UNTIL DEBT DO US PART
UNTIL DEBT DO US PART
Subnational Debt, Insolvency, and Markets
Editors
Otaviano Canuto and Lili Liu
© 2013 International Bank for Reconstruction and Development / The World Bank
1818 H Street NW
Washington DC 20433
Telephone: 202-473-1000
Internet: www.worldbank.org
Some rights reserved
1 2 3 4 16 15 14 13
This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does
not necessarily own each component of the content included in the work. The World Bank therefore does not
warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of
claims resulting from such infringement rests solely with you.
The findings, interpretations, and conclusions expressed in this work are those of the authors and do not necessarily reflect the views of The World Bank, its Board of Executive Directors, the governments they represent, or
any other institutions with which the external authors may be affiliated. The World Bank does not guarantee the
accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown
on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of
any territory or the endorsement or acceptance of such boundaries.
Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved.
Rights and Permissions
This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http://
creativecommons.org/licenses/by/3.0. Under the Creative Commons Attribution license, you are free to copy,
distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions:
Attribution—Please cite the work as follows: Canuto, Otaviano, and Lili Liu, Eds. 2013. Until Debt Do Us Part:
Subnational Debt, Insolvency, and Markets. Washington, DC: World Bank. doi: 10.1596/978-0-8213-9766-4.
License: Creative Commons Attribution CC BY 3.0
Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank
translation. The World Bank shall not be liable for any content or error in this translation.
All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818
H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org.
ISBN (paper): 978-0-8213-9766-4
ISBN (electronic): 978-0-8213-9767-1
DOI: 10.1596/978-0-8213-9766-4
Cover art: Michael S. Geller, Sea Surface Full of Clouds (1), 2009, oil on canvas, 36'' 3 36''
Cover design: Drew Fasick
Library of Congress Cataloging-in-Publication Data
Until debt do us part: subnational debt, insolvency, and markets / Otaviano Canuto and Lili Liu, Editors.
pages cm
Includes bibliographical references and index.
ISBN 978-0-8213-9766-4—ISBN 978-0-8213-9767-1 (electronic)
1. Debts, Public. 2. Bankruptcy. 3. Fiscal policy. I. Canuto, Otaviano. II. Liu, Lili (Economist)
HJ8015.U58 2013
336.3’4—dc23
2012046037
Contents
Acknowledgments
About the Editors and Contributors
Abbreviations
xv
xix
xxxi
An Overview
Otaviano Canuto and Lili Liu
1
Part 1
Subnational Debt Restructuring
31
1 Brazil: The Subnational Debt Restructuring
of the 1990s—Origins, Conditions, and Results
Alvaro Manoel, Sol Garson, and Monica Mora
33
2 Restructuring of Legacy Debt for Financing
Rural Schools in China
Lili Liu and Baoyun Qiao
81
3 Managing State Debt and Ensuring Solvency:
The Indian Experience
C. Rangarajan and Abha Prasad
109
4 Subnational Debt Management in Mexico:
A Tale of Two Crises
Ernesto Revilla
145
v
vi
Contents
Part 2 Subnational Insolvency Framework
177
5
Colombia: Subnational Insolvency Framework
Azul del Villar, Lili Liu, Edgardo Mosqueira,
Juan Pedro Schmid, and Steven B. Webb
179
6
France’s Subnational Insolvency Framework
Lili Liu, Norbert Gaillard, and Michael Waibel
221
7
Hungary: Subnational Insolvency Framework
Charles Jókay
261
8 United States: Chapter 9 Municipal Bankruptcy—
Utilization, Avoidance, and Impact
Michael De Angelis and Xiaowei Tian
311
9 When Subnational Debt Issuers Default: The Case
of the Washington Public Power Supply System
James Leigland and Lili Liu
353
Part 3 Developing Subnational Debt Markets
377
10 Transition from Direct Central Government
Onlending to Subnational Market Access in China
Lili Liu and Baoyun Qiao
379
11 The Philippines: Recent Developments in the
Subnational Government Debt Markets
Lili Liu, Gilberto Llanto, and John Petersen
417
12 Russian Federation: Development of Public
Finances and Subnational Debt Markets
Galina Kurlyandskaya
455
13 South Africa: Leveraging Private Financing for
Infrastructure
Kenneth Brown, Tebogo Motsoane, and Lili Liu
495
14 Caveat Creditor: State Systems of Local
Government Borrowing in the United States
Lili Liu, Xiaowei Tian, and John Joseph Wallis
539
Index
591
Contents
Boxes
1.1 The 1993 Federal Immediate Action Plan
1.2 The Real Plan, 1994
1.3 The Fiscal Responsibility Law (Complementary
Law No. 101, May 4, 2000)
3.1 State Borrowings
6.1 Subnational Financial Rules: Fundamental Principles,
France
7.1 Major Acts Regulating the Municipal Sector, Hungary,
1990–2011
7.2 Powers of the Trustee: Article 14 (2) of Hungary’s
Municipal Debt Adjustment Law
7.3 Restrictions on a Municipality Undergoing
Debt Adjustment in Hungary
13.1 Section 139 Intervention in the Greater
Johannesburg Metropolitan Municipality
43
44
52
113
231
264
282
283
499
Figures
1.1 National, State, and Municipal Tax Collection
and Disposable Revenue in Brazil, 1970–2010
1.2 SELIC Interest Rate in Brazil, August
1994–April 2011
1.3 Consolidated Primary Fiscal Balance in Brazil,
2001–10
1.4 Primary Fiscal Balance of States and Municipalities
in Brazil, 1991–2010
1.5 Net Public Debt in Brazil, 2001–10
1.6 Net Debt of States and Municipalities in Brazil,
1991–2010
1.7 Main Sources of State Revenues in Brazil, 2000–09
(2000 = 100)
1.8 Main Sources of Investment Financing for States
in Brazil, 2000–09
1.9 Main Sources of Revenue for Municipalities
in Brazil, 2000–09 (2000 = 100)
38
45
46
47
56
56
59
62
62
vii
viii
Contents
1.10 Main Expenditures of Municipalities in Brazil,
2000–09 (2000 = 100)
3.1 Composition of Financing Pattern of State Deficits
(as of End-March)
3.2 Deficit and Debt as Share of GDP and Interest
Payments as Share of Revenues
3.3 Extended Debt as Share of GDP
3.4 Primary Deficit as Percentage of GSDP
3.5 Box Plot Showing Debt-to-GSDP Ratio and Interest
Burden Ratio
3.6 Interest Burden in Selected States
3.7 Differential between GDP Growth Rate and Interest
Rate on State Debt
4.1 Subnational Debt in Mexico, by State, 2011
4.2 The Macroeconomic Impact of the 1994–95 Tequila
Crisis in Mexico
4.3 Deteriorating Subnational Credit Scores in
Mexico during the Global Financial Crisis
4.4 Fall in Transfers Relative to Previous Year in Mexico
during the 1994–95 Tequila Crisis and the 2008–09
Global Financial Crisis
4.5 Federal Transfers: Evolution of Expectations in
Mexico during the Global Financial Crisis, 2009
5.1 Government Structure in Colombia
5.2 Regional Transfers to Subnational Governments
as a Share of Current Central Government Revenues,
1994–2010
5.3 Subnational Government Direct Debt as Percentage
of GDP, 1990–2010
5.4 Restructuring Process under Law 550/1999
5.5 Departments and Municipalities under Debt
Restructuring Agreement, 1999–2010
5.6 Department and Municipality Debt Balance as
Percentage of GDP, 2000–10
6.1 Subnational Governments in France
6.2 Sources of Municipal Revenues in France
6.3 Sources of Departmental Revenues in France
6.4 Sources of Regional Revenues in France
64
115
117
121
123
124
124
133
150
156
161
164
165
182
184
190
200
207
211
224
228
228
229
Contents
6.5 Budgetary Control and Appeal Proceedings, France
6.6 SNG Borrowing and Debt Repayment in France
6.7 Distribution of Ratings Assigned to French SNGs
7.1 Bonds, Bank Loans, and Other Debt, Hungary,
2005–11
7.2 Cash and Liquid Financial Assets, 2004–11
7.3 The Debt Adjustment Process, Hungary
8.1 Annual Chapter 9 Filings, 1980–2011
8.2 Chapter 9 Filings by Type of Municipality, 1980–2007
10.1 Revenue and Expenditure of Subnational
Governments in China, as Percentage of Total
Government Revenue and Expenditure, 1985–2010
10.2 Top Five Countries Issuing Subnational Bonds,
2000–09 (Excluding the United States)
10.3 Rapid Urbanization in China, 1990–2010
10.4 Annual Land Transfer Fee in China, 2004–09
10.5 GDP Growth Rates and Subnational Revenue,
1991–2010
10.6 Formation of Debt Equilibrium, China
11.1 Distribution of Total Income, All Local
Government Units, 2009
11.2 Composition of Revenues by Type of Local
Government Unit, 2009
12.1 Composition of Consolidated Subnational
Borrowing in the Russian Federation, 1995–2000
12.2 Subnational Defaults by Type of Debt Operation
12.3 Federal and Subnational Debt as a Share of GDP
12.4 Regions’ Share of Total Regional Bond Debt
Outstanding at the End of 2008
12.5 Outlook on Credit Ratings: Russian Federation
Regions and Subnationals in European Countries,
End-December 2010
13.1 Trends in the Municipal Borrowing Market,
South Africa, 2005–10
13.2 Metropolitan Municipality Capital Expenditure,
South Africa, 2004/05–2009/10
13.3 Metropolitan Municipality Borrowing,
South Africa, 2004/05–2009/10
235
241
242
276
278
281
322
323
384
385
399
400
401
406
422
423
462
464
477
478
484
512
513
514
ix
x
Contents
13.4 Outstanding Debt of Metropolitan Municipalities,
South Africa, 2004/05–2009/10
13.5 Debt Composition of Metropolitan Municipalities,
South Africa, 2004/05–2009/10
13.6 Debt Service Costs, South Africa, 2004/05–2009/10
13.7 South African Municipal Infrastructure
Investment Requirements, 2010–19
515
516
517
520
Tables
1.1 Public Sector Borrowing Requirements (PSBR)
in Brazil, 1999–2010
1.2 Share of GDP, Population, and Long-Term
Debt of Borrower States in Brazil, 2009
1.3 Share of GDP, Population, and Long-Term
Debt of Municipalities in Brazil, 2009
1.4 State Revenue in Brazil, 2000–09
1.5 State Expenditures in Brazil, 2000–09
1.6 Expenditures of Municipalities in Brazil, 2000–09
2.1 Statistics on Rural Junior Secondary and Primary
Schools and Students, 2009
2.2 Assignment of Major Educational Responsibilities
among Levels of Government in China
2.3 Financing Sources for Rural Junior Secondary and
Primary Schools in China, 2008
2.4 Composition of Educational Expenditures of Rural
Primary Schools, Province of Henan, 1999 and 2002
2.5 Own Revenues as a Percentage of Total Expenditures,
by Level of Government in China, 2003
2.6 Compulsory Debt as a Percentage of Subnational
Budgetary Expenditure for 14 Regions in China, 2009
B3.1.1 Sources and Features Attached to State Borrowings
3.1 Weighted Average Spreads during 2010–11
3.2 Deposit Rate of Major Banks for Term Deposits of
More Than One-Year Maturity
3.3 States’ Vulnerability Matrix
55
57
58
58
61
63
85
89
90
91
94
95
113
116
119
125
Contents
3.4 Debt Forgiveness by Finance Commission
3.5 States’ Overdrafts and Access to Cash-Credit
4.1 Subnationals Resources in Mexico
4.2 Subnational Debt Structure in Mexico, 2011
4.3 Two Crises: Implications for the Subnational
Debt Market in Mexico
4A.1 “Ramo 28.” Nonearmarked Transfers (Participaciones
Federales), Mexico
4A.2 “Ramo 33.” Earmarked Transfers (Aportaciones
Federales), Mexico
4A.3 Mexican States’ Total Local Revenue: Own-Source and
Coordinated Federal Taxes
5.1 Total Revenues of Subnational Governments,
Percentage of Total, 2006–10
5.2 Department Revenues
5.3 Municipal Revenues
5.4 Fiscal Performance of Municipalities under 550
Debt Restructuring Agreements Compared to
the National Average for All Municipalities
5.5 Composition of Subnational Debt as Percentage
of GDP, 2000–10
5.6 Channels for Control of Deficits and Debt:
Lender-Borrower Nexus and Timing of Controls
and Sanctions
5.7 Ex-Ante and Ex-Post Fiscal Legislation
5.8 Indebtedness Alert Signals
5.9 Colombian Superintendencies
5.10 Total Debt Restructured under Law 550, 1999–2010
5.11 Entities Restructured under Law 550, 1999–2010
5.12 Entities under Law 617 and/or Law 358, 1997–2010
6.1 New Powers Devolved to SNGs in 2004, France
6.2 Deadlines for the Regular Budget Process, France
6.3 Debt Data for French SNGs Rated by Fitch
6.4 Main Ratios Used by the French Central Government
to Detect Financial Distress
7.1 Debt of Hungarian Municipalities Calculated
at Prevailing Euro Exchange Rates
128
131
148
152
159
169
170
171
186
187
188
191
192
194
195
196
205
206
206
210
225
232
243
249
275
xi
xii
Contents
7.2 Municipal Debt Adjustment Filings, Hungary,
1996–2010
7.3 Main Causes of Bankruptcy in Hungary
8.1 State Authorization of Chapter 9 Bankruptcy
10.1 Urban Infrastructure Development in China:
Selected Indicators, 1990–2009
10.2 Subnational Bonds Issued in China, 2009–11
10.3 Municipal Bonds Directly Issued by Four
Cities in 2011
10.4 Relative Shares of Expenditure at Different
Government Levels in China, 2003
11.1 Distribution of Total Expenditure, All Local
Government Units, 2001–09
11.2 Local Government Borrowing and Debt Limitations:
The Philippines
11.3 Local Government Finances: Key Ratios by
Type of Unit, 2010
11.4 Budget Surpluses of Local Government
Units, 2005–08
11.5 Structure of Philippine Local Government
Debt Markets
11.6 Outstanding Loans and Bonds of LGUs (as of
September 10, 2010)
11.7 LGU and Other Entity Outstanding Debt (with
LGUGC Guarantee), 2010, by Type of Unit
12.1 Fiscal and Debt Rules for Subnational Governments
12.2 Annual Growth of Subnational Revenue,
Including Fiscal Transfers, 2004–08
12.3 Intergovernmental Fiscal Transfers from the Russian
Federation to Regions as a Percentage of Each Type
of Transfer in Total Transfers, 2003–10
13.1 Secondary City Long-Term Borrowing,
South Africa, 2004/05–2009/10
14.1 Year of First General Law for Municipalities
and First Home Rule Law, United States
289
292
317
387
395
396
402
421
424
427
428
433
434
438
472
476
477
518
546
Contents
14.2 State Constitutional Provisions Governing
Local Debt and Borrowing Provisions, United States,
1841–90
14.3 State Constitutional Provisions on Local
Government Debt Issue, United States
14.4 Government Debt by Level of Government,
Nominal Amount, and Shares, United States,
1838–2002
14.5 State Monitoring of Local Fiscal Conditions, Home
Rule, and Local Debt Restrictions, United States
14.6 Difference-in-Differences Estimates, Local Share
of State and Local Totals, United States
14.7 Local Total Revenue, Expenditure, and Debt per Capita,
United States, 1972, 1992, 2007
14.8 State Total Revenue, Expenditure, and Debt,
United States, 1972, 1992, 2007
14.9 Combined State and Local Total Revenue,
Expenditure, and Debt, United States, 1972,
1992, 2007
550
551
556
563
572
574
575
576
xiii
Acknowledgments
This volume is the result of a team effort. We have many people to thank.
First, we thank the authors. Many of them are leading practitioners and
experts in public finance in the context of multilevel government systems.
We are indebted to them not only for the quality of their contributions but
also for the quests they have pursued tirelessly on the challenging questions that this volume tries to address.
Second, the findings of the volume have been shaped by the Sub-
Sovereign Finance Forum – Debt, Insolvency, Markets held in June 2011 at
the World Bank Headquarters in Washington, DC. The Forum comprised
senior officials and practitioners from developed and developing countries—including the countries covered by this volume—as well as academicians, international technical assistance providers, and World Bank and
International Monetary Fund managers and staff. The discussions at the
Forum have informed the development of the volume.
Third, this volume is a part of the global knowledge program of the
Economic Policy and Debt Department, PREM Network, of the World
Bank. We thank Jeffrey D. Lewis, Director; Carlos Braga, former Director;
and Sudarshan Gooptu, Sector Manager of the Department, for their
support.
Fourth, our special thanks go to the Public Private Infrastructure Advisory Facility’s (PPIAF)–Subnational Technical Assistance Program, which
financed significant portions of the Sub-Sovereign Finance Forum, the production of this volume, and the background research. We thank Adriana de
Aguinaga de Vellutini, Manager; and James Leigland and Paul Reddel, former
Managers of PPIAF, for their support.
xv
xvi
Acknowledgments
Fifth, we are grateful to colleagues and external reviewers who have
helped the authors sharpen the messages and distill the lessons shared in
this volume. Various chapters have also drawn from discussions with stakeholders in the field and during World Bank missions. In particular, we thank
the following colleagues and external reviewers for their comments.
Part 1 Subnational Debt Restructuring
José Roberto Afonso, Economist of the National Bank of Economic and
Social Development and Economic Advisor, Brazilian Congress, and Pablo
Fajnzylber, José Guilherme Reis, and Rafael Barroso of the World Bank
(chapter 1); Xiaoyun Zhang, Professor and Head of the Public Finance
Group, Research Institute for Fiscal Studies, Ministry of Finance, China,
Lezheng Liu, Associate Professor of Economics, Central University of
Finance and Economics, Beijing, China, and Chorching Goh and Min Zhao
of the World Bank (chapter 2); Dr. Vijay Kelkar, Chairman of the Thirteenth
Finance Commission, India, Professor D. K. Srivastava, Director, Madras
School of Economics, India, and Deepak Bhattasali of the World Bank
(chapter 3); Arturo Herrera, Paloma Anos Casero, David Rosenblat, Jozef
Draaisma, and Andrea Coppola of the World Bank, and Steven Webb, consultant (chapter 4).
Part 2 Subnational Insolvency Framework
Lars Christian Moller, Christian Yves Gonzalez, and Jose M. Garrido of
the World Bank (chapter 5); Alban Aucoin, former Special Advisor to the
Chairman, ADETEF, French Ministry of Economy and Finance, France,
Danièle Lamarque, former Director, Cour des Comptes, France, Michael
De Angelis, Lecturer of Business Law, University of Rhode Island, United
States, and Francois Boulanger of the World Bank (chapter 6); authorities
of the Hungarian government, Jose M. Garrido and Riz Mokal of the World
Bank, Michael De Angelis, and Mihaly Kopanyi, a consultant and former
World Bank staff member (chapter 7); Joel Motley, Managing Director, Public Capital Advisors, United States, James E. Spiotto, Partner, Chapman and
Cutler LLP, United States, and Jean-Jacques Dethier, Jose M. Garrido, and
Matthew D. Glasser of the World Bank (chapter 8); and Michael De Angelis,
and Professor John Joseph Wallis of Economics Department, University of
Maryland, United States (chapter 9).
Acknowledgments
Part 3 Developing Subnational Debt Markets
Xiaoyun Zhang, Professor and Head of the Public Finance Group, Research
Institute for Fiscal Studies, Ministry of Finance, China, Lezheng Liu, Associate Professor of Economics, Central University of Finance and Economics,
Beijing, China, and Chorching Goh, Catiana Garcia-Kilroy, Zhi Liu, and Min
Zhao of the World Bank (chapter 10); and Yan Zhang, Victor Vergara, José M.
Garrido, and Lawrence Tang of the World Bank (chapter 11); Kaspar Richter,
Stepan Anatolievich Titov, and Pavel Kochanov of the World Bank (chapter
12); Sandeep Mahajan and Matthew Glasser of the World Bank (chapter 13);
Michael De Angelis, Lecturer of Business Law, University of Rhode Island,
United States, Martha Haines, former Municipal Finance Director, U.S.
Securities and Exchange Commission, Noel Johnson, Assistant Professor,
Economics Department, George Mason University, United States, Isabel
Rodriguez-Tejedo, Assistant Professor, University of Navarre, Spain, and
Jean-Jacques Dethier and José M. Garrido of the World Bank (chapter 14).
We thank Joel Motley, Managing Director, Public Capital Advisors, United
States, for his comments on the chapters in Part 3.
Sixth, we would like to thank various government officials and other
people with whom we had discussions in the course of developing this
volume. Chapters 2 and 10 benefited from discussions with the Ministry of
Finance, China. The authors of chapter 3 would like to thank Ying Li for her
valuable analytic and technical inputs. The author of chapter 4 would like
to thank the extraordinary research assistance of Fernanda Márquez-Padilla
and Rodrigo Sanchez-Gavito, and Tuffic Miguel and Emilio Pineda for sharing in fruitful discussions on the subject. Chapter 5 on Colombia benefited
from discussions with central and local government officials, bankers, rating
agency analysts, and superintendency officials during a World Bank mission
to Colombia, October 4–6, 2010, and follow-up discussions on July 13, 2011,
with Ministry of Finance and Public Credit officials on an earlier draft. Chapter 6 on France draws on a World Bank mission to France in July 2009. The
mission team met national and subnational officials, representatives from
Agence Française de Développement and from the rating agencies, and university professors. Chapter 7 on Hungary benefited from contributions by
Laszlo Osvath, Gabor Peteri, and Miklos Udvarhelyi of LGID Ltd, and from a
series of interviews with national and subnational government officials and
members of the financial services community in Hungary. Chapter 11 on the
xvii
xviii
Acknowledgments
Philippines draws from a World Bank mission to the Philippines in January
2011, where the mission team met government officials, public financial institutions, and private banks. Chapter 12 on Russia benefited from dialogue
with analysts of rating agencies Fitch and Standard & Poor’s. The author
would like to thank Yulia Gerasimova for her research assistance. Chapter
13 benefited from background work by DNA Economics of South Africa.
The authors of chapter 14 would like to thank Jessica Hennessey and Isabel
Rodriguez-Tejedo for sharing their research findings. Finally we would like to
thank Ying Li for her extraordinary efforts on ensuring data consistency in
various chapters.
Finally, we are also grateful to colleagues who helped us move this complex project from manuscript to publication. They are Diane Stamm, our
outstanding language editor; Stephen McGroarty, Mary Fisk, and Alejandra
Viveros, who guided us through each stage of the production process;
Mayya Revzina, who guided us on the assignment of copyrights; Detre
Dyson, Debbie L. Sturgess, and Tania Tejeda, who helped format the volume; and Ivana Ticha, who efficiently put together the Sub-Sovereign
Finance Forum and managed various stages of the process.
About the Editors and Contributors
Otaviano Canuto is Vice President and Head of the Poverty Reduction and
Economic Management Network of the World Bank, a division of more than
700 economists and public sector specialists working on economic policy
advice, technical assistance, and lending for reducing poverty in the Bank’s client countries. He assumed his position in May 2009, after serving as the Vice
President for Countries at the Inter-American Development Bank since June
2007. Dr. Canuto provides strategic leadership and direction on economic
policy formulation in the area of growth and poverty, debt, trade, gender, and
public sector management and governance. He is involved in managing the
Bank’s overall interactions with key partner institutions including the International Monetary Fund. He has lectured and written widely on economic
growth, financial crisis management, and regional development. He has published more than 70 articles in economic journals and books in English, Portuguese, and Spanish, and he has spoken often on development policy and
global economic issues. Dr. Canuto holds a PhD in Economics from the University of Campinas in São Paulo, Brazil, and an MA from Concordia University
in Montreal in Canada. He speaks Portuguese, English, French, and Spanish.
Lili Liu is a Lead Economist in Public Sector and Institutional Reform, Europe
and Central Asia, at the World Bank. Until the fall of 2012, she was with
the Economic Policy and Debt Department as the cluster leader on public
finance at the subnational government level. She sits on the World Bank
Urban and Transport Sector Boards, and co-chairs the Decentralization and
Subnational Regional Economics Thematic Group, a Bankwide network
with over 350 members. Previously, she led high-level policy dialogue and
lending operations for India and other countries, covering development
xix
xx
About the Editors and Contributors
strategy, macroeconomic policy, public finance, trade, and infrastructure.
She has authored numerous publications on public finance, subsovereign
finance and their linkages to macroeconomic frameworks, intergovernmental fiscal systems, capital market development, and infrastructure finance.
She has also led advisory services to the World Bank’s operations in many
developing countries. Dr. Liu is a frequent speaker at international conferences and to visiting government delegations to the World Bank. She has a
PhD and an MA in Economics from the University of Michigan, Ann Arbor,
and a BA from Fudan University, Shanghai.
Kenneth Willy Brown is Deputy Director-General: Intergovernmental Relations Branch of the South African National Treasury. Before entering
the public sector, Mr. Brown had a career in teaching. Mr. Brown joined
the National Treasury in 1998 as a Deputy Director: Financial Planning, and
in 2001, he assumed the position of Director: Provincial Policy, which underpins the national transfers to provinces. He has also served as Chief
Director: Intergovernmental Policy and Planning, and oversaw sector policies that affect provinces and local governments. As Deputy DirectorGeneral, he oversees provincial and local government finances including
subnational transfers. Mr. Brown holds an MA in Economics from the
University of Illinois, Urbana-Champaign; a BA with Honors in Economics
from the University of the Western Cape in South Africa; and a Primary
Teacher’s Diploma.
Michael A. De Angelis is a Lecturer on the Faculty of the College of Business
of the University of Rhode Island. He is a former partner in the New York
law firm Mudge Rose Guthrie & Alexander, where he specialized in public
finance. Mr. De Angelis’s expertise is in U.S. and transitional and developing economy government finance (sovereign and subsovereign), including
general obligation, utility, infrastructure, education, health care–related
nonprofit organizations, bond funds, guaranties, and other project and
securitized financing. He has served as counsel to issuers, underwriters, borrowers, guarantors, and lenders, and he participatied in the development
of financial products for governmental infrastructure financing transactions. The recent emphasis of Mr. De Angelis’s work has been on the development of the regulatory and institutional legal framework necessary
for capital markets for sovereign and subsovereign government borrowers
About the Editors and Contributors
in transitional and developing economies, including over 25 countries and
the Western Africa Economic and Monetary Union. Mr. De Angelis holds a
JD and BA from Boston College, Boston, Massachusetts, and an MPA from
Cornell University, Ithaca, New York.
Azul del Villar has worked since 2002 in the Public Sector Group for Latin
America at the World Bank on policy development and technical assistance
loans with national and local governments, mostly to support their regional
development, decentralization, fiscal sustainability, and public financial management. She has participated in economic studies related to fiscal discipline,
budget reform, debt management, and public spending efficiency. While at
Fitch Ratings’ global infrastructure and project finance group from 2007 to
2010, she conducted analyses for project finance rating of debt issuances in
the transportation and energy sectors. She worked in the Mexican Ministry of
Foreign Affairs’ Organisation for Economic Co-operation and Development
Department from 1999 to 2001, where she analyzed the recommendations
of the Fiscal Affairs, Public Governance and Management, and Economic
Growth Committees to be implemented in Mexico. Ms. del Villar earned a
BS in Economics from the Universidad Iberoamericana in Mexico City and an
MA in Economic Policy Management from Columbia University in New York.
Norbert Gaillard is a French economist and independent consultant and is
Visiting Professor at the Graduate Institute in Geneva. He has served as a
consultant to the International Finance Corporation, the World Bank, the
State of Sonora (Mexico), the Organisation for Economic Co-operation
and Development, and the European Parliament. He has authored several
research articles and book chapters on sovereign debt and credit rating
agencies, the most recent ones of which are “The End of Gatekeeping:
Underwriters and the Quality of Sovereign Bond Markets, 1815–2007,”
coauthored with Marc Flandreau, Juan H. Flores, and Sebastián NietoParra; and “To Err is Human: Rating Agencies and the Interwar Foreign
Government Debt Crisis,” coauthored with Marc Flandreau and Frank
Packer. Dr. Gaillard has published two books: Les Agences de Notation
(La Découverte, Paris, 2010), and A Century of Sovereign Ratings (Springer,
New York, 2011). He received his PhD in Economics in 2008; his dissertation
dealt with sovereign rating methodologies. He was a Fulbright Fellow at
Princeton University, Princeton, New Jersey, during 2004–05 and 2006–07.
xxi
xxii
About the Editors and Contributors
Sol Garson teaches Financial Management of States and Municipalities in
the Public Policies, Strategies, and Development Program of the Federal
University of Rio de Janeiro, Brazil. She is also a consultant specializing in
public sector finances and budgeting. From 1975 to 2001, Dr. Garson was
an economist at the National Bank of Economic and Social Development,
during which she was seconded to the City of Rio Janeiro in 1993, and was
Secretary of Finance of the City from 1996 to 2000. As President of the
Association of Secretariats of Finance of the Capital Cities from 1997 to
2000, she coordinated discussions about tax reform, the Fiscal Responsibility Law, and other subjects of interest to Brazilian cities. In 2007, she served
as Undersecretary of Fiscal Policy at the Secretariat of Finance of the State
of Rio de Janeiro. Dr. Garson holds a PhD in urban and regional planning, and
her publications include the book Metropolitan Regions: Why Don’t They
Cooperate? (Letra Capital, Rio de Janeiro, 2009).
Károly (Charles) Jókay is an expert in municipal finance and bankruptcy and
teaches courses in municipal finance, public budgeting, and public management in the Department of Public Policy at Central European University in Budapest, Hungary. He has municipal finance and creditworthiness
experience in Central and Eastern European countries, including Bosnia and
Herzegovina, Hungary, the former Yugoslav Republic of Macedonia, Romania, and Serbia. He specializes in policy reform; municipal finance and budgeting; operational implementation of fiscal decentralization, such as the
regulation and management of municipal debt; and the improvement of
grant-based policies. A regular consultant to the World Bank, Dr. Jókay has
completed projects on municipal bond disclosure standards, public utility
transformation and regulation in the municipal services sector, and municipal debt regulation. He has a PhD and an MA in Political Science from the
University of Illinois, Urbana-Champaign and a BA in Economics from the
University of Michigan, Ann Arbor. Dr. Jókay established a family foundation
to support the education of poor, rural children in the High School of the
Reformed Church in Pápa, Hungary.
Galina Kurlyandskaya is a Russian expert in the field of public finance and intergovernmental relations. She is the Director General of the Center for Fiscal
Policy in Moscow, a Russian think tank on government finance and intergovernmental relations in transitional economies. Dr. Kurlyandskaya is providing
research-based policy advice and technical assistance to central, regional, and
About the Editors and Contributors
local governments in the area of fiscal policy, public finance management, and
intergovernmental relations both in Russia and in other developing countries.
She received her PhD in Economics from the Institute for World Economy
and International Relations, Russian Academy of Sciences, Moscow.
James Leigland is the Technical Advisor to the Private Infrastructure Development Group, where he manages the Technical Assistance Facility. Prior
to retiring from the World Bank in 2010, Dr. Leigland was Team Leader for
Subnational Technical Assistance at the Public–Private Infrastructure Advisory Facility (PPIAF), a global, multidonor trust fund managed by the World
Bank. His other positions at PPIAF since 2005 include Program Leader for
Africa and Acting Program Manager. Before joining the World Bank, Dr. Leigland was Consulting Team Leader at the Municipal Infrastructure Investment Unit, a project development company created by the South African
government, which structured 45 private financing deals from 1998 to 2005,
including public–private partnerships, privatizations, and municipal bond
issues. Dr. Leigland served as Senior Urban Policy Adviser for East Asia at
the U.S. Agency for International Development in the mid-1990s, and is a
former faculty member at Columbia University in New York City, where he
earned his PhD in Political Economy. He is the author of 35 professional
publications on infrastructure, privatization, and subnational finance.
Gilberto M. Llanto is Senior Research Fellow at the Philippine Institute for
Development Studies. He was formerly Deputy Director General (Under
Secretary) of the National Economic and Development Authority and
Executive Director of the Agricultural Credit Policy Council. He was president of the Philippine Economic Society during 2004–05 and was on the
editorial board of the Philippine Journal of Development. He is currently a
member of the Konrad Adenauer Medal of Excellence Committee, which
gives awards to outstanding local government units in the Philippines. His
research interests include local finance, housing finance, public finance, and
growth economics. He has a PhD in Economics from the School of Economics, University of the Philippines, Manila.
Alvaro Manoel is Senior Economist in the World Bank’s Economic Policy and
Debt Department where he advises on a wide range of issues, including debt
management, debt sustainability analysis, subnational government public finances, macro vulnerabilities, and fiscal policy. Before joining the Bank, he
xxiii
xxiv
About the Editors and Contributors
was Senior Economist at the Fiscal Affairs and African Departments at the
International Monetary Fund (IMF), where he participated in IMF program
discussions and technical assistance on fiscal issues in more than 20 countries. Dr. Manoel also served at the Ministry of Planning and Budget in Brazil
where he was team leader of the Economic Advisers Council and coordinated a government task force that designed and drafted the Fiscal Responsibility Act (1998–2000) and then negotiated its provisions with Members of
the Brazilian Congress. Dr. Manoel also held the position of Deputy Secretary of the National Treasury at the Ministry of Finance in Brazil. He holds a
PhD in Economics from the University of São Paulo, Brazil, and is a former
professor of Economics at the University of Brasília, Brazil.
Mônica Mora is an economist with the Institute for Applied Economic
Research (IPEA) in Brasilia, Brazil. She joined the IPEA in 1998, working in the
Macroeconomic Department. Her research and expertise are in the areas
of public finance, fiscal policy, and macroeconomic coordination. Ms. Mora
has also taught economics at the Brazilian Institute of Capital Markets since
2004. She holds both a BA and an MA in Economics from the Universidade
Federal do Rio de Janeiro, Brazil.
Edgardo Mosqueira has been a Senior Public Sector Specialist at the World
Bank since 2003. His expertise and experience are focused on public policy,
public management, decentralization, and land policies and management,
and he has experience in public sector and governance reforms in Argentina,
Bolivia, Colombia, Croatia, Ecuador, Mexico, Peru, and Serbia. Mr. Mosqueira
served as Peruvian Minister of the Presidency, Minister of Labor, and in other
high-level government positions in Peru. Before joining the Peruvian government, he was senior researcher at the Hernando de Soto Peruvian Institute
for Liberty and Democracy. He was also Dean of the Universidad Peruana de
Ciencias Aplicadas School of Law and an expert member of the United
Nations Development Programme Commission on Legal Empowerment of
the Poor. He holds an MA in International Public Policy from the Johns
Hopkins University School of Advanced International Studies, Baltimore,
Maryland and a JD from the Pontifical Catholic University of Peru, Lima.
Tebogo Motsoane is a Senior Economist with the South African National
Treasury (SANT). He joined SANT in 2004, at which time he worked on
About the Editors and Contributors
s ubnational fiscal transfers. His research and expertise are in the areas of
local government finance and municipal debt finance. Mr. Motsoane has
contributed to or authored several papers and has presented papers at various forums on municipal debt finance. His publications include “Leveraging Private Finance,” a chapter he contributed to the 2008 and 2011 Local
Government Budget and Expenditure Review, published by SANT; and “Private Sector Investment in Infrastructure,” a chapter he contributed to the
May 2009 Municipal Infrastructure Finance Synthesis Report, published by
the World Bank. Mr. Motsoane holds a Diploma in Cost and Management
Accounting from the University of Johannesburg, South Africa.
John E. Petersen was an expert in public finance and a leading economist analyzing the municipal market. He was a member of the Municipal Securities
Rulemaking Board, professor at the George Mason School of Public Policy,
Fairfax, Virginia, and a highly regarded expert on the municipal bond market.
Dr. Petersen began his career as a capital markets economist at the Federal
Reserve Board. He was Director of Public Finance for the Securities Industry
Association and then Director of the Center for Policy Research and Analysis
for the National Governors Conference in the 1970s, after which he became
Senior Director of the Government Finance Officers Asscciation’s Government Finance Research Center. In 1992, Dr. Petersen founded the Government Finance Group, a consulting firm that provided international financial
advisory and research services. From 1998 through 2002, he was a division
director at ARD/Government Finance Group. He had a BA in Economics from
Northwestern University, Evanston, Illinois, an MBA from the Wharton School
at the University of Pennsylvania, Philadelphia, and a PhD in Economics from
the University of Pennsylvania. Dr. Petersen passed away in early 2012.
Abha Prasad is Senior Debt Specialist with the Economic Policy and Debt
Department of the World Bank. At the World Bank, as part of a core team,
she has developed and has been leading the work on the Debt Management Performance Assessment tool, including for subnationals. She has
been working on debt and debt management issues in developing countries
(Bangladesh, Bhutan, Guyana, the Maldives, Moldova, Nicaragua, Nigeria,
Tanzania, and Vietnam,) and in subnationals (Brazil, India, Indonesia, Nigeria,
and Peru). Previously, she was Director of the Internal Debt Management
Department of the Reserve Bank of India, where she helped manage the
xxv
xxvi
About the Editors and Contributors
debt of India’s federal government and 28 subnationals. She provided policy
advice for framing India’s fiscal responsibility legislation and restructuring
state government debt in 2004. She has published on many issues including debt, subnational finances, banking, and operational risk management in
debt management. Ms. Prasad holds Master’s degrees from the University
of Delhi, India; and Georgetown University, Washington, DC.
Baoyun Qiao is a Professor of Economics and Dean, China Academy of Public
Finance and Policy, Central University of Finance and Economics, in Beijing,
China. Dr. Qiao has been invited by many governments and international organizations, including China, Indonesia, Pakistan, and Thailand, and the Asian
Development Bank and the World Bank, to deliver keynote speeches and
presentations on intergovernmental fiscal relations, fiscal decentralization,
local governance, subnational debt and management, and subnational infrastructure financing. Dr. Qiao has published numerous books and papers,
and his research concentrates on the areas of public finance, taxation, intergovernmental fiscal relation, budget management, local governance, and
economic development. He served as the Vice President of the Chinese
Economist Society during 2007–08. Dr. Qiao graduated from Georgia State
University, Atlanta, where he earned a PhD in Economics.
C. Rangarajan is the Chairman of the Prime Minister’s Economic Advisory
Council, India. Formerly, he held several distinguished positions as Chairman
of the Twelfth Finance Commission, the Governor of Andhra Pradesh, and
the Governor of the Reserve Bank of India. He was part of the core team
that was responsible for initiating the liberalization process in India. He is
a distinguished former member of the Indian Parliament. Dr. Rangarajan
has taught at several institutions including the University of Pennsylvania
in Philadelphia, New York University in New York City, and the prestigious
Indian Institute of Management, in Ahmedabad. In 2002, the Government
of India awarded him the Padma Vibhushan, India’s second-highest civilian award. His work on the Finance Commission for the distribution of
resources between the center and state governments in India is an exemplary example for decentralizing countries in managing intergovernmental
relations and transfers. He has published on many subjects including fiscal
federal transfers, monetary policy, agricultural growth, banking, macroeconomics, and forecasting.
About the Editors and Contributors
Ernesto Revilla is Chief Economist of the Mexican Ministry of Finance.
Before assuming his current position, he was Chief of the Tax Policy Unit
at the Ministry, in charge of the design and implementation of revenue
policy and programs, economic analysis for tax policy decisions, and
estimates for the budget and fiscal policy choices. Mr. Revilla also served
as Director General of Fiscal Federalism at the Ministry, where he was
in charge of federal transfers to subnationals and the design of policy
related to the Mexican intergovernmental fiscal relations framework;
as Chief of Staff of the Undersecretary of Revenues; and as Deputy
Director General of Securities Markets. Prior to joining the Ministry,
Mr. Revilla served at the World Bank Headquarters in Washington, DC. As
a member of the Bank’s Young Professionals Program, he was an economist in the Financial Sector Development Vice Presidency and then in
the East Asia and Pacific Region. Mr. Revilla holds degrees in economics
from the Instituto Tecnológico Autónomo de México (ITAM) and the
University of Chicago in Illinois. He currently teaches Advanced Macroeconomics at ITAM.
Juan Pedro Schmid is currently the Country Economist for Jamaica at the
Inter-American Development Bank (IDB), responsible for a broad range
of issues including monitoring of the macroeconomy, preparation of
the country strategy, and programming and analytical work. Before joining
the IDB, Dr. Schmid worked for three years as an economist at the Economic Policy and Debt Department of the World Bank. While his main area
of expertise was debt relief under the Heavily Indebted Poor Country initiative, he also contributed to work in debt management, growth analytics,
and fiscal policy. Dr. Schmid holds an MA in Economics from the University
of Zurich and a PhD in Economics from The Federal Institute of Technology,
both in Zurich, Switzerland.
Xiaowei Tian is a staff member in the Department of Finance of Gansu
Province, China. He has worked on public expenditure, budgetary management, and fiscal policies on agriculture and ecological protection.
Mr. Tian was a consultant in the Economic Policy and Debt Department of
the World Bank’s Poverty Reduction and Economic Management Network.
He worked on public finance, the impact of the global crisis on government
finance, particularly at the subnational level, and on the subnational debt
xxvii
xxviii
About the Editors and Contributors
market development in major economies. He holds an MS in Economics
from the University of Illinois at Urbana-Champaign.
Michael Waibel is a University Lecturer in Law and a Fellow of the
Lauterpacht Centre for International Law at the University of Cambridge,
Cambridge, United Kingdom. His main research interests are public international law and international economic law, with a particular focus on finance
and the settlement of international disputes. At Cambridge, he teaches international law, World Trade Organization law, and European Union law. He
has also taught at Harvard College in Boston, Massachusetts; the London
School of Economics (LSE) in London, United Kingdom; and the University
of St. Gallen in St. Gallen, Switzerland. He is the author of the monograph
“Sovereign Defaults before International Courts and Tribunals,” published
by Cambridge University Press in 2011. Dr. Waibel holds an LLM and a JD
from the Universität Wien, Vienna, Austria; an MSc in Economics from LSE;
and an LLM from Harvard Law School. He is a member of the New York Bar
and holds a diploma from the Hague Academy of International Law in the
Netherlands. He has worked for several international organizations including the European Central Bank, the International Monetary Fund, and the
World Bank.
John Joseph Wallis is Professor of Economics at the University of Maryland,
College Park, and a Research Associate at the National Bureau of Economic
Research. He is an economic historian and an institutional economist who
focuses on the dynamic interaction of political and economic institutions
over time. As an American economic historian, he has collected large data
sets on government finances and on state constitutions and has studied how
political and economic forces changed American institutions in the 1830s and
1930s. In the past decade, his research has expanded to cover a longer period,
a wider geography, and more general questions of how societies use institutions of economics and politics to solve the problem of controlling violence
and, in some situations, sustain economic growth. Dr. Wallis is the co-author
of Violence and Social Orders: A Conceptual Framework for Interpreting
Recorded Human History, written with Douglass North and Barry Weingast
(Cambridge University Press 2009). Dr. Wallis received his PhD from the University of Washington, Seattle, in 1983 and spent two years as a postdoctoral
fellow at the University of Chicago.
About the Editors and Contributors
Steven B. Webb has for the past 24 years worked for the World Bank as
lead economist, adviser, and consultant in research, evaluations, and operations in Africa, Latin America, the Caribbean, and other regions. His specialties include political economy, public finance, central banks and monetary
policy, decentralization, and economic history. His publications include numerous articles and several books: Hyperinflation and Stabilization in Weimar Germany; and with co-authors, Achievements and Challenges in Decentralization: Lessons from Mexico; Public Sector Reform—What Works
and Why?; Voting for Reform: The Politics of Adjustment in New Democracies; and In the Shadow of Violence: Politics, Economics and the Problems
of Development. Dr. Webb has also taught in the Economics Department
at the University of Michigan, Ann Arbor, and was a visiting senior economist at the U.S. Department of State. He received a BA in economics from
Yale University in New Haven, Connecticut, and an MA in History and a PhD
in Economics from the University of Chicago in Illinois.
xxix
Abbreviations
ARI
average regular income
ARO
anticipated revenue credit operations
BLGF
Bureau of Local Government Finance
BPA
Bonneville Power Administration
CGCT General Code of Territorial Entities, Code Général des
Collectivités Territoriales
DAF
Fiscal Support Agency, Dirección General de Apoyo Fiscal
DBSA
Development Bank of Southern Africa
DCED Department of Community and Economic Development,
Pennsylvania
DGCL Direction Générale des Collectivités Locales, General
Directorate of Subnational Entities
DGCP General Directorate of Public Accounting, Direction Générale
de la Comptabilité Publique
DOF
Department of Finance
a
euro
ESF
Emergency Social Fund
EU
European Union
FC
Finance Commission
FEIEF Fund for the Stabilization of the Federal Revenue for the
Federal Entities, Fondo de Estabilización de los Ingresos de las
Entidades Federativas
FIFA
Fédération Internationale de Football Association
FINDETER Subnational Development Financial Entity, Financiera de
Desarrollo Territorial, SA
xxxi
xxxii
Abbreviations
fisc the combination of a government’s fiscal activity, which
includes revenues, expenditures, and debts
FONPET National Pension Fund, Fondo Nacional de Pensiones en las
Entidades Territoriales
FPE
State Participation Fund
FRL
Fiscal Responsibility Law, Brazil
FRL
fiscal responsibility legislation, India
FUNDEF
Financial Fund for Educational Services
GDP
gross domestic product
GFI
government financial institution
GJMC
Greater Johannesburg Metropolitan Council
GNP
gross national product
GSDP
gross state domestic product
GTS
General Transfer System, Sistema General de Participaciones
IBGE
Brazilian Institute of Geography and Statistics
ICMS value-added tax on goods, intermunicipal transportation, and
communications services
IFI
International Financial Institution
IGP-DI
General Price Index, Domestic Availability
IMF
International Monetary Fund
INFIS Institutes for Territorial Promotion and Development, Institutos de Fomento y Desarrollo Territorial
IPI
value-added tax on industrialized products
IRA
internal revenue allotment
IRRF
income tax withheld at source
ISS
service tax
KWK
Kloha, Weissert, and Kleine
LGC
Local Government Commission, North Carolina
LGU
Local Government Unit
LGUGC
Local Government Unit Guarantee Corporation
LWUA
Local Water Utilities Administration
MAV
national railways
MDFO
Municipal Development Fund Office
MEC
Member of the Executive Council
MFMA
Municipal Finance Management Act
MFPC Ministry of Finance and Public Credit, Ministerio de Hacienda
y Crédito Público
Abbreviations
MHCP
Ministry of Finance, Ministerio de Hacienda y Credito Público
MOF
Ministry of Finance
MSRB
Municipal Securities Rulemaking Board
MVM
power grid
NCR
net current revenue
NPD National Planning Department, Departmento Nacional de
Planeación
NSSF
National Small Savings Fund
OCIP
Orange County Investment Pool
OECD
Organisation for Economic Co-operation and Development
OTB
Off-Track Betting Corporation
OTP former monopoly state-owned savings bank privatized in
1995, Hungary
PACER
Public Access to Court Electronic Records
PAF Incentive Program for the States’ Restructuring and Fiscal
Adjustment
PAI
Federal Immediate Action Plan (1993)
PFI
private financial institution
PICA
Pennsylvania Intergovernmental Cooperation Authority
PPP
public-private partnership
PSBR
public sector borrowing requirements
PSUs
public sector undertakings
PUDs
public utility districts
RBI
Reserve Bank of India
RCA
Regional Chambers of Accounts
RCL
net current revenue
RDF
rainy day funds
RLR
real net revenue
S&P
Standard & Poor’s
SEC
U.S. Securities and Exchange Commission
SELIC
Special Settlement and Custody System
SELIC rate prime interest rate: average charged on daily operations
backed by treasury bills and bonds, registered at SELIC
SEMs
public-private partnerships, sociétés d’économie mixte locales
SGP
Central Government Transfers
SNG
subnational government
SOC
Superintendency of Corporations
xxxiii
xxxiv
Abbreviations
SOE
state-owned enterprise
STN
National Treasury Secretariat
TSS
Tax Sharing System
UDC
Urban Development Corporation
UDIC
Urban Development and Investment Corporation
URV
real value unit
USAID
United States Agency for International Development
VAT
value-added tax
WPPSS
Washington Public Power Supply System
WMA
ways and means advances
ZAC Centers of Urban Development, Zone d’Aménagement
Concerté
Note: All dollar amounts are U.S. dollars unless otherwise stated.
An Overview
Otaviano Canuto and Lili Liu
Subnational Debt and Insolvency
State and local debt and debt of quasi-public agencies have grown in
importance. Three structural trends have contributed to the rising share
of subnational finance, including subnational debt, as a share of general
public debt (Canuto and Liu 2010a).
First, decentralization in many countries has given subnational governments (SNGs)1 certain spending responsibilities, revenue-raising
authority, and the capacity to incur debt. With sovereign access to financial markets, SNGs are seeking access to these markets as well.
Second, rapid urbanization in developing countries requires largescale infrastructure financing to help absorb influxes of rural populations.2 Borrowing enables SNGs to capture the benefits of major capital
investments immediately, rather than waiting until sufficient savings
from current income can be accumulated to finance them. Infrastructure investments benefit future generations, which should bear a portion of the cost. Subnational borrowing finances infrastructure more
equitably across multigenerational users of infrastructure services
because the debt service can match the economic life of the assets that
1
2
Until Debt Do Us Part
the debt is financing. Infrastructure services thus can be paid for more
equitably by the beneficiaries of the services.
