433–6, 444, 568
Friedman, Milton
‘Chicago School’ 545–8
Friedman critique 433–6
optimal inflation 429
permanent income hypothesis 199
Phillips curve 338
speculation 184
Frisch, Ragnar
business cycles and 428
fundamentals
definition of 568
G
GDP See gross domestic product
GDP deflator
calculating 33
definition of 568
GDP per capita
China 60, 201
definition of 568
economic growth and 58, 68
Greece 504
GDP per capita (continued)
life satisfaction and 5–7
population growth and 74–5
price levels and 409
small countries and 30
general equilibrium
concept of 273, 277, 299
definition of 568
demand disturbances, effect of 318
flexible prices and 330–3
See also dynamic stochastic general equilibrium
Germany
reunification 379
unemployment 117–18
global imbalance
budget constraints and 162
definition of 569
globalization
asset markets and 171
openness and 305
gold exchange standard
Bretton Woods system and 517
definition of 569
Triffin paradox and 518, 520
gold standard
automatism, limits of 513
benefits of 512
bimetallism 511
definition of 569
dollarization/euroization 532
domestic operation of 510
ephemeral 515
fixed exchange rate 510–11, 538
Gresham’s Law 511
Hume mechanism 511–12
international operation of 510–11
inter-war period 514
metallic standard, limits of 513
golden rule
capital accumulation/economic growth 70–2
definition of 569
dynamic inefficiency/efficiency 71–2
goods market
aggregate demand and 277–83
consumption function 278
demand determinants 278–80
desired demand function 280
endogenous/exogenous variables
283
equilibrium 280–1
equilibrium GDP 280
export function 279
investment function 278
IS curve and the 283–5
Keynesian demand multiplier 281–3
goods market equilibrium 280–1, 569
Gordon, Robert J 556–7
Great Recession 2008–2009, 18, 25, 238, 293, 297–8, 440–2, 468
Gresham’s law
across countries 36–7
definitions of 5, 29–30, 569
final sales 30
flow variable 29
household disposable income 42
income and output 5
intermediate sales 30
interpreting 33–4
monetary policy objective 251–2
measuring 33–4
nominal GDP 31
price deflators 33
price indices 33
real GDP 32
underground economy 35–6
value added 30
value subtracted 31
what it measures 32
gross disposable national income 41, 42
gross investment
definition of 569
depreciation/capital stock and 151
net investment distinguished 66–8
gross national income (GNI)
concept of 41, 47
definition of 569
net national income (NNI) and 573
growth accounting
capital accumulation 79
employment growth 79–80
Solow’s decomposition 78–9
technological changes 80–1
growth traps
definition of 569
poor health services 85
poorer countries and 81–2
H
Hansen, Alvin 556–7
hard pegs
currency boards 533
definition of 569
fixed exchange rate arrangements 383–4
varieties of 531–2
headline deficit 154
health
externality, as 483
human capital and 85
human/property rights 86–7
public infrastructure and 85
social infrastructure and 85–6
hedging
definition of 569
risks 181, 392
‘helicopter money’ 444, 574
herds 185
hollowing-out hypothesis
capital controls 531
definition of 569
household labour supply curve
definition of 569
wage rise effect 95
human capital
definition 569
education 84–5
effects 84–7
fiscal policy and 449
health 85–6
positive externality, as 449, 482
human rights
definition of 569
productivity and 86, 87
Hume mechanism
‘gold specie mechanism’ 511
balance of payment imbalances and 511–12
credibility and 512
definition of 569
hyperinflation
Central Banks and 252
concept of macroeconomics, as 10
definition of 569
demand management policies and 426
Zimbabwe and 128
I
IFM line See international financial markets (IFM) line
immigration
supply shock, as 117
imperfect competition
economic rents and 476–8
implementation lag 433, 569
import function
definition of 127–8, 569
determinants of demand 279
impossible trinity
definition of 569
IS-TR-IFM model, implications of
523–25, 529, 530–1
impulse-propagation mechanism
AS-AD model, in the 430–3
definition 569
demand shocks 431–2
persistence 432
propagation framework 432
supply shocks 431–2
income and output
concept of macroeconomics 5
primary current account and 299
underground economy 35
income effect
definition 569–70
higher wages 95–6, 100
income/expenditure flows
circular flow diagram 37–9
indebtedness
explosive process 470
government 158, 298, 460–2
inheriting 149
public 420, 457
index
consumer price 4, 8, 20, 33, 94,
128, 347
Economic Freedom Index 86–7
Laspeyres 34
Paasche 34
price 33, 34, 133, 296
producer price (PPI) 34
share price 9
stock market 8
wholesale 34
indexation
clauses 347
inflation and 383, 428
schemes 376, 383
wage/price 376, 422, 428
indifference curves
consumption-leisure trade-offs 93
definition of 570
optimal consumption and 195–6
trade unions 105
individual labour supply schedule
household labour supply curve 95
household wage increase 95
income effect 95
labour force participation 95
substitution effect 95
industrial policies
concept of 486
definition of 570
inelastic
definition of 570
labour demand/supply 98, 119, 498
inflation
aggregate supply 351
aggregate supply curve 330
average/unit costs 341
‘battle of the mark-ups’ 343–4
capacity utilization rate 10
central banks and 329
concept of 8–10
definition of 570
differential See inflation differential
hyperinflation 10
income and wealth redistribution 424–6
labour costs 343
market power 342
mark-ups, cyclical behaviour of 345–6
monetary neutrality and 329
monetary policy and 251–2
neoclassical assumption in the long run 330
Okun’s law See Okun’s law
optimal 429
Phillip’s curve See Phillips curve
price change by firms, when and how 342
price costs 341
price setting 340–1
prices as mark-up on labour costs 343
product differentiation 342
product differentiation 342
productivity/labour share 344–5
rate See inflation rate