Third, the subnational debt market in developing countries has been
going through a notable transformation. Private capital has emerged to
play an important role in subnational finance, and subnational bonds
increasingly compete with traditional bank loans. Notwithstanding
the temporary disruption of the subnational credit markets during the
2008–09 global financial crisis, the trend toward more diversified subnational credit markets is expected to continue. SNGs or their entities
in various countries have already issued bond instruments (for example, in China, Colombia, India, Mexico, Poland, the Russian Federation,
and South Africa). More countries are considering policy frameworks
for facilitating subnational debt market development (for example,
Indonesia), while others are allowing selected subnational entities to
pilot-test transaction and capacity-building activities (for example, Peru).
With debt comes the risk of insolvency. When SNGs follow unsustainable fiscal policy, it can jeopardize the ability to service their debt,
the services they manage, the safety of the financial system, their
country’s international creditworthiness, and overall macroeconomic
stability. Too often, the central government gets dragged in to provide
bailouts, which can disrupt its own fiscal sustainability and reward the
populist fiscal tactics of the recipient SNGs.
Several major emerging markets experienced subnational debt crises
in the 1990s. Newly decentralized countries face potential fiscal risks.
To many observers, runaway provincial debt in the Provinces of Mendoza and Buenos Aires was a factor behind Argentina’s sovereign debt
default in 2001. Brazil experienced three subnational debt crises in the
1980s and 1990s. In India, many states experienced fiscal stress in the
late 1990s to the early 2000s, with increases in fiscal deficits, debt, and
contingent liabilities. The 1994–95 Tequila Crisis in Mexico exposed the
vulnerability of subnational debt.
Subnational insolvency is a recurring event in history. In 1842,
eight U.S. States and the Territory of Florida defaulted on their
debt, and three other states were in perilous financial condition
( Wallis 2005). During the Great Depression, 4,770 local governments
defaulted on US$2.85 billion of debt (Maco 2001). As capital markets and their regulatory framework matured, the default rates of U.S.
An Overview
local governments declined. Yet there are recent episodes, including
the default of the Washington Public Power Supply System in 1983,
and the bankruptcy of Orange County, California, in 1994 and of
Jefferson County, Alabama, in 2011.
The 2008–09 global financial crisis has had a profound impact on
subnational finance across countries, as a result of slowing economic
growth, the rising cost of borrowing, and deteriorating primary balances. The impact has been mitigated in various countries by fiscal stimulus, monetary easing, and increasing fiscal transfers. However, looking
forward, pressures on subnational finance are likely to continue—from
the potentially higher cost of capital, the fragility of global recovery, refinancing risks, and sovereign risks (Canuto and Liu 2010b).
Aligning Fiscal Incentives
Subnational debt crises have led governments across countries to search
for frameworks to restructure subnational debt and to undertake legal,
regulatory, and institutional reforms that will sustain subnational debt
finance in the long run. In a multilevel government system, the reforms
need to resolve three challenges (Liu and Webb 2011). The first challenge applies to governments at any level, whereas the second and third
are relevant mainly in countries with multilevel governments.
The first challenge is the short time horizon of public officials, who
have shorter terms of office than citizens’ life spans. Public officials face
the risk of being forced out of office if results are painful in the short
term. The mobility of citizens and businesses between local jurisdictions means that excess borrowing could drive residents away and leave
those remaining with more debt per person than they had anticipated.
The second challenge is free riders. The interests of individual subnational governments may diverge from the common national interest
when factors such as electoral pressures motivate subnational governments to follow unsustainable fiscal policy. An individual government
would bear only part of the cost of its misbehavior, but would still
receive all of perceived benefit accrued, only if (most of) the other governments continued to follow good fiscal behavior. So, there might be a
prisoners’ dilemma—a situation where the equilibrium of isolated individual choices leads to suboptimal outcomes for all.3
3
4
Until Debt Do Us Part
The third challenge is moral hazard. Subnational borrowers might
have an incentive not to repay their creditors, and creditors might lend
without risk differentiations, if they perceive that defaulting debtors
could be bailed out by the central government.
In a country with multilevel governments, the national government exists for the purpose (among others) of protecting the common
interest, and typically has special powers such as running the central
bank and regulating the financial sector. The national government
also provides transfers to the SNGs, giving it additional leverage over
SNGs and their fiscal behavior. However, the constitution and rules
(such as on revenue sharing) may constrain the national government’s
power over the SNGs. Political considerations, such as the national
political cycle or subnational political cycles, may bias the decisions of the national government away from the optimal (Braun and
Tommasi 2004). For instance, when a state government of the same
political party as the national government faces a close election, the
national government might be inclined to “condone” the state’s fiscal misbehavior by offering a debt bailout or rescheduling guarantee.
Also, under some configurations of political institutions, the national
executive might “purchase” blocks of legislative votes by giving SNGs
fiscal favors.
The incentives in the political system affect the need for effective
subnational fiscal control institutions. To the extent that the constitution and party system lead to more centralized power, the country may
have less need for special institutions to coordinate fiscal discipline
across governments over time and among SNGs. Decentralization and
market decontrol, however, increase the need for coordination of fiscal
discipline.
The subnational debt crises or fiscal stress of the 1990s in several
major developing countries led to reforms in subnational borrowing
frameworks, including the development of ex-ante fiscal rules and debt
limits. The search for insolvency resolution has also intensified, since
ex-ante rules have been shown not to be sufficient on their own without
ex-post mechanisms. Insolvency mechanisms should increase the pain
of circumventing ex-ante regulation for creditors and debtors, thereby
enforcing preventive rules.
An Overview
Key Design Issues in Subnational Debt Restructuring
The country experiences in this volume reveal several design issues with
respect to debt restructuring frameworks: (a) how to balance the tension
between the contractual rights of creditors and the need for maintaining
public services in the event of subnational insolvency; (b) how to define the
respective roles of different levels and branches of government in resolving insolvency; (c) how to develop a collective framework for debt resolution; and (d) a basic choice among a judicial, administrative, or hybrid
approach. The country cases show that country-specific circumstances—
historical, constitutional, and economic context, and entry points for
reform—influence the framework design in each country.
The framework design ultimately needs to address the challenges
of fiscal incentives facing SNGs in a multilevel government system.
A sound framework should reduce the moral hazard of subnational
defaults, discourage free riders, bind all SNGs to pursue sustainable fiscal policies, and extend the short-term horizon of SNGs to minimize
the impact of unsustainable fiscal policy on future generations.
Public and Private Insolvency
Insolvency of subnational governments differs from that of private corporate entities. The core difference is the public nature of the services
provided by SNGs. Thus, debt restructuring inevitably involves a difficult balance between interests of the debtor (and the citizens it serves)
and the creditors (and savers). While a corporation can be dissolved, this
route is barred for SNGs. When a private corporation goes bankrupt, all
of its assets are potentially subject to attachment. The ability of creditors
to attach assets of SNGs is constrained in many countries. In the United
States, a judicial doctrine typically holds that only proprietary property
is attachable. “Proprietary property,” subject to debt foreclosure, was
defined by the U.S. Supreme Court as “held in (the municipality’s) own
right for profit or as a source of revenue not charged with any public
trust or use”4 (McConnell and Picker 1993).
Who Has the Authority over What?
Fiscal adjustment by debtors requires difficult political choices to
bring spending in line with revenues and to bring borrowing in line
5
6
Until Debt Do Us Part
with debt service capacity. In a decentralized system, tension exists
between the role of the national government in enforcing collective fiscal discipline of SNGs and the fiscal autonomy of SNGs. Can
a higher-tier government force spending cuts and tax increases in a
lower-tier government? Can courts influence spending priorities and
tax choices that are normally reserved for legislative and executive
branches? How do a country’s legal framework and political reality
define the roles of different tiers and branches of the government?
These are among the key issues that the case studies in this volume try
to address.
Subnational fiscal adjustment is also complicated by the legislative mandates of the central government vis-à-vis subnational
governments and the intergovernmental finance system (Ianchovichina, Liu, and Nagarajan 2007). Unable to issue their own currency,
subnationals cannot use seigniorage finance. SNGs may not freely
adjust their primary balance due to legal constraints on raising their
own revenue, dependence on central government transfers, or the central government’s influence on key expenditure items such as wages
and pensions. Many other policies that affect economic growth and
fiscal health of the subnational economy may also be determined
largely by the central government.
Debt restructuring and debt discharge are complex processes but
can be distilled into two basic questions: whether the creditors and
the debtor can reach agreement on debt resolution; and who holds the
cramdown power5 when both sides fail to reach an agreement (Liu and
Waibel 2009). In Brazil and Mexico, the national government led SNG
debt restructuring, and there were no debt write-offs. In Hungary and
the United States, the courts hold cramdown power when local governments and creditors negotiate.
Clarity of Rules and Collective Enforcement
Without an insolvency framework, subnational debtors and their
creditors resort to ad-hoc restructuring negotiations. The need for a
collective framework for resolving debt claims is driven not only by
conflicts between creditors and the debtor, but also by competing
interests among creditors and competing demands by constituents of
the debtor. Individual creditors may have different security provisions
An Overview
for the debt owed to them and may demand preferential treatment and
threaten to derail debt restructurings voluntarily negotiated between a
majority of creditors and the subnational debtor—the “holdout problem.” Individual ad-hoc negotiations can be costly and harmful to the
interests of a majority of creditors (McConnell and Picker 1993). The
holdout problem is not as serious if debts are concentrated in a few
banks. A collective framework for restructuring takes on more importance as the subnational bond market develops and grows to include
thousands of creditors.
Lack of clear rules for insolvency is likely to raise borrowing costs,
and may limit market access for creditworthy borrowers. South African
policy makers viewed clear rules for insolvency as critical to the growth
of a broad-based competitive subnational capital market. In the United
States, utilization of Chapter 9 of the Bankruptcy Code has carried a
strong stigma for a defaulting municipality, to offset debtor moral hazard. Municipalities are thus wary that capital markets would interpret
the filing for federal bankruptcy protection as a strong signal of financial mismanagement, to which lenders are likely to react by charging a
risk premium.
The tension between maintaining essential services and creditors’
contractual rights would imply that the pain of insolvency needs to be
shared between creditors and the debtor. The insolvency mechanism
needs to balance these competing interests and guide the priority structure of settling competing claims. The priority structure will depend,
first, on the distributional judgment of the society concerned and, second, on the effect of a chosen priority structure on the capital market
and its impact on new financing (Liu and Waibel 2009).
Judicial vs. Administrative Approach
The two approaches to subnational insolvency procedures discussed
in this volume are the judicial and the administrative.6 Various hybrids
also exist. In judicial procedures, courts make decisions to guide the
restructuring process. The judicial approach has the advantage of
neutralizing political pressures during the complex restructuring.
However, the courts’ ability to influence fiscal adjustment of SNGs
is limited because mandates for budgetary matters usually rest with
the executive and legislature. In some administrative interventions,
7
8
Until Debt Do Us Part
by contrast, a higher-level government intervenes in the entity concerned, temporarily taking direct political responsibility for many
aspects of financial management and restructuring the subnational’s
debt obligations into longer-term debt instruments.
The choice of approach varies across countries. In Hungary, the desire
to neutralize political pressure for bailing out insolvent subnationals
favors the judicial approach. South Africa’s legal framework for municipal bankruptcy is a hybrid, blending administrative intervention with
the role of courts in deciding debt restructuring and discharge. Colombia has a formal administrative process, where central government representatives facilitate restructuring negotiations between subnational
borrowers and their creditors, and supervise the implementation of the
agreement on fiscal adjustment and debt workouts. In Brazil, the federal
government restructured the subnational debt in the late 1990s conditional on the SNG undertaking fiscal reform and adjustment packages.
Similarly, the federal government in Mexico restructured states’ debts
after the Tequila Crisis, and a few years later introduced regulations on
the lenders that effectively constrained the borrowers as well. In India,
the federal government used a debt swap instrument as an incentive to
encourage states to enact their own fiscal responsibility laws.
Reforms to Align Fiscal Incentives and
Develop a Robust Framework
Reforms in subnational borrowing frameworks and debt restructuring mechanisms have gathered momentum in developing countries
since the late 1990s. The objectives of reforms are broadly similar—
strengthening fiscal management and preventing future insolvency.
Often, these proceed in tandem with broader public finance reforms,
macroeconomic stabilization, and the development of a robust
medium-term fiscal framework and transparency. The reform paths
and sequences that these countries chose reflect their own historical
context, legal framework, and reform dynamics.
This volume surveys the reform experience of selected countries in
strengthening subnational fiscal discipline and developing a framework
for the resolution of subnational debt stress. The first two sections of
the volume focus on two types of debt restructuring. One type is the
An Overview
national-government-led debt restructuring, which includes the experiences of Brazil, India, and Mexico. The review also includes China’s
experience in central-government-led restructuring of the rural education legacy debt, so that local governments could gain stronger fiscal capacity for education service delivery. Another type focuses on the
framework that spells out, in advance, the procedure in place in the
event of a subnational default. It compares the experiences of Colombia,
France, Hungary, and the United States. Subnational insolvency is not
limited to developing countries. The reform experience of developed
countries offers important lessons.
The third section of the book discusses the experiences of China, the
Philippines, Russia, and South Africa in developing their subnational
credit markets. This topic is highly relevant to aligning fiscal incentives for SNGs and developing a robust regulatory framework. When
the central government refrains from bailouts, creditors serve as an
enforcer of fiscal discipline on SNGs by pricing risks of defaults. Reducing default risks is not the same as minimizing the use of debt instruments. As already noted, debt instruments are essential for financing
large-scale infrastructure and supporting economic growth. Competitive supply of subnational credits lowers borrowing cost and extends
loan maturity.
We also include the United States, which has the largest subnational capital market in the world, with outstanding SNG (states and
local governments and their special purpose vehicles) debt of US$3.4
trillion and an annual average issuance of US$450 billion. However,
the United States was not endowed with a mature, well-functioning
market from the outset. Over its long history, the U.S. subnational capital markets experienced episodes of widespread defaults in the 1840s,
1870s, and 1930s. The reforms of legal frameworks and institutions
have been gradual and path dependent, in the sense that later reforms
built on earlier reforms. The United States experience offers lessons
for developing countries, but a developing country could not simply
duplicate the institutions that currently govern subnational borrowing in the United States. Nonetheless, the United States does offer
relevant lessons including the importance of tying revenue sources to
borrowing, transparency in markets for government credit, and creating interest among creditors in strengthening borrowing rules.
9
10
Until Debt Do Us Part
National-Government-Led Subnational Debt Restructuring
Part 1 of the volume reviews subnational debt restructuring in Brazil,
China, India, and Mexico. Brazil, India, and Mexico all have a federal
system, but the origins of their SNG debt crises differ, as do the formulations of their restructuring programs. The chapter on China focuses on
the restructuring of rural school debt for better alignment of the intergovernmental fiscal system and supporting inclusive economic growth.
These countries offer lessons on the common-pool and moral hazard
problems inherent in debt restructuring and how they have moved
toward rule-based transparent frameworks.
SNGs in Brazil experienced debt crises during the 1980s and
1990s, a period of macroeconomic instability, oil shocks, and a balance-of-payments crisis. Circumventing the strict controls imposed
on SNGs, public sector borrowing increased substantially to finance
capital investments. The federal government provided bridge loans
to assist SNGs in rolling over their external debt, but SNG debt stress
was unabated. The federal government restructured SNG external
debt in 1989 and SNG debt owed to federal entities in 1993. With
a substantial part of SNG debt unresolved and persistent pressures
placed on the primary balance, the fiscal and debt position of SNGs
continued to deteriorate. As hyperinflation was brought under control, the SNG debt obligations rose rapidly in real terms. The federal
government initiated a third round of debt renegotiations in 1996.
During the first two rounds of debt restructuring, state politicians
suffered minimal consequences and their creditors suffered almost
none. The cycle of failure in discipline and cooperation came to a halt
in the third round of debt restructuring, as the deeper political and
economic incentives had changed after a national macroeconomic
adjustment program ended hyperinflation and stabilized the economy.
Beyond macroeconomic stabilization, the federal Real Plan sought to
reorganize the entire public sector and privatize banks and public enterprises. The federal government offered SNGs incentives to restructure
their debt, but also required them to undertake comprehensive structural and fiscal reforms, including privatization of state-owned banks
and enterprises. Three of the four largest debtor states supported
the reforms and formed the core of a critical mass of states ready to
An Overview
cooperate in fiscal restraint, making it worthwhile for additional states
to join the reform.
The success of the Real Plan, together with the third round of SNG
debt restructuring, created the political and economic conditions for
enactment of the Fiscal Responsibility Law (FRL) in 2000. The law
established limits and placed restrictions on key fiscal variables and
assigned responsibility for enforcing the obligations and fiscal transparency requirements. Throughout the 2000s, state and municipal finances
improved significantly. Total public net debt as a share of gross domestic product (GDP) declined from 52 percent in 2001 to 39 percent in
2010, with the decline at all three levels of government. The improvement in the SNG fiscal accounts is associated with Brazil’s improved
macroeconomic fundamentals, but has come at a cost in reduced SNG
infrastructure investment.
Local governments in China resorted to borrowing to finance school
facilities to meet the goal of universal nine-year compulsory education. In
2000, China achieved the goal, a historic accomplishment. However, the
rural education debt became a significant fiscal burden on local governments. With the public policy goal in the late 1990s of inclusive economic
growth, the debt financing of nine-year compulsory education in the rural
areas was replaced by grant financing for all children. The new policy
needed to address the legacy debt and its write-offs.
With a strong fiscal position, the central government could easily
have written off the entire debt. This option was not chosen because it
would have encouraged moral hazard. The debt restructuring program
in 2007 shared the fiscal responsibility for debt write-off among three
tiers of government. The central government grants used an outputbased rather than an input-based formula, which took into account
both the required expenditure to achieve basic provision of education
and the local government fiscal capacity. This output-based approach
was designed to prevent perverse incentives for local governments
to increase the size of their debt or to reduce their service of debt in
anticipation of more grants or bailouts. A local government that borrowed excessively would not gain extra advantage, and another local
government that borrowed less or paid off its debt would not be in an
unfavorable position.
11
12
Until Debt Do Us Part
The Constitution of India forbids states from borrowing abroad and
requires them to obtain central government permission for domestic
borrowing. The central government places limits on states’ borrowing
through the annual discussions with states on financing state development plans. While limiting explosive growth and systemic insolvency
of state debt, the system did not prevent deterioration of fiscal conditions as indicated by high levels of debt over gross state domestic
product in many states in the late 1990s. The outstanding state debt
to GDP peaked at 32.8 percent during 2003–04, up from 20 percent
during 1997–98, and interest payments as a share of revenue receipts
increased from 16.9 percent to 26 percent over the same period. Factors contributing to the deteriorating fiscal accounts across Indian
states in the late 1990s include the rapid increase in expenditures on
salaries, retirement benefits, pensions, and subsidies; increased borrowing to support the growing revenue deficit (current expenditure
in excess of revenue including fiscal transfers); and growth in contingent liabilities associated with fiscal support to state-owned public
enterprises.
Since the early 2000s, fiscal reform has focused on moving toward
a more flexible, market-linked borrowing regime within sustainable
overall borrowing caps imposed by the central government and selfimposed state-level deficit caps. The federal government enacted the
Fiscal Responsibility and Budget Management Act in 2003, which
applies to the national government only, but some states had also
adopted their own FRLs before the enactment of the federal FRL (for
example, Karnataka and Punjab in 2002), and many states have since
2003 adopted FRLs in line with the national law. FRLs became mandatory after the Twelfth Finance Commission, and the federal government offered a sizable incentive to restructure high-cost debt to states
for passing FRLs.
During the centralized system in Mexico before the 1990s, subnational debt was implicitly guaranteed by the federal government. Important controls and consequences were outside the formal rules and were
based on political party connections. In the 1990s, Mexico’s federal government inadvertently involved itself in the decision making for subnational borrowing through pledged transfers and the implicit guarantee
of bailouts that came with them. Accordingly, creditors took little time
An Overview
to conduct thorough evaluations of subnational finances, and some
local governments borrowed beyond their means. The main vulnerabilities of the subnational debt profile were the high ratio of debt over
the shared revenues received by the states, and refinancing risks stemming from short debt maturity and floating interest rates. The 1994–95
Tequila Crisis resulted in a rapid currency depreciation, a sharp rise in
interest rates, and sharp declines in the pool of shared revenues, all of
which led to a state debt crisis. The development necessitated a costly
federal bailout program that forced a rethinking of subnational lending
parameters.
The federal bailout program rescheduled subnational debt into longterm inflation-indexed debt at affordable but positive real interest rates
and granted four years of assistance payments. To avoid a recurrence of
the fiscal crisis, each state had to agree to a fiscal adjustment program
designed by the Secretariat of Finance, which monitored compliance
prior to disbursement of the annual tranches of assistance, and brought
most states to a good financial situation by the end of the 1990s. The
indexed debt that the banks were forced to accept helped them avert total
ruin and collapse of the system, but illiquidity of the assets and low return
inflicted a penalty on the borrowers as well. The federal government also
ended its policy of formally guaranteeing subnational debt, although as a
transition it agreed to accept and execute contractual mandates by which
the borrowers pledged their revenue-sharing transfers as collateral for
the debt service. During 1999–2000, the federal government effectively
required credit ratings for subnational governments and brought in a
new subnational borrowing framework through tightened regulations on
the lending side. The federal constitution left little scope for direct regulation of the subnational borrowers.
We have learned from the experience of Brazil, China, India, and
Mexico that each debt restructuring regime needs to be based on the
origin of the debt problem and the specific historical and institutional
context of the debt stress. Debt restructuring plans must pay attention
to their incentive effects. Rule-based debt restructuring reduces ad-hoc
bargaining and adverse incentives; hard budget constraint prevents
moral hazard; and burden sharing provides proper incentives and
avoids free-riding behavior, while also recognizing the incentive role
played by higher levels of government to leverage reform.
13
14
Until Debt Do Us Part
Subnational Insolvency Systems
Part 2 of the book discusses the development of subnational insolvency systems in Colombia, France, Hungary, and the United States.
All four countries developed a framework for insolvency proceedings
in response to subnational debt crises. But the frameworks differ across
the countries, reflecting historical contexts, constitutional frameworks,
entry points for reform, and institutional developments that are path
dependent. While Hungary and the United States opted for court proceedings for insolvency, Colombia and France chose to use administrative proceedings. All four countries confronted key design issues,
whether they were federal (the United States) or unitary (Colombia,
France, and Hungary). Part 2 concludes with the 1983 default case of
the Washington Public Power Supply System in the United States, the
largest subnational bond default in modern U.S. history. This case illustrates the dynamic interactions among stakeholders that have ramifications for regulatory reform and market development.
In the late 1980s and 1990s, the trend toward political decentralization in Colombia was accompanied by more freedom for subnational
borrowing. SNGs experienced debt stress in the late 1990s to early 2000s,
exacerbated by the economic downturn. Contributing factors included
weak bank lending supervision, excessive reliance on transfers, and permission to borrow for current expenditure, which blunted incentives
for fiscal discipline. Although the SNG debt level was not high by international comparison, the arrears had been increasing by the late 1990s,
and SNG capacity for debt service had weakened, due primarily to the
decline in SNG own revenues and fiscal transfers.
The SNG debt stress led to substantial public finance reform.
Colombia enacted several laws, mostly between 1998 and 2003, that
regulate the origination of SNG debt and encourage fiscal responsibility.
Law 550 (1999) deals explicitly with bankruptcy proceedings for SNGs.
The essence of the proceedings is to evaluate and reconcile competing claims against subnational debtors, according to a defined priority
structure. The procedures in Colombia are administered by the Superintendency of Corporations (SOC) in coordination with other central
government institutions. Created in the 1930s, the SOC’s unique role
arose within a historical context in which the court system was weak.
An Overview
The SOC administers bankruptcy procedures for both corporations and
most government entities.
The implementation of Law 550 and other fiscal legislation has taken
place in the context of improving macroeconomic performance of the
country since 2003. There has been little divergence between the law
and its practice. The protection offered by bankruptcy Law 550 enables
insolvent SNGs to reach orderly debt restructuring agreements with
creditors. Focusing on debt workouts, Law 550 has limited ability to
address the root causes of fiscal stress and debt. Other complementary
laws—mainly Laws 358, 617, and 819—work in several ways by limiting
borrowing, promoting fiscal transparency, strengthening the budgetary
process, and helping to finance debt restructuring.
During 1982–83 and 2003–04, two waves of decentralization in
France devolved more powers to the three levels of SNGs: the municipalities, the departments, and the regions. This new institutional
framework has enabled SNGs to enjoy a greater degree of autonomous
expenditures, to raise their own taxes, and to borrow from financial
markets, within ex-ante rules established by the central government.
However, SNGs are subject to ex-post controls by the Prefect and the
Regional Chambers of Accounts, and to ongoing controls by the Public
Accountants.
The ex-ante fiscal rules and the regulatory framework for managing
SNG fiscal risks were established after a period of unregulated borrowing by SNGs following the decentralization and subsequent debt stress
experienced by some SNGs in the early 1990s. The regulatory framework combines the laws and regulations with three sets of institutions,
while preserving considerable SNG fiscal autonomy. The laws and prudential rules regulate debt, liquidity, and contingent liabilities. The state
exercises strong supervision and monitoring of SNG financial accounts
through the Prefect, the Regional Chamber of Accounts, and Public
Accountants. By law, SNGs cannot go bankrupt and public assets cannot be pledged as collateral. If an SNG is insolvent, the central government will intervene, enforcing fiscal adjustment and facilitating debt
negotiations among the creditors and the borrower. SNG accounts may
be placed under the control of the Prefect and the Regional Chamber
of Accounts for several reasons, including failure to present a balanced
budget, deficits exceeding 5 percent of operating revenues, and failure to
15
16
Until Debt Do Us Part
make provisions in the budget for compulsory expenditures including
debt services.
State supervision has helped to substantially reduce SNG insolvency risks, although several debt restructurings occurred in the last
two decades. Nonetheless, the lack of a clear, established legal structure
for priority payments creates uncertainties. Off-budget entities, such
as public-private partnerships, pose contingent fiscal risks, a common
challenge across countries.
The 1990 Law on Local Government in Hungary granted local governments unfettered freedom to manage their finances. They borrowed
for commercial activities and long term to finance short-term operating
deficits. The macroeconomic deterioration in the mid-1990s exposed
the seriousness of subnational financial distress. Imprudent lending by
public banks without proper evaluation of SNG creditworthiness was
attributed to the assumed central government guarantee for subnational debt. Several local governments successfully lobbied for central
government grants. This threatened to set a bailout precedent, raising
concerns of adverse incentives for debtors and creditors.
Several options were debated at the time, including informal
restructuring negotiations between creditors and local governments.
The deteriorating financial performances of local governments caused
a concern about creating contingent liabilities for the central government. The government eventually opted for a formal insolvency mechanism. Transparency and predictability were viewed as central to an
effective subnational insolvency mechanism. The Law on Municipal
Debt Adjustment, approved by the Hungarian Parliament in 1996, gives
courts the central role in fiscal and debt adjustment for insolvent local
governments.
The implementation experience has exceeded the expectations of
the framers of the law. The legal procedure is transparent, moral hazard of bailouts has been minimized, and essential services have been
maintained. The debt adjustment procedures have given participating
municipalities a clean slate to move forward. However, many insolvency
cases were resolved through informal negotiations. While bilateral negotiations are an integral part of all insolvency regimes, the nontransparency and potential asset stripping could negatively affect less-informed
or smaller creditors and the public interest. Discussions are ongoing
An Overview
among stakeholders on making the pre-bankruptcy negotiated restructuring more transparent.
In the United States, after the initial refinancing of national and state
debts incurred during the Revolutionary War, when the national government assumed the existing state debts, national government involvement in state and local government finances was minimal until the Great
Depression and New Deal programs of the 1930s. It was not until 1933
that the national government began significant grant and transfer programs to the states (Wallis 1984). Since the 1930s, national, state, and
local finances have been more closely intertwined, but the national government has generally maintained a no-bailout policy and has left the
structure and regulation of subnational borrowing to state governments.
As chapter 14 on U.S. state systems of local government borrowing shows,
there are 50 different subnational finance systems in the United States.
In response to widespread municipal defaults during the Great
Depression, the U.S. Congress in 1937 adopted a municipal insolvency
law known as Chapter 9 of the U.S. Bankruptcy Code. Chapter 9 is a
debt restructuring mechanism for political subdivisions and agencies
of U.S. states. The widespread subnational financial distress during the
Great Depression revealed the practical drawbacks of the mandamus
(a court order obliging municipalities to service debt obligations) and
informal protracted negotiations between municipal debtors and creditors (McConnell and Picker 1993). Chapter 9 delineates the procedures
whereby a debt restructuring plan acceptable to a majority of creditors
can become binding on a dissenting minority.
The design of Chapter 9 was guided by the U.S. constitutional provisions that reserve the control over state and local government finances
completely to the states. State consent is a precondition for municipalities to file for Chapter 9 in federal bankruptcy court. Chapter 9 is not
the primary subnational insolvency mechanism in the United States.
Only about half of states authorize their political subdivisions to file for
Chapter 9 relief. The unique federal structure of the United States also
profoundly influences the specific design of Chapter 9, where the federal
courts have limited ability to impose conditions on the debt adjustment
plan of an insolvent municipality. Most of the institutions that govern
subnational government borrowing in the United States are embodied
in state constitutions, state laws, and state administrative agencies.
17
18
Until Debt Do Us Part
The U.S. corporate insolvency laws have had influence on other
countries. While the U.S. municipal insolvency framework offers a valuable reference for other countries, the framework itself cannot be copied without care. Chapter 9 was conceived with the narrow objective of
resolving the holdout problem. It is based on a respected, independent,
and competent judiciary that has the authority to reject a municipality’s
Plan of Adjustment. In many developing countries, intergovernmental systems are still evolving, lending to SNGs may still be dominated
by a few public institutions, and judicial systems may lack capacity.
The development of a subnational insolvency mechanism must be
sequenced with other reforms.
To develop a legal framework to resolve financial distress, many
countries face similar objectives and challenges, namely, the interest in
the functioning of local government autonomy, safeguarding essential
public services and the assets that provide such services, transparent
procedures, the interests of creditors, and functioning subnational capital markets.
Part 2 of the volume also includes the largest municipal bond default
in the United States since the 1950s. In 1983, the Washington Public
Power Supply System (WPPSS) defaulted on US$2.25 billion in outstanding bonds. The debt issued by the WPPSS was ruled by the State
Supreme Court as invalid and unenforceable. Therefore, filing for
Chapter 9 was never an option. The default, a rare event in scale and
frequency in the modern U.S. subnational debt market, offers a window into the interactive roles of market, the courts, the regulators, the
debtor, the creditors, the federal and state governments, and taxpayers.
The WPPSS default shows that, even in a developed country, the
issuance of debt for infrastructure has endemic risks such as the lack of
transparency and disclosure, poor project management, and construction delays. But even if the debt issued is valid and legally enforceable,
the mounting problems in construction delays, cost escalations, and difficulty in refinancing existing debt would have made it difficult for the
WPPSS to pay back bondholders.
None of the bailout proposals was seriously considered by Congress.
The government took minimal enforcement action against actors in
the WPPSS drama, because the principle of self-regulation outweighed
the cost of enforcing regulations. Although there was little government
An Overview
action, the amount of private damage litigation was unprecedented and
resulted in many failed careers and business collapses. Very few individual market participants gained from the WPPSS disaster, but the
market not only weathered the storm—it became stronger. The U.S.
municipal market showed little evidence of damage resulting from the
WPPSS default. Not only did the market quickly return to normal after
the WPPSS default, but the period during which the WPPSS drama
unfolded, from 1975 to 1985, was one in which total annual municipal
bond issuance grew tenfold—a dramatic decade of growth in the history of the modern market.
Subnational Credit Market Development
Part 3 of the book focuses on subnational credit market development
in China, the Philippines, Russia, South Africa, and the United States.
Developing countries face long-term challenges in developing liquid,
deep, and competitive subnational credit markets. In general, bank
loans continue to dominate the supply of credit to SNGs in developing
countries, and public financial institutions continue to dominate credit
supply in some countries. Subnational securities markets in developing
countries in general are small in scale and lack liquidity and secondary markets. The United States has the largest, most liquid, and most
competitive subnational capital market, but the market development
has interacted with a series of institutional reforms through its history.
Although the lessons from the U.S. experience are highly specific to the
history of the American states, there are some general lessons for developing countries.
China has been investing about 10 percent of its GDP annually in
infrastructure, with SNGs taking on a large share of investments and
rapidly transforming the urban infrastructure landscape. SNGs relied
on central government onlending and their own off-budget vehicles—
Urban Development and Investment Corporation (UDIC), borrowing directly from the financial markets mainly through loans but also
bonds—and land assets-based finance to develop urban infrastructure.
The limitations of these financing instruments became evident to policy
makers in the mid-2000s. With central government onlending, SNGs
have no market interaction with creditors, and the borrowing power
19
20
Until Debt Do Us Part
and payment obligations are not linked. UDIC’s off-budget debt is nontransparent. Financing infrastructure through land lease is not sustainable in the long run, because of the up-front collection of leasing fees.
China has undertaken reforms since 2009 to allow the issuance of
provincial bonds and later the piloting of municipal bonds. Policy
makers recognized that important preconditions for the issuance of
bonds by provinces did not exist in 2009. It takes time to develop credit
rating systems, and SNGs had no market access experience. The reform
thus took a learning-by-doing approach. The central government acted
as the issuing agency, with SNGs participating in the auctions. From
2009 to 2011, RMB 600 billion (US$90 Billion) of provincial bonds
was authorized and issued. In 2011, reform took a further step: the
State Council approved piloting of direct bond issuance by four cities
(RMB 23 billion) without the central government acting as the issuing
agency.
The reform helped SNGs finance the subnational matching part of
investment projects in which the central government co-invested in
response to the 2008–09 global financial crisis. The new debt instrument significantly lowered the financing costs for SNGs, enabled them
to start acquiring market access skills, and linked the SNGs as debtors
with their debt service obligations. Piloting municipal bonds without
the central government as the issuer is one step further for SNGs to
access the market.
The issuance of SNG bonds has been supported by developing legal,
institutional, and market infrastructure. The reforms in fiscal management (including the single Treasury account and expenditure reforms)
and separating management from ownership of public enterprises have
laid the groundwork for the piloting of provincial bonds. The new bond
instrument to finance capital outlays under newly developed budgeting
procedures will facilitate the development of a framework for mediumterm capital budgeting for infrastructure investments. The audit of, and
the ongoing efforts to better classify, UDIC debt will facilitate the development of different bond instruments with different risks and securitization profiles. Further regulatory reforms can support sustainable
market access, as would complementary reforms in strengthening intergovernmental fiscal systems, enhancing fiscal transparency, and deepening financial markets.
An Overview
The Philippines is an emerging economy that continues to chart its
own course in developing its subnational debt markets. The Philippines
has been innovative in its efforts to extend the legal possibilities for local
governments to take initiative in the use of credit and in the design of
credit market techniques to make that possible. The Local Government
Code, with its broad array of borrowing powers granted to local government units, and the creation of the Local Government Unit Guarantee
Corporation to bolster local credits, are pioneering efforts.
The subnational debt market is small and levels of indebtedness are
low. The risk of default is minimized by an intercept mechanism used
to secure such debt. As a result, the lack of a formal insolvency system
is not a key challenge for developing a competitive subnational credit
market. There are more fundamental structural challenges, including
institutional and political economy factors, which deter subnational
governments from accessing private sector credit markets, and there is a
lack of competition for subnational debt instruments.
The development of competitive subnational credit markets needs
to address both demand- and supply-side constraints. On the demand
side, it is critical to strengthen the local finance and accountability systems for citizens to demand better services. On the supply side, removing constraints to private bank participation in subnational credit
markets will increase competition and help lower the cost of financing.
Some financing instruments may help forge closer links between local
governments’ own revenues and their capacity to access the market,
which in turn strengthens local accountability. The recent experiments
encouraging greater partnerships between the local governments and
the private sector credit markets could pave the way for a more competitive and diversified subnational credit market.
The subnational debt market in the Russian Federation began to
develop in the early 1990s. Unfunded federal mandates and political
decentralization contributed to the growing demand for debt instruments
including foreign currency debt. At the same time, there was a complete
lack of debt regulation, and SNGs lacked experience in managing debt
risk. Debt was issued to finance recurrent expenditures, mostly with
short-term maturities. With a rapidly deteriorating macroeconomic environment in Russia in the late 1990s, refinancing risks facing SNGs rapidly
rose. Fifty-seven of 89 regions defaulted on their debt from 1998 to 2000.
21
22
Until Debt Do Us Part
Improved macroeconomic fundamentals during 2000–08 and substantial legislative reforms—significant amendments to the Tax Code,
the adoption of the Budget Code, and the 2006 legislation on local selfgovernment—contributed to positive changes in intergovernmental
relations and incentives to formulate new principles of financial management for the regions and municipalities. The Budget Code contains
provisions for regulating the subnational debt, including the provisional
limits on deficit, debt and debt service, regulations of external borrowing, guarantees, and structure and types of debt instruments. With
revenue growth, the financial positions of the regions and municipalities strengthened considerably. The debt load of the Russian regions
remained low at the end of 2007.
The 2008–09 global financial crisis struck Russian public finances in
2009, though the impact varied across SNGs. There were, however, no
regional defaults owing to support from the federal government and the
liquidity accumulated by the regions in prior years. Since 2011, subnational fiscal positions have improved along with a gradual recovery of
oil prices and the Russian economy. The debt markets have recovered
and borrowing costs have declined. However, activity in the domestic bond market remained moderate until 2011, when the market
expanded. There are continuing challenges in subnational debt market
development. For example, most SNGs have short-term debt profiles
dominated by one-year bank loans, implying higher refinancing risk;
bank financing of the regions is dominated by a few state-controlled
banks; and there is a lack of comprehensive accounting for the contingent liabilities of government enterprises.
In the post-apartheid era, South African municipalities faced the
challenges of large-scale infrastructure investments to make up for
huge backlogs left by the apartheid regime, rapid urbanization, and
the need to accelerate economic development. The government’s 1998
White Paper on Local Government stressed the importance of leveraging private sector finance to meet the infrastructure requirements
of municipalities. This was followed by extensive stakeholder consultation between 1998 and 2003, leading to enactment of the landmark
Municipal Finance and Management Act (MFMA). As part of the financial management, the act provides a comprehensive set of ex-ante rules
regulating municipal borrowing. The act also spells out a procedural
An Overview
approach for dealing with municipalities in financial distress, which is
important for lenders.
South Africa engaged in lengthy political consultations to develop
insolvency procedures. Two constitutional amendments paved the way
for a municipal insolvency mechanism. The South African case demonstrates the complexity of subnational borrowing and insolvency legislation and the path dependency of reforms. It illustrates the importance
of building political consensus among various interest groups. Broad
support may require concerted effort over a number of years. South
Africa took two years to develop the basic policy framework (1998–
2000), another year for cabinet approval (2001), followed by two years
of parliamentary debate on the constitutional amendments and on the
Municipal Finance Management Act (2001–03).
All metropolitan municipalities have, in the last decade, borrowed
funds from the banking sector, capital markets, or both to finance
infrastructure development. Since 2005, activity in municipal credit
markets has risen quickly. Long-term borrowing increased rapidly in
the run-up to the 2010 FIFA World Cup, changing the landscape of
municipal finance from a high level of dependency on fiscal transfers
to one where borrowing plays an increasingly important role in financing capital expenditure. Private lenders credit the MFMA as the most
important factor in promoting market activities. There are continuing
challenges, however, including the lack of a fully developed secondary
market, the incompatibility of short-to-medium-term debt maturities
with long-term assets of infrastructure, and the need to crowd-in more
private financing in the market.
The United States has by far the largest local government capital
market in the world, with the longest history of market development,
achieved through a series of incremental changes in institutions over a
long period of time. All the governments below the state level—what
Americans call “local government”—are not sovereign, but rather are
created by and subject to the laws of each respective state. Local governments borrow significant amounts of money to finance infrastructure
investment and have very low rates of default. Local governments in most
states face restrictions on how they borrow and what they can borrow
for, and, in some states, how much they can borrow. For the most part,
these restrictions are on the procedures that local governments must
23
24
Until Debt Do Us Part
f ollow to approve borrowing and how debt service obligations are related
to specific revenue sources (particularly in the case of revenue bonds).
A central feature of the American experience is the importance of
ex-ante and passive insolvency systems. Twenty-three of 50 states prohibit their municipalities from filing Chapter 9 in federal courts. Only
one-third of the states have a system in place for monitoring local governments, and less than 20 percent have institutions and policies that
enable or require state action in the face of a local government fiscal
crisis. The lack of active state programs does not mean that local government borrowing and debt servicing are not actively monitored by
the larger society. Instead, it highlights how the interaction of ex-ante
institutional rules, voters, capital markets, and courts play key roles in
monitoring and limiting local government borrowing.
A distinctive feature of the American state systems is that they establish a close relationship between borrowing and taxation. Most of the
states’ constitutional reforms in the North that followed the states’ debt
crisis in the 1840s required states (and local governments after the 1870s)
to raise current taxes when they issued debt (Wallis 2005). Forcing voters
and taxpayers to simultaneously raise taxes when they borrowed money
increased the burden on borrowing, and led voters to pay closer attention
to the benefits of the expenditures and debt that local governments proposed. A similar set of incentives was set in motion with the development
of special districts and revenue bonds at the end of the 19th century. The
project that the bond proceeds will finance is securitized by the revenue
streams of the project, and the beneficiaries (including future generations) of the project will pay user fees to finance the debt service.
The U.S. experience shows the importance of creating clear interest among creditors in strengthening both the rule of law and incentives for private market development. This is in marked contrast to
a system where lenders assume that the central government would
be, and often was, ultimately responsible for repaying debts. Passive
insolvency systems also clearly require the existence of a strong and
credible rule of law in order to work. Establishing the legal precedent
that local taxpayers were not responsible for servicing debts that were
incurred in an unauthorized manner or through defective procedures
was a long, drawn-out process undertaken at the end of the 19th
century.
An Overview
The result of the framework for debt issuance has not been that local
governments borrow wildly in unauthorized ways and then default,
but rather a steady increase in the capacity of private capital markets to
assess the creditworthiness of local governments and inform potential
borrowers of the actual conditions under which local debt is issued and
will be repaid. The institutional developments such as ex-ante debt rules
(1840s), the bond counsel (the late 19th century), and the Municipal
Securities Rulemaking Board (the mid-1970s) all make the provision of
information to private market participants more credible and transparent. The national government has not violated the sovereign powers of
states to tax, spend, and borrow as they wish, nor have they impaired
the ability of states to establish systems for their local governments.
In principle, states possess the authority to unilaterally change the
structure of any local government. In practice, however, states moved
toward “general” laws governing local governments. This was an institutional change that arose endogenously in the American setting. If individual local governments could approach the state for special treatment, or if
the state can single out individual local governments for special treatment
(either positive or negative), then the incentives to create and enforce
credible rules would be eroded. If all cities know that the same rules apply
to all of them equally, then all cities collectively have a strong incentive to
make sure that a state enforces the rules equally across all cities.
Lessons Learned
Structural trends of decentralization and urbanization are likely to continue in developing countries, requiring massive infrastructure investments at the subnational level. A range of middle-income countries, and
low-income countries in transition to more open market access are contemplating expanding subnational borrowing and debt financing for
infrastructure investments. The experiences of countries covered in this
volume offer valuable lessons.
As shown by Canuto and Liu (2010a), subnational credit risks are
intertwined with broader macroeconomic and institutional reforms.
Macroeconomic stability and sovereign strength set an effective cap
on the credit ratings of SNGs and influence the availability and cost of
funds. Debt sustainability of SNGs is determined by the interplay of the
25
26
Until Debt Do Us Part
existing debt stock, economic growth, cost of borrowing, and primary
balance. Macroeconomic framework and policies strongly influence
the interplay of all these factors. The history of subnational debt crises
shows that unregulated borrowing, particularly in an unstable macroeconomic environment, is extremely risky; unfettered market access by
subnational borrowers can outpace the development of sound revenue
systems and adequate securitization.
Deficits and debt arise from the joint decision of governments making fiscal policy and their creditors. These decisions are made in light of
not only the rules governing issuance of the debt, but also the expectations about what will happen to the debtor and the creditors if payment
difficulties arise—who will lose money or who will be forced into painful
adjustment. The decisions of that lending moment become a fait accompli conditioning the subsequent decisions. This points to two important dimensions of control of government borrowing. First, the type or
timing—ex-ante controls or ex-post consequences; and second, whether
the ex-ante controls and ex-post consequences act on borrowers or lenders.
Ex-ante constraints on subnational borrowers include procedural
rules for incurring debt, limits on debt and deficit ceilings, rules for
borrowing in international markets, and regulation of subnationals’
borrowing based on fiscal capacity criteria. To complement the exante constraints and to make them credible, there need to be ex-post
consequences for failures in fiscal prudence. Without lenders there is
no borrowing or debt, so their constraints and incentives deserve equal
attention. Relying on constraints only on borrowers means that lenders
still have incentives to push loans and may find reckless officials willing
to borrow despite the rules.
Relying only on ex-ante constraints, without ex-post consequences,
gives irresponsible borrowers and lenders an incentive to get around
the ex-ante rules and execute transactions that will later get bailed out.
Relying only on ex-post consequences allows irresponsible (and large)
entities to build up such large debts that they could threaten macroeconomic stability.
Debt restructuring needs to pay close attention to its incentive
effects. Rule-based debt restructuring reduces ad-hoc bargaining and
adverse incentives; hard budget constraint prevents moral hazard;
and burden sharing provides proper incentives and avoids free-riding
An Overview
behavior, while also recognizing that higher levels of government can
create incentives for reform.
The purpose of borrowing and insolvency controls is not to minimize the use of debt financing, but rather to promote sustainable debt
financing through a competitive and diversified subnational credit system. Such a system can help ensure the lowest cost of capital and a sustainable supply of credit. Debt financing is valuable for infrastructure
development where the maturity of assets is generally longer than the
current terms of taxation and transfers.
The dynamics of subnational-central government interaction provide reform momentum. On the one hand, one or a few subnational
governments can serve as catalysts for fiscal reform and as a demonstration for national reform. On the other hand, the national government
can offer fiscal incentives to encourage subnational fiscal adjustment.
One common trait of successful debt restructuring for SNGs is the commitment of the central government to its own fiscal prudence.
The design for regulating debt and insolvency needs to be consistent
with the broader cultural, economic, legal, constitutional, and social
context of the country. Subnational fiscal adjustment and debt restructuring operate within a country’s specific intergovernmental system
that defines the respective authority of each level of government, and
within a country’s political system that defines the respective authority
of each branch and level of government. Capacity and entry point for
reform matter. The maturity of the legal system and the capacity of the
judiciary influence the choices of procedure.
Regulations on debt and insolvency cannot compensate for inadequacies in the design of overall intergovernmental fiscal relations. The
intergovernmental fiscal system underpins the fundamentals of the
subnational fiscal structure. Without increased fiscal autonomy and
greater own-source revenues, subnationals will rarely be in a position
to borrow sustainably on their own. In addition, an intergovernmental fiscal transfer system that routinely fills deficit gaps will undermine
the incentives for a balanced budget. The regulations on debt and
insolvency cannot substitute for other reforms such as budgetary and
financial management, taxation reform, and governance reforms. The
incentive signals of insolvency mechanisms require a more competitive
subnational capital market.
27
28
Until Debt Do Us Part
It is critical to understand the interaction of rules, enforcement,
and capital markets. In the case of government borrowing, decisions
to spend in the present must be matched with decisions to tax and
service debt in the future. Well-functioning capital markets are a way
for societies to pool the best information about conditions today and
changes tomorrow. When governments possess the discretionary ability
to change the rules between today and tomorrow, it becomes difficult
for the capital markets to assess either the returns from financing infrastructure spending or the risks that debts will not be repaid.
The importance of closely tying borrowing decisions to revenue
decisions as a feature of good institutional design cannot be overstated.
Debts have to be repaid, and debt issuance that is tied to tax increases or
dedicated revenue sources is much more likely to be repaid. The experiences discussed in this volume show the importance of moving to rulesbased systems in which the higher-level government treats all lowerlevel governments according to the same rules. No matter what the
rules, their ad-hoc or discretionary application is likely to be plagued
with moral hazard and common-pool problems.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. The term subnational in this book refers to all tiers of government below the
federal, or central, government. The category also includes special purpose
vehicles or investment companies created by SNGs.
2. At the national level, estimations of future infrastructure investment requirements vary greatly by income level. Estache and Fay (2010) discuss methodologies for quantifying these requirements and estimate that low-income countries
should spend 12.5 percent of GDP on investment and maintenance to meet
demand, whereas lower-middle-income and upper-middle-income countries
should spend 8.2 and 2.3 percent, respectively.
3.
Inman (2003) develops the prisoners’ dilemma model formally for this
situation and shows how restrictive are the conditions under which the market
successfully establishes subnational fiscal discipline if the central government
takes a hands-off, no-bailout approach. The conditions include competitive
An Overview
suppliers of local public services, a stable central government, clear and
enforceable accounting standards, a well-managed aggregate economy, and an
informed and sophisticated local government bond market.
4. This might include, for example, an unused vacant lot outside the corporate
limits or a private residence taken for failure to pay taxes (McConnell and
Picker 1993, 432).
5. To “cramdown” is the ability to force dissenting minority creditors to accept an
agreement between a majority of creditors and the debtor.
6. In some places, there is no system, so “ad hoc” is a third system. In other places,
defaults are dealt with as political problems, and there is no (or little) judicial or
administrative capacity to deal with the problem.
Bibliography
Braun, Miguel, and Mariano Tommasi. 2004. “Fiscal Rules for Subnational Governments: Some Organizing Principles and Latin American Experiences.” In Rules
and Practice in Intergovernmental Fiscal Relations, ed. G. Kopits. Washington,
DC: International Monetary Fund and World Bank.
Canuto, Otaviano, and Lili Liu. 2010a. “Subnational Debt Finance: Make It Sustainable.” In The Day after Tomorrow: A Handbook on the Future of Economic Policy
in the Developing World, ed. Otaviano Canuto and Marcelo Giugale, 219–238.
Washington, DC: World Bank.
———. 2010b. “Subnational Debt Finance and the Global Financial Crisis.” Premise
Note, Poverty Reduction and Economic Management Network, World Bank,
Washington, DC.
Estache, A., and M. Fay. 2010. “Current Debates in Infrastructure Policy.” Commission
on Growth and Development Working Paper 49, World Bank, Washington, DC.
Ianchovichina, Elena, Lili Liu, and Mohan Nagarajan. 2007. “Subnational Fiscal
Sustainability Analysis: What Can We Learn from Tamil Nadu?” Economic and
Political Weekly 42 (52): 111–19.
Inman, Robert P. 2003. “Transfers and Bailouts: Lessons from U.S. Federalism.”
In Fiscal Decentralization and the Challenge of Hard-Budget Constraints, ed.
J. Rodden, G. Eskeland, and J. Litvack, 35–83. Cambridge, MA: MIT Press.
Liu, Lili, and Michael Waibel. 2009. “Subnational Insolvency and Governance:
Cross-Country Experiences and Lessons.” In Does Decentralization Enhance
Service Delivery and Poverty Reduction?, ed. Ehtisham Ahmad and Giorgio
Brosio, 333–75. Cheltenham, U.K.: Edward Elgar.
Liu, Lili, and Steven Webb. 2011. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper 5587,
World Bank, Washington, DC.
Maco, Paul S. 2001. “Building a Strong Subnational Debt Market—A Regulator’s
Perspective.” Richmond Journal of Global Law and Business 2 (1): 1–31.
29
30
Until Debt Do Us Part
McConnell, Michael, and Randal Picker. 1993. “When Cities Go Broke: A Conceptual Introduction to Municipal Bankruptcy.” University of Chicago Law Review
60: 425–35.
Wallis, John J. 1984. “The Birth of the Old Federalism: Financing the New Deal.”
Journal of Economic History 44 (March): 139–59.
———. 2005. “Constitutions, Corporations, and Corruption: American States and
Constitutional Change, 1842 to 1852.” Journal of Economic History 65 (1):
211–56.
Part 1
Subnational Debt Restructuring
1
Brazil: The Subnational Debt
Restructuring of the 1990s—
Origins, Conditions, and Results
Alvaro Manoel, Sol Garson,
and Monica Mora
Introduction
During the 1980s and 1990s, Brazilian subnational governments (SNGs)
experienced extreme fiscal difficulties that resulted in increasing debt,
giving rise to major renegotiations with the federal government near
the end of the 1980s, in 1993, and during 1997–2000. In each period,
the federal government undertook massive debt restructuring with
states and municipalities. Each crisis had different underlying macroeconomic conditions and political economic dynamics, and the subsequent restructuring thus produced different results. The conditions and
results are related to the evolving system of fiscal federalism in Brazil.
The lessons drawn from these crises and the resulting debt restructuring are relevant not only for Brazil, as it continues to recover from the
2008–09 global financial crisis, but also for developing countries where
fiscal decentralization is under way and debt continues to be an important instrument for financing economic growth.
The importance of Brazilian SNGs—that is, states and municipalities1—
in the provision of goods and services gained ground in the 1980s and 1990s,
as decentralization and urbanization deepened. In 1970, 56 percent of the
total population of 93 million lived in urban areas. In 2000, 81.2 p
ercent
33
34
Until Debt Do Us Part
of the almost 170 million people did so, and the rate of urbanization
reached 90.5 percent in the Southeast region, the most developed in the
country. Throughout this period, SNGs accounted for nearly half of public sector spending and for most spending in education, health care, infrastructure, and public security.
The debt crises in the 1980s and 1990s evolved in the context of
development and macroeconomic policies of the military govern
ment (1964–85). Public sector borrowing increased substantially to
finance capital investments associated with rapid urbanization. These
investments were undertaken mainly by newly created institutions
such as public agencies and enterprises at both the federal and subnational levels. Federalism in Brazil revived in the 1980s with the return
to democracy. The 1986 Congress, with strong representation of SNGs,
crafted provisions for the 1988 Constitution that gave states significant
authority and resources, including a much broader revenue base for the
state-level value-added tax (VAT), but did not specify their spending
responsibilities or set rules for fiscal prudence.
SNGs experienced debt stress during the 1980s, a period of macroeconomic instability that included oil shocks and a balance-of-payments
crisis. Although the federal government provided bridge loans to assist
SNGs in rolling over their external debt, the debt stress of SNGs continued unabated. The three subsequent debt restructurings (1989, 1993,
and 1997–2000) dealt with different types of debt—external debt, debt
owed to federal entities, and all other debt including market securities.
The macroeconomic stabilization program in the early-to-mid-1990s,
which was centered around the Real Plan, shaped the direction of the
third debt restructuring. The Real Plan sought to reorganize the entire
public sector, including the privatization of banks and public enterprises, and to reduce hyperinflation.
The most notable feature of the third round of debt restructuring was
that it dealt with the underlying reasons for the subnational fiscal imbalance by focusing on the quality of fiscal adjustment and reforms. The
federal government offered SNGs incentives to restructure their debt,
but also required them to undertake comprehensive structural and fiscal reforms, including control over personnel spending and privatization of state banks and state-owned public enterprises. The success of
the Real Plan, together with the third round of SNG debt restructuring,
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
created the political and economic conditions for enactment of the
Fiscal Responsibility Law (FRL) in 2000. The FRL established limits and
placed restrictions on key fiscal variables and assigned responsibility for
enforcing the obligations and fiscal transparency requirements.
Throughout the first decade of the 2000s, state and municipal
finances improved significantly. SNGs began to generate a primary surplus of about 1 percent of gross domestic product (GDP), reversing the
deterioration of the previous two decades. The limits imposed by the
FRL on public spending, debt, and debt services were crucial to achieving these improved fiscal results and are central to the subnational debt
restructuring agreements and the rules of budgetary and financial execution. Total public net debt as a share of GDP declined from 52 percent
in 2001 to 39 percent in 2010, with the disaggregation declining in all
three levels of government. The improvement in the SNG fiscal accounts
is associated with Brazil’s improved macroeconomic fundamentals.
This chapter reviews the Brazilian SNG debt restructuring implemented at the end of the 1980s and during the 1990s, and assesses its
evolution within the macroeconomic context of crises, stabilization, and
reforms. Section two provides a historical context for the origins of the
SNG debt crisis. Section three defines the macroeconomic framework
that gave rise to the high indebtedness of SNGs and details the successive rounds of debt renegotiation. Section four analyzes the evolution of
SNG fiscal performance in the post-renegotiation era. Section five gives
special attention to the 2007–10 period, and assesses the impacts of the
2008–09 global economic crisis on subnational debt and finance and the
main fiscal federalism challenges emerging in the recent period. Section
six draws some conclusions.
Fiscal Federalism, Centralization, and Decentralization:
Historical Context
Fiscal federalism in Brazil has developed in alternating waves of centralization and decentralization of power. In 1891, the first constitution
of the republic adopted a federal structure that decentralized revenues
and gave states control of export taxes (Varsano 1996). It also gave
governors, supported by regional oligarchies, control of the electoral
process. The period 1930–45 was marked by a nondemocratic regime
35
36
Until Debt Do Us Part
that favored fiscal centralization. The Constitution of 1946 marked the
return to democracy and a more decentralized fiscal structure. This was
followed by a military dictatorship from 1964 to 1985—a period of centralization and authoritarianism—and redemocratization after 1985,
with SNGs increasing their share of revenues, but also accumulating
mounting debt.2
Economic and Political Crises in the 1960s and 1970s
and Authoritarian Rule
In 1964, a military government took power in the midst of a deep economic and political crisis and adopted a strategy based on two essential
elements. The first comprised measures to stabilize the economy and
promote fiscal and financial reforms. These reforms created a new tax
system, which raised the overall tax burden and improved tax collection,
and adopted modern financial instruments, creating (a) special funds
made up of earmarked revenues, (b) the Central Bank of Brazil and the
National Monetary Council, and (c) a mechanism for indexing financial
assets to inflation.
The second strategic element reshaped the public administration by
granting autonomy to the so-called indirect administration—agencies,
foundations, public enterprises, and mixed public-private companies.
A large part of the production of goods and services was transferred
to them. Centralized planning and allocation of resources covered key
sectors relating to national integration (for example, transportation and
telecommunications) and basic materials (for example, petrochemicals,
paper, and cellulose).
After the initial period of economic stabilization in the mid-1960s,
Brazil entered a phase of accelerated growth of around 11 percent a
year between 1968 and 1973, lower inflation, higher exports, and less
volatile exchange and interest rates. The accelerated growth, however,
highlighted the strong dependence of the country on capital goods and
imported inputs, particularly petroleum and its derivatives. The growth
of foreign debt and direct foreign investments increased debt servicing
and the repatriation of profits abroad. To finance large-scale projects,
the federal government also established special funds such as the Federal Fund for Urban Development and special public agencies such as
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
the National Housing Bank. The use of foreign capital to finance not
only investment but also working capital initiated a cycle of increasing
indebtedness, which eventually culminated in a national debt crisis in
1982.
A political crisis arose in 1974, when the government-backed
National Renewal Alliance Party suffered significant defeats in state and
national elections. On the economic side, oil prices skyrocketed: the first
oil shock increased oil prices by more than 300 percent from 1973 to
1974. The pressure on the balance of payments was magnified by a drop
in exports that resulted from a sharp fall in global economic growth
and rising interest rates. In mid-1979, the second oil shock led to rising interest rates, followed by higher costs of borrowing in the London
market. Brazil, which had obtained foreign resources at floating interest
rates, experienced a sharp drop in the supply of foreign credit and rising
costs to roll over debt.
A huge balance-of-payments crisis was followed by tight domestic
fiscal and monetary policies with significant impacts: (a) the federal
tax burden increased from 18.4 percent of GDP in 1980 to 20.6 percent
in 1983; (b) strict control of public spending negatively affected public
investments; (c) GDP fell 6 percent during 1981–84; and (d) inflation
accelerated from about 100 percent in 1982 to more than 200 percent
annually during 1983–85, which resulted in an erosion of public revenues and a significant increase in the stock and service of public debt,
which was indexed to inflation (see Hermann 2005).
Redemocratization in the Early 1980s: Rise in SNG Share
of Revenue and Debt Distress
The elections of 1982 marked the first step in the redemocratization
of Brazil. The return to direct elections at the state level in 1982, after
22 years of governors being appointed indirectly, allowed the new governors to recoup their sources of power, creating alliances with local
governments due to the coincidence of state and municipal elections.
The results strongly favored the opposition—10 of 22 opposition governors were elected—including in the largest and most powerful states.
While the central government lost political influence, the governors
benefited from the economic recovery in 1984 and 1985.3 As a result,
37
Until Debt Do Us Part
they were able to fulfill their mandates with high levels of investment
(see Villela and Rezende 1986, 215). A 1983 constitutional amendment increased the transfers to SNGs and raised the tax basis of states.
The states, which received 21.3 percent of disposable fiscal revenues in
1983 (corresponding to 5.7 percent of GDP), received 27 percent of
revenues in 1986 (7.1 percent of GDP). At the same time, the municipalities increased their participation from 8.9 to 12.1 percent of available revenues, jumping from 2.4 to 3.3 percent of GDP, as shown in
figure 1.1.
The wave of democratization in the context of the macroeconomic
crisis of the late 1970s to early 1980s encouraged the population to turn
to the SNGs for services, even those that were the responsibility of the
federal government. During discussions in the National Constituent
Assembly during 1987–88, governors and mayors fought for a larger
share of public revenues. The 1988 Constitution greatly expanded the
Figure 1.1 National, State, and Municipal Tax Collection and Disposable Revenue
in Brazil, 1970–2010
40
35
% of GDP
30
25
20
15
10
5
0
19
70
19
72
19
74
19
76
19
78
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
0
20 0
0
20 2
04
20
0
20 6
08
20
10
38
Year
Total tax collection
States’ tax collection
Federal disposable revenue
Municipal disposable revenue
Federal tax collection
Municipal tax collection
States’ disposable revenue
Sources: Afonso 2011; National Bank of Economic and Social Development; Secretariat of Fiscal Affairs.
Note: GDP = gross domestic product.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
social responsibilities of the government; most of the expanded services
provision was left to states and municipalities, however, which increased
their expenditure obligations and, as a consequence, the borrowing
requirements. As a result, in 1991, SNG participation in disposable fiscal revenue rose as high as 29.8 percent for states (7.5 percent of GDP)
and 15.9 percent for municipalities (4 percent of GDP). Later developments, mainly higher tax rates in the form of higher social contributions
(which are not shared with SNGs), allowed the central government to
increase its share of disposable revenue.
Despite the increase in revenues, the SNGs’ fiscal difficulties deepened, as demand for services rose and debt accumulated. The credit
crunch, the difficulty of rolling over debt, and the strong indexation of
debt to inflation deepened the fiscal vulnerability of the SNGs, which
resorted to short-term credit called anticipated revenue credit operations (ARO), which by law is expected to be repaid in the same fiscal
year, based on estimated tax revenue for the period. Nevertheless, this
type of credit operation was often rolled over and accumulated year
after year at much higher interest rates than regular credit or loans. In
1989, credit based on ARO accounted for 97 percent of state and municipal debt (see Lopreato 2000, 127).
In this difficult environment, the fiscal stance of the states and
municipalities deteriorated significantly. In addition to their weak planning and fiscal management capacity, states and municipalities could
not count on adequate sources of financing for their increased spending. Thus, several SNGs, in particular the states, ended the decade in
massive fiscal distress.
The subnational debt crisis, which erupted in the 1980s and the
1990s, was largely the result of the institutional reforms of the military government, which allowed debt to become a key source of funding for governments. The regulations, which facilitated the access of
public (and private) sectors to external resources in an environment of
strong international liquidity, led to an “economy of debt.” While the
term “economy of debt” explains a good part of the subnational debt
story during this period, the responsibility of some states and municipalities must also be understood. The most notorious example is that of
the Municipality of São Paulo, whose debt was greater than that of the
majority of states.
39
40
Until Debt Do Us Part
Until the mid-1960s, the ability to raise funds was hampered by a
legal ceiling on the nominal interest rate, which discouraged lenders in
an environment of high inflation rates. Military government reforms
included adjusting the value of bond debt for inflation (Almeida 1996).
Removing the legislative cap on nominal interest rates was one of the
most important reforms initiated by the military regime, because it
helped develop a modern capital market in the country.
Borrowing became a mechanism to circumvent the strict controls
imposed on states and municipalities. The 1967 Constitution empowered the Senate to authorize SNG domestic and external credit operations. The rules for contracting operations were defined by the central
bank, which was responsible for regulating the financial and banking
systems. In 1975, Senate Resolution No. 93 set strict rules for contracting credit operations. However, foreign operations and the so-called
“extra limit” were not regulated. To induce the states and municipalities to act in line with the federal government’s interests, the social
and urban projects considered as priorities by the federal government
were funded by those “extra limit” operations. Having passed through
the arbitrary sieve of the federal government, and authorized by the
Senate, those operations became significant. After the second oil shock
in 1979, regardless of the payment capacity of SNGs, the state and
municipal public companies were induced to borrow from abroad as
part of the federal strategy to reverse the public sector’s balance-ofpayments deficit.
To deal with the debt crisis unleashed by the Mexican moratorium
in 1982 and the consequent need to turn to the International Monetary Fund (IMF) as the balance of payments deteriorated and foreign reserves dwindled, the federal government adopted a tight fiscal
policy in 1983–84. The tax burden was increased to 27 percent of GDP,
and public investments were severely reduced. Nevertheless, inflation
continued to erode tax revenue and swell public debt. SNG budgetary and financial management was strictly conditioned by macroeconomic policies defined at the federal level. During the 1980s, SNGs
were awarded “bridge loans” through the National Treasury so that
they could roll over their external debt. In this way, SNG external debt
was transformed into “domestic” debt, with the federal government as
the main creditor.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
SNG Debt Renegotiation and Its
Macroeconomic Context
Brazil experienced three major subnational debt renegotiations from
the late 1980s to the late 1990s. These renegotiations must be understood within the broader context of macroeconomic and political
reform. The focus of each renegotiation grew progressively broader.
The first round focused narrowly on changing the terms of debt repayment. The second round accompanied major efforts to stabilize the
macroeconomic situation and make the government more efficient, as
the Brazilian economy moved from state control to open markets. The
third round consolidated the fiscal adjustment with enactment of the
FRL in 2000. This section briefly describes the main rounds of renegotiation and analyzes their links with macroeconomic stabilization and
fiscal consolidation.4
Addressing SNG Debt Distress: Three Major Renegotiations
The first round of SNG debt renegotiation in 1989 focused on resolving
SNG external debt. This problem had been fueled by the federal government itself, which had sought to raise funds to fill the balance-ofpayments gap. Based on Law No. 7976 of December 27, 1989, the federal
government refinanced SNG debt, which included federal government
loans in order to honor foreign debt guaranteed by the National Treasury or contracted until the end of 1988.
In 1991, following the inauguration of new governors and the failure
of the President Collor Plan II,5 authorities from all levels of government began to discuss a new approach, which led to the adoption of
Law No. 8388 of December 30, 1991. However, the institutional crisis
resulting from the impeachment of President Collor in 1992 postponed
implementation of the law until 1993, when the second round of debt
renegotiation started.
The second round of SNG debt renegotiation in 1993 concentrated
on the debt owed to agencies and entities controlled by the federal government. Law No. 8727 of November 5, 1993, included outstanding
amounts of loans contracted during the military period of 1964–85
for investments coordinated by the federal government.6 The new law
restructured the terms of payment by extending the maturity of loans to
41
42
Until Debt Do Us Part
as long as 20 years and limiting debt service to no more than 11 percent
of real net revenue (RLR).7 In the event that debt service obligations
exceed the 11 percent limit, the federal government would provide additional loans to cover the excess, and the state would pay back the loan
in 10 years. Furthermore, the law stipulated that interest rates would be
kept as originally established.
Although states requested that public securities (state Treasury bills
and bonds) be included, they were excluded. In fact, both federal and
subnational governments were concerned about the rapid growth of
public securities and the increasing difficulties of refinancing them in
the late 1980s and early 1990s. Constitutional Amendment No. 3 of
March 17, 1993, limited the ability of SNGs to issue public securities
until the end of 1999. SNGs were only allowed to issue public securities
in the amount necessary to refinance the principal indexed by inflation,
with one exception related to the issuance of SNG securities to pay for
judicial writs.8
The third round of debt renegotiations was initiated in 1996. The
first and second rounds restructured only part of subnational debt.
With a substantial part of SNG debt unresolved and persistent pressures placed on the primary balance, the fiscal and debt position
of SNGs continued to deteriorate. As hyperinflation was brought
under control, the SNG debt situation worsened, with debt obligations rising in real terms. SNG debt as a share of GDP increased from
9.2 percent in 1993 to 10.7 percent in 1996. The debt service pressures
were even more severe. Law No. 9496 of September 11, 1997, authorized the third round of renegotiations by establishing criteria for
consolidating, refinancing, and assuming the national debt and other
securities held by states and the Federal District. The main difference
between this round and the previous two was that a major fiscal consolidation program underpinned the debt renegotiation. Although
a bill relating to state debt was passed in 1997, legislation related to
municipal debt was enacted only in 1999.
Macroeconomic Stabilization
The decade-long effort to renegotiate subnational debt unfolded
within the broader context of national efforts to achieve macroeconomic stabilization and put Brazil onto a more sustainable growth
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
path. The second round accompanied two complementary macroeconomic adjustment programs: the Federal Immediate Action
Plan (PAI) in 1993 (box 1.1), which sought to reorganize and reorient the entire public sector including SNGs; and the Real Plan in 1994
(box 1.2), which aimed to control inflation (monthly inflation rates
were about 45 percent in the second quarter of 1994). The implementation of these two macroeconomic programs helped shape the direction and substance of the third, and most consequential, round of debt
renegotiation.
The main elements of the Real Plan in 1994 included the introduction
of a new currency (the real), the de-indexation of the economy, an initial freeze on public sector prices, the tightening of monetary policy, and
the floating of the currency, with a floor specified for its value vis-à-vis
the dollar. The Real Plan managed to break the vicious cycle of high
inflation and to de-index the economy.
Eichengreen (2007) argues that defending a particular exchange rate
with capital mobility implies implementing domestic policies aimed at
stabilizing the system. In Brazil, it was no different. The monetary policy
during the Real Plan sought to maintain exchange rate stability, contributing to the success of the stabilization plan. Thus, in addition to
the expected effects on production and consumption associated with a
Box 1.1 The 1993 Federal Immediate Action Plan
The 1993 Federal Immediate Action Plan (PAI) reorganized the public sector by untying earmarked
revenues and reducing expenditures. The PAI was based on the assumption that inflation was
caused by, among other factors, the financial and administrative disarray of the Brazilian state. The
main provisions of the PAI provide for the following:
• Raising tax revenue by implementing a financial transactions tax and efforts to combat tax
evasion
• Controlling public spending and using resources more efficiently
• Changing the nature of the relations between the national government and the SNGs, reducing
unconstitutional transfers, and restructuring SNG debt
• Restructuring the public banks (both federal and state), which aimed to privatize government
shares
• Deepening the privatization process, including the electricity and railroad sectors, and completing the privatization of state-owned enterprises in the petrochemical and steel sectors
• Creating the Emergency Social Fund, which temporarily reduced the share of earmarked revenues at both the federal and subnational levels.
43
44
Until Debt Do Us Part
Box 1.2 The Real Plan, 1994
The implementation of the 1994 Real Plan, with the PAI as its backbone, consisted of two steps.
The first step encompassed monetary reform to combat inflation by introducing the real
value unit (URV) to disrupt the indexing system of prices and incomes. Implemented on January 3,
1994, the URV followed the daily pro-rata variation of a set of three price indexes. All contracts
were to be expressed in the new unit of account. To coordinate the inflationary expectations, the
URV was pegged to the dollar. Use of the URV prior to adoption of the new currency was intended
to realign relative prices, reduce the redistributive effects, and coordinate agent expectations. On
July 1, 1994, the currency reform was completed, transforming the URV into the new currency, the
real. The entire money stock was replaced, and the real became the medium of exchange. Wages,
in turn, were converted to the new currency, calculated based on the average in URV between
March and July.
The second step adopted an anchor with the purpose of coordinating expectations. After
a first attempt to use a monetary anchor, it became clear that a stronger instrument was needed
to stabilize expectations. The choice fell naturally to the exchange rate. To meet the goal of
coordinating inflationary expectations, the exchange rate regime was redesigned three times
between 1994 and 1999:
• When the exchange rate targeting regime was adopted, the exchange rate was fixed between
October 1994 and February 1995, despite an official system of bands.
• When the Mexican crisis arose in December 1994, the significant reduction in capital flows to
emerging markets and the loss of foreign reserves by the central bank led to the adoption of a
more widely fluctuating band.
• When the huge appreciation of the real against the dollar became evident, the exchange rate
band was systematically adjusted to ensure the gradual devaluation of the real.
Despite the changes, the central bank essentially set the exchange rate during this period,
coordinating expectations and preventing the resurgence of inflation. Despite the evidence of
excessive valuation, it was believed that a possible devaluation would negatively affect inflation.
So even when the system showed signs of exhaustion, the central bank declined to redraft its
essence. Only under a speculative attack in early 1999 did the central bank, unable to sustain the
parity, drop out.
restrictive monetary policy, higher interest rates also attracted speculative capital in search of greater profitability. As shown in figure 1.2, the
Special Settlement and Custody System (SELIC)9 rate rose to more than
80 percent a year in February 1995.
While the Real Plan focused on price stabilization, the PAI and the
Emergency Social Fund (ESF) introduced greater budgetary flexibility,
generated proceeds from privatization, and created more sources of revenue. Therefore, the PAI and the ESF helped create the conditions for
the primary surpluses obtained by the consolidated public sector in the
2000s (figure 1.3).
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
Figure 1.2 SELIC Interest Rate in Brazil, August 1994–April 2011
90
80
Annual rate, %
70
60
50
40
30
20
10
Au
g/
1
Ju 994
n/
Ap 1995
r/
Fe 1996
b/
De 199
c/ 7
O 199
ct 7
/
Au 199
g/ 8
Ju 1999
n/
Ap 200
r/ 0
Fe 200
b/ 1
De 200
c/ 2
O 200
ct 2
/
Au 200
g/ 3
2
Ju 004
n/
Ap 200
r/ 5
Fe 200
b/ 6
De 200
c/ 7
O 200
ct 7
/
Au 200
g/ 8
2
Ju 009
n/
2
Ap 010
r/2
01
1
0
Month/year
Source: Central Bank of Brazil.
Note: SELIC = Special Settlement and Custody System.
However, the PAI and ESF were insufficient to ensure short-term fiscal balance, partly because a substantial share of SNG debt remained,
even after the second round of debt renegotiation.
Following the adoption of the Real Plan, especially during 1995–98,
SNGs, particularly the states, began to run primary fiscal deficits
(figure 1.4).Importantly, state representatives ended their terms in
1994, while municipal representatives were in the middle of their fouryear terms, which implied loose fiscal policy at the state level. Following
the political cycle, state public finances worsened, likely due to (a) the
end-of-inflation factor (see next paragraph), and (b) the fact that state
governments granted generous wage increases at the end of 1994 and
especially in 1995, with full financial impact on the following years.
The primary deficits during 1994–98 highlighted the structural
imbalance of the SNGs. The sharp reduction in inflation in 1994
removed a public sector instrument that governments at all levels had
used to delay payments, which reduced their real value while revenues
were indexed. In addition, the rise in interest rates accelerated the
growth of state and municipal securities debt, which was charged the
45
Until Debt Do Us Part
Figure 1.3 Consolidated Primary Fiscal Balance in Brazil, 2001–10
4.0
3.7
3.5
3.0
3.2
3.2
3.8
3.3
3.2
3.3
3.4
2.8
2.5
% of GDP
46
2.0
2.0
1.5
1.0
0.5
0
01
20
02
20
03
20
04
20
05
06
20
20
07
20
08
20
09
20
10
20
Year
Source: Central Bank of Brazil.
Note: GDP = gross domestic product.
SELIC rate—that is, the growth of securities debt was mainly due to the
accumulation of interest on the principal. The primary surplus had to
cope with higher interest rates, and aggregate SNG debt reached about
15 percent of RLR in 1995 and 10 percent of RLR in 1996 (Mora 2002).
The majority of state and municipal issuers of public securities, who
had serious difficulties generating adequate primary surpluses, could
not even afford to pay the interest on the public securities. As a consequence, they began to roll over the entire debt based on federal Senate
authorizations,10 and to use short-term ARO to finance normal budgetary operations. The amounts were large in relation to the fiscal year
budget. Even those that could pay at least a portion of interest flocked
to the Senate to request the rollover of debt; they had no incentive to
reduce their debt. In this context, SNG debt became explosive, requiring
a third round of renegotiation that encompassed all the debt stock that
had not yet been refinanced.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
47
Figure 1.4 Primary Fiscal Balance of States and Municipalities in Brazil, 1991–2010
1.5
1.4
1.0
0.8
0.8
% of GDP
0.6
0.7
0.8
0.9
1.1
1.0
1.0
0.8
0.5
0.5
0.7
0.6
0.2
0.1
0
–0.2
–0.5
–0.2
–0.5
–0.7
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
19
19
91
–1.0
Year
Source: Central Bank of Brazil.
Note: GDP = gross domestic product. Does not include state- and municipal-owned companies.
Third Round of SNG Debt Renegotiation
and Fiscal Consolidation
The central role of fiscal balance in the design of the Real Plan required
SNGs to adhere to federal efforts to stabilize the economy and bring
down inflation. Fiscal policy was a key ingredient, and the third round
of debt restructuring sought to induce SNGs, particularly states, to generate the primary surplus required to ensure long-term fiscal sustainability and macroeconomic stability.
The renegotiation opened up the possibility that the federal government would give more incentives to SNGs to adopt the principles of the
PAI plan (box 1.1). The 1996 renegotiation took place within the context
of a discussion of the role of government in the economy. The privatization process at the federal level questioned the prevailing conception
of state interventionism. Telecommunications and electricity, until then
public monopolies, began to be managed by private companies, subject
to regulatory agencies. As a condition for debt refinancing, the states
would need to undertake fiscal and financial restructuring, reorienting
48
Until Debt Do Us Part
the role of the government to make it compatible with the changes
occurring at the federal level.
In addition, the restructuring programs were aimed at tackling the
potential sources of imbalance, in order to create the conditions for the
gradual repayment of debt. The contracts between the federal government and the states included a rigorous fiscal and restructuring adjustment program (the Incentive Program for the States’ Restructuring and
Fiscal Adjustment, PAF),11 encompassing fiscal targets relating to the
following indicators: (a) the ratio of debt to RLR, (b) the primary fiscal balance, (c) the public sector wage bill, (d) own-revenue collection,
(e) privatization and concession of public services and utilities,
(f) administrative reform, and (g) the ratio of capital expenditures to
RLR. The state debt would be indexed to the general price index (General
Price Index, Domestic Availability, IGP-DI)12 and charged an annual interest rate of 6 percent, provided that the state paid 20 percent of the debt
within two years. These resources would be obtained by privatizing SOEs.
Refinancing agreements and contracts covered almost all existing
debts:
•
•
The total amount renegotiated reached 11.2 percent of GDP and
was highly concentrated in the four largest debtors. The states of São
Paulo and Rio de Janeiro accounted for 5.9 and 1.7 percent of GDP,
respectively. The states of Minas Gerais and Rio Grande do Sul also
had significant financial liabilities. These four large debtors absorbed
about 90 percent of the amount renegotiated.
Depending on the initial conditions in each state and the possibility of making an important down payment with proceeds from the
privatization of state assets, the terms of the contracts varied: limits
on the ratio of debt service to revenue, for example, could vary from
12 to 15 percent. Most of the states signed a 30-year contract, some
of which had 15-year terms to maturity. The interest rate in most of
the contracts was IGP-DI plus 6–7.5 percent a year.
To fulfill their contractual obligations, state governments conducted
a rigorous fiscal adjustment and began to generate primary surpluses.
This extensive process of fiscal and financial restructuring also included
the following:
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
•
•
•
The privatization of SOEs, especially in the electricity sector
The restructuring and privatization of public banks owned by the
states, which adhered to the Incentive Program for the Reduction of
Participation of the States in Public Sector Banking
The reorganization of state public finances, with greater fiscal
responsibility.
In addition to spending cuts, states were encouraged to take loans to
modernize tax administration, contributing to an increase in tax collection, especially the VAT on goods, intermunicipal transportation, and
communications services (ICMS). The ICMS revenue collection also
benefited from an increase in the rate levied on the transport, electricity, and telecommunications sectors, characterized by low elasticity of
demand. The fiscal adjustment strategy also increased the fiscal space
for debt payment. If the states did not fulfill their commitments, the
federal government was authorized to withhold transfers from the State
Participation Fund (FPE) and even the states own tax, the ICMS.
The PAF and its mechanisms for monitoring and enforcement were
critical to the success of the renegotiation. The PAFs, signed by the governors of 25 states,13 had annual targets for the following three years. Levels of
compliance have been high because the debt renegotiation contracts have a
specific clause that allows the National Treasury to stop making legal transfers to states that do not comply, and even to sequestrate states’ own tax revenue in case of nonpayment of the agreed portion of their debt. Each year
the goals and commitments of the previous year are evaluated and the targets updated where appropriate. Targets not achieved are subject to fines.
These procedures will follow the refinancing contracts until their discharge.
Therefore, it is reasonable to conclude that after the implementation
of the Real Plan, the federal government had a clear interest in renegotiating the state’s debt because it was linked to other economic measures
such as privatization of public utilities and state financial institutions,
and implementation of programs to modernize tax administration at
the state level. In addition, considering the macroeconomic context, the
debt renegotiation had a crucial impact on improving the solvency and
liquidity of the Brazilian banking system: large commercial bank assets
which were held against the states (previously classified as high risk)
were exchanged for high-yield federal assets (National Treasury bonds).
49
50
Until Debt Do Us Part
The negotiation with municipalities was conducted separately starting in 1998. Unlike state debt, municipal debt was not addressed in Law
No. 9496 in 1996, because the financial situation of municipalities was
far less worrying than that of states. Nevertheless, municipalities such
as Campinas, Guarulhos, Osasco, Rio de Janeiro, and São Paulo, which
held debt securities, saw their situation worsening, since the assumption
of state debt by the federal government meant that municipal securities
now had to compete “alone” with federal securities in the market.
Research conducted by the city of Rio de Janeiro14 indicated that the
50 municipalities with the highest RLR encompassed 29 percent of the
population, 55 percent of RLR, and 82 percent of long-term municipal
debt. The ratio of debt to RLR on average was 0.72. Only 28 municipalities
had a ratio higher than 1. Furthermore, in April 1998, the securities debt
of the two capital cities—São Paulo and Rio de Janeiro—constituted
25 percent of total SNG debt still to be renegotiated.15
The federal government issued Provisional Measure No. 1891 in January
1999, which extended to the municipalities the general conditions granted
to the states without requiring fiscal programs, but tightened restrictions
on debt contracting. The main terms of restructuring the municipalities’
debt were16 as follows:
•
•
An interest rate of 9 percent a year, which could be reduced to
7.5 percent a year, or 6 percent a year if the municipality extraordinarily amortized an amount equivalent to 10 or 20 percent of the
outstanding balance within 30 months of signing the contract
Exemption of municipalities from fiscal adjustment programs, which
the states were obliged to participate in under the PAF. Around 180
municipalities signed contracts with the federal government restructuring their debt under these new conditions, which affected about
10 percent of all operations with the states. For larger municipalities, the possibility of extraordinary amortization only postponed the
imposition of the 9 percent interest rate.17
In the macroeconomic sphere, the devaluation of the domestic currency in 1999, amid a currency crisis,18 implied the end of the exchange
rate anchor. This was followed by the adoption of an inflation-targeting
regime to shield the Real Plan. When this system was introduced, SNG
fiscal accounts were already improving. Still, policy makers wanted to
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
ensure that the changes would be lasting and to avoid any retreat from
maintaining fiscal balances. After all, the main objective of the new
macroeconomic policy was to stabilize the macroeconomic situation.
The introduction of inflation targeting was accompanied by the adoption of a floating exchange rate regime and a vigilant effort to maintain
fiscal austerity.
The 2000 FRL and Long-Term Fiscal Sustainability
The success of the Real Plan and the third round of subnational debt
restructuring created the economic and political conditions for the
approval and implementation of the FRL in 2000.19 The FRL established
fiscal limits and restrictions on key fiscal variables, such as personnel
expenditures and debt stock, and defined the responsibility for enforcing the fulfillment of obligations and transparency requirements (see
box 1.3). The FRL was essential to strengthening the process of fiscal
and financial restructuring of the government, including SNGs, and
ensuring long-term fiscal sustainability.
The approval of the FRL was the culmination of a series of key and
incremental reforms in which sequencing and pace were critical.20
Among the most important were the following:
•
•
•
•
The creation in 1986 of the National Treasury Secretariat (STN)
in the Ministry of Finance, with broad responsibilities for public finances, especially with regard to managing federal assets and
liabilities21
The 1988 Constitution, which expanded social responsibilities of the
government, particularly for subnationals
The closing or privatization of state commercial banks in early 1990
The 1998 Constitutional Amendment No. 19, which established new
rules related to public administration and which made public the
debate about issues like wage ceilings for public servants and pension
systems.
Also, the implementation of the Real Plan strongly influenced the fiscal path to be followed by states and municipalities.
Furthermore, according to Liu and Webb (2011, 10), “Even without
a strong party system, a powerful president can enforce subnational fiscal discipline. President Cardoso in Brazil became a strong president in
51
52
Until Debt Do Us Part
Box 1.3 The Fiscal Responsibility Law (Complementary Law No. 101, May 4, 2000)
The FRL was part of the efforts to strengthen fiscal institutions. The law applies to the federal
government, the states, the Federal District, and the municipalities; the legislature (including the
audit courts); the judiciary and the attorney general’s office; and to their respective agencies,
foundations, and funds. The FRL addresses the following issues:
• Planning and budgeting. Besides defining basic parameters for the annual budget bill to be
prepared by the executive, the Budget Guidelines Law defines fiscal targets relating to the primary balance and debt, and to rules for fiscal management. It includes a detailed assessment of
the government’s contingent liabilities. The Multiyear Plan, the Budget Guidelines Law, and the
Annual Budget Law must be consistent.
• Debt. The FRL presents a detailed definition of consolidated long-term public debt, public securities, credit operations, and guarantees; sets strict provisions on indebtedness and issuance of
public debt by the central bank, and prohibits creditor debt-restructuring operations among the
various levels of government. In accordance with Article 52 of the Federal Constitution, the Senate passed a resolution establishing limits on indebtedness by level of government. If an SNG is
in breach of the debt ceilings, new financing and discretionary transfers to the SNG are banned.
• Personnel expenditures. The FRL sets separate ceilings for personnel spending, including pensions
and payment of subcontractors. Spending is limited to 50 percent of net current revenues for
the federal government and 60 percent for states and municipalities. The law also establishes
limits for the executive, legislative, and judicial branches. In case of noncompliance, the jurisdiction will not be allowed to engage in new credit line operations, and SNGs will not be allowed to
receive transfers or credit guarantees from the federal government.
• Control and transparency. Budget reports and compliance with the limits set by the law must
be reported every two or four months. The legislative branch of each level of government, supported by its respective court of accounts, monitors the fiscal targets and ceilings. Procedures
for record keeping and consolidation of accounts were significantly improved to reveal compliance with the provisions of the law.
Article 167, Section III, of the Federal Constitution establishes a “golden rule” to prevent the use
of borrowing to finance current expenditures: the amount of new loans contracted is limited to
the amount of capital expense. Law No. 10028 (October 31, 2000) establishes penalties for public
officials not complying with the FRL.
the late 1990s even in the context of weak party loyalties and used his
office (and reputation as an inflation fighter, from when he was minister
of finance) to press successfully for fiscal discipline at the national and
subnational levels.”
All of these institutional changes informed the debate and helped
the main agents agree to more fiscal responsibility and transparency,
which were consolidated in the 2000 FRL. The preparation, discussion,
approval, and implementation of the FRL followed a strategy in which
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
all key players—federal and subnational governments, the Parliament,
the financial sector, the media, and public finance experts—were educated about the need to change and enforce the framework for public
finances in Brazil. The past “traumas” of fiscal distress, especially at the
subnational level, also made the agents more open to introducing more
controls and limits.22
SNG Fiscal Performance after Debt Renegotiation
Agreements
Since the first round of negotiations in 1989, the Brazilian economy
has undergone profound changes, including a substantial improvement
in macroeconomic indicators, especially in the fiscal stance. The institutional reforms allowed, among other things, a new pattern of intergovernmental relations in which states and municipalities were able to
achieve sound fiscal outcomes.
As noted, following implementation of the Real Plan, the country
adopted an inflation-targeting, flexible exchange rate regime and began
to generate significant primary surpluses. Inflation was reduced and has
remained below 5.9 percent since 2005. Starting in 2004, the economy
experienced solid growth, with annual rates above 5 percent (except in
2009 when GDP fell 0.6 percent). In 2010, following the accumulation
of foreign reserves, the public sector became a net creditor in foreign
currency. Not only was SNG debt reduced, but federal government
debt also fell, from 38 percent of GDP in 2002 to 27.5 percent in 2010.
The rescheduling of SNG debt and the institutional changes that
occurred in the post-FRL period led the fiscal adjustment occurring
in the past decade. Since the debt renegotiation in 1999 and 2000, the
ratio of SNG debt to GDP declined steadily, from more than 20 percent
in the beginning of 2000 to around 13 percent in 2010. To assess the
factors that contributed to this performance, we examine revenue and
expenses during 2000–09.23 It is critical to examine the consolidated
public finance data for both states and municipalities and to include
indirect institutions such as SOEs, foundations, agencies, and autarchies.24 Excluding them runs the risk of underestimating the extent of
contingent liabilities.
53
54
Until Debt Do Us Part
Fiscal Turnaround: Deficit and Debt Trend
Throughout the first decade of the 2000s, state and municipal finances
improved significantly. SNGs generated a primary surplus of about
1 percent of GDP, reversing the deterioration of the previous two decades
(table 1.1). The limits imposed by the FRL on public spending, debt, and
debt service were crucial to these results. In addition, increases in bank
credit operations for the public sector were subjected to controls and
limits established by the National Monetary Council.
Total public net debt as a share of GDP declined from 52 percent
in 2001 to 39 percent in 2010, with all levels of government showing
improvement in this indicator: for the federal government, it declined
from 31.7 to 26.4 percent; for states, from 18.1 to 11 percent25; and for
municipalities, from 2.2 to 1.8 percent (figure 1.5). These improvements
were the result not only of GDP growth, but also of rising tax revenues
and declining ratios of SNG net debt to GDP (figure 1.6).
States have benefited significantly from the increase in revenues occurring after 1998. This is due in part to the fact that ICMS collection has
been positively affected by higher taxes on certain products, such as
telephony in several states; to the higher prices of petroleum products
generally; and to improved tax collection efficiency (see Giambiagi 2008).
Municipalities experienced a substantial real increase in tax revenues, particularly the tax on services (ISS). Total municipal tax revenues increased
50 percent over inflation from 2001 to 2010. The ISS grew 105 percent
during the same period, while national GDP grew 40.7 percent.
The reduction in state public debt can also be assessed by the ratio of
net debt to net current revenue. This ratio was greater than 1 in 17 states
in 2000, but in only 7 states in 2010. States with the majority of SNG
debt, such as São Paulo and Rio de Janeiro, lowered their ratio of debt to
net current revenue, respectively, from 1.9 and 2.1 in 2000 to 1.5 and 1.4
in 2010. In general, SNG budgetary execution became stronger, and the
burden of debt service became looser. As seen earlier, the distribution
of state debt is highly concentrated, favoring the richer states, which
were the major beneficiaries of the debt renegotiation process. The three
richest states—São Paulo, Minas Gerais, and Rio de Janeiro—accounted
for about two-thirds of total debt in 2009 (table 1.2).
Like state debt in Brazil, municipal debt is also highly concentrated.
The four state capitals and five cities with medium and large populations
Table 1.1 Public Sector Borrowing Requirements (PSBR) in Brazil, 1999–2010a
% of GDP
Consolidated nominal balanceb
States and municipalities
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
5.3
3.4
3.3
4.4
5.2
2.9
3.6
3.6
2.8
2.0
3.3
2.6
3.2
2.0
2.0
3.8
1.5
1.8
0.2
0.6
0.5
1.2
-0.1
1.3
State governments
2.7
1.8
1.9
3.2
1.2
1.4
0.2
0.4
0.5
1.0
-0.1
1.1
Municipal governments
0.5
0.3
0.1
0.5
0.2
0.3
0.0
0.1
0.1
0.2
0.0
0.2
Consolidated nominal interestc
8.2
6.6
6.7
7.7
8.5
6.6
7.4
6.8
6.1
5.5
5.4
5.3
c
3.4
2.7
3.0
4.7
2.5
2.8
1.4
1.6
1.7
2.3
0.7
1.9
State governments
States and municipalities
2.9
2.3
2.7
4.0
2.1
2.4
1.1
1.4
1.4
1.9
0.5
1.6
Real interest
n.a.
n.a.
n.a.
0.3
1.1
0.7
0.8
0.8
0.5
1.0
0.5
0.6
d
Monetary correction
n.a.
n.a.
n.a.
3.8
1.0
1.7
0.3
0.5
0.9
1.0
0.0
1.1
Municipal governments
0.5
0.4
0.4
0.7
0.4
0.4
0.2
0.3
0.2
0.3
0.1
0.3
Real interest
n.a.
n.a.
n.a.
0.2
0.2
0.2
0.2
0.2
0.1
0.2
0.1
0.1
Monetary correction
n.a.
n.a.
n.a.
0.5
0.1
0.2
0.0
0.1
0.1
0.2
0.0
0.2
Consolidated primary balance
-2.9
-3.2
-3.4
-3.2
-3.3
-3.7
-3.8
-3.2
-3.3
-3.4
-2.0
-2.8
States and municipalitiesc
-0.2
-0.6
-1.1
-1.0
-1.0
-1.0
-1.1
-1.1
-1.1
-1.1
-0.8
-0.6
-0.2
-0.5
-0.8
-0.8
-0.9
-0.9
-0.9
-0.9
-1.0
-0.9
-0.6
-0.5
0.0
–0.1
–0.3
–0.1
–0.1
–0.1
–0.2
–0.1
–0.2
–0.2
–0.1
–0.1
b
State governments
Municipal governments
55
Source: Central Bank of Brazil.
Note: GDP = gross domestic product. n.a. = not applicable.
a. Includes federal companies Petrobrás and Eletrobrás during 1999–2001, not affecting data for states and municipalities.
b. Surplus in terms of PSBR concept.
c. Includes companies.
d. The breakdown of monetary correction/real interest for SOEs is not available, so the total amount of nominal interest was added to monetary correction.
56
Until Debt Do Us Part
Figure 1.5 Net Public Debt in Brazil, 2001–10
70
60
% of GDP
50
40
30
20
10
0
1 2 2 3 3 4 4 4 5 5 6 6 6 7 7 8 8 9 9 9 0 0
/0 y/0 t/0 r/0 g /0 /0 /0 v/0 r/0 /0 /0 l/0 c/0 y/0 t/0 r/0 /0 /0 /0 v/0 r/1 p/1
p b u
n n
g n n
p e
a c a
a c a
p
M O M Au Ja Ju No A Se Fe J De M O M Au Ja Ju No A S
Month/year
c
De
Total debt
States
Federal government and central bank
Municipalities
Source: Central Bank of Brazil.
Note: GDP = gross domestic product.
Figure 1.6 Net Debt of States and Municipalities in Brazil, 1991–2010
25
% of GDP
20
15
10
5
0
91 992 993 94 995 96 997 98 99 00 001 002 003 04 05 06 07 08 09 010
1 19 1 19 1 19 19 20 2 2 2 20 20 20 20 20 20 2
1
Year
19
State and municipal government net debt
State and municipal net debt (includes companies)
Source: Central Bank of Brazil.
Note: GDP = gross domestic product.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
Table 1.2 Share of GDP, Population, and Long-Term Debt of Borrower States
in Brazil, 2009
State
São Paulo
% of GDPa
% of populationa
% of states’ long-term
debt, 2009b
33.1
21.6
38.5
Minas Gerais
9.3
10.5
14.3
Rio de Janeiro
11.3
8.4
12.9
Rio Grande do Sul
6.6
5.7
9.4
Goiás
2.5
3.1
3.2
Bahia
4.0
7.6
2.4
Alagoas
0.6
1.6
1.6
Mato Grosso do Sul
1.1
1.2
1.6
Other states
Total
31.5
40.2
16.1
100.0
100.0
100.0
Sources: STN; Federal Institute of Geography and Statistics (IBGE).
Note: GDP = gross domestic product.
a. GDP and Population: 2008.
b. Long-term debt stock on December 31, 2009.
(which are home to 13.9 percent of the total population and generate
22.5 percent of GDP) account for 84.2 percent of long-term public debt
and around 20 percent of judicial writ debt (table 1.3). The Municipality of São Paulo alone accounts for two-thirds of the total. In general,
the liabilities from judicial writs are greater for municipalities than for
states. Currently, municipal debt with tax and legal or judicial obligations in arrears contributes 9 percent of the outstanding balance of
long-term debt, while judicial writs account for 27.7 percent.
State Revenues and Expenditures
The improvements experienced in the 2000s in state deficits and debt
were the result of strong revenue growth and the consolidation of
expenditures. On the revenue side, the growth of states’ own revenue
surpassed the growth of real GDP by 30 percent, contributing to a generally upward trend in the ratio of total revenue to GDP, as shown in
table 1.4, which also identifies the source of revenues.26
The volatility of the main sources of budgetary revenues (including
loans and other credit operations) is worth considering because it helps
explain why investments are unstable at the subnational level (figure 1.7).
57
58
Until Debt Do Us Part
Table 1.3 Share of GDP, Population, and Long-Term Debt of Municipalities
in Brazil, 2009
Municipality/state
% of GDPa
% of populationa
% of municipal long-term
debt, 2009b
5.8
63.3
São Paulo/SP
11.8
Rio de Janeiro/RJ
5.1
3.3
11.0
Campinas/SP
1.0
0.6
2.4
Belo Horizonte/MG
1.4
1.3
2.1
Salvador/BA
1.0
1.6
2.0
Guarulhos/SP
1.1
0.7
1.4
Contagem/MG
0.5
0.3
0.8
Joinville/SC
0.4
0.3
0.7
Mauá/SP
Other municipalities
Total
0.2
0.2
0.5
77.5
86.1
15.8
100.0
100.0
100.0
Sources: STN; IBGE.
Note: BA = Bahia, GDP = gross domestic product, MG = Minas Gerais, SP = São Paulo.
a. GDP and Population: 2008.
b. Long-term debt stock on December 31, 2009.