short to long run 352–3
supply shocks 347–8
targeting See inflation targeting
underlying inflation and the long run 349–51
underlying inflation rate 346–7
wages as mark-up on prices 343–4
inflation differential
definition 570
foreign/domestic differences 137
inflation rate
core 564
procyclical nature of 10
target 257, 579
underlying 564
inflation targeting
definition of 570
monetary policy and 254–7
inflation-targeting strategy 384, 386, 570
inflation tax
definition of 570
seigniorage and 464–6, 470
information asymmetry
bank regulation/supervision 265, 267, 268
definition of 570
insiders/outsiders workers
distinction between 570
unions and 107–8
installation costs
definition of 570
geometry of 212–14
Tobin’s q 212
instruments
definition of 570
financial 171
monetary policy 245–50, 251–7
monetary policy instrument, as 248, 254, 296
definition of 570
interbank market
definition of 570
mistrust in the 264, 265
See money market
interest rate
determination under fixed exchange rate See fixed exchange rate regime
determination under flexible exchange rate See flexible exchange rate regime
domestic 399–400, 412–13, 523–4
foreign 306, 399–400, 410, 413,
523
negative 236
nominal 299, 309
parity See interest rate parity condition
procyclical behaviour of 17
real 299
target See target/neutral interest rate
term structure of 175
interest rate channel
interest rate parity conditions
bridge to exchange rate in the long run 408–10
capital controls and 308–9
capital loss 392
covered/hedged 391–3, 565
definition of 570
exchange rate determination in the short run 396–400
exchange rate overshooting 399
forward premium 393
IFM line 307
implications of the 396–9
international financial flows and 306–7
international Fisher equation 395
nominal 132
real 132
real interest rate arbitrage 394–5
risk premia 393
short-run open economy 400
uncovered 393
intermediate sales
final sales contrasted 568
GDP and 30
internal terms of trade
definition of 570
external terms of trade contrasted 135
international borrowing
budget constraints, violations of 163–4
monitoring of 164
past 413
sovereign borrowing distinguished 164–5
international financial disturbances
fixed exchange rate regimes
313–14
flexible exchange rate regime
319–20
international financial flows
arbitrage 306
capital controls 304
capital mobility 307–9
exchange rate regimes 309
financial account liberalization 308
financial account restrictions 307–9
fixed exchange rates See fixed exchange rate regime
flexible exchange rates See flexible exchange rate regime
general equilibrium 310
IFM line 307
interest parity condition 306–7
international rate of return 306
market equilibrium 307, 309
international financial markets (IFM) line
definition of 570
shifts of 313–14, 319–20
international Fisher equation 395, 570
See also Fisher principle/equation
International Monetary Fund
assistance 520–1
conditionality 518, 520–1
definition of 570
history of 509
management of 522
monitoring international borrowing 164
quotas 517, 522
reengineered 517–19
special drawing rights 521
surveillance 521–3
voting rights 522
internet
measurement of GDP 32
new industrial revolution 78
single global market, creation of 171
unemployment and 99
intertemporal budget constraint, firms
capital stock 151
cost of investment 150
depreciation 151
fixed capital formation 150
gross investment 151
opportunity cost 150
production function 150
productive technology 151
unproductive technology 152
wealth, effect of 152
intertemporal budget constraint, household
autarky 146
budget line 148
consumption and 146–8
discounting 147–8
definition of 571
endowment 146–8
indebtedness, effect of 149
inherited wealth, effect of 149
intertemporal price 147
intertemporal price 147
present discounted value 149
real interest rate 147–8
intertemporal budget constraint, private sector
wealth gain 152–3
intertemporal price
budget constraint 147
definition of 571
intertemporal trade
consumption and 146–7
definition of 571
real interest rate 147–8
interventionism
definition of 571
laissez-faire compared 15
investment
accelerator principle 208–9, 215
aggregate demand component 204
capital depreciation and 215
examples of investment goods 204
firms 150
marginal cost 205–6
marginal productivity of capital 205–6
opportunity cost of 205–6
optimal capital investment 207
optimal capital stock 204–5, 206
rate of depreciation 207
real interest rate 208
user cost of capital 207
investment function
definition of 571
determinant of demand 278
macroeconomics and 214–15, 216
investment funds
definition of 571
involuntary unemployment
definition of 571
real wage adjustment and 102–3, 119
voluntary See voluntary unemployment
Ireland
supply-side reforms 502–4
IS curve
demand disturbances and 311–13, 318–19
economy off the 284–5
excess demand/supply 285
explained 283–4, 571
goods market and 283–7
movements along 285–7
real disturbances and 293–5
shifts of the 285–7, 318–19
IS–TR model
macroeconomic equilibrium in the 293, 295–7
monetary policy/fiscal policy in the 297–8
Italy
Mussolini and the public debt 467
public/private borrowing 159
unemployment problem 107
J
Jevons, William 221
job finding
duration of unemployment 113
job-finding rate
definition of 571
unemployment and 111–13
job matching
definition of 571
efficiency of 492
worker 113
job separation
incidence of unemployment and 112–13
K
Kaldor’s Stylized Facts of Economic Growth
steady state and 65
stylized fact No 1, 63–4, 65, 75, 76
stylized fact No 2, 64, 75, 77, 208
stylized fact No 3, 64
stylized fact No 4, 64, 99
stylized fact No 5, 64–5
Keynes, John Maynard 420, 543–5
Keynesian assumption
definition of 571
inflation, incorporating into model 334
neoclassical assumption distinguished 330–1
prices and 276
sticky prices 330
supply determined outcome under 331–2