Table 1.4 State Revenue in Brazil, 2000–09
% of GDP
Total revenue
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
12.9
12.7
13.1
12.4
12.5
13.0
13.0
12.7
13.2
13.2
Taxes
7.8
8.2
8.1
8.1
8.2
8.4
8.3
8.2
8.6
8.5
ICMS
6.7
7.1
7.0
6.9
7.0
7.1
7.0
6.9
7.2
7.1
Other taxes
1.0
1.1
1.2
1.2
1.2
1.3
1.3
1.3
1.4
1.5
Grants from other
governments
2.5
2.6
2.9
2.3
2.4
2.6
2.7
2.6
2.8
2.6
Current grants
2.4
2.5
2.6
2.2
2.3
2.6
2.6
2.5
2.7
2.5
Capital grants
0.1
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.2
Credit operations
0.3
0.1
0.3
0.2
0.1
0.1
0.1
0.1
0.1
0.4
Sale of assets
0.7
0.1
0.1
0.1
0.0
0.1
0.1
0.1
0.0
0.1
Miscellaneous
revenues
1.7
1.6
1.8
1.8
1.7
1.8
1.7
1.7
1.6
1.6
Sources: STN; state financial statements; IBGE.
Note: GDP = gross domestic product.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
Figure 1.7 Main Sources of State Revenues in Brazil, 2000–09 (2000 = 100)
4.0
3.5
3.0
% of GDP
2.5
2.0
1.5
1.0
0.5
0
1
0
20
02
20
03
20
04
20
05
20
06
20
Year
07
20
08
20
09
20
10
20
Source: STN.
During the 2000s, states’ own taxes and current grants from the federal government grew steadily until 2009 and the onset of the global
financial crisis (figure 1.7). States’ own taxes collected grew 7 percent
on average annually.27 The ICMS represented more than 80 percent
of states’ own tax revenues. The growth in ICMS was partially due to
the growth of sectors such as telecommunications, and to the price of
petroleum-related products, in addition to the states’ efforts to improve
tax collection.
Current grants from the federal government increased 36.2 percent
during 2000–09 as a result of the decentralization of the provision of
health services and the expansion of the FPE, through which 21.5 percent
of the federal income tax and the value-added tax on industrialized products (IPI) is granted to the states.28 During 2000–09, the FPE increased its
share of current transfers from 48.4 to 56.9 percent. Although benefiting
a few oil-producing states such as Rio de Janeiro, the rise in oil prices
brought a large increase in the federal transfer of royalties. Total transfers
were almost twice as high in 2008 as they were in 2003.
59
60
Until Debt Do Us Part
The sources of budgetary revenue directly related to investments—
capital grants and credit operations—declined until 2007 (figure 1.7).
Capital grants generally are not mandatory and may be restricted by
the targets set by the federal government,29 while current transfers
are totally earmarked to specific social areas, leaving no margin for
additional capital investments. In addition, credit operations are subject to limits on total public borrowing set by the central bank. From
2007 onward, both sources began to expand, even after the world financial crisis. The improved macroeconomic conditions and fiscal performance allowed the states to raise new debt while reducing their ratio of
debt to revenue.
On the expenditure side, the FRL limits personnel expenditures to no
more than 60 percent of net current revenue. The reduction in personnel expenditures and the simultaneous increase in revenue (especially
current revenue) improved this ratio: in 2000, 13 states were over the
60 percent limit compared with only a few states in 2009, reflecting a
temporary deterioration caused by the global financial crisis. The control of wage bill outlays, together with the reduction in investments30
as a percentage of primary revenue, allowed states to obtain primary
surpluses to fund their debt service. In addition, as a result of the reduction in debt stock, debt service fell as a share of GDP, from 1.2 percent in
2000 to 1 percent in 2009.
Debt service in a given year may not include the total debt service
incurred that year, since the debt renegotiation agreements placed
a limit on the ratio of debt service to net current revenue. The major
borrowers may have been accumulating residual amounts of debt service, which they capitalized as part of the debt stock, according to the
debt renegotiation contracts. Table 1.5 presents expenditures as a share
of GDP during the 2000s. The increase in grants to municipal governments was due not only to the growth in tax revenue, but also to the
creation of a financial fund for education services.31
At the subnational level, investments are highly dependent on fiscal year current surpluses and eventual revenues from the sale of assets,
which makes them susceptible to the economic cycle. Besides that,
new administrations generally follow a political cycle, cutting current
(mainly capital) expenses as soon as they take office and increasing
spending at the end of their term.32 For example, in 2003 and 2007, new
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
61
Table 1.5 State Expenditures in Brazil, 2000–09
% of GDP
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
12.9
13.0
13.4
12.5
12.5
12.9
13.1
12.4
12.9
13.3
5.8
5.8
6.1
5.8
5.6
5.7
5.8
5.6
5.5
5.7
Public servants
3.7
3.7
3.4
3.3
3.1
3.2
3.3
3.3
3.3
3.4
Retired/pensions
1.9
1.8
2.1
2.0
1.9
1.9
1.9
1.8
1.8
1.8
Other personnel
expenses
0.2
0.3
0.6
0.6
0.6
0.6
0.6
0.5
0.4
0.5
Grants to other
governments
2.0
1.9
2.4
2.2
2.3
2.4
2.2
2.3
2.5
2.5
Investments + acquisition
of financial assets
1.4
1.3
1.3
0.9
1.0
1.1
1.2
1.0
1.3
1.5
Debt service
1.2
1.1
1.1
1.1
1.0
1.0
1.1
1.0
1.0
1.0
2.5
2.8
2.4
2.5
2.6
2.7
2.8
2.5
2.6
2.7
Total expenditure
Personnel
Miscellaneous expenses
a
Sources: STN; state financial statements; IBGE.
Note: GDP = gross domestic product.
a. Goods and services/social security transfers.
administrations took power, which helps explain the relatively small
amount of investments. During the 2008–09 global financial crisis,
investments rose sharply due mainly to the expansion of capital grants
and credit operations as part of countercyclical macroeconomic policies
(figure 1.8). In 2009, the National Bank of Economic and Social Development offered federal loans to states to fund critical projects in the
Federal Program for Growth Acceleration.33
Municipal Revenues and Expenditures
Unlike the states, which generated a primary surplus in the early 2000s
only after the third round of debt renegotiation, municipalities began to
generate primary surpluses in 1997. This was an inaugural year of new
municipal administrations, when expenditures are generally lower. The
primary balance increased during 1999–2001, when debt renegotiations
took place, and remained positive in the following years.
Municipal revenues increased substantially during 2000–09 (figure 1.9).
Tax collection grew at an annual average rate of 4.4 percent. Since 2004,
the ISS has been growing no less than 10 percent per year in real terms,
with the exception of 2009, which was during the global financial crisis, when the rate (still) was 3.3 percent higher than in the previous
Until Debt Do Us Part
Figure 1.8 Main Sources of Investment Financing for States in Brazil, 2000–09
1.6
1.4
1.2
% of GDP
62
1.0
0.8
0.6
0.4
0.2
0
01
00
03
02
20
20
04
20
20
20
05
20
Year
06
20
07
09
08
20
20
20
Credit operations
Capital grants
Investments + acquisition of financial assets
Source: STN.
Figure 1.9 Main Sources of Revenue for Municipalities in Brazil, 2000–09
(2000 = 100)
160
140
120
100
80
60
40
20
20
00
01
20
02
20
03
20
05
04
20
20
06
20
07
20
08
20
Year
Taxes
Current grants
Sources: STN; municipal financial statements; IBGE.
Capital grants
09
20
Credit operations
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
63
year. Faster economic growth, changes in legislation, and modernization of tax management contributed to this improvement. Grants from
the federal and state governments, by far the major group, increased
during 2000–09, at an average rate of 4.8 percent per year. During the
period, municipalities counted on additional resources for education
(the FUNDEF and the Fund for Maintenance and Development of Basic
Education and Teacher Training), and for health, since the provision of
health services was decentralized.
The composition of expenditures was relatively stable, consisting
mainly of personnel and other miscellaneous expenses, most of which
are current expenses incurred for the provision of services. Table 1.6
identifies the main items of expenditure. It also includes a memorandum item—the ratio of personnel expenses to net current revenues—
that remained relatively stable throughout the period of analysis, with
a slight increase in 2009. Figure 1.10 highlights the major municipal
expenditures for this period.
At the local level, the political cycle explains the smaller share of
investments in total expenditures in the initial years of a new administration, as occurred in 2001 and 2005. The ratio of investments to total
Table 1.6 Expenditures of Municipalities in Brazil, 2000–09
% of GDP
Total expenditure
Personnel
2000
2001
2002
2003
2004
2005
2006
2007
7.0
7.0
7.4
7.0
7.2
7.1
7.7
7.6
2008 2009
7.8
8.0
3.0
3.0
3.2
3.1
3.2
3.3
3.4
3.3
3.4
3.7
Public servants
2.3
2.3
2.3
2.3
2.3
2.3
2.4
2.4
2.4
2.6
Retired/pensions
0.3
0.3
0.4
0.3
0.3
0.3
0.3
0.3
0.4
0.4
Other personnel
expenses
0.4
0.4
0.6
0.5
0.6
0.6
0.7
0.6
0.6
0.7
Investments + acquisition
of financial assets
0.8
0.7
1.0
0.8
0.8
0.6
0.9
0.9
1.1
0.8
0.2
0.2
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
2.9
3.0
2.9
2.9
2.9
3.0
3.2
3.1
3.1
3.2
43.1
42.1
43.5
44.8
43.2
41.3
42.8
42.8
41.7
45.6
Debt service
Miscellaneous expenses
a
Memo
Personnel/NCR (%)
Sources: STN; municipal financial statements; IBGE.
Note: GDP = gross domestic product, NCR = net current revenue.
a. Goods and services/social security transfers.
64
Until Debt Do Us Part
Figure 1.10 Main Expenditures of Municipalities in Brazil, 2000–09 (2000 = 100)
200
180
160
140
120
100
80
00
20
01
20
02
20
03
20
05
20
04
20
06
20
07
20
08
20
09
20
Year
Personnel
Debt service
Investments + acquisition of financial assets
Miscellaneous expenses
Sources: STN; municipal financial statements; IBGE.
Note: GDP = gross domestic product.
expenditures was 10 percent in 2001, rising to an average of 12 percent
during the three remaining years of the term. In 2005, the ratio fell to
8.8 percent, recovering to an average of 12 percent in the remaining
years of 2005–08 administrations. The lower ratio of debt to revenue
results in a lower burden of debt service, which averaged 0.26 percent of
GDP during 2000–09. The ratio of debt service to net current revenue,
per the FRL target, averaged close to 3.5 percent. Debt service may not
include the total amount incurred in the year, since limits are placed on
payments to net current revenue. The sharp increase in debt service is a
“base effect,” however, because the municipal debt renegotiations took
place mainly during 1999–2000, and debt payments were resumed in
2001, with the National Treasury as the major creditor.
To sum up, during 2000–09, except for some specificities such as the
growth of debt until 2003, public sector borrowing requirements, debt,
and personnel expenditures improved steadily. Growth in revenue and
control of personnel expenses, particularly in the states, strengthened
the fiscal accounts of SNGs, and this improvement was fundamental to
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
tackling the crisis of 2008–09.34 One important lesson is that the subnational fiscal framework, which was established at the end of the 1900s
with the debt renegotiations, has not only allowed SNGs to improve their
fiscal stance, but has also shown a strong resilience and a reasonable level
of flexibility, overcoming without major changes two economic downturns. First, during 2001–03, right after enactment of the FRL, Brazil
faced an energy supply crisis (especially electric power), which had a
significant impact on economic developments, and a market confidence
crisis associated with the 2002 election. Second, the fiscal framework also
stood firm during the 2008–09 global financial crisis, which is analyzed
in the next section.
Impact of the Global Financial Crisis and Challenges
to Subnational Fiscal Sustainability
Brazil showed strong resilience during the 2008–09 financial crisis.
Blanco, Barbosa Filho, and Pessôa (2011) conclude that Brazil’s adoption
of far-reaching structural reforms and price stabilization initiatives in the
1990s, followed by the consistent pursuit of sound macroeconomic policies in the 2000s, strengthened the country’s resistance to external shocks.
Nevertheless, the global financial crisis affected the Brazilian economy and state and municipal finances. The immediate and perhaps most
important impact was on financial markets—a reduction in external
credit and a sharp depreciation of the domestic currency. The downturn
in economic activity in the world’s major economies had a significant
impact on the price of commodities exported by Brazil: exports fell
10.2 percent in 2009 compared to 2008, reducing tax revenue, particularly the ICMS of exporter states.
The government’s active management of macroeconomic policies—
fiscal, monetary, and external policies, in particular—mitigated the effects
of the crisis. After contracting 0.3 percent in 2009, real GDP rebounded in
2010, growing 7.5 percent and, because of the continuation of the global
economic crisis, economic growth slowed in 2011 with GDP increasing
only 2.7 percent. Inflation increased to around 6 percent a year but is
still close to the upper band of the inflation targets defined by the central bank. Gross official reserves recovered, and credit to the private sector
returned to trend levels in 2010.
65
66
Until Debt Do Us Part
The decline in economic activity and the implementation of a set
of tax exemption measures aimed at sustaining aggregate household
consumption led to a fall in tax revenue in 2009. Tax as a share of
GDP declined from 34.4 percent of GDP in 2008 to 33.6 percent in
2009, as GDP contracted 0.6 percent and total tax collection fell almost
3 percent. The federal government increased its spending on salaries
for civil servants and raised the minimum wage, affecting the payment
of pension benefits. As a consequence, the consolidated primary fiscal
surplus fell to 2 percent of GDP in 2009, well below the average level of
3.5 percent during 2004–08. Inflation as measured by IGP-DI declined
1.4 percent in 2009, reducing the nominal interest rate and the need for
public sector borrowing.35
The countercyclical fiscal policy included many measures that supported state and municipal revenues: (a) an increase in capital grants
due to federal investments programmed under the Federal Program
for Growth Acceleration; (b) a reduction in income tax collection and
cuts in the IPI tax, which helped support the auto industry, prevent
job losses, and reduce SNG transfers through the FPE and Municipalities Participation Fund; (c) compensation for the loss of current
grants—states received additional credit lines from the National Bank
of Economic and Social Development to keep investments and to match
federal capital transfers related mainly to the Federal Program for
Growth Acceleration; and (d) receipt by municipalities of general grants
to keep the amounts transferred from the Municipalities Participation
Fund the same as in 2008. Public banks expanded the supply of credit to
help exporters finance their costs in foreign currency and to help corporations and consumers offset the decline in private credit.36
Impact of the Crisis on States and Municipalities
The impact of the crisis differed across states and municipalities and
across participants in each group. It is necessary to examine the impact
not only of the crisis but also of federal policies to offset the economic
downturn.37
The immediate impact of the economic downturn on the fiscal
accounts of states was to reduce ICMS collection and transfers from
the FPE, the latter due to the decrease in federal revenues from the
income tax and tax on industrialized products. For all the states, these
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
two sources of revenue have represented around 50–70 percent of disposable income (after transfers to municipalities).38 The impacts of the
global crisis on state finances began to be felt only in late 2008 and were
not enough to erase the revenue gains in 2007 and most of 2008:
•
•
I n 2008, the ICMS, the main source of states’ own revenues, grew at
an average real rate of 9.5 percent compared with 3 percent in 2007
(8 out of the 27 states had increases over 10 percent). FPE transfers from the federal government performed even better, increasing
12.9 percent compared with 3 percent in 2007 (uniform for all the
states). In 2009, the reduction in revenues from the ICMS (2.5 percent as a national average) and in the FPE (8.9 percent) was only a
reduction relative to the excellent performance of 2008. In 2010, the
ICMS tax collection had recovered to its precrisis level.
Fiscal indicators for a sample of 10 states during 2007–09 confirm
these results.39 In 2009, all but one state in the sample had revenue
losses compared with 2008, but they were in a stronger position than
in 2007 as measured by net current revenue (RCL).40 The investment
boom during 2008–09 was stupendous. In 2007, investments were
financed mostly by “other internal sources,” which included the current surplus (after payment of debt service). The expansion of investments was an important element in the expansion of capital grants
and credit operations, although the composition of sources differed
among states. The federal government, through National Monetary
Council resolutions, provided additional resources to states through
loans to compensate the loss of transfers from the FPE. The increase
in funds raised through new indebtedness was small compared with
the outstanding stock of debt; thus, the ratio of net debt to net current revenue was not seriously affected. Total credit granted in 2009
was about 2.5 percent of the outstanding contractual debt at the end
of 2009. Moreover, the growth in net current revenue and decline in
the IGP-DI helped reduce the ratio.41
The impact of the global crisis was less intense on municipalities than
on states. While states generally lost current revenues during 2008–09,
municipalities were able to offset the losses from ICMS transfers
through increases in other transfers and own taxes. Fiscal indicators for
a sample of six municipalities (Belo Horizonte, Cuiabá, Porto Alegre,
67
68
Until Debt Do Us Part
Rio de Janeiro, Salvador, and São Paulo) show that in 2009, total revenue remained constant, mainly due to the following factors:
•
•
•
•
he property tax and the tax on services grew faster than inflation
T
(5 and 3.3 percent, respectively), allowing total tax revenue to grow
in real terms.
The federal transfers targeted to health services increased 10.2 percent
in real terms, compensating the slight real reduction (0.5 percent) in
other current transfers.
The federal government adopted a policy of using voluntary transfers
to compensate the loss of transfers from the Municipalities Participation Fund (around 25 percent of current grants), in order to maintain the nominal amount received by each municipality during the
previous year.
The resources of the Fund for Maintenance and Development of
Basic Education and Teacher Training grew largely as a result of state
contributions.
In contrast to state investments, municipal investments, as reflected
by investments of the six capital cities, either decreased or grew just
slightly in 2009.42 In addition, the stability of revenues resulted in a stable ratio of personnel expenses to net current revenue.
Challenges to Subnational Fiscal Sustainability
The fiscal performance of Brazil’s SNGs during the last decade was
impressive. However, SNGs are facing challenges that may have an
impact on their long-term fiscal sustainability. Here we highlight three
issues that bear a more immediate relationship to the debt restructuring
framework discussed in previous sections:43 (a) the indexation rules for
debt that was renegotiated at the end of 1990s and the accumulation of
residuals, (b) the narrow fiscal space and low current public investment,
and (c) the potential for reducing debt service cost through more competitive subnational credit markets. We deal with each in turn.
As noted, the SNG debt renegotiation indexed the subnational
debt to the IGP-DI, plus a “real” interest rate of about 6 percent. This
arrangement was reasonable at the time of the renegotiation because of
the then prevailing macroeconomic conditions. The interest rate (or the
SELIC rate, which is the basic funding cost of the federal government)
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
was much higher than the conditions agreed. The exchange regime had
a fixed exchange rate, and the inflation targeting regime was not yet part
of the monetary instruments. Finally, there was no umbrella fiscal law,
like the FRL, establishing fiscal limits and conditions. Now, the macroeconomic context has changed with new, improved mechanisms such
as a floating exchange rate, and inflation-targeting regimes resulting in
a higher level of macro stability. More important, the SELIC rate has
steadily declined, especially since the global economic crisis, and the
cost of the renegotiated contracts is much higher than the basic interest
rate in the domestic economy. As a result, states would like to refinance
their debt in the market, but the contracts forbid doing so.44
The state authorities have initiated negotiations in the Congress in
the hope of changing the conditions of the original contracts. The IGPDI is highly volatile, however, and therefore not a good reference for
the asset and liability management of federal and SNG governments. A
related challenge that is derived from the debt renegotiation contracts is
the 13 percent cap on net revenues. Although this cap is a good countercyclical feature, which was useful for the most indebted states, it has
generated a residual in several states—building up capitalized interest,
which is a potentially serious financial problem. According to the legislation, this residual must be repaid in 10 years starting in 2030. The
residuals also depend on GDP growth and the trend of the IGP-DI. For
some states, such as Rio de Janeiro, residuals could disappear if GDP
growth averages 4 percent during 2010–20 and the IGP-DI converges to
inflation (according to Levy 2009). Official projections for the Municipality of São Paulo estimate that if the growth of net current revenue is
about 1 percent per year, the ratio of debt to net current revenue will
be about 327 percent in 2030, and the ratio of debt service to net current revenue may reach 51 percent, indicating an unsustainable path
(according to Prefeitura da Cidade de São Paulo 2011).
The narrow fiscal space and the various regulations for new indebtedness have posed challenges in financing public investments at the subnational level. The brunt of SNG fiscal adjustment has fallen on public
investments, resulting in a deterioration in infrastructure and threatening economic growth. While the limits imposed on SNG debt financing
as part of the fiscal adjustment program have been successful in turning
around the SNG fiscal imbalance, a challenge ahead is to identify ways
69
70
Until Debt Do Us Part
of increasing infrastructure investments and finance while maintaining
fiscal discipline. In the long term, economic growth is a key determinant
of fiscal sustainability.
Finally, another challenge is to develop competitive credit markets for
subnational public investments. More competitive subnational credit
markets help lower the cost of financing and extend maturity. Cost of
financing is another key determinant of fiscal sustainability. Canuto
and Liu (2010) show that the subnational debt market in developing
countries has been undergoing a notable transformation. Private capital
has emerged to play an important role in subnational finance in countries such as Poland, Romania, and South Africa. Subnational bonds
increasingly compete with traditional bank loans. Notwithstanding the
temporary disruption of credit markets during the crisis, the diversification of subnational credit markets is expected to continue. SNGs or
their entities in various countries have already issued bond instruments
(for example, in China, Colombia, India, Mexico, Poland, the Russian
Federation, and South Africa). More countries are considering policies
to foster the development of subnational debt markets (for example,
Indonesia and South Africa), while others are piloting transaction and
capacity-building activities to the same end (for example, Peru). Competitive financial markets, with a variety of buyers and sellers and a variety of financial products, can keep borrowing costs down.45
In Brazil, public financial institutions are currently the primary providers of credit to SNGs.46 As part of debt restructuring, SNGs have
been prohibited from issuing bonds. This has helped bring subnational
debt onto a more sustainable path. Looking forward, a key challenge
is to identify ways of increasing competition in the subnational credit
market to reduce the cost of financing and help fiscal sustainability in
the long term. International experience shows that prudent regulatory
frameworks for subnational capital markets are important to manage
the risks of defaults, and the process of developing such frameworks is
gradual and incremental.
Conclusions
The SNG debt crisis in Brazil was the result not only of autonomous
decisions by the SNGs, but also of decisions by the federal government
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
to implement economic development plans amid financial internal and
external constraints, mainly during the authoritarian political regime of
1964–85. The SNGs had an important role in both attracting external
resources that eased the pressure on external accounts and implementing large projects to accelerate urbanization. Coupled with the external fundraising of SNGs—led by central government macroeconomic
policy—the recourse to borrowing was a way to circumvent the strict
controls on fiscal management, with a strong negative effect on the fiscal accounts of SNGs, particularly the states.
The control of subnational borrowing in the form of debt renegotiation agreements in 1997 imposed strict rules on the financial and fiscal
management of SNGs and assured their adherence to the guidelines set
by the federal government for the conduct of macroeconomic policy.
The latest round of renegotiations during 1997–2000 paved the way to
macroeconomic stability.
Building on the success of the Real Plan, in 1999 Brazil adopted a
regime of inflation targeting and flexible exchange rates and initiated
a strong fiscal adjustment program that generated significant primary
surpluses. Inflation was reduced and has stayed below 5.9 percent since
2005, partly explained by the strong expansion of imports at progressively lower prices due to the appreciation of the Brazilian currency
during 2003–10. Starting in 2004, the Brazilian economy began a sustained path of economic growth. Brazil was resilient in the face of the
2008–09 global financial crisis, owing to the combination of sound
macroeconomic management and a highly favorable external scenario
throughout the 2000s (except for the international liquidity crisis in
2009). Both SNG and federal government debt levels were reduced,
with federal government debt falling from 38 percent of GDP in 2002
to 26.4 percent in 2010. Following the accumulation of reserves, the
public sector became a net creditor in foreign currency. The country
achieved trade surpluses, mainly between 2005 and 2007, averaging
US$43.7 billion a year.
The country’s institutional consolidation and modernization, resulting in approval of the Fiscal Responsibility Law in 2000, is crucial to
understanding the trajectory of relative fiscal consolidation and reduction in the ratio of debt to GDP achieved in the last 11 years. However,
the institutional reforms do not end with approval of the FRL. The FRL
71
72
Until Debt Do Us Part
principles, procedures, and requirements would be ineffective if not
followed by other initiatives such as continued efforts to improve rules
and procedures for registering and reporting on public accounts and
the adoption of uniform criteria to demonstrate compliance with legal
limits. These efforts have given more credibility to financial statements,
particularly from SNGs.
This chapter shows that in all government spheres—federal and
subnational—fiscal adjustments and consolidation have been greatly
facilitated by the growth of government tax revenue. Since the debt
restructuring, almost all states and a large number of municipalities
have been modernizing their tax administration. Revenue growth, however, has not translated into expanded public investment, leaving urgent
projects in urban infrastructure with insufficient funding. In contrast,
the provision of basic public services has expanded—increasing the
coverage of health services and education, for example—but this has
increased the need to fund higher current expenditures.
Notwithstanding the success of fiscal reform and debt restructuring, Brazilian SNGs face challenges that may impact the sustainability
of subnational finance and economic growth. The three challenges that
have more direct bearing on debt restructuring are the indexation rules
and potential accumulation of residuals, narrow fiscal space to finance
the public investment gap and its impact on economic growth, and
the need to foster competitive subnational credit markets for lowering
financing costs for public investments.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Brazil is a federal republic, encompassing a federal government (the union),
26 states, a Federal District (Brasília), and 5,564 municipalities. Ranked fifth
in world population, Brazil had 191 million inhabitants in 2010. States and
municipalities vary greatly in size. Unless otherwise indicated, the states include
the Federal District.
2. Serra and Afonso (2007) remind us that the federative framework in Brazil is a
system that is still evolving and point out main aspects in terms of vertical and
horizontal decentralization and government powers.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
3. The restrictive fiscal policy and reduced reliance on imports as a result of
investments made during 1974–79 (the second National Development Plan),
coupled with the world economic recovery, allowed GDP to grow 5.4 percent in
1984 and around 4.5 percent in 1985 and 1986 (Giambiagi et al. 2005).
4. Rezende and Afonso (2002) find that two important facts have shaped the way
Brazilian fiscal federalism currently looks: (a) the transition from authoritarian
rule to democracy, following the demise of the military regime in 1985; and
(b) the policies adopted in the 1990s to achieve domestic and external balances.
Rigolon and Giambiagi (1999) describe the context in which the renegotiation
took place and emphasize that it was part of the administrative measures aimed
at reducing SNG indebtness.
5. The Collor Plan II attempted to deal with the persistent inflation and increasing
difficulty in placing public bonds. The plan aimed to eliminate overnight operations and other forms of price indexation. A Financial Investment Fund was
created as a captive market for government securities. Fiscal austerity measures
included the blocking of federal spending by the ministries and state companies. Despite lower inflation, political resistance to the economic team doomed
the plan (see Gremaud, Vasconcellos, and Toneto 2010).
6. Law No. 7976 of 1989 refinanced the direct and indirect debt of SNGs derived from
loans that had been granted by the national government in order to meet commitments due to external credit operations, guaranteed by the National Treasury.
7. RLR is used to calculate the SNG’s debt payment limit and the ratio of financial debt to RLR. It is also used as a parameter for the states’ fiscal adjustment
and restructuring programs. It is calculated as a moving 12-month average
of revenue collected, excluding revenue from credit operations, sale of property, voluntary transfers, donations received to meet capital expenses, and, in
the case of states, transfers to municipalities, due to legal or constitutional
participations.
8. Judicial writs, called precatórios, are legal requests for payment of a certain
amount by the federal, state, or municipal treasuries. They cannot be appealed.
At the end of judicial enforcement, a letter is submitted to the president of the
court requiring payment of the debt. The requests received by the court until
July 1 of each year are included in the budget proposal, to be paid the following
fiscal year. SNGs often have significant amounts of these debts in arrears.
9.
The SELIC rate—Brazil’s prime interest rate—is the average interest rate
charged on daily financing of operations and is backed by government securities registered in the SELIC.
10. According to the Federal Constitution, the Senate is entitled to define amounts
and conditions of debt issued by the union, states, and municipalities.
11. The debt negotiations have adopted a “contractual approach.” See Grembi and
Manoel (2011) for an analysis of Latin America.
12. The IGP-DI is a weighted index of the Wholesale Price Index, the Consumer
Price Index, and the National Construction Cost Index, with weights equal to
73
74
Until Debt Do Us Part
6, 3, and 1, respectively. The “Domestic Availability” version was created in 1969
with the aim of isolating the effects of coffee price oscillations by assigning a
lower weight to export products.
13.
The states of Amapá and Tocantins had no significant debt so were not
included.
14. Based on Ministry of Finance data.
15. The states of São Paulo and Bahia had already completed their renegotiation.
16. At the end of 1998, the Municipality of Rio de Janeiro presented a proposal for
restructuring its debt with the federal government. The proposal suggested the
same restructuring terms that applied to the states (Law No. 9496) (see Prefeitura Municipal do Rio de Janeiro 1998).
17. Recently, the Municipality of Rio de Janeiro restructured its debt through
a World Bank loan of US$1.045 billion to repay up to 25 percent of the debt
renegotiated with the federal government. This prepayment reduced the interest from 9 to 6 percent.
18. See IMF (2003, 9) for a detailed explanation of the crisis. According to the IMF,
the Brazilian crisis, like others in emerging markets during the 1990s, would be
better “described as capital account crises to distinguish them from the more
conventional crises which have their origins mainly in the current account.”
19. The Fiscal Responsibility Law was approved on May 4, 2000, as Complementary
Law No. 101 (see box 1.3). Passage of a complementary law requires a higher
threshold of voting (75 percent of both houses).
20. Leite (2011) argues that some of these factors are key to explaining why the FRL
was passed in the Brazilian Congress.
21. Setting up the National Treasury at the Ministry of Finance was a milestone
in terms of comprehensive analysis and control of public finance in Brazil.
Among other responsibilities, it included (a) executing the budget (cash flow),
(b) overseeing subnational public finances, (c) monitoring financial aspects of
government banks and SOEs, (d) public debt management (registering, controlling, reporting), (e) managing federal government financial assets, and
(f) accounting and reporting on federal government accounts. For several years,
the National Treasury was also in charge of “internal control” activities for the
federal government.
22. Tavares, Manoel, and Afonso (1999) describe the FRL project and the antecedents of the new fiscal rules.
23. The main sources of information in this section are the Federal Institute of
Geography and Statistics (IBGE), the Central Bank of Brazil, the STN, the
Council of States Secretaries of Finance of the Ministry of Finance, and the
financial statements of several states. Data pertaining to budget execution of
the states for 2000–09, made available by the STN based on information provided by the states, were consolidated. Despite differences in methodology and
coverage among the different sources, the results obtained for the indicators are
consistent, allowing comparison of the data.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
24. Central Bank of Brazil data for net public debt and main fiscal indicators distinguish direct and indirect administration (state and municipal governments) from
the figures for SOEs. STN data consolidate figures for revenues, expenditures, and
gross debt for direct and indirect administration with those related exclusively
to dependent companies. Data for municipalities during 2000–09 cover around
95 percent of Brazilian cities.
25. The debt stock is the debt of both state governments and state companies.
26. Data on SNG revenue and expenditures are based on information made available by the STN and by states and municipalities through official sites. Time
series presenting the evolution of revenues, shown in figure 1.9, eliminate the
impact of inflation.
27. Beginning in 2000, taxes include federal income tax withheld at source and
the payment of tax debts in arrears (dívida ativa). In 2000, both accounted for
4.5 percent of total tax revenue. Therefore, comparisons can be made for only
2000–09.
28. Reports from the Brazilian Internal Revenue Service (Receita Federal) indicate that the ratio of income tax to GDP was 12.6 percent in 2009, up from
5.5 percent in 2000. The ratio of the IPI to GDP was 1.7 percent in 2006, up
from 0.9 percent in 2000, due to a temporary exemption from the IPI granted
to the automotive sector.
29. To comply with the targets for primary balance set by the Congress, the federal
government may have to reduce the amount of capital grants to SNGs.
30. Investments include the acquisition of financial assets, mostly related to the
issuance of bonds of independent state companies.
31. The Financial Fund for Educational Services (FUNDEF) was created by Constitutional Amendment No. 14/96 instituting intergovernmental financial
cooperation for improving elementary education. The fund is financed by earmarking percentages of transfers from the revenue-sharing system to guarantee
a specified minimum amount of spending per student enrolled in public elementary schools throughout the country. FUNDEF is funded by (a) 15 percent
of the municipal and state share of the ICMS; (b) 15 percent of the Municipalities Participation Fund; (c) 15 percent of the FPE; and (d) 15 percent of the
municipal and state share in the Compensation Fund for Exports. FUNDEF
funds are distributed according to the number of students enrolled in municipal- or state-owned elementary schools. If the money collected from these
sources is not enough to guarantee the minimum spending established by law,
the federal government is responsible for providing supplementary transfers.
In 2007, the earmarking of revenues was expanded, with the creation of the
Fund for Maintenance and Development of Basic Education and Teacher
Training, to 20 percent from 15 percent, and other revenues were included,
mainly state revenues. The distribution criteria were also changed to take into
account, among other things, students in higher education, a service provided
by the states.
75
76
Until Debt Do Us Part
32. In Brazil, federal, state, and municipal officials serve four-year terms. The head
of government—the president, the governor, and the mayor—can be reelected
once. Federal and state terms coincide, and municipal terms begin and end two
years thereafter. Municipal administrations took office in 2001, 2005, and 2009.
Federal and state administrations took office in 2003 and 2007.
33. In 2009, the state of São Paulo sold the state bank, Nossa Caixa, to finance
investments.
34. Piancastelli and Boueri (2008), after examining the evolution of state fiscal
accounts after 10 years of debt renegotation, conclude that the renegotiation
was a major effort to improve the fiscal stance of the public sector.
35. Central bank data; http://www.bc.gov.br.
36. See the Ministry of Finance website; http://www.fazenda.gov.br.
37. Data used in this analysis can be found on the STN website and on the websites of several states and municipalities. Information about 2010 municipal
finances came from FRL reports, available on the websites of municipalities
and the STN. Some indicators may not be available in some years for all the
states and municipalities. The Finanças do Brasil (FINBRA) database for states
and municipalities is available at http://www.stn.fazenda.gov.br/estados_muni
cipios/index.asp.
38. According to IBGE, the Brazilian territory encompasses five geographic regions:
North, Northeast, Center-west, South, and Southeast. The North and Northeast
are the poorest, and the South and Southeast are the richest. The nation’s capital, Brasília, is in the Center-west Region.
39. States in the sample are Bahia, Ceará, Goiás, Maranhão, Minas Gerais, Paraná,
Pernanbuco, Rio de Janeiro, Rio Grande do Sul, and São Paulo.
40. RCL is calculated as current revenue (taxes, intergovernmental transfers, and
other) minus constitutional transfers to municipalities (obligatory) minus the
revenue of social contributions to the public servants’ pension fund.
41. As indicated, the debt stock is subject to monetary correction by the IGP-DI,
which varied negatively in 2009 (−1.43 percent).
42. This was the first year of a four-year (2009–12) administration, and, following the political cycle, current expenditures, mainly investments, are reduced,
allowing the administration to accumulate cash to spend during the last years
of its term.
43. There are other factors influencing SNG fiscal sustainability. These include
the intergovernmental fiscal system, the expenditure framework, and taxation
reform, which are beyond the scope of this chapter.
44. For example, Article 35 of the FRL prohibits any form of debt renegotiation
between public entities. Pellegrini (2012) also analyzes several constraints, such as
legal, technical, and fiscal, which impede normal renegotiation of the current debt.
45. Various countries have been moving toward more diversified instruments,
including bonds. Total SNG bond issuance in developing countries reached
US$45.1 billion during 2000–07 and US$102.8 billion during 2008–10Q1, with
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
China being the largest and dominant issuer, followed by Russia (Canuto and
Liu 2010, based on data from DCM Analytics).
46. Central bank data on domestic debt of the four largest states in 2010 (including direct and indirect administrations) indicate that private institutions provide only 0.5 percent of total funding for domestic debt. They have a notable
presence only in Minas Gerais and São Paulo, where they hold about 20 and
10 percent, respectively, of the outstanding SOE debt.
Bibliography
Abrucio, Fernando Luiz. 2002. Os barões da federação: Os governadores e a redemocratização brasileira. 4th edition. São Paulo: Hucitec.
Afonso, José Roberto. 2011. Crise, estado e economia brasileira. São Paulo: Agir.
Almeida, Anna Ozório de. 1996. “Evolução e crise da dívida pública estadual.” Texto
para Discussão 448, Instituto de Pesquisa Econômica Aplicada (IPEA), Brasília.
Bacha, Edmar. 1995. “Plano Real: Uma avaliação preliminar.” Revista do BNDES
2 (3): 3–26.
Blanco, Fernando, Fernando de Holanda Barbosa Filho, and Samuel Pessôa. 2011.
“Brazil: Resilience in the Face of the Global Crisis.” In The Great Recession and
Developing Countries: Economic Impact and Growth Prospects, ed. Mustapha K.
Nabli. Washington, DC: World Bank.
Canuto, Otaviano, and Lili Liu. 2010. “Subnational Debt Finance and the Global
Financial Crisis.” PREM Economic Premise 13: 1–7.
Eichengreen, Barry. 2007. Globalização do capital: Uma Historia do Sistema Monetário International. São Paulo: Editora 34.
Garson, Sol. 2010. “Gestão fiscal do Espírito Santo 2002–2008: Pavimentando o
caminho para o crescimento da economia estadual.” In Espírito Santo: Instituições, desenvolvimento e inclusão social. Vitória: Instituto Jones dos Santos Neves.
Giabiagi, Fabio, Fernando Blanco, and Wagner Ardeo. 2008. “Proposta de recriação
de uma dívida estadual e Municipal com o Setor Financeiro Privado.” Revista
do BNDES 14 (29): 365–98.
Giambiagi, Fábio. 2008. “18 anos de política fiscal no Brasil: 1991/2008.” Economia
Aplicada 12 (4): 535–80.
Giambiagi, Fábio, and Ana Claudia Além. 2007. Finanças públicas: Teoria e prática
no Brasil. 3rd edition. Rio de Janeiro: Editora Campus.
Giambiagi, Fábio, André Villela, Lavinia Barros de Castro, and Jennifer Herrman. 2005.
Economia brasileira contemporânea (1945–2004). Rio de Janeiro: Editora Elsevier.
Gremaud, Amaury Patrick, Marco Antonio Sandoval Vasconcellos, and Rodinei
Toneto Jr. 2010. Economia brasileira contemporânea. 7th ed. São Paulo: Atlas.
Grembi, Veronica, and Alvaro Manoel. 2011. “Fiscal Rules for Subnational Governments? Evidence from Latin America.” World Bank, Washington, DC.
Hermann, Jennifer. 2005. “Auge e declínio do modelo de crescimento e endividamento:
O II PND e a crise externa (1974/1984).” In Economia Brasileira Contemporânea
77
78
Until Debt Do Us Part
(1945–2004), ed. Fábio Giambiagi, André Villela, Lavinia Barros de Castro, and
Jennifer Herrman, 93–115. Rio de Janeiro: Editora Elsevier.
IMF (International Monetary Fund). 2003. “The IMF and Recent Capital Account
Crises: Indonesia, Korea, Brazil.” Evaluation Report, Washington, DC. http://
www.imf.org/external/np/ieo/2003/cac/pdf/all.pdf.
———. 2011a. “IMF Executive Board Concludes 2011 Article IV Consultation with
Brazil.” Public Information Notice 11/108, Washington, DC. http://www.imf
.org/external/np/sec/pn/2011/pn11108.htm.
———. 2011b. World Economic Outlook: Slowing Growth, Rising Risks. Washington,
DC. http://www.imf.org/external/pubs/ft/weo/2011/02.
Leite, Cristiane Kerches da Silva. 2011. “Federalismo, processo decisório e ordenamento fiscal: A criação da Lei de Responsabilidade Fiscal.” Texto para
Discussão 1593, Instituto de Pesquisa Econômica Aplicada, Brasília.
Levy, Joaquim V. F. 2009. “Stability and Innovation in Brazilian Subnational Debt.”
Paper presentation at the Subnational Fiscal Reform and Debt Management
Forum, World Bank, Washington, DC, April 29.
Liu, Lili, and Steven Webb. 2011. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper 5587,
World Bank, Washington, DC, March.
Lopes, Francisco. 1986. O choque heterodoxo: Combate à inflação e reforma monetária. Rio de Janeiro: Campus.
Lopreato, Francisco Luiz C. 2000. “O endividamento dos governos estaduais nos
anos 90.” Economia e Sociedade 15 (dezembro): 117–58.
Modenesi, Andre de Melo. 2005. Regimes monetários: Teoria e experiência do Real.
Barueri: Manole.
Mora, Monica. 1997. “Processo de endividamento dos estados: Problemas e limites
à descentralização e à autonomia.” Dissertação de Mestrado, Universidade Federal do Rio de Janeiro, Rio de Janeiro.
———. 2002. “Federalismo e dívida estadual no Brasil.” Texto para Discussão 866,
Instituto de Pesquisa Econômica Aplicada, Brasília.
Pellegrini, Josué A. 2012. “Dívida Estadual.” Textos Para Discussão 110, Núcleo de
Estudos e Pesquisas do Senado, Brasília.
Piancastelli, Marcelo, and Rogério Boueri. 2008. “Dívida dos Estados 10 Anos
Depois.” Texto Para Discussão 1366, Instituto de Pesquisas Econômicas Aplicada, Rio de Janeiro.
Prefeitura da Cidade de São Paulo. 2011. Dívida com a União: MP 2185-35. Secretaria de Finanças, Report of December 31, 2011, São Paulo.
Prefeitura Municipal do Rio de Janeiro. 1998. “Proposta de reestruturação das dívidas municipais.” Rio de Janeiro.
———. 2011. “Emissão de Títulos.” Texto para Discussão, Rio de Janeiro.
Rezende, Fernando, and José R. Afonso. 2002. “The Brazilian Federation: Facts,
Challenges and Prospects.” Working Paper 149, Center for Research on Economic Development and Policy Reform, Stanford University, Stanford, CA.
Brazil: The Subnational Debt Restructuring of the 1990s—Origins, Conditions, and Results
Rigolon, Francisco, and Fábio Giambiagi. 1999. “A Renegociação das Dívidas e o
Regime Fiscal dos Estados.” In A Economia Brasiliera nos anos 90, ed. F. Giambiagi, and M. M. Moreira, 111–44. Rio de Janeiro: National Bank of Economic
and Social Development.
Serra, José, and José R. Afonso. 2007. “Fiscal Federalism in Brazil: An Overview.”
CEPAL Review 91, Santiago, Chile.
Tavares, Martus, Alvaro Manoel, and José Roberto R. Afonso. 1999. “Uma proposta
para um novo regime fiscal no Brasil: O da responsabilidade fiscal.” Compendio de Documentos, XI Seminário Regional de Política Fiscal, CEPAL, Brasília.
Varsano, Ricardo. 1996. “A evolução do sistema tributário ao longo do século:
Anotações e reflexões para futuras reformas.” Texto para Discussão 405, Instituto de Pesquisa Econômica Aplicada, Brasília.
Villela, Renato, and Fernando Rezende. 1986. “Alterações na política financeira:
Desenvolvimento urbano e problemas metropolitanos; Aspectos financeiros.”
In Alternativas político-institucionais para a região metropolitana de Salvador:
Dimensão nacional, ed. Ana Maria Brasileiro, 212–34. Rio de Janeiro: Convênio
IBAM/CONDER.
79
2
Restructuring of Legacy Debt for
Financing Rural Schools in China
Lili Liu and Baoyun Qiao
Introduction
China started to promote nine-year compulsory education in rural
areas throughout the country in the mid-1980s. To achieve that goal, the
Compulsory Education Law, enacted in 1986, mandated that the central and local governments at all levels guarantee the operational and
capital expenditures to implement compulsory education. In practice,
the local governments, especially the county-level governments, played
a main role in financing rural compulsory education. With limited fiscal resources, local governments financed rural compulsory education
funds through a multitude of channels. In particular, to achieve the
national goal of accomplishing universal nine-year compulsory education by 2000, local governments (towns and villages) resorted to borrowing to finance school facilitates to help meet the minimum facility
standards for all schools, although the borrowing was not allowed by
the 1994 Budget Law of China.
In 2000, nine-year compulsory education became universal in China,
a historic accomplishment. However, the accumulation of debt that
resulted from borrowing to finance rural compulsory education had
become a significant fiscal burden on local governments. The aggregate
81
82
Until Debt Do Us Part
rural education debt was about RMB 110 billion1 at the end of 2007,
about 3.9 percent of aggregate subnational own fiscal revenue, excluding transfers. Of the outstanding compulsory education debt, about
RMB 80 billion was used to finance capital expenditure, and about RMB
30 billion was used to finance operational expenditures. The negative
impact of rural compulsory education debt on the rural economy and
social development had become serious. The rural Tax-for-Fee Reform
in 20012 was intended to reduce financial burdens (such as taxes, surcharges, and fees) on rural households, but it also further limited local
fiscal resources. At the same time, the rural reforms increased the
demand for fiscal resources to finance rural infrastructure and social
services.
Restructuring the legacy debt for financing rural schools thus became
a priority of the central government in the mid-2000s. With the public policy goal in the late 1990s of inclusive economic growth, the debt
financing of nine-year compulsory education in rural areas would be
replaced by grant financing for all children. It was necessary to resolve
the widespread indebtedness of rural local governments to make the
debt-to-grant financing transition, and the debt-to-grant financing
would need to address the legacy debt and its write-offs.
Resolving the legacy debt and making a transition to grant finance
should be viewed as part of building a sustainable fiscal system framework to ensure the delivery of basic public services, including implementation of the Compulsory Education Law. With less fiscal capacity
as a result of the rural Tax-for-Fee Reform, the county-level governments could choose new approaches to increase the fiscal burden of
rural residents.
China’s central government, through the Ministry of Finance, initiated a program of restructuring the rural legacy school debt in 2007.
This was a pioneering effort—the first time the central government
undertook debt restructuring of subnational governments nationwide.
A key design issue in any debt restructuring is how to avoid moral hazard. International experience shows that unconditional write-offs create
soft budget constraints. To address the problem of moral hazard, the
Ministry of Finance’s plan had two distinct features.
First, the fiscal resources to finance the debt write-off were distributed equally among the central government, provincial governments,
Restructuring of Legacy Debt for Financing Rural Schools in China
and subprovincial governments; that is, the central government, provincial governments, and subprovincial governments each contributed
one-third of the financial resources.
Second, the distribution of central government grants was based on
a formula that took into account the required expenditure to achieve
basic provision of education and the local government fiscal capacity
to deliver the results. Thus, the distribution of central grants to a particular local jurisdiction was not directly related to the size of the debt
of that jurisdiction. This output-based approach was meant to prevent
perverse incentives for local governments to increase the size of their
debt or to reduce their service of debt in anticipation of more grants
or bailouts.
This chapter analyzes the moral hazard aspect of the rural debt
restructuring within China’s current intergovernmental fiscal framework. The remainder of the chapter is organized as follows. Section two
summarizes the compulsory educational system and what responsibilities are assigned to each government level in China. Section three examines the features of rural education legacy debt. Section four focuses on
China’s strategy to deal with moral hazard. Section five explains how
China restructured the legacy debt for financing rural schools. Section
six offers conclusions.
The Responsibility and Revenue Assignment for
Education in China
China has a unitary system with five levels of government. The central government is at the top under which are four tiers of subnational
government: provincial, prefecture (city), county, and township.3 The
central government determines the establishment and the geographic
division of the provincial, prefecture, and county governments, and
the provincial government determines the establishment and the geographic division of the township governments.
The provincial government, directly under the central government, consists of the governments of the provinces, autonomous
regions, and large municipalities with provincial status. A p
refecture
or a prefecture-level city under the jurisdiction of a province is an
administrative division between a province and a county. At the end
83
84
Until Debt Do Us Part
of 2009, China had 333 prefectures or prefecture-level cities. A county
or county-level city is the general administrative division at the local
level. At the end of 2009, there were 2,858 counties or county-level cities
in China. The township is the basic administrative division in the countryside. At the end of 2009, there were 40,858 townships in China.
According to the Chinese Constitution, the People’s Congress and
people’s government at the levels of province, prefecture (or city),
county, and township are the local legislative and executive organs of
power, respectively.
Law on Nine-Year Compulsory Education
The Law on Nine-Year Compulsory Education, enacted by the central
government on July 1, 1986, mandated requirements and deadlines
for attaining universal education (to be tailored to local conditions)
and guaranteeing school-age children the right to receive at least nine
years of education (six years of primary education and three years of
secondary education). People’s Congresses at various local levels were,
within certain guidelines and according to local conditions, to choose
the steps, methods, and deadlines for implementing nine-year compulsory education in accordance with the guidelines formulated by the
central authorities. The program sought to bring rural areas, which
had four to six years of compulsory schooling, into line with their
urban counterparts.
The Decision on the Reform of the Education System by the Central
Committee of the Party in 1985 divided jurisdictions into three categories: (a) cities and economically developed areas in coastal provinces and
a small number of developed areas in the hinterland (approximately
25 percent of China’s population); (b) towns and villages with an average level of development (approximately 50 percent of China’s population); and (c) economically backward areas (approximately 25 percent of
China’s population).
In fact, some jurisdictions in the first category had achieved universal nine-year education by 1985, and all jurisdictions in the category
had achieved universal nine-year education by 1990. The jurisdictions
in the second category had achieved universal nine-year education by
1995, and technical and higher education were projected to develop at
the same rate. The jurisdictions in the third category should attain the
Restructuring of Legacy Debt for Financing Rural Schools in China
85
basic education target for all students with a timetable consistent with
the pace of local economic development, and the central government
would support the educational development. The central government
also would assist educational development in minority nationality and
remote areas.
Currently, there are over 260,000 rural primary and junior secondary
schools with about 76 million students (see table 2.1).
Distribution of Educational Responsibilities among
Levels of Government
Fiscal decentralization reforms in China after the economic reform
started at the beginning of the 1980s provided local governments with
significant fiscal autonomy in various areas such as the determination
of their own spending priorities and policies on relevant aspects of local
budgets.
In 1994, the central government introduced the Tax Sharing System (TSS) reform, with the two major goals of increasing the share
of combined government revenue in total gross domestic product
and the central share in combined government revenue (World Bank
2002). This reform introduced clear and stable assignments of tax revenues between the central and provincial governments, and created
Table 2.1 Statistics on Rural Junior Secondary and Primary Schools and Students, 2009
Number of
schools (unit)
Junior
secondary
schools
Nine-year
primarysecondary
schools
Total
enrollment
(person)
30,178
22,921
7,257
19,345,061
28,590
22,240
6,350
18,498,085
Run by private institutions
1,222
557
665
646,478
Run by other departments
366
124
242
200,498
Item
Rural junior secondary education
Run by education departments
and collectives
Rural primary education
Run by education departments
and collectives
234,157
56,555,439
231,360
54,941,004
Run by private institutions
2,395
1,260,235
Run by other departments
402
354,200
Source: National Bureau of Statistics 2010.
86
Until Debt Do Us Part
separate tax administration services at both levels of government. The
TSS reform introduced the value-added tax as the major government
revenue source, and established a uniform tax-sharing system that
provided that the central government would receive 75 percent of the
value-added tax and subnational governments 25 percent.
The taxes assigned to the center and the major taxes shared with the
subnational governments were collected by the newly created National
Tax Services, which operated in all provinces. The new system provided for separate local (subnational) tax services for the collection of
the taxes assigned to local governments. The headquarters of the local
tax services, the State Administration of Taxation, was empowered to
supervise local tax services and prohibit the use of tax exemptions by
local governments.
Given the main focus of the TSS reform on the taxation side, however, there was no apparent change in either policy or practice in terms
of expenditure assignment between the central government and subnational government and among the four tiers of subprovincial governments. In fact, the TSS restated the prereform expenditure assignment
and provided only basic guidelines to define expenditure responsibilities between the central and subnational governments. For example, the
State Council Regulations on the Implementation of the TSS in 1994
defined expenditure responsibilities of central and subnational governments as follows:
Central budgets are mainly responsible for national security, international affairs, the costs of operating the central government, the needs
for adjusting the structure of the national economy, coordinating
regional development, adjusting and controlling the macro economy,
and others. Detail items include: national defense, cost of military
police, international affairs and foreign aid, administration costs of the
central government, centrally financed capital investments, the technical renovation of central-government-owned enterprises and new
product development costs, the expenditure to support agriculture, arts
and culture, education and health, price subsidies and others.
Subnational budgets are mainly responsible for the costs of running
subnational governments, and the need for local social economic development. Detail items include: the costs of running subnational government, the needs of local economic development, a part of the costs of
Restructuring of Legacy Debt for Financing Rural Schools in China
running the military police and militia, locally financed capital investments, the technical renovation of local government-owned enterprises
and new product development, the costs of support to agriculture,
urban maintenance and construction, and the expenditure to support
arts and culture, education and health, price subsidies and others.
These guidelines illustrate that both the central government and
subnational governments not only have extensive expenditure responsibilities, but also that these responsibilities overlap and are not specific. Adding to the lack of clarity in expenditure assignments is the fact
that the subprovincial governments do not have explicit expenditure
assignments. The expenditure assignments for subprovincial governments are basically at the discretion of their provincial governments. To
improve expenditure at the subprovincial government level, the M
inistry
of Finance in December 2002 issued “Suggestions on Improving Sub-
provincial Fiscal Relations,” to provide further g uidelines on subprovincial expenditure assignment. However, considerable challenges remain
in clarifying expenditure assignment.
Education, for example, is mainly the responsibility of subnational
governments. For compulsory education, the role of the central government is that of policy maker and overall planner. In addition, the central government has the responsibility for establishing special education
funds for subsidizing compulsory education in poor, minority areas,
and teacher education in all areas. The provincial government has the
overall responsibility for formulating the development plan for compulsory education and providing assistance to counties to help them meet
recurrent education expenditures. The responsibility for actually implementing compulsory education programs lies with the cities or districts
of large cities in the case of urban areas, and with counties in the case of
rural areas.
The provision of compulsory education services in rural areas is one
of the major concerns of the central government because of the generally worse service conditions, especially in poor rural areas. Some initiatives, especially the “Decision on Strengthening Rural Education,” issued
by the State Council in September 2003, expanded the expenditure
responsibilities of the central government on compulsory education.
This basic service was defined as a shared responsibility with the goal of
supporting students from poor families by waiving their payments for
87
88
Until Debt Do Us Part
textbooks, tuition, and miscellaneous fees, and by subsidizing housing
expenditures for elementary and secondary education students.
In general, several important decision-making powers were decentralized to county governments in implementing the State Council
2003 directive. For example, the county governments were able to close
schools or reduce their size in their jurisdictions. This decentralized
power eventually influenced the education sector in significant ways. In
addition, upper-level governments (the central and provincial governments) became more involved in financing education.
There were two driving forces behind the changes in the assignment of responsibilities for upper-level governments. First, as discussed
below, the reduction in revenues that resulted from the Tax-for-Fee
Reform rendered some local governments fiscally unable to finance
education (Xiang and Yuan 2008), so more transfer funds were required
to keep education services running smoothly. Second, reducing disparities in education expenditure was deemed to be a desirable public policy
goal. Therefore, more transfer funds should go to poorer areas.
County governments have become the most important—but not the
only—players in the current regime of education finance. Other levels
of government, such as the prefecture, province, and central government, also have roles to play; even township governments and village
self-governing bodies have some responsibilities in particular areas
(Huang 2006). The power to make education policy is still centralized in
the hands of provincial and central governments. Nevertheless, county
governments have been given considerable latitude for making decisions
on the daily operations of educational services, as shown in table 2.2.
Revenue Assignment of Rural Compulsory Education
Although there are various financing channels, compulsory education
in general is mainly financed by the government, particularly by the
government’s budgetary expenditure, as shown in table 2.3.
The finance system of compulsory education in China has evolved
since the early 1950s, in three major stages.
The first stage was 1950–93. During that period, primary and secondary education services were mainly provided by subprovincial governments. The central government provided financial support, mostly upon
Restructuring of Legacy Debt for Financing Rural Schools in China
89
Table 2.2 Assignment of Major Educational Responsibilities among Levels of
Government in China
Terms
Central
government
Provincial
governments
Prefecture
governments
County governments
Policy
1. Establishing
the
educational
system
2. Curriculum,
approval of
textbooks
3. Determining
teacher-pupil
ratios
1. Making policy
for education
development
2. Planning
for the
reorganization
of primary
and secondary
schools
3. Examining and
evaluating
schools
4. Approving size
of teaching
staffs
1. Coordinating
educational
planning
2. Implementating
educational
examination
and evaluation
1. Planning local
school system
structure
2. Paying teachers
3. Managing
principals and
teachers
4. Guidancing
teaching
activities
5. Evaluating rural
schools
6. Proposing
teacher-pupil
ratios
Financing
Transferring
funds to
help poor
and minority
areas
Providing support
to poor
counties
None
Purchasing
equipment and
books
Teacher
salaries
Earmarking
transfer funds
for teacher
salaries
Using transfer
funds to
support poor
counties
Providing transfer
funds to poor
areas
Paying teachers
Operational
costs
None
1. Determining
teacherstudent
ratios and
corresponding
operational
expenditures
for rural areas
2. Transferring
funds to help
poor areas
Helping county
governments
finance
operational
costs
Providing funds
for operation of
schools
Construction
costs
Earmarking
subsidies for
the repair of
dangerous
classrooms in
poor areas
Providing
subsidies
to county
governments to
finance school
facilities
Helping county
government
finance building
of school
facilities
Implementing plans
for building
schools
Source: Compiled by the authors.
90
Until Debt Do Us Part
Table 2.3 Financing Sources for Rural Junior Secondary and Primary Schools in China, 2008
RMB millions
Rural junior
secondary
schools
Percent
Rural primary
schools
Percent
Government
135,722.3
95.03
222,362.4
96.76
358,084.7
96.09
Budgetary
128,372.0
89.88
213,637.4
92.96
342,009.4
91.78
Financing sources
Total
Percent
Private school
210.4
0.15
210.7
0.09
421.1
Donations
823.0
0.58
1,233.2
0.54
2,056.2
0.55
4,412.2
3.09
4,058.6
1.77
8,470.8
2.27
School charge
Tuition
Other
Total
0.11
1,173.4
0.82
1,073.3
0.47
2,246.8
0.60
1,656.0
1.16
1,954.2
0.85
3,610.2
0.97
142,823.9
100.00
229,819.1
100.00
372,643.1
100.00
Source: National Bureau of Statistics 2010.
request of the provincial governments. Education services were treated
as local public goods, and education expenditure was financed mainly
by budgetary funds, education surcharges, donations to education, and
student fees. Minban teaching staff (nongovernmental employees) were
commonly used in rural schools as instructors, and school facilities were
financed mostly by local residents.
The second stage was 1994–2003. During that period, the education
sector continued to decentralize. The 1994 TSS reform had an important impact on education finance. As noted, the TSS reform had two
major goals: increasing the share of combined government revenue in
total gross domestic product, and increasing the central share in combined government revenue. Given that larger shares of fiscal resources
went to the upper-level governments, especially the central government, and that no changes were made in the assignment of expenditure
responsibilities, local governments, especially township governments,
began to experience increasing difficulties in providing educational services. During this period, off-budgetary resources, in particular school
fees and charges, played a bigger role in financing education than in the
first stage.
There were other policy shocks that added to the fiscal pressures facing local governments. For example, during 1996–97, many Minban
teachers became government employees, in accordance with explicit
Restructuring of Legacy Debt for Financing Rural Schools in China
policies set by the State Council. Local governments were already struggling financially, and the accompanying increases in salary payments
added to the burden. In addition, the Nine-Year Compulsory Education
plan required all schools in rural areas to comply with regulations concerning school facilities. To comply with these regulations, local governments incurred large increases in construction expenditures.
As shown in table 2.4, with the expanding expenditure per student
in rural primary schools in Henan province, salary payments increased
significantly, and construction was more concentrated in 1999 to meet
the requirement that schools implement the Nine-Year Compulsory
Education plan. Meanwhile, a 1999 survey by the Ministry of Finance
revealed that the minimum financial gap for the local government
to implement the Nine-Year Compulsory Education plan was RMB
35 billion (Hu 2002).
In short, during this period, the increases in expenditures and contracted budgetary revenues rendered some local governments unable to
provide enough resources to fulfill their educational responsibilities. Many
local governments suffered significant shortages in education funds, with
the following consequences: (a) teachers in many provinces could not
get paid on time—in fact, the total amount of unpaid teachers’ salaries
(Beijing, Shanghai, Tianjin, Zhejiang, and Tibet were not included) was
RMB 13.6 billion (Liao 2004); (b) students were charged high fees in order
to finance the regular operation of schools; and (c) local residents were
heavily taxed to finance the new school facilities (see, for example, Li 2006).
Table 2.4 Composition of Educational Expenditures of Rural Primary Schools, Province of
Henan, 1999 and 2002
RMB per student
Year
1999
2002
Change
Education expenditure
364.45
545.76
181.31
Teacher salaries
224.42
450.23
225.81
61%
82%
21%
97.16
81.59
–15.57
As share of total expenditure
Operational expenditure
As a share of total expenditure
Construction expenditure
As a share of total expenditure
Source: Henan Education Department 2003.
Note: Bolded text indicates expenditures.
27%
15%
–12%
42.87
13.94
–28.93
12%
3%
–9%
91
92
Until Debt Do Us Part
The third stage began in 2003 and continues to the present. To reduce
the tremendous financial burdens facing rural households, the central
government in 2001 initiated the Tax-for-Fee Reform. The reform canceled various charges and fees levied on rural households by rural local
governments, and improved rural tax structure and administration.4
While reducing the financial burdens on rural households, the reform
also eliminated the revenue base for township governments, since their
major revenue bases, in particular five agriculture-related taxes and
budgetary funds, were no longer available. The central government also
reassigned the responsibility for basic education services to the countylevel government, and increased the central contribution for financing
primary and secondary education.5
However, the newly assigned responsibility to the county governments was not accompanied by revenues, since the agriculture-related
taxes and other fees were also main sources of the county government
budgets. In response to these budgetary constraints, county governments first closed a number of primary and secondary schools, then
started to charge higher student fees to finance operational costs for
the schools that remained open. Under the new arrangement, teacher
salaries were increased and were more likely to be paid on time, while
the operational costs were financed jointly by budgetary funds and student fees. The financial arrangements for construction costs, however,
remained unsettled (Liao 2004).
The unevenness between revenue availability and expenditure
responsibilities since the 1994 TSS reform created serious challenges in
service provision at the local level (Jia 2008). Other central government
policies also contributed to the financial challenge faced by local governments. In particular, implementation of the Nine-Year Compulsory
Education plan required that all rural schools satisfy certain conditions,
including good physical facilities and a qualified teaching staff.6 While
this policy led to significant increases in local expenditure needs for
education and construction costs, in particular, the central government
made no new provision of funds for local governments (Zong 2010).
In response to the imbalances in their public finances, subprovincial
governments developed two strategies.
On the expenditure side, both the quality and quantity of some public goods provided by local governments decreased. For example, in
Restructuring of Legacy Debt for Financing Rural Schools in China
some areas, local governments failed to pay teachers on time. In other
areas, teachers were perhaps relatively luckier; they could receive their
paycheck on time, but with a condition, say, that their salaries would be
reduced by 20 percent. Failure to pay teachers was not an isolated occurrence; rather, it reportedly took place across the country, except in some
rich areas like Beijing and Shanghai (Zhang 2005).
On the revenue side, it appears to have been difficult for rural local
governments to increase revenue collection on their own, since the tax
bases and tax rates are largely controlled by the central government.
But local governments were able to collect additional funds to finance
education and other local public goods through a variety of fees and
charges, formally and informally. Given that no horizontal accountability mechanisms were in place, in the face of financial stress it can be reasonably expected that local governments would have increasingly and
consistently exerted their powers to obtain money from rural households (Liao 2004).
Serious pressures from this situation redounded to the central government in a variety of ways. To protest unpaid and delayed salaries,
teachers first loudly voiced their concerns through their representatives
in the People’s Congress at different tiers of government. In addition,
more teachers quit their jobs in rural schools or moved to schools in
richer areas for higher and more stable salaries (Cai 2002; Wang 2004:
Zhou, Liu, and Tian 2003). As a consequence, an adverse selection problem for teachers developed across rural schools; those teachers taking
jobs at or remaining in rural schools were in many cases not the teachers
those schools needed. The loss of qualified teachers reduced the quality
of educational services provided in rural areas and raised serious concerns among parents about the quality of education. These concerns
eventually filtered up to the central government. In particular, restructuring of legacy debt for financing rural schools needed the immediate
attention of the central government.
Fundamentally, county governments assumed the key role in budgeting education finance following the 2003 State Council directive.
County governments pool all revenues from own and transferred funds,
and decide on their use to finance education and to distribute funds to
all schools in all three categories in their jurisdictions. The roles played
by upper-level governments are mostly supportive, although in some
93
94
Until Debt Do Us Part
instances they contribute a significant share to financing general education expenditure. Higher-level governments pay special attention to
teacher salaries by earmarking funds, but do little in relation to other
categories of expenditures.7
Significant Features of Rural Compulsory Education Debt
In general, there is a mismatch between responsibility and revenue
assignment in China; in particular, the aggregated own revenue of
county and lower governments amounts to only 42 percent of their
expenditure, as shown in table 2.5. Although the central government
required subnational governments to provide rural compulsory education, the county governments, which are responsible for its implementation, and other tiers of subnational government, had limited resources
to do so, which forced them to borrow funds to finance the services
(Xiang and Yuan 2008). However, as mentioned, borrowing is not permitted under the Budget Law of China, and this further complicated the
debt issue and encouraged county governments to borrow off budget.
The result was that not only was debt concentrated in the county
governments, but there was also asymmetric information between the
county governments and the upper levels of government, especially
the central government, about the size of debt and its service cost. In
addition, the debtor-creditor relationship was informal, in general,
because the borrowing governments did not have the legal status to
borrow (Shi 2004). More important, most county governments were
not able to pay the debt by relying on own resources. These factors
increased the difficulty of solving the problem of the rural education
debt (Wang 2007).
Table 2.5 Own Revenues as a Percentage of Total Expenditures, by Level of
Government in China, 2003
Level of government (consolidated)
Average
Minimum
Provincial
53
8
75
Prefecture
55
2
85
County and lower
42
0
90
Source: Ministry of Finance.
Maximum
Restructuring of Legacy Debt for Financing Rural Schools in China
Table 2.6 presents the share of compulsory debt in subnational budgetary expenditures for 14 provinces at end-2005. Since most of the
debt is concentrated in county-level governments,8 the share of debt in
county-level expenditures would be much higher. The debt mainly consisted of arrears on funds owed to construction companies that built the
schools and to suppliers, and on funds for teacher wages and pensions.
A small part of the debt was incurred by local government off-budget
vehicles to construct schools, that is, funds borrowed from financial
institutions.
Restructuring the rural compulsory education debt became a priority of the central government in mid-2000. The main reasons include
(a) it was an important step toward improving rural human capital and
achieving equality between urban and rural education, (b) writing off
the rural compulsory debt was the prerequisite to building a sustainable
finance system for rural compulsory education, and (c) it was needed
Table 2.6 Compulsory Debt as a Percentage of Subnational Budgetary
Expenditure for 14 Regions in China, 2009
RMB billions
Provinces
Compulsory education debt
(by end-2005)
Inner Mongolia
3.92
Jilin
Helongjiang
Budgetary expenditure
Percent
192.684
2.03
2.31
147.921
1.56
2.434
187.774
1.30
Jiangsu
5.75
401.736
1.43
Anhui
3.079
214.192
1.44
Fujian
2.312
141.182
1.64
Jiangxi
4.094
156.237
2.62
Hubei
3.074
209.092
1.47
Hunan
4.803
221.044
2.17
Sichuan
9.092
359.072
2.53
Ningxia
0.901
43.236
2.08
Guizhou
3.994
137.227
2.91
Shan’xi
3.065
184.164
1.66
Guansu
2.333
124.628
1.87
Total
51.161
2,720.189
1.88
Source: Ministry of Finance.
95
96
Until Debt Do Us Part
to prevent increasing the fiscal burden on rural residents. In addition,
writing off rural compulsory education debt was relatively easy in terms
of size and complexity, and it could provide experience and knowledge
for restructuring other subnational debt (Zhang 2007).
Although the size of the aggregate debt is relatively small, in the view
of policy makers, how the debt was restructured would influence the
future behavior of subnational governments. In particular, improper
incentives could lead to moral hazard for local governments, which
could lead to more problems in the future when trying to solve debt
problems.
Key Strategy for Dealing with
Moral Hazard in Debt Restructuring9
The RMB 110 billion rural compulsory education debt accounted for an
insignificant portion of the RMB 10.7 trillion in subnational liabilities
at the end of 2010.10 However, the restructuring of the rural education
debt represented the first effort to restructure subnational debt; thus, its
design and approach would affect subsequent debt restructuring efforts.
In particular, an improperly designed framework could create negative
incentives for subnational governments concerning their future borrowing decisions.
A key issue in writing off the rural compulsory education debt concerns moral hazard. The soft budget constraint challenge exists in the
fiscal system of many countries (see, for example, Liu and Webb 2011).
An improperly designed fiscal system encourages moral hazard on the
part of local government and creates the soft budget constraint. Not
surprisingly, the soft budget constraint issue has also been a challenge
for China’s fiscal system. To some extent, the compulsory education debt
resulted from the soft budget constraint and the weakness of China’s
intergovernmental fiscal relations.
Consequently, how to deal with a potential moral hazard challenge
was a serious concern to policy makers in designing the restructuring
package of the compulsory education debt. The restructuring of this
debt served as a pilot from which to draw lessons. Finally, restructuring the rural compulsory education debt in China took place within the
framework of intergovernmental fiscal relations. The process of the debt
Restructuring of Legacy Debt for Financing Rural Schools in China
restructuring would help clarify intergovernmental fiscal relations in
delivering basic education and help inform future reforms to the intergovernmental fiscal system.
There might have been other options to resolve the rural legacy debt.
One would have the central government write off the entire debt. This
option was not chosen because it would have encouraged moral hazard.
Another option would have allowed provincial governments to resolve
the debt of their local governments. This option was not chosen because
the central government viewed the rural debt restructuring as an opportunity to realign intergovernmental fiscal relations with respect to rural
education; letting provincial governments resolve the problem was not
feasible given the existing intergovernmental fiscal relations. The chosen mechanism was based on the principles of burden sharing, transparency, and formula-based restructuring. As mentioned, the rural debt
is only a small portion of subnational government debt, and there are
broader issues concerning moral hazard and opportunistic behavior of
subnational governments. The experience of rural debt restructuring
can offer lessons on addressing these broader issues.
In restructuring the compulsory education debt, there were two types
of moral hazard. The first relates to the overall borrowing size of local
governments. As the debtors, the county governments might have an
incentive to expand the size of total rural compulsory education debt in
anticipation of seeking more central government grants to replace the
incurred debt. The second type of moral hazard concerns the behavior
of county governments in anticipation of a write-off. Instead of using
their own revenues to contribute to the write-off of debt, the county
governments might have an incentive to seek more bailout grants to
write off the existing debt.
To deal with these two types of moral hazard, the central government
designed a strategy with two distinct features.
First, the fiscal resources required for debt write-off and restructuring are distributed among three tiers of government: roughly one-third
from the central government, one-third from provincial governments,
and one-third from lower-tier governments. Thus, the fiscal burden of
debt restructuring is shared.
Second, the distribution of the central grants was based on an
output-based rather than an input-based formula. The output includes
97
98
Until Debt Do Us Part
two set of factors. The first set formed the base on which to calculate
the total grants to a jurisdiction, and the second set formed the base
to determine the performance of a local government in writing off the
rural compulsory education debt.
Regarding the first set of factors, although the overall size of the central government grant was determined by the overall size of the compulsory education debt,11 the grant going to each individual jurisdiction
was not directly linked to the size of the debt of that jurisdiction. The
total grants were only a pool that provided the grants’ source; the linkage between the pool and its distribution among local governments
went through the following two steps.
First, as mentioned, the grants would not be directly related to the
total debt of a jurisdiction; that is, the central government grants would
not be tied to actual indebtedness of county or town governments.
Second, the distribution of central government grants to local governments for write-offs was determined by a formula that considered
four factors within local government jurisdiction: (a) number of students, (b) number of schools, (c) population density, and (d) local fiscal
capacity.
By this method, a local government that borrowed excessively would
not gain extra advantage, and another local government that borrowed
less or paid off its debt would not be in an unfavorable position. By
choosing objective factors and giving full consideration to the financial
difficulties of all pilot areas, the grants to all pilot areas would be calculated uniformly according to the formula.12 The areas having more
financial difficulty would enjoy a higher proportion of subsidy. Counties and towns that had not incurred debt but needed to cover a certain number of schools and students could obtain funds as a positive
incentive. In addition, there was a penalty rule to prohibit new debt
for compulsory education. The grants to a local government would be
reduced by the central government if new rural compulsory education
debt emerged. Provincial government grants were required by the central government to follow the same principles.
The second set of factors formed the base to determine the performance of a local government in writing off the rural compulsory education debt. To encourage subnational government effort in writing off
debt, the central government established the incentive by providing
Restructuring of Legacy Debt for Financing Rural Schools in China
a subsidy for those that had made efforts to write off the compulsory
education debt, and did not provide a subsidy to those that had not
completed the work within the stipulated time limit. That is, central
government grants to a province were based on performance, and only
after the debt write-offs of lower-level governments could a province
receive funds from the central government.13 The requirements, formula, and matching methods for writing off debt were included to preclude possible rent-seeking behavior.
All pilot provinces and their local governments were encouraged to
mobilize revenues to contribute their share for writing off compulsory
education debt. Subnational governments were encouraged to raise
revenues through, for example, improving collection efficiency of local
budgeted revenue, expenditure efficiencies, revitalizing idle school facilities, and mobilizing donations.
Restructuring of Legacy Debt for
Financing Rural Schools, in Practice
Based on the above strategy, the central government launched the project to restructure the rural compulsory education debt. In December
2007, the General Office of the State Council transmitted the Advice
Notice on Pilots Working on Resolving the Debt for Rural Nine-Year
Compulsory Education, which was prepared by the Working Group
under the State Council on the comprehensive reform in rural areas.14
The preparation work included the classification and audit of existing
subnational rural school legacy debt.
Implementation of the project followed the existing framework
of intergovernmental fiscal relations. Under this framework, rural
compulsory education was planned by provincial governments and
implemented by their county governments. Thus, the pilot project
for writing off rural compulsory debt was organized by provincial
governments and implemented by county governments. According
to the State Council’s Advice Notice, all pilot provinces must have
refrained from incurring new debt to finance compulsory education.
All subnational governments were required to adjust their financial
expenditure structure to establish reliable revenue sources for servicing new debt.
99
100
Until Debt Do Us Part
Of the total rural school debt of RMB 110 billion outstanding at
the end of 2007, RMB 80 billon had been borrowed to finance capital
investments (for school construction). An additional RMB 30 billon
was used to finance operational deficits. The implementation strategy
was designed to write off RMB 80 billion of debt for financing school
construction, with two steps.
The first step was to write off the compulsory education debt of 14
provinces15 within two years (that is, by the end of 2009). During this
step, all non-pilot-project provinces would choose two or three counties
(cities, districts) to pilot the writing off of such debt. After progress was
achieved in the 14 provinces and the pilot cities and counties, the second step was to extend the restructuring exercise to all other provinces,
and to complete the write-off of the debt in all non-pilot provinces by
the end of 2010.
While the originally planned central government contribution
toward writing off the RMB 80 billion compulsory education debt had
been projected to be RMB 26.67 billion, the final central government
contribution was RMB 30 billion, slightly higher than the originally
planned one-third, and the subnational government contribution was
RMB 50 billion. By the end of 2009, the central government provided
RMB 14.5 billion in grants, about 90 percent of the total funds, for writing off the debt of 14 pilot provinces and Chongqing, consistent with
the schedule of debt write-off.
In 2009, the central government launched a project of compulsory
debt write-off in non-pilot areas. By the end of 2009, RMB 3.58 billion
in grants had been provided to the non-pilot provinces, and in 2010, the
remaining RMB 11 billion in grants was transferred to a central government account for distribution to the non-pilot provinces. By the end of
2011, the funds were almost completely disbursed.
Given the success of writing off rural compulsory education debt
relating to capital financing in the pilot areas, the central government in
2009 initiated a new program to write off debt that was used to finance
operational deficits in the pilot areas. The same output-based formula
was used. The total pool of the central grants for writing off operational
debt for compulsory education was 37.5 percent of total operational
debt related to rural compulsory education, the same percentage as that
for the capital debt.
Restructuring of Legacy Debt for Financing Rural Schools in China
The first 14 pilot provinces completed the task of writing off debt
in 2009, which benefited about 1.7 million rural creditors. The 17 nonpilot provinces actively prepared for the write-off and started writing
off debt in 2009. By the end of 2009, 31 provincial governments had
paid RMB 56.6 billion of the rural compulsory education debt, among
which the central government provided RMB 14.5 billion in grants, and
subnational governments, including provincial governments, financed
RMB 42.1 billion. By the end of 2011, almost all rural compulsory education debt had been restructured, including a RMB 30 billion operational deficit. Since 2007, the central government has contributed RMB
30 billion toward writing off rural compulsory education debt, or
37.5 percent of the debt.
To ensure implementation of the overall strategy, the central government has established a management system to supervise implementation. In principle, all central government grants were designed to
contribute to the write-off of the compulsory education debt. Meanwhile, for an individual local government, the remaining grants from
the central government after writing off compulsory education debt
could be arranged to write off other local non-education-related debt,
with priority given to the rural-education-related debt.
Subnational governments are required to establish a compulsory education debt control system to monitor grants from the provincial and
central governments and the progress of writing off debt to make sure
the grants are used effectively and follow central government requirements. In addition, the departments of finance of provincial, municipal,
and county governments are required to report the use of the grants in
their budget and final financial reports to the relevant People’s Congress
or Standing Committee. The provincial finance department must report
the progress of the write-off of compulsory education debt and the
usage of grants to the central government on a monthly base. The penalty may be imposed rule if new debt were contracted. In addition, the
Ministry of Finance retained RMB 2 billion to deal with contingencies.
It is too early to assess the impact of writing off the rural education debt, since the entire exercise was completed only at the end 2011.
Several questions will necessarily arise, including: Do local g overnments
continue to borrow to finance rural education? Given the transition from debt to grant financing of rural compulsory education, has
101
102
Until Debt Do Us Part
the grant allocation system been sufficient to finance compulsory
education?
Although China achieved universal compulsory education in 2007,
a key question is how to ensure the sustainability and quality of education. Evaluating the debt write-off and addressing these questions will
need to be done in the context of the evolving changes in the intergovernmental fiscal system. Chinese reform of its intergovernmental fiscal
system is ongoing, as are major discussions on the assignment of expenditure functions among the tiers of governments; the streamlining of
the tiers, potentially into three tiers; and the need to continue to reform
the intergovernmental revenue system to grant subnational governments revenue flexibility at the margin, which is critical to underpin
their access to financial markets.
Conclusion
International experience has shown that it is difficult to undertake debt
restructuring and to write off debt liabilities while managing moral hazard. In a unitary system of government, the design mechanism by the
central government has an important bearing on the incentive signals
to subnational governments and financial markets. Important lessons
can be drawn from China’s experience of formulating its strategy and
framework for dealing with the write-off of rural compulsory education
debt.
It is important to have proper intergovernmental fiscal relations to
assure the delivery of basic public services. Besides the proper assignment of expenditure and revenue among the tiers of government, it is
necessary to provide formal fiscal instruments for subnational governments to manage capital expenditure. One of the main reasons for the
existence of China’s subnational government compulsory education
debt was the lack of fiscal resources and fiscal instruments, such as formal debt financing.
A suitable subnational debt management framework is a critical
part of intergovernmental fiscal relations. In designing a sound subnational debt management system, it is important to have an information system and an accounting and statistics reporting system to ensure
the risks of subnational debt are transparent and reported. In addition,
Restructuring of Legacy Debt for Financing Rural Schools in China
there needs to be a functioning audit department. More important,
subnational governments in general have the incentive to overborrow
because of the soft budget constraints and the common pool problem. Thus, it is important to provide proper incentives to avoid moral
hazard.
Writing off subnational debt should proceed with the goal of
improving intergovernmental fiscal relations. In China, county governments should be responsible for servicing their debt, which was consistent with the responsibility assignment of the existing fiscal system.
However, the county governments may lack incentives to pay their debt
without the effective incentive provided by the central government.
Although the central government has sufficient fiscal resources to write
off the entire rural education debt, such a write off would lead to moral
hazard problems.
China established a system in which the distribution of grants was
based on a transparent, rule-based, and output-based formula; the distribution of the grants to a particular local government was not related
to the size of the debt of that local government. If a grant to an individual jurisdiction were linked to the size of its debt, it could undermine
efforts by that local government to prudently manage its debt service,
and at the same time could encourage accumulation of additional debt
in anticipation of larger bailouts. The system, based on the performance
efforts and standard factors of local governments, such as the size of
the student population and their respective fiscal capacity, encouraged
local governments to achieve the goal while managing moral hazard
problems.
Each debt restructuring will need to examine the origin of the debt
problem and the specific historical and institutional context of the legacy debt. Any debt restructuring will need to pay close attention to the
design and its incentive effects.
The experience and lessons learned from China’s rural debt restructuring can help generate lessons for developing a consistent strategy for
debt restructuring in general. A rule-based debt restructuring reduces
ad-hoc bargaining and adverse incentives, a hard budget constraint prevents moral hazard, and burden sharing provides proper incentives and
avoids free-riding behavior, while also recognizing the incentive role
played by higher levels of government to leverage reform.
103
104
Until Debt Do Us Part
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. The exchange rate at the time of writing of the RMB (renminbi) to the U.S. dollar was US$1 to RMB 6.30. The RMB has appreciated continuously since 2005,
at about 3–5 percent per year.
2. In 2001, a package of policies called the Tax-for-Fee Reform was enacted. See
the next section for details.
3. Based on Article 30 of the Constitution of China of 2004, the administrative division of China is as follows: (a) the country is divided into provinces,
autonomous regions, and municipalities directly under the central government; (b) provinces and autonomous regions are divided into autonomous
prefectures, counties, autonomous counties, and cities; and (c) counties and
autonomous counties are divided into townships, nationality townships, and
towns. Municipalities directly under the central government and other large
cities are divided into districts and counties. Autonomous prefectures are
divided into counties, autonomous counties, and cities.
4. The reform started with a pilot program in Anhui province and later was
implemented nationwide. Before the reform, there were five agriculture-related
taxes, including various taxes on agriculture and a slaughter tax, and various
charges such as an education surcharge. There were three core components of
the 2001 Tax-for-Fee Reform: (a) cancelations of charges and fees imposed on
rural households, including the cancelation of township general fees, the educational surcharge, and the slaughter tax; (b) two adjustments: (i) changes in
the agriculture tax made the amount of taxable land and the tax rate fixed,
and the production level should be determined by the average of the past five
years; the maximum tax rate was set at 7 percent; and (ii) changes in the special
agriculture tax. This tax is levied by the procedure applied to the agriculture
tax but with a higher tax rate; and (c) one reform: the revenue the village can
collect is essentially a surcharge on the agriculture tax, the maximum rate of
which is 20 percent. This surcharge can be used to pay for three items in village
expenditures: (i) salaries of the leaders in the village self-governing committee,
(ii) the old-age support program, and (iii) the operational costs of the governing body. After this reform, only the agriculture tax remained. In 2006, the
agriculture tax was also abolished.
5. In 2001, the State Council issued a new policy, the “Decision on the Reform
and Development of Primary Education,” which assigned the responsibility of
providing educational services to county governments. In May 2002, the State
Council issued a complementary policy change for education. In the “Notice on
Improvement of Administration of Compulsory Education in Rural Areas,” the
Restructuring of Legacy Debt for Financing Rural Schools in China
responsibility for financing compulsory education was removed from township
to county governments. The two regulations ended the township-centered system
of the previous 17 years (1985–2002), and the county governments began to play
a core role in providing educational services.
6. In 1993, the central government issued a Blueprint on Education Reform and
Development, which stated that “two basics” should be reached before 2000. In
the following years, the Ministry of Education issued several regulations on the
qualifications of teachers and the status of school facilities. To meet the requirements, township governments, administrative villages, and households made
a big effort, but sizable debt was accumulated, since outlay expenditures were
beyond the ability of local communities.
7. For example, 1,424 counties out of a total of 2,806 received earmarked transfer
funds for teacher salaries in 2003.
8. Following implementation of the State Council Directive in 2003, township
and village government debt was transferred to the accounts of their respective
county governments.
9. This and subsequent sections are based on discussions with Ministry of Finance
officials.
10. “Auditing Report 2010,” No. 35, National Audit Office of China.
11. The deadline for calculating the size of debt was December 31, 2005.
12. The grants for writing off the compulsory education debt for pilot areas were
based on a standard compulsory education input gap (standard compulsory
education debt) and the coefficient of grants. The formula was: Grants for a particular pilot area = the standard input gap (of the area) × the coefficient of grants
(of the area), where the standard input gap (of the area) = ∑[(0.85 × the number
of students in rural compulsory education [county-level jurisdiction where the
rural population stands for no less 60 percent of total population] × the standard
input gap per student) + (0.15 × the number of the schools × the standard input
gap per school × the unit cost difference)]. The standard input gap per student
and the standard input gap per school are calculated, respectively, in accordance
with the number of students and population density by provincial jurisdictions
based on the overall input gap. The unit cost difference is calculated by county
based on the factors related to construction cost, such as the geographic elevation
and weather conditions. The coefficient of grants is determined by fiscal capacity.
The coefficient grants for the middle and western regions follow the coefficient
of fiscal difficulty applied in the general transfers.
13. The annual grants for a pilot area were based on the schedule of writing off
compulsory education debt for that area and used the following formula: The
amount of grants (in year T) = the amount of debt that had been written off
(in year T) / the total amount of the compulsory education debt that should be
written off × the total grants for the particular pilot area. The grants were partially advanced at the beginning of the year based on the anticipated write-off
during the year, and the remaining portion was given at the end of every year
105
106
Until Debt Do Us Part
after completion on schedule. The funds provided by the central and subnational governments earmarked for writing off the debt were managed in the
single account system of the National Treasury and monitored regularly.
14. The General Office of the State Council, 2007, No.70.
15.
The 14 provincial jurisdictions were Anhui, Fujian, Guansu, Helongjiang,
Hubei, Huizhou, Hunan, Inner Mongolia, Jiangsu, Jiangxi, Jilin, Ningxia,
Shan’xi, and Sichuan.
Bibliography
Cai, Yue’e. 2002. Why Did Rural Teachers Quit Their Jobs? A Case Study on the Prefecture of Hengyang, Hunan Province. Statistics Bureau of Hunan Province,
Hunan. http://www.hntj.gov.cn/csdc/news/ztdc/2002917171354.htm.
Chang, Gene H. 2003. “Urbanization and Unemployment in China.” In Urbanization and Social Welfare in China, ed. Aimin Chen, Gordon G. Liu, and Kevin H.
Zhang, 96–105. Aldershot, England: Ashgate Publishing.
Henan Education Department. 2002. An Analysis of Education Finance of Henan
Province. Henan. http://www.jyck.gov.cn/system/info/showinfo.asp.
Hu, Yifan. 2002. “Difficult Issues in Rural Compulsory Education.” Caijin (11):
44–51.
Huang, Shixian. 2006. “Exploration of Function of Village and Township Governments after Reform of Tax and Fees.” Academic Journal of Changbai (4):
27–30.
Jia, Kang. 2008. “Build a Central-Local System with Matched Function and Finance.”
China Public Finance 13: 1–3.
Li, Qun. 2006. “Thinking about Funding for Compulsory Education in Rural Areas.”
Journal of Education Exploration 2: 44–45.
Liao, Hongfeng. 2004. “An Analysis of the Education Expenditure after Tax for Fee
Reform.” Journal of Guangxi School of Public Finance 2: 43–48.
Liu, Lili, and Steven B. Webb. 2011. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper Series
5587, World Bank, Washington, DC. http://ssrn.com/abstract=1792244.
National Bureau of Statistics. 2010. China Statistical Yearbook 2010. Beijing: China
Statistics Press.
Qiao, Baoyun. 2011. The Subnational Financial Platforms of China. Beijing: China
Fiscal and Economic Press.
Shi, Hongxiu. 2004. “The Decentralization of Lending and Borrowing Cannot Deal
with Problem Assessment: The Way Out of Local Government Debt.” Journal of
the Banker 5: 68–71.
Wang, Wenhong. 2007. “A Study of Transfer Payment Mode of Public Finance in
Rural Compulsory Education—Based on the Study of the Country-Level
Finance from the Perspective of Public Finance.” Journal of Dalian University
28 (1): 123–28.
Restructuring of Legacy Debt for Financing Rural Schools in China
Wang, Youwen. 2004. “Paying Attention to Loss of Teaching Staffs in Rural Areas.”
China Education Daily, November 26. http://www.jyb.cn/gb/2004/11/26/zy/
jryw/12.htm
World Bank. 2002. “China—National Development and Sub-national Finance:
A Review of Provincial Expenditures.” Public Expenditure Review 22951,
Washington, DC.
Wu, Harry Xiaoying. 1994. “Rural to Urban Migration in the People’s Republic of
China.” China Quarterly 139 (September): 669–98.
Xiang, Jiquan, and Fangcheng Yuan. 2008. “The Effect of Reducing Revenues on
Compulsory Education in Rural Areas—Current Financial Dilemma and
Policy Choice of Compulsory Education in Rural Areas, February 23.” China
Agricultural Economy and Information Net. http://www.caein. com/index.asp
?xAction=xReadNews&NewsID=31785.
Yuan, Guilin. 2004. “Research on the Current Situation and Multi-Development
Model of the ‘County-Level’ Administration Mechanism for Compulsory Education in Rural Areas.” Journal of Northeast Normal University (Philosophy and
Social Sciences) 1 (1): 115–22.
Zhang, Shaochun. 2007. Speech at the Executive Meeting of Writing-Off Rural
Compulsory Education Debt, Chongqing, June 5. http://www.mof.gov.cn/pub/
nczhggbgs/zhengwuxinxi/lingdaojianghua/200806/t20080616_45565.html.
Zhang, Shouxiang. 2005. “The Compulsory Education System in Rural China: Progress, Problems and Solutions.” Basic Education Review 1: 4–8.
Zhang, Zhuogong. 2004. “The Outstanding Issues and Policy Options in Dealing
with the ‘County-Level’ Administration Mechanism for Compulsory Education in Rural Areas.” China Agriculture Education 1: 4–6.
Zhou, Dongai, Ziling Liu, and Jianjun Tian. 2003. “Why Do Rural Teachers Quit?” People’s Daily, April 6. http://www.people.com.cn/GB/kejiao/40/20030406/964244
.html.
Zong, Xiaohua. 2010. “Empirical Test of Fiscal Decentralization, Population Movement and the Investment of the Local Government in Compulsory Education.”
Journal of Statistics and Decision 15: 86–88.
107
3
Managing State Debt and
Ensuring Solvency: The Indian
Experience
C. Rangarajan and Abha Prasad
Introduction
There has not been any repayment default among the Indian states,1
although fiscal stress and debt repayment pressures were experienced by
many states in the late 1990s with continued deterioration evidenced in
the early 2000s. The deterioration in the current account was the driving force for declining fiscal health as reflected by the worsening of fiscal
and primary balances. An analysis of the evolution of states’ debt, deficit, and interest payments reveals three distinct phases.
The first, pre-1998 phase, was characterized by low current account
(revenue balance)2 and fiscal deficits, with moderate debt levels. The
second phase, during the late 1990s to mid-2000s, reflected significant
deterioration in all key deficit indicators, with rising debt levels and
interest burden. During this period, the outstanding states’ debt to gross
domestic product (GDP) peaked at 32.8 percent in 2003–04, up from
20 percent in 1997–98, and interest payments as a share of revenue
receipts increased from 16.9 to 26 percent over the same period.
Of concern was the fiscal stress experienced by the central government over the same period (Pinto and Zahir 2004), during which the
combined center-state fiscal deficit rose from 7.3 percent of GDP to
109
110
Until Debt Do Us Part
9.4 percent. Furthermore, this reflected only the direct liabilities of
states; exacerbating the debt burden and repayment pressure were the
contingent liabilities, in the form of guarantees issued by states to support their enterprises. This was followed by the third phase and the
onset of fiscal correction and reforms from the mid-2000s onward.
This is reflected in the lowering of all key deficit indicators and debt
and interest payments as a share of GDP.
The resulting reform package had three interrelated components.
First, to reverse the fiscal decline, a fiscal adjustment package was formulated to control the growth of current expenditures (such as wages and
pension), and structural reform of the taxation system (such as moving
from a turnover tax to a value-added tax) was instituted. S econd, a rulebased institutional framework was developed to ensure the sustainability of the adjustment and consolidation. Third, there was a move from
central government onlending to states toward market-based financing, with a focus on both self-regulation (through fiscal legislation) and
market discipline.
The priority of fiscal consolidation was to restore the balance of
revenue accounts, that is, to reduce the revenue deficit to zero. It was
realized that even after lowering the primary deficit, the debt service
repayment pressure and high indebtedness would continue, because
about 80 percent of states’ borrowings during 2003–04 was at high-cost,
non-market rates. But turning states to a sustainable fiscal path implied
reducing both the stock of debt and the cost of borrowing. However,
debt restructuring, write-offs, and relief would have an inherent moral
hazard challenge. Being cognizant of this, the debt restructuring program was linked with broader institutional reforms, including providing incentives for states to undertake difficult fiscal reforms.
The Twelfth and Thirteenth Finance Commissions (FCs)3 comprehensively examined the situation for both the center and the states, highlighted their interdependence, and presented an overall strategy. Most
subsequent reforms in terms of fiscal responsibility legislation (FRL) were
based on a well-considered strategy and incentive structure. Although
the steps taken were gradual, the synergistic effect of many institutional,
fiscal, and legislative reforms was much greater. The reform efforts were
initiated and implemented by different parts of the government—the
FC, the Ministry of Finance (central g overnment), the states themselves,
Managing State Debt and Ensuring Solvency: The Indian Experience
the Planning Commission, and the Reserve Bank of India (RBI). All parties were aware that this was not “business as usual” (World Bank 2005),
and there was a sense of urgency about transforming the situation.
Among the states themselves, there was a move away from competitive populism (Kurien 1999), which included subsidies and lowering tariffs, toward coordination by ending the competitive tax rate reduction
and instituting the value-added tax, which proved to be highly buoyant.
This, coupled with increases in the states’ share of central taxes instituted
by the Twelfth and Thirteenth FCs and the high buoyancy of the center’s
direct taxes, improved state finances and led to their progress towards the
FRL goals. The coordination and consultation among all engaged entities ensured consistency of approach and moved the reforms forward.
There is ample fiscal literature on the states’ fiscal reform—fiscal
rules, the quality of fiscal adjustment, expenditure and taxation reforms,
power sector reform, and budget and financial management reforms.4
This chapter focuses on the states’ borrowing and debt restructuring process, underpinned by the move toward a rule-based framework
and market discipline.5 It concentrates on the perspective of the policy
maker during this period and reflects on the key challenge that was to
balance the provisions of debt relief with the need to avoid moral hazard and enforce fiscal discipline.
The rest of the chapter is organized as follows. Section two presents
the states’ borrowing framework as prescribed by the constitution, and
changes in the borrowing channels and lending policy for states, while
incentivizing market access with a rule-based system. Section three
summarizes the trends in states’ deficits, debts, and interest payments in
the last two decades, and highlights interstate disparities in fiscal performance. Section four presents the major policy and institutional reforms
undertaken to restructure states’ debt and discusses efforts to minimize
moral hazard. Section five presents the impact and challenges of the
ongoing global financial crisis. Section six offers conclusions.
States’ Borrowing Regime
India is a federal polity of 28 state governments and 7 union territories.
The states’ borrowing regime is defined by federalism, characterized
through the constitutional division of powers among the three levels
111
112
Until Debt Do Us Part
of government—the center, the states, and the local bodies.6 The power
to raise major taxes is allocated to the central government, while major
expenditure responsibilities are assigned to states due to their proximity to local issues and needs. While states’ own revenues constitute
37 percent of total revenue receipts, their expenditures account for
55 percent of total central government expenditure (RBI 2011a). The
imbalance is addressed through fiscal transfers from the center to the
states, mandated by the FC.
The constitutional arrangements for revenue sharing among the
Indian federation and the consultative mechanism among the c enter and
states have tended to reduce the risk of explicit state defaults.7 Regarding
the constitutional arrangement, the FC uses a formula-based approach
to allocate taxes and grants, with the objective of filling the expenditurerevenue gap (deficit financing). The vertical sharing between the center
and states is simplified by including all central taxes and excise duties
in the divisible pool of central taxes.8 For the horizontal sharing among
states, the FCs have attempted to correct the differentials in revenue
capacity and cost factors inherent in the diversity of states. The pattern
of transfers through the FC channel shows that the share in central taxes
has persistently been the predominant component of revenue sharing
since the First FC (RBI 2011a). Starting with the Ninth FC, a greater
emphasis on fiscal discipline has been added to balance the gap-filling
approach (RBI 2007).9
Residual imbalances in the fiscal accounts after the federal transfers are financed through borrowing. The main borrowing sources
are domestic, external, and issuance of loan guarantees. The borrowing channels are multiple and the process complex but are organized
around the principles of maintaining sustainability, solvency, and liquidity of the states (for a description, see box 3.1).10 The overall control is
with the center, under Article 293(3) of the Constitution, which states
that if any state government is indebted to the center, it requires the
center’s permission to borrow. Further, the Constitution forbids states
from borrowing abroad on their own. Thus, all external borrowing
must be onlent or guaranteed by the center.11
The limit on the annual amount and sources of borrowing is based
on consultations among the center, the state government, the Planning
Commission, and the RBI.12 Previously, after the delinking of plan13
Managing State Debt and Ensuring Solvency: The Indian Experience
113
Box 3.1 State Borrowings
Borrowing channels for states are multiple and the process complex; some channels are controlled
and restricted by the center and others are more autonomous.
Borrowing channels controlled by the center are the following:
• Market borrowings. Market borrowings are controlled by the center and managed by the RBI.a
The state securities issued through this channel are eligible for meeting the banks’ statutory
liquidity requirements and are thus backed by “automatic” intercepts from the state treasury
account (automatic debit). There have been no restructuring or defaults associated with these,
and investors perceive an implicit sovereign guarantee attached to them.
• Loans from the center. Historically, the center used to borrow and onlend to states. This has
now changed, with financial market developments and states’ ability to borrow on their own
behalf, onlending from the center was discontinued in May 2005.
• Loans from banks and financial institutions. The center sets the global ceiling on the amount
states can borrow from the banks and financial institutions, but the rate of interest is negotiated
directly by the state with the concerned creditor. The rate of interest depends on the perceived
credibility and fiscal position of the state.