wages sticky 331
Keynesian demand multiplier
definition of 571
goods market 281–3
Keynesian revolution 543–5
Keynesianism
definition of 571
monetarism contrasted 25
Kydland, Finn 549
L
labour
definition of 571
factor of production, as 8
labour demand
definition of 571
labour demand curve 97–8
marginal productivity of labour 97
real wages 97–8
shifts in 98–9, 100
technical change and 99
labour demand curve
elastic/inelastic 98
labour force
labour force participation
definition of 571
men 95–6
women 95–6, 102
labour market
states and transitions 110
unemployment and See unemployment
labour market equilibrium
labour supply/demand interaction 99–101
trade union and 106
labour market institutions
definition of 571
labour markets and 103
labour market policy
active labour market policy 492
Beveridge curves 492
consumer/producer surplus 498
dismissal, regulation of 495–6
expenditures on programmes 494
heterogeneity 491
imperfect contracts 493–6
imperfect information 491–2
imperfect mobility 494–5
job matching 492
labour market regulations 493–6
labour tax wedge 498
labour taxation 497–8
social policies incentives 496–9
social safety net 496–7
taxation 496–9
welfare trap 497
labour share
definition of 571
exogenous increase of 348
productivity 344–5
share of income in manufacturing 8–9
wage mark-up 344
labour supply
aggregate labour supply 96–7
consumption-leisure trade-off 93–7
demand See labour demand
elasticity of 119
households, reaction to wage increase 95
individuals See individual labour supply schedule
person-hours 96
labour tax wedge
definition of 571
labour unions
collective bargaining 103–8
economics of 105–6
European 104
labour markets and 103
rationale for 104–5
LAD line
definition 571
fixed exchange rate 362, 378
flexible exchange rate 372, 378
monetary policy strategy, determination by 372
Laffer curve
definition of 571
theory and reality 489–90
laissez-faire principle
definition of 15, 20, 571
freshwater economists and 25
interventionism contrasted 15
market efficiency and 476
output and employment stabilization 453
large economies
closed/open 320
Laspeyres index 34
law and order 482–3
Law of One Price
definition of 571
purchasing power parity and 136, 138
Layard, Richard 552–3
legal tender
banknotes 226
credibility and 237
dollarization 532
euro banknotes 127
fiat money 221
Lehman Brothers collapse 23, 379, 443
leisure
consumption and 94, 95, 99, 102
definition of 572
lender of last resort
central banks, as 267–8
definition of 572
lending
bonds 178
economics/sociology of 147
foreign 164
individuals 149
mortgages 199
net lending 45, 48, 50
overview of 144
risk and 160, 175
stocks 179
US Financial Crisis 2008–2009 181
life-cycle consumption
permanent income 199, 201
principle of 572
liquidity service
definition of 572
liquidity squeeze 226
LM curve 290–1, 572
logarithmic scale
definition of 572
London Club 164
long-run equilibrium
short-run equilibrium distinguished 362–4
definition 572
M
macroeconomics
aggregate activities 21
business cycles and 13–15
concepts of 5–11
definition of 572
demand and supply 20–1
Europe as birthplace of 557–8
genesis of 15–20
Keynesian Revolution 543–5
long run, in the 12–13
methodology of 21–4
microeconomics and 20
microfoundations of 549–50
neoclassicals See neoclassicals, neoclassical economist
New Keynesian 550
overview of 4
science, as a 15–20
short run, in the 13–15
macroeconomic equilibrium
forecasting 23–4
general approach to 295–7
IS–TR model, use of 293–7
micro foundations of 549
modelling 22–3
monetary policy disturbances/TR curve shifts 295
new Keynesian 550
real disturbances/IS curve shifts
293–5
macroprudential
definition of 572
measures 265–6
malfunctioning markets
industrial policies 486
monopolies 484
privatization 484
regulation 484–5
subsidies 485–6
trade policies 486
marginal cost of capital
definition of 572
investment 207, 213–5, 251
marginal productivity of capital
installation costs and 216
optimal capital stock, relationship with 574
marginal productivity of labour
definition of 572
labour input increase and 97
real wage and 119
marginal rate of intertemporal substitution
consumption and 196
description of 572
marginal rate of substitution
consumption-leisure trade-off and 94
definition of 572
market-clearing
definition of 572
market efficiency 476
neoclassicals and 444
structural unemployment and 115
market efficiency
arbitrage 182–3
bid-ask spread 183
information and 182–9, 191
irrational behaviour 184–5
market-clearing 476
speculative manias and 184–9
market equilibrium assumption 277–8
market failures
definition of 572
externalities 478– 80
increasing returns 480
information asymmetries 480–2
public goods 480
market interest rates
market liquidity
definition of 572
market makers and 183
market maker
definition of 572
financial institutions as 183
market power
definition of 572
firms and 476–7
mark-up pricing and 342
minimum wages and 108
product differentiation and 342
mark-up pricing 342, 572
‘mattress operation’ 236
maturity
consol and 178
definition of 572
derivatives and 180
loans and 173
maturity premium
definition of 572
loans and 173
maturity transformation
business model of banks 226–7, 239
definition of 572
media 27
median voter 436–7
median voter theorem 573
microeconomics
consumption and 195
macroeconomics contrasted 20
perfect competition and 438
production function and 61
minimum wages
definition of 573
market power and 494
reasons for enactment 108–9
youth unemployment and 109, 119
misalignment
definition of 573
exchange rate 412, 504, 519
pricing 182
monetarist
derivation of term 25
Keynesians contrasted 25
Monetarist Revolution 545–8
monetary aggregate 223, 573
monetary arrangements, international
Bretton Woods See Bretton Woods International Monetary System