• External loans. Previously, the center would onlend the proceeds in rupees at harder terms,
adjusting exchange exposure and elongating maturities. With the recent change in lending policy,
the entire loan proceeds are passed through directly by the center to states at the same terms
(currency, maturity, and amortization) given by the creditor. The states bear the currency and the
refinancing risk, but most do not undertake an impact evaluation of the cost-risk trade-offs of
such transactions on their total debt portfolios (see table B3.1.1).
Table B3.1.1 Sources and Features Attached to State Borrowings
Amount controlled
by the center
Automatic
intercepts
Creditor
perceives guarantee
External loans
Yes
Yes
Yes
Loans from center
Yes
Yes
—
Market borrowings
Yes
Yes
Yes
Loans from bank and
financial institutions
Yes
No
Partially
Provident funds
No
No
No
NSSF
No
No
No
Contingent liabilities
No
No
Partially
Note: NSSF = National Small Savings Fund. — = not applicable.
Borrowing channels not controlled by the center are the following:
• Small savings loans and use of state provident funds.b Prior permission from the center is not
required for these. The small savings schemes are run by the center with a social security objective to encourage household savings. Eighty percent of the collections within a state’s territory
are automatically passed on by the National Small Savings Fund to that state. The rate of interest
(continued next page)
114
Until Debt Do Us Part
Box 3.1 (continued)
paid by states is currently fixed at 9.5 percent. The money is available for 25 years with a five-year
grace period.
• Special purpose vehicles. States issue loan guarantees to special purpose vehicles, which borrow
in the market with the backing of these guarantees. Anecdotal evidence suggests that loan proceeds have been sometimes used to finance current state expenditures.
Liquidity management:
• Ways and means advances (WMA) from the RBI. These are designed to meet temporary liquidity shortfalls. They are formula based and depend on the state’s total expenditures. If the shortfall is higher than the WMA amount, the state gets into overdraft, which is extended at a penal
rate of interest to be cleared within 10 days or the account of the state is frozen. If there is surplus cash in the single treasury account, it is invested in 14-day intermediate treasury bills.
In the interest of transparency, the number of days a state uses the facility during a fiscal year
is published in the RBI’s Annual Report. Access to this short-term credit facility disciplines states to
manage liquidity shortfalls prudently to avoid closure of accounts, and benefits them in avoiding
arrears and payment defaults.
a. The RBI manages domestic borrowings for each of the 28 states through separate agreements
with each.
b. A provident fund is a retirement benefit scheme; employees contribute 12 percent of
monthly wages and can withdraw funds on retirement or on reaching age 55. Contributions are an
unfunded liability in the public account, but balances are available to the state (see Rao, Prasad, and
Gupta 2001).
borrowing and plan grants, states tended to revise their objective of
maximizing plan assistance by arguing for higher plan sizes, thereby
committing to higher borrowing. This changed considerably with the
enactment of FRL targets. Key decision parameters on the demand side
include the states’ financing needs, developmental needs, repayment
profile, and, since the early 2000s, its debt sustainability. On the supply
side, an important factor is the absorption of liquidity from the market
by both the center and states, without impinging on the supply of credit
for the private sector for productive purposes.14 Since the mid-2000s,
the ceiling on borrowing by a state has been capped by the fiscal targets
under the state-level FRLs.15
Figure 3.1 illustrates the changing financing pattern of states’ debt
during the 1990s and 2000s. This mirrors three phases, with a decline
in the center’s loan intermediation and onlending (since 1998–99), an
increase in the National Small Savings Fund (NSSF)16 and small savings
borrowings (1999–2000), and the move toward market-based financing
since mid-2000.
Managing State Debt and Ensuring Solvency: The Indian Experience
Figure 3.1 Composition of Financing Pattern of State Deficits (as of End-March)
80
Percent
60
40
20
0
19
90
–9
5(
av
er
ag
e)
19
97
–9
8
19
99
–2
00
0
20
01
–0
2
20
03
–0
4
20
04
–0
5
20
05
–0
6
20
06
–0
7
20
07
–0
8
20
08
–0
20
9
09
–1
0
(R
E)
20
10
–1
1(
BE
)
–20
Year
Market borrowings
Loans from center
NSSF
Small savings, provident fund, etc.
Source: Reserve Bank of India.
Note: BE = budget estimates, RE = revised estimates, NSSF = National Small Savings Fund (described in box 3.1).
Traditionally, loans from the center were the dominant source of
funding for states. In keeping with the trend of financial sector liberalization, the center’s loan intermediation role has been reduced since
1999–2000. The other notable change has been the rising share of NSSF
and small saving loans (see box 3.1 for a description). The share of
NSSF increased sharply to 69 percent during 2004–05 from 39 percent
during 2001–02. This characterized a move from center-controlled
borrowings to the autonomous NSSF but at higher cost (NSSF loans
at 9.5 percent compared with cheaper market loans, weighted average
of 8.39 percent during 2010–11). There is an inflexibility related with
NSSF borrowings, since these are based more on availability and collection within the territory of the state than the requirement by the
state to borrow. The NSSF loans are also at higher interest costs and
have been more asymmetrically beneficial for the center (Thirteenth
FC, 144).
Given the Twelfth FC recommendations for greater autonomy and
discontinuation of the financial intermediary role for the center, the
lending policy was changed, with more market access for states.17 Thus,
states got more freedom, but also greater responsibility to manage their
115
116
Until Debt Do Us Part
debt. A consequence of the new lending policy was the move to market discipline and transparency to enhance credibility among the market participants. Competition gradually increased among states to avail
themselves of the best market terms and obtain credit ratings. There has
been evidence of some variation in the spreads among states, with some
states borrowing at slightly lower rates, although the overall range of the
spreads has been narrow (table 3.1).
Cross-country evidence shows that spreads over central government securities should be linked to debt and deficit (fiscal) indicators
of states. For example,18 Schuknecht, von Hagen, and Wolswijk (2009)
concluded this for the European Union member states, and Lemmen
(1999) analyzed similar issues for the subnational governments in Australia, Canada, and Germany. Poterba and Rueben (1999) found that
states with tighter antideficit rules and authority of state legislatures
can issue debt at a lower interest burden. However, somewhat counterintuitive is the case in India. Bose, Jain, and Lakshmanan (2011)
indicate that the conventional deficit indicators have not been significant in determining the yield spreads during 2006–07 to 2010–11. The
study, however, concludes that since the period is characterized by the
prevalence of rule-based fiscal policy, it appears to have provided confidence to investors regarding states’ commitment to fiscal discipline.
Although the impact of FRLs cannot be directly determined, it cannot
be undermined.
Table 3.1 Weighted Average Spreads during 2010–11
Weighted average
spreadb (basis points)
General category states/
union territories
Special category statesa
30–40
Puducherry, Gujarat, Goa,
Rajasthan, Tamil Nadu, Bihar
Manipur, Nagaland, Meghalaya,
Tripura
40–50
Kerala, Andhra Pradesh, Madhya
Pradesh, Punjab, West Bengal,
Uttar Pradesh, Karnataka,
Haryana, Orissa
Assam, Himachal Pradesh,
Jammu and Kashmir
50–60
Sikkim, Uttarakhand
Source: Rakshitra, various issues, The Clearing Corporation of India Ltd.
a. Special category states are all the North-eastern states, along with Jammu and Kashmir, Himachal Pradesh,
and Uttarakhand. They have distinct characteristics: a low resource base, cost disabilities due to their
physical geography, sparse terrain, remoteness, and historical circumstances. These states account for only
5–6 percent of all states’ gross domestic product.
b. Over the center’s benchmark.
Managing State Debt and Ensuring Solvency: The Indian Experience
117
This shifting in the sources and methods of state borrowing has had
a bearing on the interest payments, deficits, and debts of the states. The
next section presents the changing trends of states’ fiscal deficit, debt
composition, and the interest burden, and details on interstate variability in these key indicators.
Trends and Composition of States’ Deficit, Debt,
and Interest Burden
An analysis of the evolution of states’ deficit, debt, and interest burden (defined as the ratio of interest payments to current receipts)
during the 1990s and 2000s reveals three distinct phases (as depicted
in figure 3.2). The first phase, in the early to mid-1990s, was characterized by low current account and fiscal deficits, moderate debt
levels, and a tolerable interest burden. The second phase, during the
late-1990s to mid-2000s, was characterized by a significant deterioration in state finances, with all key deficit indicators, debt levels, and
interest burden rising. The third phase, from the mid-2000s, was
Figure 3.2 Deficit and Debt as Share of GDP and Interest Payments as Share of Revenues
35
30
4
25
3
20
2
15
10
1
Debt/GDP and IP/RR
GDF/GDP and RD/GDP
5
5
0
19
90
–
19 91
91
–9
19 2
92
19 –93
93
–
19 94
94
19 –95
95
–
19 96
96
19 –97
97
19 –98
9
19 8–
99 99
–2
20 000
00
20 –01
01
20 –02
02
20 –03
03
20 –04
04
20 –05
05
20 –06
06
20 –07
07
2 –0
20 008 8
09 –0
– 9
20 10 (
10 RE
–1 )
1(
BE
)
0
Year
GDF/GDP
RD/GDP
Debt/GDP
IP/RR
Source: RBI Handbook of Statistics on State Finances, various issues.
Note: BE = budget estimates, GDP = gross domestic product, GFD = gross fiscal deficit, IP = interest payments,
RD = research and development, RE = revised estimates, RR = revenue receipts.
118
Until Debt Do Us Part
c haracterized by the onset of fiscal correction and reforms and manifests with improvements in key fiscal indicators.
Until the mid-1990s, states’ finances were relatively stable, characterized by low current and fiscal deficits, averaging below 1 and 3 percent
of GDP. Debt levels remained moderate at about 20 percent of GDP, and
the interest burden hovered close to 15 percent of revenue. The turning
point came during 1998–99, with a significant deterioration in the current account, which became a key driving force for the declining fiscal
health of the states, with increased spending on administrative services
and interest payments.
The next phase, 1998–99 to 2003–04, saw the steep rise in the fiscal deficit as a ratio of GDP—from 2.8 to 4.2 percent; the revenue
deficit more than doubled from 1.1 percent of GDP to 2.5 percent. As
a result, the states’ outstanding debt to GDP grew from 21.7 percent
during 1997–98 to its peak of 32.8 percent during 2003–04. Interest payments as a share of revenue receipts (repayment burden) rose
from 17.9 to 26 percent over the same period, and the primary deficit
grew from 0.9 to 1.5 percent. This period, until 2003, was also characterized by higher interest rates, with the interest rates being gradually liberalized; the average market interest rate on states’ borrowing
was over 10 percent during this period. Concomitantly, the average
interest burden, at 23.4 percent, was significantly higher than the
15 percent considered tolerable for a sustainable debt level (Dholakia,
Mohan, and Karan 2004).19
There is vast fiscal literature on the factors leading to the deterioration of state finances in the late 1990s. Factors considered critical to
the fiscal deterioration include the impact of the wage revisions; inability to contain wasteful expenditure, including subsidies; reluctance to
raise additional resources; and competitive reduction in taxes. Mohan
(2000) pointed to the increasing debt service payments and inadequate returns on government spending as important factors behind
the deterioration in states’ fiscal conditions. Acharya (2001) and Rao
(2002) attributed the worsening of revenue (current) balance during
this period to the implementation of the Fifth Pay Commission recommendations.20 The RBI Study of State Budgets, 2002–03, while drawing
attention to the growing fiscal and revenue deficit and high debt levels
of states, pointed to the following causes of the deterioration in states’
Managing State Debt and Ensuring Solvency: The Indian Experience
fiscal condition: (a) an inadequate increase in tax receipts, (b) negative
or negligible returns from public investments due to losses in public
sector undertakings (PSUs), (c) large subsidy payments, (d) increased
expenditure on salaries due to pay revisions, and (e) higher pension
outgo. Another study, by Prasad, Goyal, and Prakash (2004), concludes
that interest payments played a prominent role in the deterioration of
state finances.
Until the mid-1980s, interest rates on government borrowing were
highly subsidized, indicative of the degree of financial repression. After
the 1980s, the rates on government bonds became progressively aligned
with market interest rates; during the 1990s there were increases in both
bank deposit rates and policy rates (table 3.2). During the 1990s, average interest rates rose, and those on state government bonds averaged
Table 3.2 Deposit Rate of Major Banks for Term Deposits of More Than
One-Year Maturity
percent
Year
Average interest rate
Bank rate/repo rate/reverse repo
1
2.0
3.0
Mar-91
10.0
10.0
Mar-92
12.5
12.0
Mar-93
11.0
12.0
Mar-94
10.0
12.0
Mar-95
11.0
12.0
Mar-96
12.5
12.0
Mar-97
12.0
12.0
Mar-98
11.3
10.5
Mar-99
10.3
8.0
Mar-00
9.5
8.0
Mar-01
9.3
7.0
Mar-02
8.0
6.5
Mar-03
5.3
5.0
Mar-04
4.8
4.5
Mar-05
5.8
4.75
Mar-06
6.5
5.5
Mar-07
8.3
6.0
(continued next page)
119
120
Until Debt Do Us Part
Table 3.2 (continued)
Year
Average interest rate
Bank rate/repo rate/reverse repo
8.3
6.0
Mar-09
8.3
5.0
Mar-10
6.8
5.0
Mar-11
8.6
6.75
Mar-08
Source: RBI 2011c.
Note: Average interest rate refers to the midpoint of interest rates charged by commercial banks on demand
deposits. In column 3, the policy rate used is the relevant policy rate at that time. The bank rate was used
for the period prior to 2003, when it was in active use. For the subsequent period, the repo/reverse repo
rate was used depending on the prevailing liquidity conditions in the system.
over 10 percent during the 1990s (RBI). At the same time, the reliance on market borrowing to finance the fiscal deficits increased from
11 percent in the 1980s to 16 percent in the 1990s. Significant changes in
the structure and cost of state government debt contributed to a sharp
increase of about 60 percent in the repayment burden from the beginning to the end of the 1990s. Interest rates started softening in the mid2000s, and these were taken advantage of in formulating the debt swap
scheme for states (discussed in “Debt Restructuring and Institutional
Reform” section).
The data in figure 3.2 capture only the direct and explicit state liabilities; exacerbating the debt burden and repayment pressure were
the contingent liabilities, in the form of guarantees issued by states to
support their enterprises. During the mid-to-late 1990s, there was a
rapid increase in the issuance of guarantees by states to support their
public enterprises, many of which could not borrow on their own
credit strength.21 Although the latest data indicate that loan guarantees issued by states were lower at 2.8 percent of GDP by end-March
2009 compared to 3.3 percent of GDP in 2008, this does not incorporate the unfunded pension liabilities or the losses of the state PSUs.
The Thirteenth FC estimated that by the end of 2007–08, about 1,160
state PSUs had accumulated losses of about Rs 659.24 billion (almost
1.3 percent of GDP), particularly the implicit liabilities associated with
power utility companies, because their large accumulated losses represent a huge exposure for states.22, 23 The probability that these liabilities will
devolve are not identical for each state, and thus cannot be treated uniformly in terms of their fiscal impact.24 As a rule of thumb, assuming that
Managing State Debt and Ensuring Solvency: The Indian Experience
121
about one-third of such liabilities devolve to the states to service, figure
3.3 presents a broader concept of “extended” debt, that is, debt inclusive of the likely devolvement of outstanding guarantees, to provide an
assessment of the exposure and fiscal risk for the states. Extended debt
is calculated as direct debt (explicit) plus one-third of the contingent
liabilities25 extended by the state governments (as reported by them).
This adds to the stress scenario being faced by the states.
Along with the deterioration in state finances during the second
phase (1998–99 to 2003–04) was the fiscal stress experienced by the
central government. The combined center-state fiscal deficit rose from
7.3 percent of GDP in 1997–98 to an average of 9.3 percent over the
period. Studies indicate that although India has had primary deficits, it
has avoided an explosive rise in debt, mainly because of high economic
growth rates relative to the interest rate paid on government debt. Milan
(2011) analyzed the decomposition of India’s public debt trajectory
using the method of debt dynamics and concludes that the strong rate
of economic growth compared to the interest rate paid on debt helped
avoid an explosive debt trajectory. The situation was similar for states’
finances, where the lower rate of interest on debt and the higher revenue
buoyancy (Thirteenth FC, 126) (from both their own taxes and their
share in central taxes) enabled improvements in the fiscal stance.
40
5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
0
35.5
Debt/GDP, %
35
24.5
30
33.4
32.9
30.1
28.4
27.7
27.2
25
20
15
20
09
08
20
07
20
20
06
20
05
20
04
20
00
19
92
10
Year
Debt/GDP
Extend debt/GDP
GFD/GDP
Sources: Reserve Bank of India and author’s calculations. Latest data on guarantees are available only until 2009.
Note: GDP = gross domestic product, GFD = gross fiscal deficit.
GFD/GDP, %
Figure 3.3 Extended Debt as Share of GDP
122
Until Debt Do Us Part
Fiscal correction set in after 2004–05, with the onset of reforms that
went beyond the “realm of fiscal space” (World Bank 2004, 11). These
included reforms on the expenditure side to contain spending, restrict
recruitment, and curb growth in administrative expenditures; and some
states cut the cost of pension schemes and reduced subsidies (through
power sector reforms), including closure of and privatization of selected
PSUs. On the revenue side, reforms aimed to enhance revenue receipts
by revising tax rates and broadening the base, while focusing on
improving tax compliance. Institutional reforms reflected a paradigm
shift, with the adoption of medium-term fiscal frameworks and FRL at
the state level (Howes, Lahiri, and Stern 2003).
Much has been written about the reforms to correct the fiscal imbalances and sectoral improvements, including improving the business
climate to facilitate growth (World Bank 2003b). The reforms undertaken specifically to restructure or reduce the debt and interest burden,
along with those to enhance the credibility of states and ensure sustainability of debt, are discussed in the next section. Since the implementation
of reforms in the mid-2000s, the declining fiscal/debt trends have been
reversed. However, the 2008–09 global financial crisis has posed challenges.
Another aspect to consider is that, at the macro level, the states’ aggre
gate analysis masks state-level disparities in fiscal performance. The
differentiation among state performance persists, but the dynamics
change over time, with some states reversing their fiscal decline from
the second phase to the third phase (for example, Karanataka and
Orissa), while the record of some states continued to deteriorate (for
example, West Bengal). Assessing the performance of individual states
against the median for the period reveals that in terms of the primary
deficit, among the nonspecial category states,26 Bihar, C
hattisgarh,
Gujarat, Haryana, Karnataka, Madhya Pradesh, Orissa, Punjab, and
Uttar Pradesh have improved fiscal performance (primary balance)27
since 2004–05 (until 2009–10) compared with the deterioration d
uring
1998–99 to 2003–04 (compared with median values), while Goa,
Jharkhand, Kerala, Maharashatra, Uttar Pradesh, and West Bengal continued to have persistently high deficits even during the fiscal correction
phase (from 2004 onward) (see figure 3.4).
As expected, most of the states that reflected weak fiscal performance
are also plagued with high debt and repayment burdens across the three
Managing State Debt and Ensuring Solvency: The Indian Experience
123
Figure 3.4 Primary Deficit as Percentage of GSDP
10
Mizoram
8
Nagaland
Percent
6
4
Sikkim
Himachal
Pradesh
75 percentile
2
0
Himachal
Pradesh
Jharkhand
Uttarakhand
Median
25 percentile
–2
0
–1
05
20
–2
99
19
19
91
–9
00
8
4
–4
Years
Source: The box plot was created by Stata using data from Reserve Bank of India.
Note: The line in the middle of each box indicates the median (50th percentile); the top of the box indicates the 75th
percentile and the bottom the 25th percentile. The lines above and below the box represent the adjacent values,
which are within 1.5 times the interquartile range (iqr) of the nearer quartile (75th percentile for the line above and 25th
percentile for the line below). The iqr is calculated as the value of the 75th percentile minus that of the 25th percentile. Thus, the largest value (upper end of the upper line) is identified by the 75th percentile plus 1.5 times the iqr, and
the smallest value (the lower end of the lower line) is the 25th percentile minus 1.5 times the iqr. GSDP = gross state
domestic product.
periods. As can be seen from the box plot in figure 3.5, a large number
of states have both debt and interest burden above the 75th percentile
across the three periods under study. A case in point is West Bengal,
which has persistently had high debt (an average of 45 percent during
2005–10) and a large interest burden (39.3 percent) continuing relentlessly, even during the current period (figure 3.6).
Analyzing vulnerability in terms of debt as a ratio of gross state
domestic product (GSDP) and interest burden (payments as a share of
each state’s revenue receipts) provides a useful indication of the susceptibility that states face. Table 3.3 plots states in a matrix that highlights
states facing more vulnerability. Gujarat, Himachal Pradesh, Kerala,
Until Debt Do Us Part
Figure 3.5 Box Plot Showing Debt-to-GSDP Ratio and Interest Burden Ratio
120
Mizoram
100
Percent
Arunachal
Pradesh
Mizoram
80
60
75th percentile
40
West
Bengal
Median
West
Bengal
20 25th percentile
0
–1
05
20
19
99
19
–2
91
–9
00
8
4
0
Years
Debt/GSDP
IP/RR
Source: The box plot was created by Stata using data from Reserve Bank of India.
Note: GSDP = gross state domestic product, IP = interest payments, RR= revenue receipts.
Figure 3.6 Interest Burden in Selected States
50
45
40
35
Percent
30
25
20
15
10
5
0
–1
20
05
4
–2
19
99
91
–9
00
8
0
19
124
Years
Bihar
Orissa
Uttar Pradesh
Gujarat
Punjab
West Bengal
Source: Authors’ calculation using data from RBI reports.
Kerala
Rajasthan
Median
Mean
Managing State Debt and Ensuring Solvency: The Indian Experience
125
Table 3.3 States’ Vulnerability Matrix
Debt
Debt-GSDP Ratio
Very high
(above 50%)
Interest payment
Very high
(above
25%)
Ratio of
interest
payment to
revenue
receipts
High (30–50%)
Medium
(20–30%)
Low (below
20%)
West Bengal
High
(15–25%)
Himachal
Pradesh
Gujarat, Kerala,
Punjab, Rajasthan
Maharashtra
Medium
(10–15%)
Jammu and
Kashmir
Andhra Pradesh, Bihar,
Goa, Jharkhand,
Madhya Pradesh,
Orissa, Uttar Pradesh,
Uttaranchal
Karnataka,
Tamil Nadu
Haryana,
NCT Delhi
Low
(below
10%)
Arunachal
Pradesh,
Manipur,
Mizoram,
Nagaland,
Sikkim
Meghalaya, Tripura
Assam
Chhattisgarh
Source: Authors compilation using data from RBI reports.
Note: GSDP = gross state domestic product, NCT = National Capital Territory.
Punjab, Rajasthan, and West Bengal reflect both debt levels of over
30 percent of GSDP and a high interest burden. The combined GSDP
of these states accounts for over 12 percent of the national GDP. Their
continued struggle with fiscal adjustment poses a challenge, which is
further compounded by the global financial crisis (discussed in “Impact
of the Global Financial Crisis and Going Forward” section).
In keeping with the diverse fiscal situation in states, the Thirteenth
FC recommended a state-specific approach for adjustment based on
past fiscal performance (with 2007–08 the base year), and prescribed
differentiated adjustment paths for different groups of states. It was
estimated that to attain the aggregate target of states’ debt-to-GDP
ratio of 25 percent, the aggregate fiscal deficit of states should be maintained at 3 percent of GDP. Being an aggregate, however, this target
indicator does not reflect the specific realities of individual states. For
example, an abrupt reduction in fiscal deficits in states that also had
high revenue deficits would lead to undesirable compression in capital
expenditures.
126
Until Debt Do Us Part
Thus, the Thirteenth FC, while keeping a balance between the
need for customization with the requirement for adopting a uniform
approach for determining targets for all states, recommended a differentiated approach. It was recommended that the nonspecial category
states that had a revenue surplus or balance in the base year 2007–08
adopt a road map that eliminated their revenue deficits by 2011–12,
and target fiscal deficit to 3 percent of GSDP. Other states with a higher
revenue deficit in the base year were to adjust following a gradualist
approach to avoid sudden cutbacks in capital expenditures, and eliminate the revenue deficit by 2014–15 and achieve a 3 percent fiscal deficit
by 2013–14.28
Debt Restructuring and Institutional Reform
The structural deterioration in states’ finances led to intense deliberations among stakeholders—parliamentarians, policy makers, think
tanks, and other interested parties—about reform options to not
only reverse the fiscal decline and lower debt levels, but also to put
state finances on a more sustainable path going forward. The fiscal
correction in state finances since the mid-2000s thus resulted from
interrelated reforms on multiple fronts, helped in large part by the
higher reveune buoyancy (Thirteenth FC) and the overall strong
economic growth in India. The priority of fiscal consolidation was
to restore the balance of revenue accounts—that is, reducing the
revenue deficit to zero. The reforms included the standard fiscal consolidation measures through expenditure and taxation reforms. But,
importantly, efforts were taken to develop a rule-based institutional
framework, including fiscal responsibility laws, to ensure the sustainability of the consolidation. Such a rule-based system complemented
the move from central government onlending to the market-based
financing mechanism for meeting the states’ financing requirements.
It was realized, however, that even after lowering the primary deficit, the debt service repayment pressure and high indebtedness would
continue, since about 80 percent of states’ borrowing in 2003–04 was
at high-cost, nonmarket rates.
Research indicates that in addition to the important elements of
fiscal consolidation, such as controlling the rapid growth of current
Managing State Debt and Ensuring Solvency: The Indian Experience
expenditures and implementing structural taxation reforms, fiscal
consolidation must include the objective of reducing repayment pressure by reducing interest costs (Dholakia, Mohan, and Karan 2004;
Prasad, Goyal, and Prakash 2004). The Twelfth FC had also viewed the
large interest payments as a major factor leading to the outstanding
debt of states, and felt that reducing these payments was integral to
attaining debt sustainability. With regard to the broad approach on the
issue of debt sustainability, the Twelfth FC was of the view that debt
relief measures were required as a prerequisite to achieve revenue balance. Moreover, international experience showed that given the high
indebtedness of states, it would be difficult to adhere to the fiscal targets when established by the states’ fiscal responsibility law (Liu and
Webb 2011). To achieve this would imply reducing both the stock of
debt and the cost of borrowing.29
However, it was also recognized that debt write-offs, relief, and
restructuring alone cannot ensure the sustainability of state finances.
Policy makers were cognizant that waivers of loans and interest
should be restricted to avoid moral hazard problems and encourage
debt repayment discipline. The debt restructuring was thus linked
to states undertaking reforms to increase revenue efforts, controlling
expenditure, and reorienting expenditures toward supporting growth
(Twelfth FC). This section focuses on the debt restructuring program
and its links to incentive packages offered to states for undertaking
institutional reforms.
Debt Relief and Fiscal Responsibility Legislations
Debt relief had been provided by the waiving of repayment and/or
interest payments due, altering the terms of repayment, reducing interest rates, and consolidation of loans. In the 1980s and 1990s, successive
FCs had given unconditional debt relief to states, although the relief
had been provided only periodically, and the amount of relief was not
significant (table 3.4).30 Thus, states have had to repay most of the debt
they incurred. The Tenth and Eleventh FCs started to link debt relief
with fiscal performance.31 However, it was not until the Twelfth FC
that debt relief was linked explicitly to rule-based legislative reforms.
In a pathbreaking move, the Twelfth FC recommended debt relief for
states contingent upon the enactment of fiscal responsibility laws and
127
128
Until Debt Do Us Part
Table 3.4 Debt Forgiveness by Finance Commission
Finance Commission
Year of report
Rs (billion)
1974
20
Seventh
1979
Eighth
1984
Ninth
1989
Tenth
1995
Eleventh
Twelfth
Sixth
GDP Rs (billion)
% of GDP
667
2.95
22
1,025
2.11
23
2,223
1.03
10
4,357
0.22
5
10,672
0.05
2000
34
20,050
0.17
2005
535
31,494
1.70
Sources: McCarten 2001; Report of Thirteenth Finance Commission 2009.
Note: GDP = gross domestic product.
incorporation of a fiscal correction path, with milestones for attaining
fiscal targets while improving the current (revenue) balance (reducing
the deficit to zero by 2008–09).
To implement the recommendations of the Twelfth FC, the Debt
Consolidation and Relief Facility was introduced during 2005–06,
which provided debt relief through consolidation, rescheduling repayments for a fresh term of 20 years, and lowering of the interest rate
on the debt to 7.5 percent. All states were eligible to obtain relief from
the year they enacted FRL. This amounted to Rs 187 billion in terms
of lower interest payments, and Rs 211 billion in terms of lower repayments, totaling Rs 398 billion (US$8.9 billion32) during 2005–06 to
2009–10. In addition, repayments due during 2005–10 on central loans
contracted up to March 31, 2004, (after consolidation and rescheduling)
were eligible for write-off subject to the reduction in revenue deficits.
The debt write-off would also be subject to containment of the fiscal
deficit to the 2004–05 level. Subject to these provisions, if the revenue
deficit were brought down to zero by 2008–09, all repayments during
2005–10 would be written off.
Carrying forward the momentum to support states toward urgent
fiscal correction, the Thirteenth FC worked out a differentiated fiscal
adjustment road map (described in the previous section), with a statespecific approach based on past fiscal performance (using 2007–08
as the base year) for different groups of states. A key requirement is
that all states eliminate their revenue deficits (the deficit on current
balance), but they can have a fiscal deficit of 3 percent of GSDP by
Managing State Debt and Ensuring Solvency: The Indian Experience
2014–15, along with a reduced debt target of 24.3 percent of GDP in
the same year (from 27 percent in 2008–09). The debt relief granted
was similar to the Twelfth FC; all loans to states from the Government of India outstanding as of 2009–10 would be written off if the
state enacted or amended its FRL. Moreover, interest on past NSSF
loans (contracted during 2006–07) was reduced to 9 percent from
9.5 percent.
The center enacted the Fiscal Responsibility and Budget Management Act in 2003, with applicability only to the national government.
Some states had also enacted their own FRLs before the center (for
example, Karnataka and Punjab, in 2002), and many states had since
2003 adopted FRLs in line with the national law. The Twelfth FC subsequently mandated that states pass FRLs to avail themselves of the
benefit of debt relief, with revenue deficits (total revenue minus current
expenses) to be eliminated and fiscal deficits to be reduced to 3 percent
of GSDP by fiscal year 2009.33 Since then, all 28 states have passed FRLs,
most of which require the state to present a medium-term fiscal plan
with multiyear rolling targets for key fiscal indicators, along with the
annual budget, to the state legislature. Some of the FRLs, passed by
states, also place limits on guarantees; others mandate the disclosure of
contingent liabilities and other borrowing. Most FRLs require disclosure of significant changes in accounting policies.
Fiscal targets adopted by Indian states are remarkably similar to each
other with respect to fiscal and revenue deficits. Some states adopted
additional legislation on fiscal targets, such as the Kerala Ceiling on
Government Guarantee Act (2003), which was enacted the same year
as its FRL. According to the Guarantee Act, the guarantee outstanding for any fiscal year shall not exceed Rs 140 billion,34 no government
guarantee shall be given to a private entity, and the Guarantee Redemption Fund shall be established. Other initiatives included the setting up
of (a) the Consolidated Sinking Fund (1999) to provide a cushion for
repaying market loans of states (20 states have established this), (b) the
Guarantee Redemption Fund (2001) to provide a cushion for servicing
any contingent liabilities because of guarantees issued by state governments to its PSUs (11 states have established this), and (c) several technical committees and working groups on topical issues of cash and debt
management.35
129
130
Until Debt Do Us Part
Debt Swap and Securitization: A Move toward
Market-Based Financing
The fiscal correction was given an impetus with the introduction of a
“debt swap scheme” during 2003–04 to lower the existing interest burden and increase market access. Loans from the center amounting to
Rs 1,000 billion (US$23 billion36) with interest rates in excess of 13 percent were substituted with new market loans and small savings proceeds
at lower rates of interest; the outstanding debt remained unchanged.
The market conditions prevailing were fortituous and the rates were
significantly lower, at 7.5 percent (RBI State Finances Study 2004–05,
p. 24), enabling an interest savings for states of Rs 310 billion (US$7.1
billion37) and 0.75 percent per year in revenue (Twelfth FC). This direction toward the market was reaffirmed by the Twelfth FC in conjunction
with state debt relief, where it stated, “As regarding the future lending
policy, the central government should not act as an intermediary and
allow the states to approach the market directly” (Twelfth FC, 236).
States issued “power bonds” to securitize the fiscal risks emanating from the losses of electricity utilities arising from the gap between
the cost of producing power and the tariff charged. This gap between
the cost and tariff had resulted in significant losses and an accumulation of arrears. With the securitization, arrears and accrued interest at
about 1.5 percent of GDP were cleared by states through the issuance
of 15-year tax exempt “power bonds.”38 Cognizant of the moral hazard
issue, this was clearly announced as a one-time settlement measure and
was supplemented with reforms to ensure discipline going forward. Participating states qualified for funds on the basis of reform milestones
and improvements in the reduction of commercial losses. Although
state liabilities had increased by 22.8 percent during 2003–04 at the time
of issuance of these bonds, many states have prepaid, and only Rs 144.23
billion (US$3.23 billion39) remained as of end-March 2011.
In addition to the above debt restructuring program to link with
institutional reform and move toward market access, the role of the
RBI is also important. First, as the regulator of the banking sector, the
RBI sets the statutory requirements for banks to hold state debt. This
increases the acceptability of state securities by the market. Second,
the RBI tightened the regulation for use of the overdraft facility by
states. Previously, states had resorted to the facitily as a way to roll
Managing State Debt and Ensuring Solvency: The Indian Experience
over short-term borrowing to finance structural deficits. The terms
and conditions for facilty use were formula based and specified. Moreover, the use of the facility by the states is disclosed to the market on
an ex-post basis. For example, the market has information on the better performers compared to the chronic-deficit states (table 3.5 shows
that Punjab, Uttarkhand, and West Bengal depended on this facility
during 2010–11 to meet their temporary resource gap). Such information influences market sentiment and spreads, while lowering credit
ratings.
The intent of the FRL, debt swap, securitization, and the move
toward market operation was to support the fiscal discipline reform at
Table 3.5 States’ Overdrafts and Access to Cash-Credit
Number of days
Special WMA
Andhra Pradesh
Normal WMA
Overdraft
2009–10
2010–11
2009–10
2010–11
2009–10
2010–11
1
3
—
—
—
—
Haryana
7
10
5
10
—
8
Kerala
18
—
2
—
—
—
Madhya Pradesh
11
—
11
—
—
—
Maharashtra
—
—
—
—
—
—
Karnataka
—
—
—
—
—
—
Nagaland
69
—
45
—
13
—
Punjab
130
133
128
132
29
13
Rajasthan
—
—
—
—
—
—
Uttar Pradesh
8
4
8
4
—
—
West Bengal
95
195
15
113
8
62
Himachal Pradesh
—
—
—
—
—
—
Manipur
—
—
—
—
—
—
Mizoram
29
25
15
15
—
—
Goa
—
—
1
—
—
—
Uttarakhand
69
35
26
12
9
10
Meghalaya
—
1
—
—
—
—
Jharkhand
—
—
—
—
—
—
Source: Reserve Bank of India, Annual Report 2010–11.
Note: WMA = ways and means advances. Normal WMA is formula based, special WMA is after access to
normal WMA but is collateralized. — = no access to facility and strong cash management position.
131
132
Until Debt Do Us Part
the state level to reverse the structural decline of state finances from the
late 1990s to the early 2000s. It will be difficult to precisely evaluate the
direct impact of these reforms. In this context, a study by Liu and Webb
(2011) concludes it would be difficult to precisely separate and measure the effects of the FRL given the lender-borrower nexus and various
channels that would influence government fiscal deficits and indebtedness. Nonetheless, it was noted that to the extent the FRL intends to
improve government finance and avoid over-indebtedness, it is worthwhile ascertaining whether FRL has been associated with improved fiscal outcomes.40
Liu and Webb (2011) choose growth of public debt before and after
passing subnational FRL in several countries, including India. The
measurement of the fiscal improvement or deterioration was normalized, since each state government might have passed its FRL in different years. The paper shows that in Indian states, the growth of debt to
GSDP was slower in the post-FRL period than in the pre-FRL period
for 24 of 26 states. Twenty-one of these 24 states had reversed the trend
of increasing debt to GSDP in the pre-FRL period.
A study on the “Dynamics of Debt Accumulation in India” (Rangarajan and Srivastava 2008) pointed to the fact that accumulation of
debt can be seen as the result of the balance between cumulated primary
deficits and the cumulated weighted excess of growth over interest rate.
Decomposing the change in the central government’s liabilities relative
to GDP shows that a significant part of the cumulated primary deficit
could be absorbed due to the excess of growth over interest rates. However, this cushion is not always available, and the sharp increases in debt
relative to GDP during 1997–2003 were because of both factors, that
is, cumulated primary deficit and excess of effective interest rate over
growth rate.
One study (Milan 2011) shows that strong economic growth relative to the interest rate paid on government debt helped India avoid
an explosive rise in debt despite the existence of successive primary
deficits. The same holds true for the aggregate performance of states;
figure 3.7 shows that the fiscal correction phase in states also coincides with a higher GDP growth rate and lower rate of interest paid
on state debt. This was also in part the result of fiscal correction,
which led to a reduction in government dis-savings and debt. This,
Managing State Debt and Ensuring Solvency: The Indian Experience
Figure 3.7 Differential between GDP Growth Rate and Interest Rate on State Debt
25
Percent
20
15
Onset of
deterioration
where
g<r
Period of
correction with g > r
10
5
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
2
20 009
10
20 (RE)
11
(B
E)
0
Year
Nominal GDP growth (g)
Average interest rate on debt (r)
Source: Authors’ calculation using base data from RBI.
Note: BE = budget estimates, GDP = gross domestic product, RE = revised estimates.
however, masks the varied fiscal performance of individual states,
which has been mixed over the period, causing concern in the context
of the global crisis.
Impact of the Global Financial Crisis and Going Forward
Although the immediate impact of the global financial crisis on state
finances was somewhat subdued, there are implications going forward.
The challenges will be more on the resource side through reduced central transfers, accentuated by the low-cost recovery by states. These may
well render difficult the achievement of the Thirteenth FC road map.
Going forward, sustainable finances require states to undertake reforms
to maintain solvency via, among other things, increases in own taxes,
implementing a goods and services tax, and revising tariffs.
Assessing the impact of the crisis on the center, the immediate
impact was relatively “muted” (Milan 201141; RBI 2011a). Although there
was a setback in the growth of the economy, the bounce back was swift
and impressive (Reddy 2011). The countercyclical fiscal and monetary
policy actions and, more critically, the high-growth trajectory that was
maintained at over 7 percent during 2009–10 and over 8 percent during
2010–11 helped minimize the impact (“World Bank Economic Update,”
133
134
Until Debt Do Us Part
September 2011). The initial conditions—the relatively low external
debt, the high foreign exchange reserves, and selective capital controls—
helped reduce the impact of the external shocks. Nevertheless, worries
remain because of the high general government deficit and public debt
levels.42 It is widely acknowledged that high levels of deficit and debt
reduces “elbow room” and the ability to borrow and respond to such
shocks and extreme events.
On the growth front, a slowdown in the next two years is anticipated
(World Bank 2011),43 as a result of uncertainties weighing down investment, tighter macroeconomic policies intended to contain inflation,
and the base effect of the strong agricultural rebound during 2010–11.
Slow growth in core Organisation for Economic Co-operation and
Development countries implies that the domestic drivers for growth will
need to be strengthened. Moreover, there is a realization that concerted
efforts will be necessary to avoid fiscal slippages by the center during
2011–12, especially if the rise in commodity and fuel prices continues
at an elevated level. The sustainability of the fiscal stance will, however,
need measures to control, if not compress, expenditures along with revenue augmentation.
Turning to the states, the impact of the global crisis got intertwined
with the wage rise, and the fiscal situation deteriorated during 2009–10.
On the revenue front, there was a reduction in revenue receipts during
2008–09 and 2009–10, reflecting a fall in the state share of central taxes,
which had been affected by the economic slowdown. There was also a
deceleration of agricultural output that coincided with the crisis and
could in part explain the revenue falls in some states. Expenditures rose
primarily because of the revision of pay and salary arrears, which coincided with the crisis years. Of 17 nonspecial category states, 11 had current balance deficits, while the overall fiscal deficit widened in all states
except Jharkhand and Kerala.
The impact on the management of state debt was, however, insignificant, reflecting in part the strengths of the state borrowing regime,
with its ban on borrowing abroad; the limited history of bailouts; and
the enactment of FRLs. During 2008–09 and 2009–10, countercyclical
measures were taken by states to mitigate the impact of the crisis on
domestic economic activity. These included relaxing the deficit levels to
3.5 percent of GDP (from 3 percent legislated under the FRLs). Further,
Managing State Debt and Ensuring Solvency: The Indian Experience
the center allowed states a larger share of market borrowings to compensate for the unprecedented impact of exogenous factors on the fiscal situation. Concomitantly, states increased market borrowings by
34.6 and 28.6 percent during 2008–09 and 2009–10, respectively, compared with the increase of 23 percent during 2007–08. Interestingly, a
positive impact of this was an improvement in the interest profile of
states, with the share of high-cost market loans (with an interest rate
over 10 percent) declining further during 2009–10.
It needs to be emphasized, however, that macroeconomic stabilization and countercyclical policy actions are the key responsibility of the
center and not of subnational governments. Accordingly, the Thirteenth
FC recommended that instead of raising the borrowing limits for states
in the event of such shocks, the center should assume the entire responsibility for the additional resource mobilization and pass these to states
in the form of increased devolution. This (formula-based) devolution
will meet the differential requirements of the states in terms of both fiscal capacity and fiscal need.44
Much of the deterioration in the fiscal position of the states during
that period was temporary, and thus could be attributed to deterioration in the share of central taxes because of the crisis and arrears of
pay revision (Reddy 2011). Although the fiscal deficit had deteriorated
to 3.3 percent of GDP during 2009–10, indications were positive for a
turnaround in 2010–11, as reflected in the study of aggregate state budgets by the RBI. It also appears that fiscal discipline at the state level may
have acted as a source of comfort for the market. This is corroborated
by the fact that after witnessing some stress during the initial period of
the global financial crisis, most states reverted to the path of fiscal consolidation, with lower deficit ratios during 2010–11.
In sum, even though the immediate impact of the global financial
crisis on state finances was somewhat subdued, this has implications
going forward. The challenges are likely to emerge through the reduced
impact of central transfers, given that the center’s deficit has not shown
signs of abatement. The overall current transfers to states are budgeted to decline by 0.4 percentage points of GDP during 2010–11 (RBI
2011a). Going forward, sustainable state finances requires reforms to
increase states’ own tax revenues by speedily implementing the goods
and services tax. Implementation of this tax is expected to reduce
135
136
Until Debt Do Us Part
v ertical imbalances, with states being able to tax the services sector (the
fastest-growing sector, which accounts for over 65 percent to GDP), and
provide gains to India’s GDP of 0.9 –1.7 percent (Thirteenth FC).
States also need to review their tariff polices, particularly in power and
irrigation, to ensure that the gap between costs and recovery is narrowed,
if not closed. It is estimated that for the power tariff, even for the best-performing states, increases of 7 percent per year are required to bridge the
cost-to-recovery gap, while the not-so-good performers require increases
of almost 19 percent per year (Thirteenth FC). If not rectified, these issues
will render the achievement of the Thirteenth FC road map difficult.
Another critical issue is that, although in the aggregate, states have
contained their fiscal accounts, the impact needs to be evaluated in
the context of the contingent liabilities, which include not only guarantees, letters of comfort, and liabilities of state-owned enterprises, but
also implicit contingent liabilities arising due to vulnerabilities in the
state PSUs and pensions. Especially if tight liquidity conditions impact
the health of the state-owned enterprises and PSUs, fiscal risks could
spill over onto states’ fiscal positions. Although these must be addressed
more from the perspective of the fiscal risks that arise from such contingent liability, it is important to keep them in mind.
Conclusions
Although states have faced fiscal stress, systemic insolvency and defaults
have not occurred because of a mix of factors. The significant growth
rates of the Indian economy in the late 1990s and 2000s have also played
a critical role in alleviating the interest burden on debt and ensuring
that the debt does not grow in an explosive trajectory. The serious
efforts at fiscal consolidation and institutional reforms enabled states to
get on the path toward fiscal correction. In addition, the restriction on
borrowing and the constitutional arrangements enabled the onset of fiscal correction in an appropriate manner.
Furthermore, lowering the interest burden on debt was important in
enabling the states to pursue a sustainable course. The approach to debt
relief, linked with incentives to implement reforms, has greatly helped
avoid moral hazard problems. However, while the focus has been mainly
on direct debt obligations, contingent liabilities pose a serious fiscal risk
Managing State Debt and Ensuring Solvency: The Indian Experience
on states finances, unless monitored and adequately controlled. Moreover, the aggregate picture masks interstate disparities and vulnerabilities, which require customized reforms and correction packages rather
than a “one-size-fits-all” approach.
The change in lending policy and patterns of borrowing has provided greater flexibility, but also more responsibility to states, with
market discipline becoming an important plank. The greater access to
resources from the market requires more active debt management and
strategy development, using robust analysis to ascertain the cost risk
of the debt portfolio. Strengthened debt management capacity institutional arrangements at the state level, with a more active risk management approach, will be required to meet future challenges.
Although the global financial crisis has had a relatively insignificant
impact on Indian states, policy makers must always be cognizant of the
fact that despite an absence of systemic insolvency and defaults, high debt
reduces the maneuverability and flexibility of policy to respond. However, it needs to be emphasized that countercyclical policy is the responsibility of the federal government and not of subnational governments. If
the fiscal deficit targets are to be relaxed at all to overcome cyclical downturns, then that should be done by the federal government, which can
increase its borrowing and pass it on via higher devolution and grants
to the subnational governments. This means that the subnationals’ fiscal
deficit targets are unchanged. Fiscal challenges remain and are likely to be
critical if the rise in commodity and fuel prices continues at an elevated
level. There are concerns that growth might slow to 7–8 percent in the
next two years (World Bank 2011). This could generate the dilemma of
needing to compress expenditures for ensuring fiscal sustainability while
simultaneously needing countercyclical spending to boost growth.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. The word “states” is used interchangeably in this chapter with “state governments” and refers to the total data and performance of the 28 state governments
in India.
137
138
Until Debt Do Us Part
2. In Indian fiscal accounting, revenue balance refers to current balance, that is,
total revenue minus current expenditures.
3. The FC is a constitutional body appointed every five years or sooner to review
the finances of the center and state governments and recommend devolution
of taxes and other proceeds from the center to the states (vertical transfers)
and among the states themselves (with the objective of horizontal equity).The
FC uses a formula-based approach, assigning weights for various relevant factors such as population, income disparity, area, tax effort, and fiscal discipline.
These weights have changed over time. There have been thirteen FCs since
independence.
4. See Ianchovichina, Liu, and Nagarajan 2007; Pinto and Zahir 2004; Rangarajan
and Srivastava 2008; Reddy 2000; and World Bank 2004.
5. The share of debt of the local governments—the third tier of government—is
not covered in this chapter. Local government debt in India is small. Local governments, with limited fiscal autonomy, are largely dependent on fiscal transfers and onlending from higher levels of government.
6. These were added in the 73rd and 74th Amendments to the Constitution. The
Seventh schedule to the Constitution specifies the legislative, executive, judicial,
and fiscal domains of Union and State governments in terms of Union, State,
and concurrent lists.
7. The “financial relations between the Centre and the States are designed with
great care and circumspection … to forestall precisely the kind of difficulties
that even the older federations do not appear to have overcome in securing
closer correspondence between resources and functions of the different layers
of Government” (Sixth FC).
8. Over time, the FCs have taken various approaches to addressing state concerns
regarding the composition of the divisible pool of central taxes and inter se
allocation criteria between and among states. Although FCs have aimed to foster fiscal stability among the states, an empirical analysis reveals that although
transfers have helped to reduce the overall gross fiscal deficit of the states, horizontal fiscal inequity is yet to be addressed (Kannan et al. 2004).
9. RBI, “State Finances: A Study of Budgets,” 2007, p. 58.
10. Solvency refers to the government’s ability to service its debt without defaulting.
It is defined as the condition that the state government’s net stock of debt does
not exceed its ability to pay off that debt at some time in the future (measured
by the present discounted value of its future primary surpluses). Liquidity is the
ability to meet short-term cash needs (within the year); that is, in each period
the state government has enough resources (flows) to cover expenditures plus
debt service. Debt is considered unsustainable if it will lead to insolvency in the
future (see Ley 2010). Also important is the concept of credibility, or the confidence of investors that solvency and liquidity will be maintained.
11. External loans are project-based loans, except for some structural adjustment
loans, usually from multilateral development banks at concessional terms.
Managing State Debt and Ensuring Solvency: The Indian Experience
12. Since 2006, the Standing Technical Committee, with representation from the
center, states, and the RBI, has been making annual projections of states’ borrowing requirements. The committee considers several factors including the
macroeconomic and financial conditions, sustainability of debt, provisions of
FRL, and fiscal risks from issuance of guarantees.
13. India used to have a system of planned development under which grants and
other assistance was provided to states.
14. It may be argued that because India still has a large fiscal deficit, this effectively crowds the private sector. But banks are not required to hold more than
25 percent of their net deposits in liquid liabilities such as government securities. Banks, however, hold a higher percent, which goes to the question of
demand for credit offtake from business.
15. The central government temporarily allowed states to increase the fiscal deficit
to 3.5 percent of GSDP during 2008–09 and to 4 percent during 2009–10 in
response to the global financial crisis (Canuto and Liu 2010).
16. The NSSF was established in April 1999; small savings collections are invested
in central and state government special securities. At present, 80 percent of all
small savings collections within a territory of a state are invested in the same
state securities. This adds to the debt of the state but is not controlled by the
center. Moreover, the inflows are autonomous and depend on the spread
between the small savings rate and the deposit rate. See “Report of the Committee on the Comprehensive Review of the National Small Savings Fund,”
June 2011. There is an inflexibility associated with NSSF borrowing, since
these are based on availability and not necessarily on the requirement by
the state to borrow, and are at higher interest costs and with an asymmetry
toward the center (Thirteenth FC, 144).
17. The Twelfth FC stated that “. . . as regards the future lending policy, the central
government should not act as an intermediary and allow the States to approach
the market directly.” The practice of onlending from the center has been discontinued since then (recommendations of the Twelfth FC were accepted in
May 2005).
18. These studies are quoted in Bose, Jain, and Lakshmanan (2011).
19. Interest payments as a ratio of revenue receipts provide an explanation of the
interest burden and the level of “tolerable” debt. Debt is said to be tolerable
if servicing it does not impose an intolerable burden on the fiscal position.
Dholakia, Mohan, and Karan (2004) analyzed what interest burden a state can
tolerate as a proportion of its revenue receipts. In 2004, Dholakia, Mohan, and
Karan used one-fifth of revenue receipts as a tolerable ratio. The FC also considered the same.
20. In India, government wages are reviewed and revised periodically, usually every
10 years.
21. “The Report of the Group to Assess the Fiscal Risks of State Government Guarantees,” (RBI, July 2002) reported that guarantees grew at an average rate of 16
139
140
Until Debt Do Us Part
percent per year during 1992–2001 (paragraph 5). To avoid an escalation of the
fiscal risks, the RBI had issued regulatory guidelines to banks (which were the
major investors) to only invest in state public sector undertakings (PSUs) if there
was a clear revenue stream from the PSU/project, rather than that accruing from
the state government budget. In addition, all the PSU issues were to be rated by
at least two domestic credit rating agencies if banks were to invest in them. Also,
such investments had credit risk weights and provisioning requirements.
22. Thirteenth FC, p. 103.
23. A 2011 note by Citi Investment Research and Analysis indicates that the total
losses of state electricity boards in 2010–11 were Rs 635 billion, and those from
five states (Bihar, Madhya Pradesh, Rajasthan, Tamil Nadu, and Uttar Pradesh)
account for about 71 percent of the total losses.
24. “The Report of the Group to Assess the Fiscal Risks of State Government
Guarantees,” (RBI, July 2002), calculated the sectoral default ratios in 2001.
The power sector was 15.09 and industry was 39.19. The total default ratio
was 3.7.
25. These relate only to those that are reported by the states. Indirect and implicit
liabilities, although a source of fiscal risk, are not included here. A comprehensive review is difficult because of inconsistencies and gaps in data coverage and
definitions, and is not the remit of this chapter.
26. The focus of this analysis is on the general or nonspecial category states, since
they account for almost 95 percent of the total of all states’ GSDPs and over
92 percent of India’s population.
27. Analyzed as the average of the indicator with reference to the median values
during the period of study.
28. All special category states with a base fiscal deficit of less than 3 percent of
GSDP during 2007–08 could incur a fiscal deficit of 3 percent during 2011–12
and maintain it thereafter. Manipur, Nagaland, Sikkim, and Uttarakhand should
reduce their fiscal deficit to 3 percent of GSDP by 2013–14.
29. The Twelfth FC states that “[L]arge interest payments have been a major factor leading to the increase in the outstanding debt of state governments . . .
and therefore, the reduction in interest payments is integral to attaining debt
sustainability . . .” (p 226). Dholakia, Mohan, and Karan (2004) conclude that
the reduction in effective interest rates was an important factor for the interest
burden (interest payment to revenue receipt [IP/RR]) to be at a tolerable level
and debt to be sustainable in the states. This required a reduction in both the
stock of debt and its costs.
30. The share of debt relief in GDP has declined from 2.95 percent in the Sixth FC
to 0.17 percent in the Eleventh FC, indicating a decrease in the relative commitment to central debt forgiveness over time (McCarten 2001).
31. The Eleventh FC linked a portion of the untied central grants to the fiscal correction of the individual states as part of the Fiscal Reforms Facility. Although
the grants were small, at Rs 106.07 billion (US$2.43 billion equivalent), they
Managing State Debt and Ensuring Solvency: The Indian Experience
helped trigger useful reforms (using an exchange rate at the end of the 1st quarter of 2000 of Rs 43.62 = US$1.0).
32. The average of the quarterly average exchange rates from the 2nd quarter of
2005 to the 1st quarter of 2010 is used (from the IFS), which is Rs 44.64 =
US$1.0.
33. The summary of Fiscal Responsibility Laws in Indian states is based on Liu and
Webb (2011) and the RBI State Finances report.
34. About US$3 billion, assuming an exchange rate of Rs 46.7 = US$1.00.
35. These include issues relating to debt sustainability, model FRL, pension liabilities, state-government-guaranteed advances and bonds, fiscal risk of state government guarantees, voluntary disclosure norms for state governments, state
government guarantees, and methodology for compilation of outstanding liabilities and periodic revisions in the ways and means advances limit.
36. Using the end of March 2004 exchange rate at Rs 43.445 = US$1.0. (Source: IFS)
37. At an exchange rate of Rs 43.445 = US$1.0.
38. Settlement of state electricity boards dues, May 2001; and “State Fiscal Reform
in India: Progress and Prospects,” World Bank (2005). The balance accrued
interest at Rs 100 billion, which was written off (see chapter by A. Rastogi, in
the India Infrastructure Report 2004, 24).
39. Exchange rate as of end-March 2011 at Rs 44.65 = US$1.0 (Source: IFS)
40. Corbacho and Schwartz (2007) discuss the problems of determining the direction of causality. Their study compared national fiscal deficits in countries with
and without FRLs, and found that the former had smaller deficits. Data on subnational deficits for such cross-country comparisons, however, are not readily
available.
41. “Fiscal Policy for Growth and Development in India: A Review,” Milan et al.
2011 (forthcoming).
42. The high fiscal deficits for the general government, which averaged around
8 percent of GDP in the 1990s and 2000s, are expected to have reached
10 percent of GDP in 2009–10, with debt averaging over 80 percent of GDP in
the 2000s (World Bank, September 2011).
43. “World Bank India Economic Update,” September 2011.
44. Thirteenth FC, 136.
Bibliography
Acharya, S. 2001. India’s Macroeconomic Management in the Nineties. New Delhi:
Indian Council of Research in International Economic Relations.
Ahluwalia, M. 2000. “Economic Performance of States in Post-Reforms Period.” Economic and Political Weekly 6: 1637–48.
Bose, D., R. Jain, and L. Lakshmanan. 2011. “Determinants of Primary Yield Spreads
of States in India: An Econometric Analysis.” Working Paper Series 10, Reserve
Bank of India, Mumbai.
141
142
Until Debt Do Us Part
Canuto, O., and L. Liu. 2010. “Subnational Debt Finance and the Global Financial
Crisis.” Economic Premise 13, World Bank, Washington, DC.
Corbacho, A., and G. Schwartz. 2007. “Fiscal Responsibility Laws.” In Promoting Fiscal Discipline, ed. M. S. Kumar and T. Ter-Minassian, 58–77. Washington, DC:
International Monetary Fund.
Dholakia, R. K, T. T. Ram Mohan, and N. Karan. 2004. “Fiscal Sustainability of Debt
of States.” Report submitted to the Twelfth Finance Commission, Indian Institute of Management, Ahmedabad, India.
Government of India. 1952. Report of the First Finance Commission. New Delhi:
Government of India.
———. 1988. Report of the Ninth Finance Commission. New Delhi: Government of
India.
———. 1993. Report of the Tenth Finance Commission. New Delhi: Government of
India.
———. 1999. Report of the Eleventh Finance Commission. New Delhi: Government
of India.
———. 2004. Report of the Twelfth Finance Commission. New Delhi: Government
of India.
———. 2009. Report of the Thirteenth Finance Commission. New Delhi: Government of India.
———. 2011. India Public Finance Statistics 2010–2011. New Delhi: Government of
India.
Howes, S., and S. Jha. 2004. “State Finances in India: Towards Fiscal Responsibility.”
In Fiscal Policies and Sustainable Growth in India, ed. Edgardo M. Favaro and
Ashok K. Lahiri, 258–81. New Delhi: Oxford University Press.
Howes, S., A. Lahiri, and N. Stern, eds. 2003. State-level Reforms in India: Towards
More Effective Government. Delhi: Macmillan.
Ianchovichina, E., L. Liu, and M. Nagarajan. 2007. “Subnational Fiscal Sustainability Analysis: What Can We Learn from Tamil Nadu?” Economic and Political
Weekly 42 (52): 111–19.
Kannan, R. K., S. M. Pillai, R. Kaushaliya, and Jai Chander. 2004. “Finance Commission Awards and Fiscal Stability in States.” Economic and Political Weekly 39 (5):
477–91.
Kishore, A., and A. Prasad. 2007. “Indian Subnational Finances: Recent
Performance.” IMF Working Paper 07/205, International Monetary Fund,
Washington, DC.
Kopits, G. 2001. “Fiscal Policy Rules for India?” Economic and Political Weekly 36 (9):
749–56.
Kurien, N. J. 1999. “State Government Finances: A Survey of Recent Trends.” Economic and Political Weekly 34 (10): 1115–25.
Lemmen, J. 1999. “Managing Government Default Risk in Federal States.” Special
Paper 11, Financial Markets Group, Nijmegen, the Netherlands.
Managing State Debt and Ensuring Solvency: The Indian Experience
Ley, E. 2010. “Fiscal (and External) Sustainability.” World Bank, Washington, DC.
http://econpapers.repec.org/paper/pramprapa/13693.htmhttp://mpra.ub.unimuenchen.de/23956/.
Liu, L., and M. Waibel. 2008. “Subnational Insolvency: Cross-country Experiences and
Lessons.” Policy Research Working Paper 4496, World Bank, Washington, DC.
Liu, L., and S. Webb. 2011. “Laws for Fiscal Responsibility for Subnational Discipline.” Policy Research Paper 5587, World Bank, Washington, DC.
Milan, B. 2011. “Fiscal Policy for Growth and Development in India: A Review.”
Forthcoming. PREM Network Fiscal Policy for Growth and Development,
World Bank, Washington, DC.
McCarten, W. J. 2001. “The Challenge of Fiscal Discipline in the Indian States.” In
Decentralization and Hard Budget Constraints, ed. J. Rodden, G. Eskeland, and
J. Litvack. Cambridge, MA: MIT Press.
Mohan, R. 2000. “Fiscal Correction for Economic Growth: Data Analysis and Suggestions.” Economic and Political Weekly 35 (24): 2027–36.
Pinto, B., and F. Zahir. 2004. “India: Why Fiscal Adjustment Now.” Policy Research
Paper 3230, World Bank, Washington, DC.
Poterba, J., and K. Rueben. 1999. “State Fiscal Institutions and the U.S. Municipal
Bond Market.” Working Paper 6237, National Bureau of Economic Research,
Cambridge, MA.
Prasad A., R. Goyal, and A. Prakash. 2004. “States Debt and Debt Relief.” Economic
and Political Weekly 39 (26): 2726–36.
Rangarajan, C., and D. K. Srivastava. 2008. “Reforming India’s Fiscal Transfer System: Resolving Vertical and Horizontal Imbalances.” Economic and Political
Weekly 43 (23): 47–60.
———. 2011. Federalism and Fiscal Transfers in India. New Delhi: Oxford University Press.
Rao, H., A. Prasad, and A. Gupta. 2001. “A Study of State Public Accounts in India.”
Discussion Paper 1, Reserve Bank of India, Mumbai.
Rao, M. G. 2002. “State Finances in India: Issues and Challenges.” Economic and
Political Weekly 37 (32): 3261–71.
Rastogi, A. 2004. “The Infrastructure Sector in India.” In India Infrastructure Report
2004, ed. S. Morris, 23–52. New Delhi: Oxford University Press.
RBI (Reserve Bank of India). 2002. Report of the Group to Assess the Fiscal Risks of
State Government Guarantees. Mumbai: RBI.
———. 2003. Report on Currency and Finance, 2001–02. Mumbai: RBI.
———. 2004. Annual Report, 2003–04. Mumbai: RBI.
———. 2007. Handbook of Statistics on State Finances and Study of State Budget.
Several Issues. Mumbai: RBI.
———. 2011a. Annual Report 2010–11. Mumbai: RBI.
———. 2011b. Report of the Committee on the Comprehensive Review of the National
Small Savings Fund. Mumbai: RBI.
143
144
Until Debt Do Us Part
———. 2011c. “RBI Deregulation of Savings Bank Deposit Interest Rate.” Discussion Paper, RBI, Mumbai.
Reddy, Y. V. 2000. “Fiscal Reforms at the State Level: Review and Prospects.” Address
presented at the India States Reform Forum, organized by the World Bank,
New Delhi, November 24.
———. 2011. “Subnational Finance: Indian Experience.” Address presented at the
Sub-Sovereign Finance Forum, “Debt, Insolvency and Market: Lessons Learned
and Emerging Issues,” World Bank, Washington, DC, June 20.
Schuknecht, L., J. von Hagen, and G. Wolswijk. 2009. “Government Risk Premiums
in the Bond Market EMU and Canada.” European Journal of Political Economy
84 (1): 371–84.
World Bank. 2003a. “India, Rural Access to Electricity.” World Bank office, New
Delhi.
———. 2003b. India Sustaining Reform, Reducing Poverty. New Delhi: World Bank.
———. 2005. State Fiscal Reform in India: Progress and Prospects. New Delhi:
Macmillan.
———. 2011. World Bank India Economic Update. Washington, DC, September.
4
Subnational Debt Management in
Mexico: A Tale of Two Crises
Ernesto Revilla
Introduction
Mexico has experienced two major macroeconomic crises in the last
two decades. The 1994–95 Tequila Crisis and the 2008–09 global financial crisis had important implications for the functioning of subnational debt markets. In the Tequila Crisis, the macroeconomic shock
affected subnationals through higher interest rates on their debt and the
simultaneous reduction in their federal transfers. These shocks made
their debt unsustainable, and the federal government intervened with
an ambitious restructuring program. In 2009, the global crisis did not
affect interest rates in Mexico, but the slowdown in economic activity
and the decline in the price of oil reduced federal transfers to subnationals considerably, dramatically affecting their repayment capacity. In this
context, an innovative mechanism was designed to smooth the shock
and ensure the sustainability of local public finances. These episodes
hold important lessons for policy makers interested in designing debt
management mechanisms for subnational debt in developing countries.
They also shed light on the behavior of subnational debt markets in
periods of stress, and policy responses that can be used in dealing with
recovery during a crisis.
145
146
Until Debt Do Us Part
This chapter contributes to the study of the Mexican Fiscal Federalism Framework in Mexico. While Giugale and Webb (2000) and Revilla
(2012) presented general overviews of Mexican intergovernmental relations, this chapter is part of a new wave of efforts to study aspects of
intergovernmental fiscal relations in the country. In particular, it adds
to the few studies that have been done on subnational debt in Mexico.
Among these, Giugale, Korobow, and Webb (2000) describe the “new
subnational regulatory framework in Mexico,” in place since 2000. In
addition, Hernández, Díaz-Cayeros, and Gamboa (2002a) study the
determinants and consequences of the 1995 bailout, while Giugale,
Hernández, and Oliveira (2000) give an overall overview of the subnational debt market at the dawn of the century.
This chapter more closely relates to the literature on subnational
debt restructuring, as in Liu and Waibel (2009); Prasad, Goyal, and
Prakash (2004); and Ter-Minassian and Craig (1997). In particular, it
adds a developing country dimension to those studies of subnational
debt markets after the global crises, such as the ones in Canuto and Liu
(2010a, 2010b). Together with the other chapters in this volume, this
chapter sheds light on the very difficult questions and dilemmas that
policy makers face when dealing with subnational debt markets after
macroeconomic crises, especially in developing countries.
The chapter is structured as follows. Section two describes the
Mexican fiscal federalism framework. Section three describes the
restructuring of subnational debt after the Tequila Crisis. Section four
describes the response after the global crisis. Section five discusses the
similarities, differences, lessons, and conclusions for subnational debt
management.
The Fiscal Federalism Framework in Mexico
All intergovernmental fiscal relations systems are different and constrained by the local culture, politics, and the economics of the institutional setup. Mexico’s fiscal federalism is defined by a very large vertical
imbalance, an enormous dependence of subnationals on federal transfers, and on a low level of subnational debt, all of which influenced the
objectives and constraints of the policies implemented during the two
crises discussed in this chapter.
Subnational Debt Management in Mexico: A Tale of Two Crises
Mexico is a federal country divided into 31 sovereign states and one
federal district. Each state is composed of municipalities, which are the
basic political unit and which have some sovereign autonomy over their
political and fiscal development. Being political subdivisions of states,
municipalities are extremely heterogeneous in their level of development.1 The fiscal federalism framework in this three-tier government
structure consists of the set of laws, rules, and institutions that allocate
spending and tax responsibilities, and the transfers and the institutional
framework for subnational debt.
A salient feature of Mexico’s fiscal federalism framework is the strong
dependence of state and municipality finances on federal transfers
(see, for example, Giugale and Webb [2000], Revilla [2012], and references therein). On average, the share of resources from federal sources
accounts for 85 percent of total revenues for subnationals. This strong
dependence on federal resources has remained mostly constant over
time. This feature of Mexico’s federalism is the main characteristic that
determines the politics and economics of current and future reforms on
the subject.
Table 4.1 shows the composition of the annual flow of resources
to subnationals in Mexico. As can be seen, around 85 percent of revenues for states and municipalities come from federal transfers, around
11 percent from own-source revenues, and borrowing accounts for only
5 percent of annual flow, on average. Federal transfers can be grouped
into three main channels: (a) earmarked transfers; (b) nonearmarked
transfers; and (c) a smaller, but growing, component of new transfers
for specific purposes and infrastructure.2
Nonearmarked transfers, called participaciones, are the biggest item
in states’ budgets and the biggest line item in the federal budget. They
consist of a set of funds that vary in size and composition. The biggest
one accounts for 86 percent of the total and is called the General Participation Fund (Fondo General de Participaciones). It is calculated as
20 percent of a federal pool of revenues that are shared3 and distributed according to a formula that correlates per capita transfers to the
level of economic activity (measured by growth in states’ gross domestic product [GDP]) while giving incentives to increase own-source
revenue.4 The rest of the funds are smaller and include a fund for
municipalities, one to give incentives to improve tax administration
147
148
Until Debt Do Us Part
Table 4.1 Subnationals’a Resources in Mexico
billion pesos 2006–10
2006
2007
2008
2009
2010 Compositionb (%)
895
911
1,080
1,100
1,169
100
8.9
785
781
943
936
973
83
7.4
Nonearmarked transfers
(participaciones)
329
333
423
421
437
37
3.3
Earmarked transfers
(aportaciones)c
388
379
420
439
461
39
3.5
Total resources
Federal transfers
As % of GDPb
Other
68
70
99
76
74
6
0.6
Convenios
44
56
73
76
74
6
0.6
Excedentes
24
13
26
—
—
0
0.0
Own-source revenue
93
103
120
116
133
11
1.0
Financinge
17
26
17
49
63
5
0.5
2,264 2,486
2,861
2,817
2,960
100
23
1,358
1,508
1,492
50
11
d
Memorandum items
Federal budget
Federal nonoil revenue
1,015
1,205
Sources: Ministry of Finance; Mexico and states’ public accounts.
Note: — = not available, GDP = gross domestic product.
a. States and municipalities.
b. Values correspond to 2010.
c. Includes resources for education expenses in the Federal District, where education has not been decentralized.
d. Includes special decentralization agreements and excess revenue surplus.
e. Corresponds to registered debt with the Ministry of Finance and includes all debt approved by local legislatures; it does
not include short-term loans (for cash management) or contingent liabilities (that is, pensions).
at the local level, compensatory funds for states where oil is extracted,
and a redistributive fund for the 10 poorest states. For a complete
description, see table 4A.1.
Earmarked transfers, called aportaciones, consist of eight funds and
are itemized in “Ramo 33”5 of the federal budget. There are special
funds for education, health, social development, and public security.
The biggest one is the Fund for Basic Education, which accounts for
59 percent of the total. This fund, and the Fund for Health Services
(Fondo de Aportaciones para los Servicios de Salud), is meant to cover
the wage bill for paying teachers and medical professionals who were
transferred to states in the 1990s with the decentralization of education
and health. There is widespread agreement that the large amount of
money spent through these funds has not contributed to more efficient
Subnational Debt Management in Mexico: A Tale of Two Crises
service delivery, that the assignments among states are extremely
inefficient, and that there is some level of corruption in the spending
of these resources (see IMCO 2010). For a complete description of earmarked transfers, see table 4A.2.
Regarding own-source revenues of subnationals in Mexico, the first
salient fact is the low level of own tax effort by states and municipalities.
The level of subnational own revenue is low by international standards
and relative to its potential. The main tax handle of municipalities is the
property tax. Mexican municipalities collect 0.2 percent of GDP. This figure is the lowest in Latin America (Bolivia collects 0.3 percent of GDP,
Brazil 0.7 percent, Colombia 1.3 percent, and Argentina 1.7 p
ercent)
and one of the lowest in the world (the Organisation for Economic Cooperation and Development average is 2 percent of GDP) (ECLAC 2009;
OECD 2010). In practical terms, only the Federal District and some big
municipalities collect the property tax efficiently, and the vast majority of
local governments in the country do not collect it at all. This remains one
of the biggest challenges for the Mexican fiscal federalism framework.6
For states, the main taxes are the payroll tax7 and the administration of federal taxes on vehicles and gasoline, from which the states
are allowed to keep the revenue. States have other local taxes such as a
lodging tax (important in states with high rates of tourism), taxes on
the use of old motor vehicles, and taxes on local lotteries and games.
However, the revenue collected from these taxes does not represent significant resources. For details on states’ local revenues in Mexico, see
table 4A.3.
Subnational debt in Mexico is low by all accounts and relative to international standards; the stock of subnational debt in Mexico accounts
for 79 percent of annual nonearmarked transfers, or 2.9 percentof
GDP. Figure 4.1 shows the stock of debt for subnationals in Mexico in
2011. Although subnational debt as a share of nonearmarked transfers
increased from 50.7 percent in 2008 to 79.2 percent in 2011—the debt
as share of GDP increased from 1.7 percent in 2008 to 2.9 percent in
2011—subnational debt in Mexico is low when comparing with countries such as Brazil, China, and India.8
The debt in figure 4.1 includes all direct liabilities that are incurred
by subnationals that are registered with the Federal Ministry of Finance
and that were approved by their local legislatures. It does not include
149
150
Until Debt Do Us Part
Figure 4.1 Subnational Debt in Mexico, by State, 2011
a. As a share of nonearmarked transfers
300
250
Percent
200
150
100
Average: 79 percent
50
0
ila oo ón ua rit án ra ral uz nia tal co as as go ma co es as ur osí la to oa ca los go ro án ro he co ala
hu a R Le uah aya oac onoede racrifor To Jalis hiap ulip ran oli éxi ientatec ia SPot uebajua inal axaore idal erre cat rétapec bas axc
a
C ma Du C M cal ac ornuis P an S O M H Gu Yu ue am Ta Tl
Contan evo hih N ich Sto F Ve Cal
s Z lif L
Q C
u
i
C
i
M
Ta
a
r
Gu
j
a n
u
ua
st Ba
Q N
Ag ja C Sa
Di
a
B
b. As a share of state’s GDP
9
8
7
Percent
6
5
4
3
Average: 2.9 percent
2
1
0
ila rit oo as án uz ra ua ón as co a nia tal co go ca as ral ur es la oa osí los go ro to án ro co he ala
hu aya a R hiap oac racr onouah Le atec éxi olimifor To Jalis ran axa ulipede ia S ient uebinal Pot ore idal erreajua cat réta bas pec axc
a
l
l
Du Oamato F fornsca P S uis M H Gu uan Yu ue Ta am T
Co Nntan C ich Ve S hih evoZac M CCal
Q
C Nu
L
C
M
T tri ali ua
G
ja
ui
n
a
s
g
C
Q
B
Sa
Di aja A
B
Source: Ministry of Finance, Mexico.
Note: GDP = gross domestic product. Data include municipal debt.
Subnational Debt Management in Mexico: A Tale of Two Crises
short-term loans (incurred and paid in full within the fiscal year and
used mostly for cash management), or contingent liabilities such as
pensions or supplier’s credit, both of which can pose risks. If shortterm debt is not officially registered as debt, it could potentially be
used to finance current expenditures. In addition, short-term debt was
exempted from a higher risk rating and the need to establish prudential
reserves. Although state retirement plans represent only 2.3 percent of
total retirement accounts in the country, there is a risk that they might
become unsustainable in the next decade.
There is a great diversity in terms of the structure and financial sustainability of state retirement schemes. A majority of state retirement
plans operate as defined benefit plans, which are, in general, unfunded
liabilities of state governments. There are no recent studies that estimate the amount of these liabilities. However, according to the conclusions of a meeting of the National Fiscal Convention (Convención
Nacional Hacendaria), Hewitt Associates estimated that states’ pensions
in 1998 accounted for around 25 percent of GDP. Nearly one-third of
state retirement plans operate as funded defined benefit plans, but only
7 percent of states have defined contribution schemes based on individual retirement accounts.
As can be seen in figure 4.1, the stock of subnational debt in Mexico
is only 79 percent of annual nonearmarked transfers, or 2.9 percent of
GDP. At first glance, this low level of debt represents a puzzle from the
point of view of economic theory, given the shocks that subnationals faced during the crisis and the need for infrastructure investment.9
Some observers of the Mexican fiscal federalism framework have
concluded that some kind of hidden bailouts must exist in Mexico
simultaneously with incentives for subnationals to rent-seek from the
federation as an instrument to smooth fiscal shocks and close year-end
budgets (see Hernández, Díaz-Cayeros, and Gamboa 2002a, 2002b).
These grants would make debt unnecessary as a mechanism to balance
the fiscal accounts.
Table 4.2 shows the structure of the stock of subnational debt in
Mexico. The total stock of subnational debt is collateralized with federal
transfers or with a future flow of local taxes. That is, no lender gives an
unsecured loan to a subnational in Mexico. Three-quarters of the total
stock of state and municipality debt in Mexico has federal transfers as
151
152
Until Debt Do Us Part
Table 4.2 Subnational Debt Structure in Mexico, 2011
billion pesos, and percentage
Collateral
Creditor
Federal transfers
a
Own revenue
Total
% of GDP
Commercial banks
186
15
201
1.5
Development banks
86
6
92
0.7
Securitizations
18
40
58
0.4
Trust funds
8
11
19
0.1
Other
19
2
21
0.2
Total
317
74
391
2.9
% of GDP
2.4
0.6
2.9
b
Source: Ministry of Finance, Mexico.
Note: GDP = gross domestic product.
a. Includes debt collateralized with nonearmarked transfers (participaciones) and with FAIS and FAFEF, which
are earmarked funds (aportaciones) that may be collateralized; see table 4A.2.
b. Includes Sofoles (Limited Purpose Financial Institutions), Sofomes (Multiple Purpose Financial Institutions),
and suppliers.
collateral. The income pledged is usually the nonearmarked transfers,
but some earmarked funds are starting to be used as well.10 As regards
the creditors, private commercial banks hold 51 percent of the debt,
government development banks hold 24 percent, and the rest is placements with the markets (mainly bonds or securitized notes). Since 2005,
this composition has remained almost unchanged.
In Mexico, the institutional framework for subnational debt starts
with the 1917 Constitution, which mandates a “golden rule” for state
and municipal debt: all indebtedness must be used to finance “productive public investments.” What this means in practice (and whether it
includes modern debt operations such as refinancing or debt buybacks) has been the subject of much debate among lawyers, including
the Supreme Court, state treasuries, and investment bankers who advise
subnationals on the flexibility of the constitutional rule.11 The Constitution also prohibits states and municipalities from borrowing in foreign
currency or from a foreign creditor.12
Below the constitutional level, Mexico’s subnational debt framework
was reformed in 2000. The old framework was based on the concept
of the mandato (mandates). This meant that the federal government
Subnational Debt Management in Mexico: A Tale of Two Crises
acted as a trustee in servicing state debt that had been collateralized
with participaciones.13 What happened in practice was that the mandato
was perceived by the markets as a guarantee by the federal government
on subnational debt. Not surprisingly, as argued in Giugale, Korobow,
and Webb (2000), this perception of a federal bailout created two problems: (a) banks had the incentive to make loans, since they perceived
them to be risk free; and (b) subnationals also had the expectation of
a bailout since it was not credible that the federal government would
in fact reduce transfers.14 To eliminate these problems, several reforms
were implemented from 1997 to 2000 (for a detailed account of these
reforms, see Guigale, Hernández, and Oliveira [2000]).
The reforms regarding the new regulatory framework for subnational debt, in place since 2000, were based on two main concepts:
an explicit renunciation of federal bailouts and a new system aimed
toward a correct evaluation by lenders of idiosyncratic subnational risk. These objectives were pursued through (a) the elimination of the mandatos; (b) establishment of a link between the capital
risk weighting of bank loans to subnationals and their credit rating;
(c) and a requirement to register subnational loans with the Ministry of Finance, conditional on being current on fiscal transparency
requirements.15
Ten years after the establishment of the new regulatory structure,
it can be said that Mexico’s subnational debt framework is more of a
hybrid between a rules-based and a market-based system. Indeed, it can
be described as a quasi-market-based system that rests on the following
three distinct characteristics.16
The first characteristic is the credible threat of no federal bailout. This
was accomplished with the elimination of the mandato (the instruction
that subnationals gave to the federal government to service their debt
for them, out of their transfers) and the creation of intercepts (which,
in practice, are set up as trust funds established by the subnational and
their creditors).
The second characteristic is the increased transparency of the subnational debt market. All collateralized debt must be registered with the
Ministry of Finance (conditional on having been approved by the local
congress, and the state being up-to-date in transparency requirements).
If it is not registered, then the loan is automatically risk weighted by
153
154
Until Debt Do Us Part
regulators at the penalty rate of 150 percent, which not only raises the
cost of the loan directly but also makes the bank credit committees
reluctant to lend at all. States have found that there is a strong incentive
to register loans that are not legally required to be registered, since this
often results in better credit conditions from the lenders. Therefore, the
Mexican registry of subnational public debt is quite accurate in listing
all outstanding claims.17 This process has resulted in increased transparency of the Mexican subnational debt market. Thus, the general public
and opposition parties have imposed a certain amount of fiscal discipline on local governments with this mechanism.
The third characteristic that defines the regulatory regime is that
many of the constraints on the market are the result of the prudential regulation of banks, rather than the result of direct fiscal rules on
subnationals.18 In particular, a capital risk weight is assigned to loans
to subnationals depending on the credit rating of the loan. Therefore,
the pricing of credit should be a function of the creditworthiness of the
state or municipality. Almost all of them now get credit ratings, since
not having a credit rating also leads to the penalty capital weighting of
150 percent.19
The combination of the described rules and mechanisms implemented in Mexico has ensured an orderly and functional subnational
debt market. Notwithstanding the low stock of subnational debt, as
in any comparison among Mexican states, there is a wide heterogeneity across states in their indebtedness level, and some states continue to
face fiscal adjustment challenges.20 Nonetheless, the level of subnational
debt does not appear to pose a significant systemic or macroeconomic
problem.
In fact, the relevant policy question might very well be the opposite: is subnational debt in Mexico suboptimal, given increased needs
for development and infrastructure? The answer is beyond the scope of
this chapter. However, the low amount of subnational debt in Mexico
(as a percentage of GDP) is the second salient fact of its fiscal federalism
framework, and frames the policy responses that were taken under the
extreme macroeconomic shocks suffered during the 1994–95 Tequila
Crisis and more recently with the “great recession” of 2008–09. The different policy responses regarding the safeguarding of the subnational
debt market are detailed in the following two sections.
Subnational Debt Management in Mexico: A Tale of Two Crises
The 1994–95 Tequila Crisis and the
Restructuring of Subnational Debt
For Mexico, 1994 was a disastrous year. It included the assassinations of
the official party’s presidential candidate and of its leader, the rise of an
armed insurrection in the southern state of Chiapas, and the continual
deterioration of foreign investors’ perceptions. With a fixed exchange
rate, these events led to a massive run on foreign reserves. On December 19, 1994, Mexico suffered one of its greatest macroeconomic shocks
in its history when the fixed exchange rate regime was abandoned. In
1995, the GDP dropped 6.2 percent in real terms compared to the previous year. Inflation reached 52 percent that same year. The Mexican
peso lost 49.7 percent of its value in December 1994, and throughout
1995 the currency depreciated an additional 49 percent. The nominal value of the exchange rate, which was 3.4 pesos per dollar at the
beginning of December 1994, reached 7.7 pesos per dollar by the end
of December 1995. International reserves at the central bank dropped
from US$30 billion at the beginning of 1994 to only US$6 billion in
December 1994. The impact on interest rates was astounding, as well:
interest rates of a one-month Treasury bill reached more than 80 percent
during 1995.
The crisis brought painful costs in terms of increased poverty, a costly
bank restructuring, and a difficult economic environment for firms and
families. It is no surprise that under these conditions, states and municipalities faced dire financial circumstances, and, given that their debt was
mostly contracted at a variable rate, their obligations became unsustainable overnight. Figure 4.2 shows the interest rate of a one-month
Treasury bill, and the impact of the crisis on GDP and consumption,
employment, and the exchange rate.
Subnationals faced two main direct shocks that made them unable
to service their debts. First, the extraordinary rise in interest rates made
their debt untenable, since most of it was contracted at variable rates.
Second, given that the main source of income was the participaciones
(which fluctuate with the federal taxes that are shared), the impact
of the crisis on federal revenue implied that in 1995, federal nonearmarked transfers were 22 percent lower in real terms than in 1994.
With lower income sources and higher interest payments, the specter
155
a. GDP and consumption growth
in Mexico, 1992–95
b. Unemployment rate, 1993–95
8
7
Percent
Percent
8
4
0
–4
–8
–12
–16
6
5
4
2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5
–9 –9 l–9 –9 –9 –9 l–9 –9 –9 –9 –9 –9 –9 –9 l–9 –9
aJ n Apr Ju Oct Jan Apr Ju Oct Jan Apr Jul Oct Jan Apr Ju Oct
GDP
3
3 3 3 3 4 4 4 4 5 5 5 5
–9 r–9 l–9 t–9 –9 r–9 l–9 t–9 –9 r–9 l–9 t–9
Jan Ap Ju Oc Jan Ap Ju Oc Jan Ap Ju Oc
Consumption
c. Mexican peso, 1994–96
d. Interest rate of a 28-day Cete, 1994–96
7
5
Percent
9
Pesos per dollar
156
Figure 4.2 The Macroeconomic Impact of the 1994–95 Tequila Crisis in Mexico
3
4 4 4 4 4 5 5 5 5 5 6 6 6 6 6 6
–9 r–9 –9 –9 t–9 –9 r–9 y–9 g–9 t–9 –9 r–9 –9 –9 t–9 –9
Jan Ma May Aug Oc Jan Ma Ma Au Oc Jan Ma May Aug Oc Dec
Source: Banco de México.
Note: GDP = gross domestic product. A Cete is a credit title issued by the federal government.
100
80
60
40
20
0
4 4 4 4 5 5 5 5 6 6 6 6
–9 r–9 l–9 t–9 –9 r–9 l–9 t–9 –9 r–9 l–9 t–9
Jan Ap Ju Oc Jan Ap Ju Oc Jan Ap Ju Oc
Subnational Debt Management in Mexico: A Tale of Two Crises
of default loomed larger. In this context, the federal government intervened to engineer an important restructuring process that was based on
the following four main pillars. The following series of interventions
did not occur as a single event, but were spread over the recovery period
of the crisis.
First, the federal program included a direct restructuring mechanism. In this way, the federal government, through the Ministry of
Finance, restructured around 90 percent of the outstanding subnational
debt (in an amount equivalent to US$8 billion at 2009 prices). The
restructuring lowered the interest rate to a fixed 10.5 percent n
ominal
rate and increased the maturity from an average of 6.6 years (see
Fedelino and Ter-Minassian 2010) to 15 and 20 years. It was structured
by Banobras, the federal government’s development bank that lends to
subnational governments.
Second, to help states deal with the decrease in federal transfers (participaciones) caused by the lower federal collection of shared taxes, the
federal government gave an extraordinary transfer to all states in 1995
and 1996 of approximately US$1 billion (2009 prices) for each year,
equivalent to 10 percent of annual transfers.
Third, the federal government, again through Banobras, engineered
an extraordinary loan for states collateralized with nonearmarked
transfers. The loan was equivalent to US$500 million (2009 prices) or
5 percent of annual transfers. It would be paid out of one-year transfers
and at the federal government’s cost of financing.
Fourth, the federal government resorted to extraordinary discretional transfers to some states that were negotiated independently and
usually not reported. By definition, this “hidden bailout” is difficult to
quantify because there are no data and it does not appear in traditional
accounting or reports of subnationals. However, Hernández, DíazCayeros, and Gamboa (2002b) try to quantify these “secret” transfers21
using reductions in debt stocks that are unmatched by state government
surpluses, and differences in interest rates before and after debt renegotiations, since interest rates negotiated after the crisis varied among
states. Hernández, Díaz-Cayeros, and Gamboa also argue that some of
the new credit obtained via official development banks was used for
current expenditures and not investment (as the law mandates), which
would amount to an indirect bailout. Finally, when considering the
157
158
Until Debt Do Us Part
determinants of these hidden bailouts, they find that the size of the bailout was related to the size of the state and to the previous level of fiscal
indiscipline (with states that had bigger deficits getting more support),
but not to political variables.
The bailout worked in preventing the meltdown of subnational debt
markets, thus preserving the functioning of local governments and service delivery. In studying the consequences of the bailout (both the open
and hidden parts), Hernández, Díaz-Cayeros, and Gamboa (2002b) find
two important consequences. First, there were distributional effects,
with higher per capita extraordinary transfers given to states with higher
per capita GDP. Hence, poorer states (less indebted) received less in
extraordinary support. Second, as with any bailout, some moral hazard
problems were created, since the bailout did not resolve structural fiscal imbalances. It consisted basically of a one-year relief program. After
the crisis, subnational governments kept incurring deficits because they
anticipated they would be bailed out again.
These special bailouts came from a large discretionary account for the
presidency, which had traditionally been in the budget. After the ruling
party lost control of Congress in 1997, however, this practice stopped,
and that contributed to the decision to move to the hybrid rules- and
market-based system described earlier.
The “Great Recession” of 2008–09 and
Subnational Debt in Mexico
As in most countries, the global crisis of 2008–09 caused deep macroeconomic management problems for Mexico. The impact was severe:
growth slowed to a painful minus 6.1 percent in 2009, and the public
finances of all levels of government suffered accordingly. However, a
few things had changed since the Tequila Crisis. One decade of sound
macroeconomic management that achieved much needed fiscal and
monetary space, combined with a different transmission channel and
the external origin of the crisis, produced a very different effect on the
subnational debt market. Table 4.3 summarizes the similarities and differences of both crises.
What was fundamentally different in the 2008–09 crisis for subnationals was the absence of an interest rate shock. The one-month
Table 4.3 Two Crises: Implications for the Subnational Debt Market in Mexico
Tequila Crisis, 1995
Global financial crisis, 2008–09
Origin
Domestic
Foreign
Cause
Reversion of large capital inflows together with some
financial vulnerabilities:
• Semifixed exchange rate
• Large current account deficit resulting from a huge credit
expansion
• Substantial rise in interest rates in the United States
• Accumulated political tensions during 1994a
•
•
•
•
Global asset price bubbles and low interest rates
Subprime mortgage crisis in the United States
Excessive leveraging leading to serial defaults
Weak regulation of financial markets
Macroeconomic impact
for Mexico
•
•
•
•
•
•
•
•
Currency depreciation of 49 percentd
GDP dropped 6.1 percent in real terms
Inflation of 4 percent
Interest rate fluctuations between 4 and 8 percentc
Impact on Mexican credit
markets
• Complete dry-up of local credit
• Banking crisis
• Dry-up of foreign credit, but less impact on local credit markets since local banks remained strong throughout the crisis
Impact on subnational credit
markets
• Severe dislocation
• States unable to repay debt service because of:
° Higher interest payments
° Less capacity for repayment, as revenues dropped
• Significant effort to contain the impact
• States suffered only through a lower capacity to service
payments, but Rainy Day Funds were used to smooth the
shock.
Currency depreciation of 117 percentb
GDP dropped 6.2 percent in real terms
Inflation exceeded 50 percent
Interest rates reaching 80 percentc
Note: GDP = gross domestic product.
a. Gil-Díaz 1998.
b. From December 1994 to December 1995.
c. Rate on one-month Treasury bill.
d. Maximum depreciation during 2009. However, the Mexican peso recovered part of its value throughout 2009.
159
160
Until Debt Do Us Part
Treasury bill fluctuated from 4 to 8 percent from 2008 to 2010. This is
in contrast to what happened during the Tequila Crisis, when the rate
increased from 10 percent at the beginning of 1994 to above 80 percent
by the first quarter of 1995. This meant that there was no immediate
increase in the cost of servicing the debt for states and municipalities. The shock, however, came through a different set of channels that
affected the revenue of subnational governments and, hence, the possibility of servicing that debt.
First, a dramatic decrease in the price of oil meant significantly lower
oil revenues, which are shared among levels of government.22 In 2009,
federal government oil revenue dropped 32 percent in real terms. Second, the slowdown of economic activity implied a significant reduction
in federal tax revenue, which is also shared. Federal nonoil tax revenue
during 2009 decreased 10.5 percent in real terms. The combined impact
of these shocks on federal revenue meant significantly reduced transfers
for subnationals. Without the use of Rainy Day Funds (RDFs) (see discussion below), transfers in 2009 would have decreased 15 percent in real
terms. Given their almost complete dependence on federal resources, this
implied a momentous reduction in their capacity to service their debt
and finance government operations. In the absence of federal intervention, many states would have defaulted on their debt. Figure 4.3 shows
the deterioration in subnational credit ratings, and therefore on credit
conditions, during the crisis.23
Under this scenario, the federal government could have provided a
direct bailout of the states via extraordinary transfers or a combination
of the mechanisms discussed in the previous section, or it could have
forced or been instrumental in a system-wide restructuring of subnational debt. The solution, however, came from a different and innovative mechanism: the coordinated sale of future federal surplus revenues
that belong to the states.
The coordinated, collective mechanism has been developed to
smooth the shock on local public finance. The Mexican macroeconomic management framework was significantly improved in 2005
with the approval of a Federal Fiscal Responsibility Law. Among
other things, the law mandated a balanced federal government budget
and created RDFs for the federation and for subnationals. The funding of the RDFs was through annual federal surplus revenue (both oil
Subnational Debt Management in Mexico: A Tale of Two Crises
161
Figure 4.3 Deteriorating Subnational Credit Scores in Mexico during the
Global Financial Crisis
Number of increases in subnational ratings minus decreases
15
10
5
0
–5
–10
–15
05 06 06 06 06 06 06 07 07 07 07 07 07 08 08 08 08 08 08 09 09 09 09 09 09 10 10 10 10
c– b– r– n– g– ct– c– b– r– n– g– ct– c– b– r– n– g– ct– c– b– r– n– g– ct– c– b– pr– n– ug–
De Fe Ap Ju Au O De Fe Ap Ju Au O De Fe Ap Ju Au O De Fe Ap Ju Au O De Fe A Ju A
Sources: Fitch Ratings; Standard & Poor’s.
Note: Increases in graph (+1 for each positive change in rating), decreases (-1 for each negative change in rating), changes
in economic outlook (+/-0.25).
and federal tax) when, in any given year, receipts exceed the p
rogram.
Although the size of the funds in terms of GDP was small (reaching 0.7 percent of GDP for the federal RDF, and 0.2 percent of GDP
for the subnational RDF), by 2008 the federal fund had accumulated
86 billion pesos and the subnational fund—the Fund for the Stabilization
of the Federal Revenue for the Federal Entities (Fondo de Estabilización
de los Ingresos de las Entidades Federativas, FEIEF)—had accumulated
25 billion pesos.24 The funds were designed to be used to smooth out
temporary decreases in federal revenues, which was the case in 2009.
To understand the size of the macroeconomic shock for Mexico
caused by the global crisis, consider the difference between expectations
for 2009, formed in the fall of 2008, as reflected in the macroeconomic
forecasts included in the budget, and the observed data for the close of
that fiscal year. The federal budget for 2009 included both a real GDP
162
Until Debt Do Us Part
growth forecast of 1.8 percent and an average price of oil of US$70
per barrel for the year. When 2009 ended, growth was a full 6 percentage points lower, while the price of oil averaged US$53 per barrel. This
implied a gap of 480 billion pesos in the federal government balance,25
and a reduction in federal nonearmarked transfers for subnationals of
70 billion pesos (or 15 percent) relative to the budget. In the absence
of a smoothing mechanism, chaos would have ensured for states and
municipalities. As in other parts of the world,26 services would have to
be cut dramatically; taxes raised; subnational workers would have been
laid off, with the associated political cost; and defaults would have been
inevitable.
The first line of defense to smooth the decrease in federal transfers
and prevent problems in the subnational debt market was to use savings
in the state’s RDF, the above-mentioned FEIEF. Soon it became clear that
the entire available balance in the fund (25 billion pesos) would not be
enough to cover the expected decrease in transfers for that year (a gap of
70 billion pesos between state aggregate budget transfers and expected
transfers was projected by June 2009). The federal government, under
pressure from states and municipalities, and under financial stress of its
own, was considering more traditional avenues for restructuring subnational obligations as described in the previous section, to close the
projected gap: a generalized extraordinary transfer, a restructuring of
subnational debt to lower payments, and giving much needed space to
local treasuries, or a direct loan to states. They all had their drawbacks.
An extraordinary transfer would put additional pressure on the federal government’s finances, and would completely shift the cost of the
crisis onto the federation. A restructuring of subnational debt would
have been difficult to achieve given the decentralized nature of the
market, the heterogeneity of lenders, and the diverse exposure of states.
A direct loan by the federation to subnationals had the disadvantage
that the federal government would have to put the asset on its books at a
time when its fiscal position was weak—in relative terms—and the loan
would have had to be standard in the sense that each state would have
had to get local legislative approval (a difficult process that was complicated by federal and local politics and slow and difficult timing, and
would not have been successful because some states were already at their
locally established debt limit27). Also, the proceeds from the financing
Subnational Debt Management in Mexico: A Tale of Two Crises
would have had to conform to the constitutional golden rule and be
used for infrastructure.28
Instead of using one of the traditional avenues for restructuring,
the federal government, together with the states, engineered an innovative mechanism that satisfied the following criteria: (a) involvement
of the subnationals’ own balance sheets in the smoothing of shocks;29
(b) giving subnationals a direct substitute of nonearmarked transfers;30
and (c) making it fast, credible, and efficient. The process was as follows:
the federal government used its coordination powers to harmonize the
needs of all subnationals for additional financing and put them on a
path to access the market collectively at a low cost of finance. The specific mechanism used was the leveraging of the RDF for states, that is,
the FEIEF.
Since the FEIEF belongs to the states, and is funded by a future flow
of income (the sequence of future annual excess surplus that corresponds to subnationals), it was an effective vehicle to bring to present
value future resources. Essentially, the correct response to a transitory
fiscal gap is to use debt financing to avoid increasing taxes or reducing
expenditure.31 However, no state by itself would have had access to the
markets, or would have done so at high prices, given the deterioration of
liquidity in the credit markets at the time. The federal government coordinated the states—and municipalities—to agree to the selling of future
flows of their RDF for a present value amount to be received and used as
nonearmarked transfers in 2009.
States and municipalities, through the Mexican National Association of State’s Secretaries of Treasury,32 (Comisión Permanente de
Funcionarios Fiscales) orchestrated the operation with the advice and
coordination assistance of the federal government. The whole structuring process, from initial design to its closing, took four months.
Subnationals obtained 40 billion pesos in the market33 (equivalent
to 10 percent of annual nonearmarked transfers), to be paid back in
13 years (or sooner if the future flows toward the RDF are larger than
expected) at a cost of financing similar to that of the federation—and
about 200 basis points lower than the average cost of finance for subnationals in Mexico.34 This substantial amount of resources almost
completely closed the gap in nonearmarked transfers, bringing it to
minus 2.2 percent (compared to the budget forecast), an astoundingly
163
Until Debt Do Us Part
small shortfall given the worst crisis since the Great Depression in the
1930s. In fact, the federal government had a substantially bigger fiscal
gap to close that year, and for all practical purposes, the Mexican subnationals did not suffer the impact of the global crisis in their finances.
Debt continued to be served on time, and there was no dislocation in
the subnational credit market.
Figure 4.4 compares the fall in nonearmarked transfers in each of
the crises. Whereas the Tequila Crisis reduced nonearmarked transfers
by 22 percent relative to the previous year (and hence the restructuring program described in the previous section was implemented), the
global crisis, in the absence of policy intervention, would have reduced
transfers by 15 percent in 2009 relative to 2008. However, with the
mechanism described (using the subnational RDF, current and future),
transfers were reduced only 5 percent in real terms relative to 2008. This
shock was then easily absorbed by subnational governments.