currency crises See currency crises
exchange rate regime See exchange rate regime choices
gold standard See gold standard
inter-war period See monetary arrangements, inter-war
International Monetary Fund See International Monetary Fund
monetary unions See monetary unions
monetary arrangements, inter-war
ephemeral gold standard 515
free float period 514–15
managed float 515
monetary base 225, 573
monetary disturbance/shock
definition of 573
TR curve shifts 295
monetary economy
definition of 573
real economy contrasted 10
monetary neutrality principle
assertion of 55
definition of 573
evidence for 127
exchange rates and 132–5
hyperinflation 127
money and 125–6
money and prices 126–9
nominal interest rate See nominal interest rate
overview of 124
principle 576
purchasing power parity 136–7
real interest rate See real interest rate
monetary policy
central banks and 245–7, 270
definition of 573
demand management instrument, as 20
disturbances 295
duress, under 296
expansionary 295
financial stability See financial stability
floating exchange rates and 400
GDP growth 251
high employment 251
information asymmetry 265, 267, 268
innovation and 270
instruments of 251–3
lessons for the future 442–4
low and stable inflation 251
money multiplier 253
neutrality principle 251, 254, 270
objectives of 250–1
open market operations 249, 270
preventing crisis 270
price stability 251
refinance 245, 249
strategy, choice of 252
targets of 251–7
Taylor rule 257–9, 270
TR curve and 289–92
transmission channels 259–63
monetary policy autonomy
definition of 573
loss of 310–11, 314–15, 533
swiss franc and 322–3
monetary policy shock
definition of 573
expansionary 567
monetary targeting
definition of 573
monetary policy type, as 289
strategy of 253–4
monetary unions
European Monetary Union 536–7
N–1 problem 533
optimum currency areas 534–5
TARGET-2 controversy 535
money
banking/credit creation, as a by-product of 237
Cambridge equation 126–7
commodity monies 221
creation process 228–9, 245–7
definitions of 221–3, 239
demand for 233–5
fiat 221
free-banking 237
hyperinflation 127–8
interest rate 233
market See money market
medium of exchange, as 125
multiplier 229, 232
necessity for 137
negative interest rates 236
payment systems, new innovations 239
positive 238
prices and 126–9
prices and output 130–1
private convenience, as 236–7, 239
public good, as 237, 239
sight deposits 223, 239
wealth, as form of 221, 239
money growth targeting
definition of 573
money illusion
definition of 573
workers/firms suffering 338
money market
definition of 573
demand for money 233–5
derived demand 234
equilibrium 235
interbank interest rate 233
short-run equilibrium 231–3
TR curve and 289
money multiplier 229, 232, 253, 573
money neutrality principle
assertions of 251
dichotomy principle and 137
monopolies 477, 484, 505
Mortensen, Dale 553–4
mortgage securitization
US Financial Crisis 2008-2009, 181–2
Mundell-Fleming model
definition of 573
small open economies and 304–5, 320
Mussa’s stylized facts
floating exchange rates 397
N
N-1 problem
definition of 573
monetary union and 533–4
national income accounts
central role of 29
variations of estimates 35
natural interest rate
definition of 573
natural monopoly
definition of 573
NDP See net domestic product (NDP)
negative externalities
costs imposed on society 478, 505
definition of 575
negative interest rates
zero lower bound 236
neoclassical assumption
definition of 573
flexible prices 330
Keynesian assumption distinguished 330–1
monetary neutrality 331
nominal wages/prices adjustment of 332
output/labour market 333
supply determined outcome 331–3
neoclassicals, neoclassical economists
assumption See neoclassical assumption
cost of inflation 424–5
definition of 573
demand management 422–3, 428
general equilibrium with flexible prices 330–3
growth model 555
Keynesian economists and 329, 330–1, 421–4, 428, 438–9, 445, 544
Keynesian economists and
long run and 330–2
market-clearing 444
rational expectations 423–4
real business cycle research and 549
right-leaning governments 436–7
synthesis 545
uncertainty and 445
net borrowing 44, 45, 48
net domestic product (NDP)
net export function 279–80
net exports 573
net investment
definition of 573
net lending 44, 45, 48, 49, 50
net taxes
definition of 573
taxes/transfers, difference between 38
Netherlands
minimum wages 109
supply-side reforms 502–3
neutral interest rate See natural interest rate
neutrality of money See monetary neutrality principle
New York Mercantile Exchange (NYMEX) 145
Nickell, Stephen 552–3
NN1
definition of 573
no-profit condition
arbitrage See arbitrage
definition of 573
stocks 179
uncovered interest rate parity 393, 412
noise traders
definition of 574
market efficiency and 185
nominal disturbance
real contrasted 577
nominal exchange rate
appreciation of currency 132
British terms 132
depreciation 133
effective exchange rate 134
European terms 132
movements in 133–4
nominal GDP
calculating 31
definition 574
German 37
real distinguished 31–2
nominal interest rate
definition of 574
Fisher equation 370–1
monetary neutrality and 131–2
real interest rate distinguished 132, 370–1
zero lower bound 296
non-excludable
definition of 574
public goods 449, 480
non-pecuniary externalities
definition of 575
government intervention 505
training and education 480
non-performing loans
banks and 248
banks’ trust 264
definition of 574
recession and 442
non-price competitiveness
definition of 574
real exchange rate determinant 403–4
non-rivalrous
definition of 574
public goods 449, 487
non-standard/unconventional policies
definition of 574
quantitative easing 576
normative analysis
methodology of macroeconomics 22
normative economics
definition of 574
Northern Rock, bank run 228
O
objectives
central banks 291–2
definition of 574
oil shocks 339, 574
Okun’s law
definition of 574
inflation/output relationship 335
unemployment gap/output gap link 335–6
theory of 336
reality of 337
Phillip’s curve/Okun’s law combined 338
open market See money market
open market operations
definitions of 574
types of 249
open position
definition of 574
foreign investment risk 393
openness
external trade in goods/services 11
financial markets and 305–6
definition of 574
opportunity cost
capital, of 164
definition of 574
investment 150, 151, 205–7
investment function 214
leisure and 94
optimal capital stock
definition of 574
investment 204–7, 208
optimal sequencing
financial liberalization 531
optimum currency area
definition of 574
theory of 534
Ostmarks
short-lived market for 184
output
aggregate supply 351
factors shifting Phillips/aggregate supply curve 351–2
gap See output gap
GDP 8
inflation/unemployment and 348–53
macroeconomic concept, as 5
production function and 61
short to the long run 352–3
supply-determined in the long run 331–3
underlying inflation and the long run 349–51
output cost of disinflation 381, 574
output gap
central bank reaction to 289
GDP deviations 13
meaning of 14, 574
monetary policy ad 257
Okun’s Law and 354
Taylor rule 257, 270, 272, 288
unemployment gap and 335
output determination
fixed exchange rates, under 310–14
flexible exchange rates, under 315–20
output-labour ratio
definition of 574
overvalued currency
Britain and 515
P
Paasche index 34
Pareto principle
definition of 574
market intervention 478–9
Paris Club 164
parity/central parity
change 314–15
definition of 574
realignment 577
partisan business cycle 436, 445, 574–5
payments system
commercial banks 225, 226, 231
definition of 575
economic importance, of 238
new innovations 239
threats to 128
PCA function
definition of 575
pecuniary externalities
definition of 575
property rights and 506
perfect competition
laissez-faire view 476
market failures 476
market-clearing 476
policy failures 476
perfect foresight 146, 386
permanent income
China and the hypothesis of 201
definition of 575
life-cycle consumption and 198–9
perpetuity
consol and 178
definition of 575
persistence
definition 575
expectations of 423
inflation, of 438
low interest rates, of 298, 556
supply side 432
person-hours
aggregate labour supply 96
definition of 575
Phillips curve
challenges for 338
factors that shift 351–2
definition of 575
disappearance/re-emergence of 339–40
inflation and 329–30
long-run 350–1
long-run aggregate supply curve 338
money illusion 338
oil shocks 339
Phillips curve/Okun’s law combined 338
Phillips trade-off 334
reality of 335
rehabilitation of 348–9
stagflation 338
wage inflation/rate of unemployment relationship 334
Phillips trade-off 334, 575
physical capital
accumulation of 84–5
productive input, as 61
stock 66
Pissarides, Christopher 553–4
Poland
current income and spending 204
policy failures
definition of 575
efficiency and 476
policy lags 433
political business cycles 436–7, 575
poor countries
growth traps 81–3
investment and 68
Solow model and 88
See also rich countries/poor countries
population growth
capital-widening 73
GDP per capita and 74–5
portfolio investment
financial accounts and 44, 47
positive analysis 22
positive economics
definitions of 22, 575
positive externalities 478, 506, 575
positive money/Vollgeld
definition of 575
future money, as 238
potential output
definition of 575
monetary policy 257
Prescott, Edward 549
present discounted value
budget constraint 149
definition of 575
price deflators 33, 34
price flexibility
definition of 575
demand/supply adjustments and 331
price indices 33, 34
price level
consumer price index 20
price stability
monetary policy 251
primary budget deficit
definition of 575
primary budget surplus
definition of 575
public budget constraint 154
primary current account
budget constraint of the Nation 161–2
function 402
real exchange rate and 401–3
primary current account function
definition 576
real exchange rate/primary account balance 402
primary deficit
public budget constraint 154–5
primary income
consists of 45
definition of 576
primary international income 45, 562
balance on 562
private income
definition of 576
net interest on loans from banks 39
rent/royalties 39
wages/salaries 39
private sector demand
consumption See consumption
investment See investment
privatization
definition of 576
malfunctioning markets and 484
procyclical
definition of 576
inflation rate 10
producer price index (PPI) 34
producer surplus
definition of 576
product differentiation 342, 576
product market policies
externalities, dealing with 482–4
health 483
human capital accumulation 482–3
industrial policy 486
law and order 482
malfunctioning markets, dealing with 484–6
monopolies 484
privatization 484
public ownership of firms 486
regulation 484–5
subsidies 485–6
trade policies 486
production function
budget constraints 150, 152
Cobb-Douglas 61
definition of 576
intensive form 63
labour demand curve and 97
public infrastructure and 85
productive efficiency
definition of 576
fiscal policy 450
property rights
definition of 576
productivity and 86–7
public budget constraint
European Debt Crisis 155
government budget line 154
primary budget surpluses 154
primary deficit 154
public debt
debt stabilization 457–9, 464–9
default 467
deficit cutting 464–5
European sovereign debt crisis 468–9
growth/no inflation 460–2
inflation tax 464, 465–7
inflationary finance 463–4
interest rate relief 468
Italy and 467
long-run economic growth 468–9
no growth/no inflation and 459–60
seigniorage 457–64, 465–7
stability/growth pact, euro area 466
public goods
consumption smoothing 452
definition of 576
fiscal policy and 449–50
non-excludable 449, 480
non-rival 449
public infrastructure
definition of 576
productivity and 85
public ownership 486
public/private consolidated budget constraint 156–7
purchasing power parity (PPP)
aggregate demand 359
absolute PPP 136–8
definition of 576
exchange rate in the long run 136–8
GDP and 30
relative 136–8
purchasing power parity (PPP) line
definition of 575
Q
q-theory of investment
definition of 576
relationship between aggregate investment to Tobin’s q, 209, 210
Tobin’s q and 211
quantitative easing
definition of 576
monetary policy and 261–3, 443
‘quants’ 180
quota
definition of 576
IMF and 517, 522
R
Ramsey principle
definition of 576
funding public goods 487
random walk
explanation of the variable 576
theory of consumption 198
rate of depreciation
definition of 576
rational expectations hypothesis
definition of 576
future expectations and 145–6
perfect foresight 146
reasons for adopting 145
rational expectations revolution 548
real business cycle (RBC) theory 440, 576–7
real consumption wage
definition of 577
leisure and 94
real disturbance/shock
definition of 577
effect of 293–5
nominal disturbance contrasted 577
real economy
definition of 577
financial markets and 10–11
introduction of euro 127
monetary economy contrasted 11
monetary neutrality and 124
short run, in the 137
real exchange rate
comparison of price indices 133
effective exchange rate 134
equilibrium See equilibrium real exchange rate
France/Germany 369
fundamental determinants of 403–7
inflation differential 134
long run constancy 137
measuring in practice the 134
misalignment of 405
movements in 133–4
natural resources 404
net external position 404–7
non-price competitiveness 403
real GDP
definition of 577
nominal distinguished 32
real interest rate
arbitrage 394
consumption and 199–201, 204, 215
definition 577
Fisher equation 370–1
household’s budget constraint 147–8
investment and 208, 210
nominal interest rate distinguished 132, 370–1
real wage adjustment
involuntary unemployment 102–3
real wage rigidity
collective bargaining and 103–7
definition of 577
efficiency wages and 110
social minima and 108–10
realignment
definition of 577
recapitalization
banks and 267
definition of 577
recessions 276, 577
recognition lag 433, 577
refinance
definition of 577
financial institutions and 249, 269
regulation
banks and 226, 237, 264–7, 504, 531
dismissal 496–7
worker mobility and 494–5
relative price
definition of 577
leisure, of 94
relative purchasing power parity
definition of 577
type of PPP 136
religions
lending and 147
reserve asset transactions
financial account and 44, 47
reserve multiplier
control of money supply 231
definition 577
reserves
minimum legal requirements 249–50
reserve ratio 239, 252
definition of 577
legal 249
requirement by countries 230
voluntary 250
returns to scale
constant 62
decreasing 62, 565
definition of 577
increasing 62, 570
revaluation
definition of 577
exchange rate changes 314
Ricardian equivalence proposition
consolidated budget constraint and 157–8
critical review of 158–61
defined 577
rich countries/poor countries
conditional convergence 83–4
convergence hypothesis 81–3
human capital effects 84–7
risk
allocation of 174–5
aversion 174
diversification 176
premium 175
price of 175–6, 177
risk-return curve 177
risk-return trade-off 176–8
stocks and 180
what is 177–8
risk averse
asset holders and 174
definition of 577
risk premium
definition of 577
interest parity conditions and 394
shares and 180
‘rocket scientists’ 180
Romer, Paul 555
S
sales
final/intermediate contrasted 568
saltwater economists
freshwater economists contrasted 25
saving schedule
capital accumulation and 66
definition of 577
savings
meaning of 39, 577
role in economic growth 68–70
secondary income
secondary income account 45
secondary international income 45
secular stagnation hypothesis
defined 298, 578
growth accounting and 81
securitization
definition of 578
financial institutions and 180
seigniorage
deficits, financing 466
inflation tax and 464–7
public debt and 457–63
self-fulfilling attacks
definition of 578
self-fulfilling crisis
definition of 578
equilibrium exchange rates and 412
separation rate
cyclical 113
definition of 578
incidence of unemployment 111, 112
separations
involuntary 110
reasons for 110
sequencing
definition of 578
shares See stocks
short-run equilibrium
long-run equilibrium distinguished 362–4
definition 578
short-run neutral interest rate 292, 578
sight deposits
characteristics of 223
central banks, at 229–30
Slutsky, Eugen
business cycles and 428
small open economies
capital controls 304–5
Europe and 558
fixed exchange rate See fixed exchange rate regime
flexible exchange rate See flexible exchange rate regime
Mundell-Fleming model 304
social infrastructure
productivity and 86–7
social safety net 496–7
social minima
real wage rigidity and 108–10
soft pegs 531, 540
Solow decomposition
definition of 578
growth accounting and 78–81
Solow, Robert 555
Solow growth model
capital accumulation, effect of 78
definition of 578
employment growth, effect of 79–80
technological change, effect of 80–1
Solow residual
definition of 578
growth accounting and 79
sovereign borrowing
definition of 578
enforcing sovereign loans 164
spatial arbitrage
definition of 578
special drawing rights (SDRs)
definition of 578
speculative attacks
definition of 578
fixed exchange rate regimes and 322, 526, 528
speculative bubbles
banks and 238
definition of 578
herd behaviour 185
price behaviour and 188
speculators 184–5
spot exchange rate
definition of 578
foreign exchange rate and 390
stabilization
automatic stabilizers 454–6
budget figures, interpreting 456–7
consumption and tax smoothing 452–3
output/employment 453–4
stagflation
definition of 578
Phillips curve and 338–9
policy challenge 377–8
steady state
capital accumulation and 66, 67–8
consumption, raising in 71
definition of 578
growth rates 77
population growth and 74, 81
technological progress 77, 81
sterilized intervention
definition of 578
stochastic view of business cycles
definition of 578
stock market index 8
stock variables
flow variables contrasted 29
stocks
capital gains and losses 179
dividends 174, 179
foreign firms, in 48
no-profit condition and 179
risky nature of 180
stock market index 8
valuation of 179
structural unemployment
concept of 115
definition of 578
stylized facts
definition of 578
Kaldor’s 63–5
subsidies 485–6
substitution effect
consumption-leisure trade-off 95
definition of 579
Summers, Lawrence H 556–7
supply shock/disturbance
absence of 351
deep depreciation 347
definition 579
demand shocks contrasted 431
exogenous events 347–9
labour 117
lessons from 378–9
occurrence of 376–8
oil shock 347
positive/negative 348, 354
supply side
definition of 579
persistence and 432
productive potential of economy 18, 21
supply-side policies
competition policy 477
economic rents 476
externalities 478–80
feasible 502–4
impact of 481
imperfect competition 476–8
information asymmetries 480–2
laissez-faire 476
long and variable lags of 500–1
macroeconomics of 474–5
market failures 478–82
market-clearing 476
natural monopoly 480
Pareto principle 478
perfect competition 475–6
policy failures 476
political economy of reforms 501–2
public goods 480
reforms in practice 502–4
telecommunications 485
surveillance
International Monetary Fund and 509, 521–3
swap
credit default 180
currency 132
definition of 579
Sweden
nominal and real exchange rate 134
Switzerland
safe haven, as 426
swiss franc, challenges of 322–4
systemic risk
definition of 579
T
target inflation rate
definition of 579
price stability objective 257
target/neutral interest rate 579
definition 573, 579
short-run neutral interest rate 292, 295
targets
definition of 579
exchange rate 257
inflation 254–7
money growth 253–4
tax distortions
definition of 579
tax smoothing
definition of 579
German Reunification 452
taxation
deadweight loss of taxation 487–8
distortionary taxation 488
efficiency, effect on 487–9
Laffer curve, theory and reality 489–90
necessary evil 487
non-rivalry 487
price of intervention, as 487–90
public goods 487
Ramsey principle of public finance 487–9
tax base, adverse effect on 489–90
Taylor, John 550
Taylor rule
definition of 579
TR curve and 287–9
technological innovation
financial stability and 268–9
technological progress
definition of 579
economic growth and 59, 75–7
labour-augmenting 76
living standards, rise in 76
steady state and 77
technological shocks
definition of 579
techno-pessimists/techno-optimists 78, 81
theories
realism and 21
testing 22–3
time consistency problem
definition of 579
promises and 500–1
Tobin’s q
definition of 579
installation costs 211–12
microeconomics foundations of 211
total factor productivity
definition of 579
TR curve
definition of 579
monetary policy autonomy, loss of 310–11
monetary policy disturbances 291–2, 317–18
shifts of 295, 317–18
Taylor Rule and 287–8
TR schedule, slope of 289–90
trade openness 305
trade policies
definition of 579
trade unions
definition of 579
transmission channels of monetary policy
definition of 579
trends 4
definition of 579
triangular arbitrage
definition of 579
foreign exchange markets and 183, 391
Triffin paradox
fixed exchange rates and the 518
Tulipmania
speculative bubble 188
U
ultimate bond See consol
uncovered interest rate parity
capital gain/loss 393
definition of 580
international Fisher equation 395
open/unhedged/uncovered foreign investment 393
risk premia 394
uncovered position See open position
underground economy
definition of 580
estimates of the size of 36
gross domestic product (GDP) and 35
underlying inflation rate 346–7, 564, 580
undervalued currency
definition of 580
France 515
unemployment
concept of 5–8
countercyclical in nature 10
definition of 580
dynamic interpretation of 110–13
equilibrium 113–18, 567
equilibrium rates 352
European problem 107
frictional 115, 568
interpretation of 101–2
involuntary 571
static interpretation of 102–10
structural 115, 578
technical change and 99
voluntary 101, 580
unemployment benefit
definition of 581
disincentives to employment 117
intentions of 113
social minima 108
unemployment gap
definition of 580
unemployment rate
unemployment trap 113
unhedged position See open position
United Kingdom
unpaid work
underground economy and 35
unsterilized intervention
sterilized contrasted 580
US Financial Crisis 2008-2009, 181–2, 525
user cost of capital
explanation of 580
optimal capital cost and 207
relationship to Tobin’s q, 212, 213
utility
consumption and 195
consumption-leisure trade-off and 93
definition 580
indifference curves and 195, 196
V
value added
definition of 580
examples of 31
value subtracted
examples of 31
vehicle currencies
definition of 580
foreign exchange markets 389
volatility
asset prices and 169, 172
exchange rate 387
financial markets 171
goods and services markets distinguished
Volcker rule
definition of 580
Vollgeld See positive money/Vollgeld
voluntary unemployment
definition of 580
involuntary See involuntary unemployment
von Hayek, Friedrich August 551
vulnerabilities
definition of 580
W
wage bill 343, 580
wage inflation 334, 580
Wassenaar Accord 502
wealth
consumption and 201–3, 215
inheriting 149
intertemporal budget constraint and 164
present discounted values 149
welfare trap
definition of 580
social safety net and 497
Wicksells’ cashless society 224
Woodford, Michael 550
Y
yield
definition of 580
yield arbitrage
definition of 580
yield curve
definition 580
riskless assets and 182
youth unemployment
minimum wages and 109
Z
zero lower bound
definition of 580
Great Recession and 297–8
interest rates and 261–3, 296
Zimbabwe
inflation and 128, 252

Macroeconomics

Uploaded by

Copyright:

Available Formats

Macroeconomics

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Macroeconomics

Uploaded by

Copyright:

Available Formats

Macroeconomics

The cover image shows the face of Dionysus.

1.1 Real Income per Capita (GDP in euros, 2000 prices)

3.1 The Growth Phenomenon

4.1 Annual Total Hours Worked and Average Wages, 1870–2014

16.1 Inflation Targets Set by Inflation-Targeting Central Banks in 2016

1.1 GDP per Capita and Life Satisfaction in 2014

3.1 GDP per capita in the UK 1920–2015

4.1 Household Preference

5.1 Money Growth, Inflation, and Exchange Rate Depreciation, 1975–2006

9.1 Currency in Circulation (in per cent of nominal GDP)

11.1 Cyclical Fluctuations

1.1 Burns–Mitchell Diagrams, Now and Then

2.1 Value Added and Value Subtracted: Two Examples

3.1 China and the Chinese Puzzle of Economic Growth

4.1 Technical Change and Unemployment

5.1 Evidence for Monetary Neutrality: The Introduction of the Euro

8.1 Indifference Curves, Intertemporal Substitution, and Optimal Consumption

9.1 Wicksell’s Cashless Society

10.1 How the ECB Does It

15.1 The Real Interest Rate Parity Condition

1.1 Overview of Macroeconomics

1.2 Macroeconomic Concepts

1.2.1 Income and Output

Fig. 1.1 GDP per Capita and Life Satisfaction in 2014

1.2.3 Factors of Production and Income Distribution

1.2.5 Financial Markets and the Real Economy

Table 1.2 Openness (ratio of average of exports and imports to GDP, %)

European Union 6.1 12.0

1.3 Macroeconomics in the Long Run: Economic Growth

Fig. 1.6 Quarterly Gross Domestic Product, UK, 1956:1–2015:3

Box 1.1 Burns–Mitchell Diagrams, Now and Then

1.5 Macroeconomics as a Science

1.5.1 The Genesis of Macroeconomics

Fig. 1.7 Burns–Mitchell, Now: Eight Countries, 1970–2015

1.5.2 Macroeconomics and Microeconomics

Box 1.2 All Kinds of Cycles

1.6 The Methodology of Macroeconomics

1.6.1 What is to be Explained?

Fig. 1.10 Endogenous and exogenous Variables

1.6.2 Theory and Realism

1.6.3 Positive and Normative Analysis

1.6.4 Testing Theories: The Role of Data

1.6.5 Macroeconomic Modelling and Forecasting

Box 1.3 Forecasting the Crisis Years

2009 France Germany Japan Korea UK USA

1.7 Preview of the Book

1.7.2 Controversies and Consensus

1.7.3 Rigour and Intuition

1.7.4 Data and Institutions

Box 1.4 Macroeconomic Schools of Thought: A Primer

2.2 Gross Domestic Product

2.2.1 Three Definitions of Gross Domestic Product

Box 2.1 Value Added and Value Subtracted: Two Examples

2.2.2 Real versus Nominal Quantities, Deflators versus Price Indices

Box 2.2 What GDP Measures

Nominal Real GDP

Price deflators and indices

2.2.3 Measuring and Interpreting GDP