Figure 4.5 shows the evolution of the expectations of the end-of-year
gap between observed transfers and the budget forecast, during 2009.
Each point on the lines represents the expected gap for 2009 as of the
month indicated. The lower line represents the gap without the use of
the RDF, and the upper line represents the expected gap with the innovative use of the RDF. Several conclusions can be drawn.
Figure 4.4 Fall in Transfers Relative to Previous Year in Mexico during the 1994–95
Tequila Crisis and the 2008–09 Global Financial Crisis
0
1995
2009
w.o. RDF
2009
Actual
Total
RDF
RDF
%, real terms
164
–10
Mechanism
–20
–30
Source: Ministry of Finance, Mexico.
Note: w.o. RDF = without Rainy Day Fund.
Subnational Debt Management in Mexico: A Tale of Two Crises
165
Figure 4.5 Federal Transfers: Evolution of Expectations in Mexico during the
Global Financial Crisis, 2009
Expectation of end-of-year gap, formed in each month
0
–2.2
with
RDF
–5
Percent
–10
–12.6
without
RDF
–15
–20
c
v
De
No
O
ct
p
Se
l
g
Au
Ju
n
Ju
ay
M
r
Ap
ar
M
b
Fe
Jan
–25
Source: Ministry of Finance, Mexico.
Note: RDF = Rainy Day Fund.
First, one can see the evolution of the crisis and how it was worsening
during the first half of the year. At its worst point (June 2009), nonearmarked transfers were expected to be 20 percent lower than what was
forecasted in the budget. This would have been a substantial blow to
subnational governments. Second, in the last half of 2009, there was
a slight recovery in the economy, which was reflected in improved
expectations; but still, without the RDF transfers, it would have been
12.6 percent lower than budgeted. Third, as mentioned, the RDF operation almost closed the gap completely and, by the end of the year,
transfers were only 2.2 percent lower than budgeted.
After the operation, subnational debt markets continued functioning
normally and debt continued to be serviced. As the recovery occurred,
credit conditions gradually improved, beginning in the second quarter
of 2010 (see figure 4.3 earlier). The innovative use of credit markets and
the involvement of the subnational governments’ own balance sheets in
the debt management have contributed to preserving the health and stability of the subnational debt markets. The stability of the subnational
166
Until Debt Do Us Part
debt markets has also been helped by the improved macroeconomic
management in the country prior to the crisis and the turnaround economic growth. In 2010 and 2011, the Mexican economy grew at 5.4 and
3.9 percent, respectively.
Lessons and Conclusions
Subnational debt markets perform essential functions, expanding the
resources of local governments to finance infrastructure and facilitating
the transfer of resources across time to smooth out transitional fiscal
shocks. They also pose risks, particularly if the central government has
to bail out local governments in times of stress.
In reality, however, all debt markets will fail from time to time. That
is why a well-structured regulatory framework needs to take into consideration both the ex-ante rules for getting into debt, and the ex-post
mechanisms to deal with insolvency and restructuring. Governments
will deal with crises constrained by the mechanisms in place, the nature
of the crisis, and the tools available at the time. Learning from other
times and places is of value to add to the toolkit of policy makers,
improve the current set of institutions, and prevent further dislocation
in markets.
Mexico experienced two major macroeconomic crises in the last two
decades, both of which had important bearings on the subnational debt
market. While the two episodes affected local governments substantially,
the policy responses were markedly different and therefore had distinct
consequences. This chapter explored Mexico’s approach to subnational
debt management in each of those crises.
One of the main lessons of the 1995 experience is that if a bailout of
subnational governments is necessary, it should not be addressed exclusively to closing the year-over-year deficits in primary balance. Instead,
the main focus should be solving the structural fiscal imbalances of
states. This means that the expenditure path must be determined by the
expected flow of future income. The federal government should condition the extraordinary transfers to certain results, such as reducing
unnecessary expenses (a “structural adjustment” strategy).
The global crisis introduced a different set of challenges to ensuring
the orderly functioning of the subnational credit market. In this case,
Subnational Debt Management in Mexico: A Tale of Two Crises
since interest rates remained low, the channel affecting subnational
finances was in their repayment capacity because of the lower resources
that states and municipalities had available in 2009. In this case, the federal government did not resort to a traditional bailout or to extraordinary transfers, but used its coordinator function to achieve a more
efficient outcome: the subnational governments got directly involved
to bring to present value the future flow of revenue of their RDF. This
innovative mechanism ensured that states’ own balance sheets were used
to smooth the fiscal shock. Also, the use of the (present and future) RDF
implied that subnational governments in Mexico did not suffer significant fiscal consequences from the global crisis. Given the fiscal consequences on governments around the world, of economies advanced
and developing, this is remarkable. Nonetheless, the uncertainty of
the global recovery poses challenges to macroeconomic management,
including the management of public finance at both the federal and
subnational levels.
The desired level of RDFs is a complex subject. A range of factors
influence the level, including macroeconomic and market conditions,
fiscal policy objectives, and the size and duration of macroeconomic
shocks. In the case of Mexico, the success of leveraging the RDF might
imply a lower optimal long-run level of RDFs—since one could bring to
present value future flows of the fund. (The large RDFs might become
a temptation for politicians to spend.) However, one would be averse
to having to depend on access to markets specifically at the time when
one is experiencing a fiscal shock. Mexico had a solid fiscal position
coming into the crisis, and hence had extraordinary access to markets
even in the downturn (consider also the path of interest rates and access
to credit for the 2008–09 global crisis, shown in table 4.3). But crises
come in different shapes and have different transmission channels, so
this might imply a larger optimal long-run level of RDFs. Hopefully, the
Mexican experience contributes to the larger debate on the optimal size
of stabilization funds.
Another important lesson is the consideration of the relative benefits of a rules-based mechanism for subnational debt regulation compared to a market-based mechanism. Mexico has evolved into a hybrid,
quasi-regulated market system. In this regulatory framework, the
major ingredients are the federal threat of no bailout, the transparency
167
168
Until Debt Do Us Part
of markets and, more important, the regulation of the market via the
prudential regulation of banks. This appears to have worked. Subnational debt, at its low levels, does not appear to pose a macroeconomic
or systemic threat. Indeed, it is the fact that it has a relatively low value
(at 2.9 percent of GDP) that is surprising, given the infrastructure needs
of subnationals in Mexico.
A country’s macroeconomic framework has an important bearing
on its subnational debt markets. The important elements of the macroeconomic management framework for the health and evolution of subnational debt markets are (a) the fiscal position and debt stock of the
federal government, (b) the currency regime, (c) monetary policy, and
(d) economic growth. Future challenges for Mexico include translating
the success of macroeconomic management at the federal level to create
a more dynamic and transparent subnational debt market that contributes more effectively to the financing of infrastructure at the local level
and, hence, to the economic growth and development of the country.
Annex
Table 4A.1 “Ramo 28.” Nonearmarked Transfers (Participaciones Federales), Mexico
Fund
Purpose
Funding
Distribution criteria
Destination
Share of totala (%)
FGP
Revenue sharing with states and
municipalities
20 percent of RFP
State GDP growth; local
revenue (level and growth)
State and
municipalc
FFM
Revenue sharing with municipalities
1 percent of RFPb
Municipal revenue
(water and property tax)
Municipal
4
FOFIE
Incentive for enforcement of tax laws
1.25 percent of RFPb
Measures of local effort of
enforcement of tax law
State and
municipalc
5
3.17 percentd
Resources for oil-producing
municipalities
3.17 percent of a special
oil royalty
Municipal revenue (water and
property tax)
Municipal
0.3
0.136 percente
Resources for borderline
municipalities
0.136 percent of RFPb
Municipal revenue
(water and property tax)
Municipal
0.7
FEXHI
Compensate for oil and gas
extraction
0.6 percent of main oil
royalty
Oil and gas production
State and
municipalc
1
IEPS
“Sin tax” revenue sharing with states
and municipalities
8 percent of tobacco;
20 percent of beer
and alcohol
Local consumption
of those goods
State and
municipalc
2
FOCO
Compensate the 10 poorest states
2/11 of local gasoline tax
collection
Inverse of nonoil GDP per
capita
State and
municipalc
1
b
86
169
Source: Ministry of Finance, Mexico.
Note: FGP = General Participation Fund (Fondo General de Participaciones), FFM = Fund for Municipal Aid (Fondo de Fomento Municipal), FOFIE = Tax Enforcement Fund (Fondo de
Fiscalización), FEXHI = Fund for Oil Extraction (Fondo de Extracción de Hidrocarburos), GDP = gross domestic product, IEPS = “Sin Tax” Revenue Sharing (Impuesto Especial sobre la
Producción y Servicios), FOCO = Compensation Fund (Fondo de Compensación).
a. Shares calculated based on data for 2010.
b. Shared Federal Revenue (Recaudación Federal Participable, RFP): The pool of federal revenues that is shared with states and municipalities includes the income tax, the value-added
tax, all other federal taxes, and oil revenues. It does not include revenue from public enterprises, federal government financing, or certain other sources of nontax revenue.
c. States are required by law to share at least 20 percent of these resources with municipalities.
d. 3.17 percent of special oil royalty (3.17 percent del Derecho Adicional).
e. 0.136 percent of RFP (0.136 percent de la RFP).
170
Table 4A.2 “Ramo 33.” Earmarked Transfers (Aportaciones Federales), Mexico
Purpose
Funding
Distribution criteria
Destination
Share of totala
(%)
FAEB
Elementary education
Theoretically, enough money
to cover payrollc
Student enrollment and state
spending on education
State
59
FASSA
Health services
Theoretically, enough money
to cover payrollc
Health indicators; number of
health workers
State
12
FAIS
Social and rural infrastructure
0.303 percent of RFPb
Poverty index
State
9
2.197 percent of RFP
Municipal
b
FORTAMUNDF
Municipal strengthening
2.35 percent of RFPb
Population
Municipal
9
FASP
Public security
Budget negotiation process
Population; delinquency and
criminality indexes
State
2
FAETA
Promote adult education and
literacy
Theoretically, enough money
to cover payroll
Schooling and workers
State
1
FAM
Social assistance and education
infrastructure
0.814 percent of RFPb
Social vulnerability index
State
3
FAFEF
Financial needs and pensions
1.4 percent of RFPb
Inverse of GDP per capita
State
5
Source: Ministry of Finance, Mexico.
Note: FAEB = Fund for Elementary Education (Fondo de Aportaciones para la Educación Básica), FASSA = Fund for Health Services (Fondo de Aportaciones para los Servicios de Salud),
FAIS = Fund for Social Infrastructure (Fondo de Aportaciones para Infraestructura Social), FORTAMUNDF = Fund for Municipal Strengthening (Fondo para el Fortalecimiento Municipal y de las Demarcaciones Territoriales del D.F.), FASP = Fund for Public Security (Fondo de Aportaciones para la Seguridad Pública), FAETA = Fund for Adult Education (Fondo de
Aportaciones para la Educación Tecnológica y de Adultos), FAM = Fund for Social Assistance (Fondo de Aportaciones Múltiples), FAFEF = Fund for State Strengthening (Fondo de
Aportaciones para el Fortalecimiento de las Entidades Federativas), GDP = gross domestic product.
a. Shares calculated based on 2010 data.
b. Shared Federal Revenue (Recaudación Federal Participable, RFP): The pool of federal revenues that is shared with states and municipalities includes the income tax, the value-added
tax, all other federal taxes, and oil revenues.
c. These funds were created to cover states’ education and health payrolls after the decentralization of these sectors in the 1990s. The size of these funds has usually been determined by
political forces during the federal budget negotiation process, and almost all states argue that the resources they receive from these funds are insufficient to fully cover their payroll.
Table 4A.3 Mexican States’ Total Local Revenue: Own-Source and Coordinated Federal Taxesa
million pesos
2005
2006
2007
2008
2009
2010
Total local revenue
106,762
Own revenue
81,894
124,420
138,213
160,687
165,433
184,338
92,892
103,262
119,667
115,552
132,829
Taxes
34,818
39,160
44,396
47,864
49,417
57,706
Payroll
20,178
23,276
27,567
30,227
31,523
36,466
4,761
6,065
6,729
7,113
7,851
9,684
721
1,295
1,299
1,360
1,344
1,361
788
866
1,054
1,104
1,110
1,337
992
1,170
1,275
1,299
1,267
1,235
2,259
2,734
3,101
3,350
4,130
5,752
Other taxes
Use of motor vehicles (> 10 years)
Lodging
Personal property
Otherb
Property tax and property sales taxc
Nontax revenued
Coordinated federal taxesa
9,879
9,818
10,100
10,523
10,042
11,556
47,075
53,732
58,865
71,803
66,135
75,123
24,868
31,528
34,952
41,020
49,881
51,509
Vehicle-related taxes
20,873
23,989
25,827
26,175
24,515
23,773
Federal tax on use of motor vehicles
15,262
18,814
20,245
21,100
20,448
19,093
5,611
5,175
5,582
5,075
4,067
4,680
Tax on new vehicles
Fuel tax
Other coordinated federal taxese
—
—
—
5,080
15,334
17,482
3,995
7,539
9,125
9,765
10,032
10,254
171
Sources: Ministry of Finance, Mexico; states’ public accounts.
Note: — = not available.
a. “Coordinated federal taxes” are federal taxes (the base and rate are defined by the federal government) administered and fully collected by state governments. In that sense, they
behave (and are sometimes considered as) local revenue.
b. Includes taxes on lotteries and games, special profession taxes, state tax on use of motor vehicles, etc.
c. Considers the revenue from the property tax and the property sales tax in the Federal District, where these taxes are collected at the state level. For the rest of the states, they
are collected at the municipal level.
d. Alcohol, drivers’, and other licenses; received; state-owned enterprises; fines and charges; and other.
e. Federal income tax and value-added tax for low-income firms, federal taxes in coastline areas, special fund for the tax on new vehicles.
172
Until Debt Do Us Part
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. There are 2,440 municipalities in Mexico, with a wide heterogeneity in size and
level of development. Population ranges from 1.8 million in the largest municipality (about the size of Phoenix, Arizona) to only 102 in the smallest. The most
developed municipality in Mexico has a Human Development Index close to that
of Portugal, while the least developed can be compared to Sierra Leona. Municipal budgets range from 4.2 million pesos to 4.1 billion pesos, a ratio of 1:1,000.
2. This “third channel” consists mainly of a set of new special-purpose funds
that are mostly earmarked for infrastructure. The growth of this channel can be seen in “Ramo 23” of the federal budget: in 2007 it amounted to
10.5 billionpesos, while for 2012 it is budgeted at 30.6 billion pesos (an increase of
134 percent in real terms). Examples of funds included are the “regional fund”
(for 10 states); the new “metropolitan funds,” which currently distribute
resources to 46 metropolitan areas; and other funds for specific purposes such
as natural disasters, aid to migrant workers, and for paving municipalities.
3. The pool of federal revenues that is shared with states and municipalities
includes the income tax, the value-added tax, all other federal taxes, and oil revenues. It does not include revenue from public enterprises, federal government
financing, or certain other sources of nontax revenue.
4. The formula was reformed in 2007 from an old formula that caused wide distortions in Mexico’s fiscal federalism. For a detailed description of the distribution formula and its reform, see Revilla (2012).
5. “Ramo 33” refers to line item 33.
6. Efforts have been made since 2007 to give incentives to subnationals to increase
own-source revenues. The most important was the reform of the formula for nonearmarked transfers, which started being used in 2008. The new formula is designed
to substantially increase the incentives for states and municipalities to increase their
own revenue. After only four years (and considering that the new formula provides for a gradual transition, since it was designed with a generous hold-harmless
clause), it can be seen that subnationals are greatly increasing their local tax efforts.
7. The subnational payroll tax is collected on the payrolls of businesses that operate within state lines, at a rate that is freely set by the state legislature. All states
collect the tax now at a rate that fluctuated between 1 and 3 percent in 2011.
8. See chapter 1 on Brazil, chapter 10 on China, and chapter 3 on India in this volume.
9. According to the Global Competitiveness Report published by the World Economic Forum (2006), Mexico’s “infrastructure competitiveness” is ranked 64th
among countries. Its performance stands below the world average and below
the average for Latin America. Mexico’s investment in infrastructure accounts
Subnational Debt Management in Mexico: A Tale of Two Crises
for only 3.2 percent of GDP, and compares poorly to the investment of countries like Chile (5.8 percent of GDP) and China (7.3 percent of GDP).
10. Particularly from the Fund for Social Infrastructure (Fondo de Aportaciones
para Infraestructura Social) and the Fund for State Strengthening (Fondo de
Aportaciones para el Fortalecimiento de las Entidades Federativas). Of the total
206 billion pesos collateralized with transfers, only 5 billion (2.3 percent) pesos
are collateralized with earmarked transfers.
11. See Fitch Ratings (2011); Mexican Congressional Budget Office (Centro de
Estudios de Finanzas Públicas) (CEFP 2009); and Velázquez (2005) and references contained therein.
12. If a state borrows from an international financial institution, the credit must
be channeled through federal government development banks first (so that the
forex risk is borne by the federal government and the state does not have any
direct obligation to a foreign entity).
13. In theory, the federal government could deduct subnationals’ debt service payments from the transfers to the states.
14. Especially because most of the participaciones are used for current expenditure,
so a reduction in them would leave a state unable to operate and provide basic
services.
15. The relevant laws and regulations are the Fiscal Coordination Law (Ley de
Coordinación Fiscal), the Public Debt Law (Ley de Deuda Pública), and the
Regulation of Article 9 of the Fiscal Coordination Law.
16. The author is indebted to discussions with Emilio Pineda for this interpretation.
17. The registered debt, as mentioned elsewhere in this chapter, includes all explicit
loans obtained by subnationals from private commercial banks, government
development banks, and the market that were approved by local legislatures.
It does not contain contingent (implicit) liabilities, such as pensions, or unsecure short-term loans used for cash management. The registered debt can be
accessed online at http://www.hacienda.gob.mx/Estados/Paginas/Deuda.aspx.
18. Subnational fiscal rules for the case of Mexico have the added disadvantage that
the accounting practices of local governments are widely heterogeneous and, in
some cases, deeply flawed. There is an accounting harmonization process that
was set up in 2008 with a constitutional reform that will modernize accounting
procedures at all levels of government. As of 2012, states have progressed slowly
toward accounting harmonization.
19. These banking regulations were put in place in 1999–2000, as part of the prior
actions for the 1999 Decentralization Adjustment Loan from the World Bank.
They had to be done through financial sector regulation, over which the federal government has authority, because constitutionally the federal government
could not impose such rules directly on the states.
20. In 2011, it was revealed that Coahuila, a northern state, falsified documents to
hide the true size of its debt. In reality, in the previous two years it had accumulated a debt of $35 billion pesos (295 percent of its annual nonearmarked
173
174
Until Debt Do Us Part
transfers, or 9 percent of its GDP), while reporting only $7 billion pesos. While
this case highlights the need to strengthen the transparency of subnational
financial reporting, it does not change the overall view of the Mexican subnational debt market as one of low indebtedness without significant systemic
risks.
21. In some cases, the extraordinary support could have taken the form of a direct
transfer, a renegotiation of debt with a federal development bank (including a
reduction in interest and principal or the outright forgiveness of the debt), or
support through a budgetary mechanism (for example, reducing the share of
subnational expenditure in projects that combine federal and local resources,
that is, a reduction in the pari passu of programs, and so forth).
22. The price of Mexican oil suffered a dramatic fall as the crisis hit financial markets,
dropping from a maximum of US$130 per barrel in July 2008 to a minimum of
US$28 per barrel in December of the same year. Data source: Bloomberg.
23. As in any subnational debt market, there is the question of whether credit ratings truly reflect state’s idiosyncratic credit risks (and these, in turn, fiscal risks).
While a detailed analysis of the informational content of subnational credit ratings in Mexico is beyond the scope of this chapter, we consider the observed
ratings as a good approximation of the credit quality of subnationals at a given
moment in time.
24. Total funds amounted to US$8.4 billion, at the exchange rate for December 31,
2008, of 13.82 pesos per dollar.
25. This gap for the federal government was finally closed with the use of the f ederal
RDF, with additional debt, a cut in expenditure, and nonrecurrent revenues.
26. “In the United States the estimated collective gap between states’ income and obligations for 2011 will be $55 billion dollars. This means that more than 30 states
are projecting a 2011 shortfall of 10 percent or more as a percentage of this year’s
budget. Many states have already used a big proportion of their RDFs: according
to the same report, 14 states are expected to have reserves of less than 1 percent of
their annual spending. In order to close the budget gap, states in the U.S. are supposed to make serious expenditure cuts, which might be a difficult job given the
upward pressures arising from certain areas, particularly Medicaid” (“The Other
Financial Crisis,” Time, 175 (25), June 28, 2010).
27. For example, the debt of the State of Mexico and the Federal District is at the
higher boundary, so there was no space for additional financing.
28. Which, given the fungibility of money, would not have been a problem for
those states with significant investment programs. However, some states would
not have had space to use it otherwise.
29. In the sense of imposing, or making credible, a hard budget constraint on states.
30. This criterion gives economic efficiency to the restructuring process. Since
nonearmarked transfers are the freest form of financing for subnationals, they
would use the proceeds to finance their own budgetary priorities during the
crisis, as defined by their own legislature process.
Subnational Debt Management in Mexico: A Tale of Two Crises
31. Provided, of course, that the transitory fiscal gap does not materially reduce the
present value of receipts.
32. This is the main representative body in the Mexican national fiscal federalism
system. It consists of eight states’ ministers of finance who represent the 32. It
has powers to decide, as representative of the states, and it coordinates with the
federation all relevant topics of fiscal federalism in the country.
33. Given that (according to the Federal Fiscal Responsibility Law) FEIEF is to be
used as a perfect substitute of nonearmarked transfers, the funds obtained from
the operation had the same nature: they could—and were—used legally as
nonearmarked transfers (and not for infrastructure).
34. The loan was paid in full in only two years—by mid-2011 (11 years ahead of
schedule)—because of the favorable evolution of oil prices, which increased the
repayment capacity of subnationals.
Bibliography
Ahmad, Ehtisham, José Antonio Gonzalez Anaya, Giorgio Brosio, Mercedes GarciaEscribano, Ben Lockwood, and Ernesto Revilla. 2007. “Why Focus on Spending
Needs Factors? The Political Economy of Fiscal Transfer Reforms in Mexico.”
Working Paper WP/07/252, International Monetary Fund, Washington, DC.
Canuto, Otaviano, and Lili Liu. 2010a. “Subnational Debt Finance and the Global
Crisis.” PREM Economic Premise 13, World Bank, Washington, DC, May.
———. 2010b. “Subnational Debt Finance: Make It Sustainable.” In The Day After
Tomorrow: A Handbook on the Future of Economic Policy in the Developing World,
ed. Otaviano Canuto and Marcelo Giugale, 219–38. Washington, DC: World Bank.
CEFP (Centro de Estudios de las Finanzas Públicas). 2009. La Deuda Subnacional en
México. Mexico City: CEFP.
ECLAC (Economic Commission for Latin America and the Caribbean). 2009.
Estadísticas e Indicadores Económicos CEPALSTAT. Santiago: ECLAC.
Fedelino, Annalisa, and Teresa Ter-Minassian. 2010. “Making Fiscal Decentralization
Work: Cross-Country Experiences.” Occasional Paper 271, International Monetary Fund, Washington, DC.
Fitch Ratings. 2011. Marco institucional de los gobiernos subnacionales en México.
Special Report, New York, September 14.
Gil-Díaz, Francisco. 1998. “The Origin of Mexico’s 1994 Financial Crisis.” The CATO
Journal 17 (3): 303–13.
Giugale, Marcelo, Fausto Hernández, and Joao de Carmo Oliveira. 2000. “Subnational Borrowing and Debt Management.” In Fiscal Decentralization in Mexico:
Achievements and Challenges, ed. Steven Webb and Marcelo Giugale, 237–70.
Washington, DC: World Bank.
Giugale, Marcelo, Adam Korobow, and Steven Webb. 2000. “A New Model for Market-Based Regulation of Subnational Borrowing: The Mexican Approach.”
Policy Research Working Paper 2370, World Bank, Washington, DC.
175
176
Until Debt Do Us Part
Giugale, Marcelo, and Steven Webb. 2000. Achievements and Challenges of Fiscal
Decentralization: Lessons from Mexico. Washington, DC: World Bank.
Hernández, Fausto, Alberto Díaz-Cayeros, and Rafael Gamboa. 2002a. “Determinants and Consequences of Bailing Out States in Mexico.” Eastern Economic
Journal 28 (3): 365–80.
———. 2002b. “Fiscal Decentralization in Mexico: The Bailout Problem.” Research
Network Working Paper R-447, Inter-American Development Bank, Washington,
DC.
IMCO (Mexican Institute for Competitiveness). 2010. The Black Box of Public
Spending. State Competitiveness Index 2010. Mexico City: IMCO.
Liu, Lili, Abha Prasad, Francis Rowe, and Signe Zeikate. 2009. “Subnational Debt
Management in Low Income Countries in Transition to Market Access.” In
Debt Relief and Beyond, ed. Carlos Braga and Doerte Doemeland, 343–72.
Washington, DC: World Bank.
Liu, Lili, and Michael Waibel. 2008. “Subnational Borrowing, Insolvency and Regulation.” In Macro Federalism and Local Finance, ed. Anwar Shah, 215–44. Washington, DC: World Bank.
———. 2009. “Subnational Insolvency and Governance: Cross-Country Experiences and Lessons.” In Does Decentralization Enhance Service Delivery and
Poverty Reduction?, ed. Ehtisham Ahmad and Giorgio Brosio, 333–76. Cheltenham, U.K.: Edward Elgar.
OECD (Organisation for Economic Co-operation and Development). 2010. Revenue
Statistics 2010. Paris: OECD Publishing.
Prasad, Abha, Rajan Goyal, and Anupam Prakash. 2004. “States’ Debt and Debt
Relief.” Economic and Political Weekly 39 (26): 2726–36.
Revilla, Ernesto. 2012. “Mexico’s Fiscal Federalism.” Draft. Ministry of Finance,
Mexico. Available via email: ernesto_revilla@hacienda.gob.mx.
Ter-Minassian, Teresa, and Jon Craig. 1997. “Control of Subnational Government
Borrowing.” In Fiscal Federalism in Theory and Practice: A Collection of Essays, ed.
Teresa Ter-Minassian, 156–74. Washington, DC: International Monetary Fund.
Velázquez, César. 2005. State Public Debt Laws and Fiscal Performance: The Case of
Mexico. Mexico City: Universidad Iberoamericana.
World Economic Forum. 2006. The Global Competitiveness Report, 2006–2007.
Geneva: World Economic Forum.
Part 2
Subnational Insolvency
Framework
5
Colombia: Subnational
Insolvency Framework
Azul del Villar, Lili Liu, Edgardo Mosqueira,
Juan Pedro Schmid, and Steven B. Webb
Introduction
The fiscal and debt stress of Colombia’s subnational governments
(SNGs) in the late 1990s and early 2000s, exacerbated by the economic
downturn, led to substantial public finance reform. Addressing the
insolvency of some SNGs was essential to this reform process. Colombia
has several laws, mostly enacted between 1998 and 2003, that regulate
the origination of debt by SNGs, encourage fiscal responsibility, and
provide for central government assistance in rescheduling subnational
debt when that becomes necessary. One law—Law 550 (1999)—deals
explicitly with bankruptcy proceedings for SNGs.
Although it was traditionally a centralist country, Colombia has
become the most decentralized unitary republic in Latin America
through a process that started in the early 1970s and accelerated in the
1990s. By the mid-1990s, a number of shortcomings in the decentralization framework had become evident. Besides the absence of fiscal responsibility institutions in these years to control subnational indebtedness,
intergovernmental fiscal relations also suffered from a lack of institutions
to ensure adequate allocation and use of transfers and to motivate SNGs
to generate own revenues and provide required matching funds.
179
180
Until Debt Do Us Part
During 1997–2003, the Colombian government passed several laws
to discourage excess spending and borrowing. In 1999, it passed the
first bankruptcy law (Law 550) in the country, which focused primarily on private, public, and mixed-ownership corporations, but Law 550
also included provisions under Chapter V for bankruptcy procedures
of highly indebted SNGs. In 2000, Law 617 modified some features in
the application of Law 550, addressing SNGs and decentralized services
entities (not covered by a sector-specific superintendency). Regulation
(Reglamento) 1248 in 2001 also clarified debt restructuring and central
government guarantees.
This chapter concerns Colombia’s bankruptcy or insolvency framework—its provisions and the actual experience of its implementation
in the broader context of reforms to strengthen subnational fiscal discipline in the country.
Most of the bankruptcy procedures were initiated in the early 2000s,
to deal with subnational debt problems accumulated during the 1990s,
which were further compounded by the general fiscal and economic
crisis in the country from the late 1990s to early 2000s. The development and implementation of the bankruptcy proceedings were helped
by enactment of several other fiscal reform laws (1998–2003) that
encouraged subnational fiscal responsibility and discipline. Strong economic growth after 2003 also helped the fiscal position of governments
at all levels.
The Colombian bankruptcy procedures for SNGs differ from those
in countries such as Hungary and the United States. The procedures in
Colombia are administrative within the legal framework,1 led by the
Superintendency of Corporations (Superintendencia de Sociadades,
SOC) in coordination with other institutions such as the Ministry of
Finance and Public Credit (Ministerio de Hacienda y Crédito Público,
MFPC) of the central government. In contrast, the courts take the center seat in local government insolvency proceedings in Hungary and the
Unites States.2 The unique role of the SOC in Colombia arose in an historical context where the court system was weak, and thus an alternate
arrangement was created, in which the SOC administers bankruptcy
procedures for both corporations and most government entities.3
Increasingly, SNGs in Colombia used the Law 550 process not
because they borrowed too much from lenders, but because other
Colombia: Subnational Insolvency Framework
claimants (wage earners, suppliers, and so forth) have gotten court
judges, outside the SOC, to recognize their claims to the unpaid SNG
bills. The embargos by courts—using intercepts of fiscal transfers and
bank accounts—force subnationals to pay these bills, with added penalties and interest payments. By initiating a bankruptcy process under
Law 550, SNGs can obtain a halt to the embargoes, past and prospective,
and go through orderly restructuring of their debts. The essence of the
550 proceeding is to evaluate and reconcile competing claims against
the subnational debtor, according to a defined priority structure.
There has been little divergence between the law and actual practice
for dealing with subnational insolvency, in the sense that essentially
all the debt restructuring and adjustment operations have been done
according to procedures prescribed in the laws. Nonetheless, for any
one subnational situation, each of the laws, and (even more), the group
of laws, provides a variety of options for how to address the problems.
Thus, understanding the actual practice requires seeing which options
are usually chosen and why.
To understand Law 550—its origins and its practice—we review
the evolution of the intergovernmental fiscal policy and context of the
other laws that regulate it. The remainder of the chapter is structured as
follows. Section two presents the structure of the Colombian decentralization framework and its development since decentralization started.
Section three shows how the borrowing framework developed in order
to provide both ex-ante fiscal rules and debt limitations and ex-postbankruptcy proceedings, as well as to enhance transparency in the context of SNGs’ medium-term fiscal frameworks. Section four describes
the Law 550 framework for insolvency proceedings. Section five reviews
the law’s implementation and evaluates its effects. Section six summarizes and concludes.
Structure and History of Subnational Governments4
Colombia is a unitary republic composed of 32 regions (departamentos) and around 1,100 municipalities (municipios). Ten of these
municipalities have the status of districts, which also manage the expenditures of the department in which they are located. Each department
has a governor (gobernador) and a department assembly (asamblea
181
182
Until Debt Do Us Part
departamental), both of which are elected by popular vote for a fouryear term. The municipal governments are headed by a mayor (alcalde)
and administered by a municipal council (consejo municipal), which
are also elected for four-year terms (see figure 5.1).
Decentralization History and Challenges in the 1990s
Until the early 1970s, Colombia was a strongly centralist country.
National agencies controlled most of the subnational spending programs. This situation started to change with the constitutional reforms
adopted in 1968 that obliged the central government to share its current revenues with SNGs through the so-called situado fiscal.5 In addition, the reforms allowed SNGs to provide local services through public
companies and decentralized service entities that are independent from
the central government, and municipalities were granted autonomy to
plan and coordinate local development and to provide services under
the supervision of the departments (Bird 1984).
Political considerations led to important extensions of decentralization in the 1980s. The Constitutional Reform of 1986 (Acto Legislativo 01) introduced popular elections for mayors starting in 1988, and
Decrees 77 to 80 of 1987 transferred to municipalities the responsibilities for spending on basic infrastructure and social services.6 Law
14/1983 widened the tax base for municipal and departmental taxes
and also prescribed the ranges within which the SNGs could set their
taxes, in order to avoid destructive tax competition or a negative
Figure 5.1 Government Structure in Colombia
Central government
Decentralized
service sector
(entidades
descentralizadas
del nivel
territorial)
32 regions (departamentos)
Governor (gobernador)
Regional assembly (asamblea departamental)
1,100 municipalities (municipios)
Mayor (alcalde)
Municipal council (consejo municipal)
Source: Authors, based on the Colombian constitution.
Colombia: Subnational Insolvency Framework
impact on Colombia’s international competitiveness. Revenue sharing also increased substantially with Law 12/1986, which intended for
the national government to increase the transferred share of the valueadded tax up to 50 percent by 1992. Until the 1991 Constitution, the
president appointed the governors of departments, making them more
like deconcentrated branches of the national government than autonomous subnational entities. Thus, the decentralization of the 1980s was
largely to the municipal level. This experiment was deemed a failure,
however, because too many municipalities lacked the administrative
capacity to deliver services, and some were being overrun by guerilla
insurgencies (Dillinger and Webb 1999; Sánchez and Gutiérrez 1995;
Rojas 2003).
Decentralization accelerated substantially with the 1991 Constitution, which made the office of governor an elected post and (together
with Law 60/1993) committed the national government to increase the
amount of transfers assigned to subnational entities each year until it
reached 46.5 percent of the central government’s current revenues
by 2002. These transfers were complemented by a system of natural
resource royalties (regalías) that would remain mostly in the producing
and transit localities, and a system of cofinancing in which the central
government would transfer funds conditional on the participation of
local governments for projects in the areas of urban and road infrastructure (Ahmad and Baer 1997; Dillinger and Webb 1999). The government plans to revise the rules for the royalties in 2012.
The transfer system that resulted from the 1991 constitutional
changes focused on the financing of education, health, and water and
sanitation in order to equalize the provision of these services across
regions. In addition, municipalities with a population of less than 30,000
could use up to 28 percent of these transfers to pay for working expenditures. However, the rapid increase of transfers, which could serve as collateral, also stimulated the growth of expenditures and debt in territorial
governments7 and diminished the incentives for SNGs to raise their
own revenue. On top of the growing transfers, SNGs ran current fiscal
deficits and new municipalities were created to gain access to transfers.
The number of municipalities increased from 745 to 998 between 1994
and 1999. At the same time, the transfer system constrained the possibility of balancing central government finances, because 46 percent of
183
Until Debt Do Us Part
any increase in current revenues went out to subnationals as additional
transfers. Similarly, countercyclical fiscal policy by the central government became less effective, because subnational expenditures were cyclical with current revenues (Dillinger and Webb 1999).
Together with enhanced political autonomy and the responsibility for local public service delivery, the transfer of current revenues to
SNGs increased from 13 percent of national government revenue in
1973 to 49 percent in 1999. Since then, transfers as a share of total SNG
revenues have declined, to less than 40 percent by 2010, partly because
SNGs were increasing their own revenues (see figure 5.2).
The increasing expenditure responsibilities of SNGs were not
matched by adequate own-resource instruments, and transfers were
excessively earmarked. Smaller SNGs struggled to cover the share of
operating expenses that were not funded by transfers but had little
incentive to manage other expenditures effectively. Besides the absence
of fiscal responsibility institutions in these years, intergovernmental
fiscal relations also suffered from the lack of institutions to ensure
Figure 5.2 Regional Transfers to Subnational Governments as a Share of Current
Central Government Revenues, 1994–2010
60
55
50
45
40
Percent
35
30
25
20
15
10
5
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
0
19
184
Year
Source: Balance Fiscal Gobierno Nacional Central 1994–2010, MFPC, http://www.minhacienda.gov.co.
Note: Colombia fiscal year = calendar year.
Colombia: Subnational Insolvency Framework
adequate allocation and use of transfers and to incentivize SNGs to
generate own revenues and provide required matching funds (Rojas
2003).
The above situation blunted the SNGs’ incentives for fiscal discipline.
Various factors had contributed to the fiscal and debt crisis at the subnational level in the 1990s, compounded by the economic recession in
the late 1990s to early 2000s, which added to the fiscal problems at the
national level.
Starting in the late 1990s, the national government introduced a
series of measures to bring subnational finances under control: increasing their own revenue collection, making fiscal transfers to SNGs more
predictable in real terms, and introducing a legal and regulatory framework for fiscal responsibility in SNGs. The framework includes procedures to deal with insolvency of subnational entities; stronger limits on
current expenditures, especially the wage bill; and procedures to implement adjustment plans and overcome insolvencies for SNGs. These
measures, along with, among other things, the economic growth since
the mid-2000s, helped Colombia constrain unsustainable subnational
debt accumulation and contributed to the relatively healthy fiscal situation today in most of Colombia’s subnationals.
Subnational Responsibilities and Resources
Expenditures by SNGs averaged about one-third of total government
expenditures from 2005 to 2010,8 with the combined spending accounting for about 27.5 percent of gross domestic product (GDP) during the
same period. The share of SNGs in public spending overstates the fiscal
autonomy of SNGs, however, because nearly half of subnational spending consists of earmarked transfers in the education and health sectors.
This leaves the SNGs only limited control over resource allocation.9
SNGs depend heavily on transfers from the center, which represent
around 58 percent of their total revenues. Currently, there are three
main transfers: the General Transfer System (Sistema General de Participaciones, GTS), direct royalties, and rentas cedidas (central government taxes earmarked for certain local administrative activities). The
share of these transfers account for 47, 9, and 2 percent of total SNG
revenues, respectively, and much of the GTS is earmarked for education
and health services (table 5.1). The share of transfers varies widely, with
185
186
Until Debt Do Us Part
Table 5.1 Total Revenues of Subnational Governments, Percentage of Total,
2006–10
Total revenue
Tax revenue
2006
2007
2008
2009
2010
100
100
100
100
100
29
30
28
29
29
Nontax revenue
8
7
5
6
6
Transfers (funcionamiento)
2
2
2
3
2
Royalties
CGT
7
8
10
8
9
45
47
46
46
47
Cofinancing
2
3
1
2
2
Others
7
4
7
7
5
Source: NPD, Desempeño Fiscal de los departamentos y municipios, 2010.
Note: CGT = Central Government Transfers; data may not tally due to rounding of decimals.
large municipalities being mostly self-financed and small municipalities
and poor departments depending almost entirely on transfers. Some
SNGs with hydrocarbon and other mineral exports are well financed
with royalties.
Before 2001, transfers made subnational revenues and expenditures
strongly procyclical, because the transfer formula was directly linked to
current central government revenues. The constitutional reforms in 2001
and 2007 and Laws 715/2001 and 1176/2007 delinked transfers from
central government current revenues and clarified the distribution of
competences among different layers of government. The 2001 reforms
aggregated most of the previous transfers10 into the GTS and set it to
grow on a real basis unrelated to central government revenues. As a result
of the reforms, transfers to SNGs have followed a predictable path without the volatility of the late 1990s.11 Today, the GTS is the largest transfer,
amounting to 4.2 percent of GDP or 34 percent of central government
current revenues and accounting for 47 percent of subnational revenues.12
Under the 1991 Constitution, SNGs of regions producing minerals
(mainly oil and coal) and serving as ports for exports keep the main
share of natural resource royalties. Law 756/2002 specifies the royalty
rate, which depends on the type of natural resource and the value of
production in the entity. Up to 32 percent of the value of production
is reserved for the National Royalty Fund (Fondo Nacional de Regalías),
which was established to finance mining development, environmental
Colombia: Subnational Insolvency Framework
protection, and regional development projects nationwide. The remainder is distributed among producing departments, producing municipalities, and port municipalities, and these funds may be used only for
investment in the National Pension Fund (Fondo Nacional de Pensiones
en las Entidades Territoriales, FONPET), education, health (infant mortality, and health-for-the poor projects), or water supply and sewerage.13
The oil price boom since the mid-2000s—even since 2008, prices have
remained high by historic standards—has kept revenues in these subnationals above trend, and for those that have borrowed on that basis, a
sustained fall of oil prices would bring a debt problem.
Besides transfers and royalties, departments and municipalities also levy
taxes. Tables 5.2 and 5.3 disaggregate them by departments and municipalities. On average, own revenue as a share of total revenue is 26 and
31 percent for departments and municipalities, respectively. While these
values are relatively low compared with the expenditures responsibilities,
Table 5.2 Department Revenues
thousands of millions of Colombian pesos
Revenue
2010
Departmental taxes
Percent
% of tax revenue
Car
374
7
Registry
392
8
Liquor
936
19
1,435
28
565
11
Beer
Cigarettes and tobacco
Fuel sobre tasa
280
6
Others
1,054
21
Tax revenue
5,036
26
Transfers
9,749
50
Royalties
2,845
14
1,108
6
339
2
616
3
% of total
Nontax revenue
Cofinancing
Others
Total revenue
19,693
Source: NPD, Desempeño Fiscal de los departamentos y municipios, 2010.
Note: Data may not tally due to rounding of decimals.
187
188
Until Debt Do Us Part
Table 5.3 Municipal Revenues
thousands of millions of Colombian pesos
Revenue
2010
Municipal taxes
Percent
% of tax revenue
Unified property tax
3,339
31
Tax on gross business receipts
4,522
42
Fuel sobre tasa
1,100
10
Others
1,930
18
% of total
Tax revenue
10,891
31
Transfers
17,682
50
Royalties
1,850
5
Nontax revenue
1,945
5
622
2
2,418
7
Cofinancing
Others
Total revenue
35,408
Source: NPD, Desempeño Fiscal de los departamentos y municipios, 2010.
Note: Data may not tally due to rounding of decimals.
the own-revenue base of SNGs is much higher than in other unitary
Latin American countries (World Bank 2009).
Departmental tax bases are narrow, with the bulk of resources being
raised by taxes on alcohol, liquor, and tobacco, although in some wealthier departments the tax on vehicle ownership is important. Compared
to departments, municipalities have a greater tax base, with the property tax and the tax on gross business receipts (impuesto de industria y
comercio), which each contribute 30 and 40 percent, respectively, to the
total municipal tax revenues, on average. The remainder comes from
a host of other, minor taxes. Large municipalities raise more own revenue than the average, and small municipalities raise much less, often
almost nothing in poor and remote places. Thus, own-revenue capacity of SNGs varies widely, and with it their capacity to service debt. As
shown below, the fiscally stronger SNGs borrowed heavily in the 1990s
and got into debt trouble, but in the 2000s, most of the entities with
debt problems were smaller and poorer. Their debt problems arose less
from formal borrowing than from arrears and other manifestations of
general problems with governance and financial management.
Colombia: Subnational Insolvency Framework
Subnational Debt
Subnational Borrowing Trends
The 1991 Constitution gave territorial governments substantial autonomy
over borrowing and bond issuance (Art. 287 of Carta Constitutional). The
1986 administrative regulation on debt limits still governed subnational
borrowing. SNGs could borrow, including for current expenditures, as
long as the ratio of debt service to current income was below 30 percent,
and municipalities could incur higher debt levels if they were approved by
the MFPC. Borrowing had to be approved by the departmental assembly
or the municipal council. Art. 364 of the Constitution also stated that the
debt of an SNG should not exceed its payment capacity, but the implementing law was only adopted in 1997 and applied in 1999, when the
country was already in financial and fiscal crisis.
Subnational borrowing increased in the 1990s. The increase in central
fiscal transfers through Law 60/1993 stimulated borrowing from banks,
which treated territorial entities as preferred debtors based on the formula-based transfers from the center. Banks were not sufficiently aware
of the norms for subnational budgetary rules, and their risk management did not assess the real capacity to pay. Instead, they focused on the
availability of collateral in the form of royalties and transfers, ignoring
what claims other parties might already have on the same collateral. Also,
many of the credits were used to finance current expenditures or investment projects with delayed payment streams. A central problem was that
the transfers and the competences of the SNGs were not always aligned.
Operational expenditures were growing at a higher rate than own revenues, which led to a build-up of current deficits. Current local expenditures as a share of local tax revenues increased between 1990 and 1999
from an average of 140–170 percent in municipalities and from 169 to
314 percent in departments, with the gaps covered by transfers and credit.
The combination of political autonomy, weak bank lending supervision, excessive reliance on transfers, and permission to borrow for
current expenditure blunted SNGs’ incentives for fiscal discipline. The
economic slowdown in the late 1990s to early 2000s also weakened the
subnational fiscal accounts. Subnational debt increased from about
6 percent of GDP in the mid-1990s to 7.6 percent in 2000 (figure 5.3).
The decentralized service entities—utility companies owned by SNGs—
account for a significant share of subnational debt.
189
Until Debt Do Us Part
Figure 5.3 Subnational Government Direct Debt as Percentage of GDP, 1990–2010
9
8
7
6
% of GDP
5
4
3
2
1
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
0
20 1
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
0
19
190
Year
Departments and municipalities
Total SNGs
Source: Fiscal Affairs Department (Departamento de Asuntos Fiscales, DAF).
Note: The solid line reflects SNGs’ total debt, including decentralized service entities; the dotted line sums
only department and municipality debt. GDP = gross domestic product, SNG = subnational government.
Although the debt level was not high by international comparison,
there were two problems. First, the arrears (which are not included in
figure 5.3), had been increasing. Second, the capacity of SNGs to service
the direct debt had weakened, due primarily to the decline in own revenues and fiscal transfers. The National Planning Department (Departmento Nacional de Planeación, NPD) calculated fiscal performance
indicators for all municipalities under Law 550. Table 5.4 compares the
indicators for the municipalities under Law 550 debt restructuring with
the national average, and it distinguishes municipalities according to
the status of their debt restructuring agreement as of the end of 2011.
Nineteen municipalities out of the 26 that had successfully completed
the agreement (73 percent) were above the national average in this indicator. For the municipalities that were still carrying out the agreement,
only a slight majority (41 percent) had indicators above the national
average. Most of the municipalities with failed agreements had indicators below the national average, as one would expect.
At the height of the problem in 2000, external creditors accounted
for 38 percent of total SNG debt, and this declined to 30 percent as of
Colombia: Subnational Insolvency Framework
191
Table 5.4 Fiscal Performance of Municipalities under 550 Debt Restructuring
Agreements Compared to the National Average for All Municipalities
Municipalities
Fiscal performance above
national average
Fiscal performance at or below
national average
Restructured debt fully paid off, as per
550 Agreement
19
7
550 Agreement still being carried out
as of 2011
19
27
550 Agreement still being negotiated
2
4
550 Agreement failed
2
7
Source: NPD calculations, as of 2011.
end-2010—falling from 3.5 percent to 1.3 percent of GDP, as shown in
table 5.5. Subnational debt is highly concentrated in terms of the number of borrowers. The decentralized service entities—utility companies
owned by SNGs—account for about two-thirds of SNG borrowing,
both domestically and overseas. Twenty entities, including the capital
city Bogotá and 10 decentralized service entities, account for 88 percent
of subnational debt.
At the department level, for its domestic debt composition, the
largest source of subnational credit has been commercial banks—
74 percent of the total. The rest consisted of central government onlending (8 percent), bonds (5.6 percent), and the government-owned
FINDETER INFIS14 (2 percent), as of end-2010.
Capital markets remain small in Colombia. As of June 2010, only six
subnational entities have issued bonds, of which three are SNGs, and
three are decentralized service entities (that is, utility companies).15
Most of the SNG bonds have been paid off; only Bogotá still had outstanding bonds as of June 2010.
Overborrowing from banks in the late 1990s was concentrated in
departments and bigger municipalities. Smaller municipalities more
typically did not have a lot of bank debt, but rather accumulated arrears
on salaries, contributions to pension funds, social security contributions, and payments to providers. The combination of increasing interest rates and slower growth in the late 1990s led to an unsustainable
fiscal situation for SNGs (Dillinger and Webb 1999; Rojas 2003).
Since the early 2000s, the SNGs overall have improved fiscal balances,
going from an aggregate deficit of 1 percent of GDP in 1999 to a surplus
192
Until Debt Do Us Part
Table 5.5 Composition of Subnational Debt as Percentage of GDP, 2000–10
2000
2003
2007
2008
2009
2010
5.6
4.9
3.5
2.9
2.8
3.2
Departments
1.28
0.95
0.51
0.40
0.40
0.40
Capital municipalities
1.13
1.01
0.71
0.50
0.40
0.40
Municipalities
0.41
0.27
0.22
0.20
0.20
0.30
Decentralized entities
2.76
2.68
2.09
1.80
1.80
2.10
Domestic
External
3.5
2.8
1.80
1.5
1.6
1.3
Departments
0.01
0.05
0.03
0.03
0.03
0.03
Capital municipalities
0.42
0.47
0.37
0.30
0.30
0.30
Municipalities
0
0
0
0
0
0
3.1
2.3
1.4
1.2
1.3
1.1
9.1
7.7
5.3
4.4
4.4
4.5
Decentralized entities
Total
Source: DAF, MFPC: Informe sobre la Viabilidad Fiscal de los Departamentos, Vigencia 2010.
Note: GDP = gross domestic product.
of 1 percent in 2008.16 Thus, the problem of SNG debt since the early
2000s was not large in the aggregate size of the debt, but rather that the
debt was concentrated in a small number of places, often those with
guerrilla and drug problems. Thus, having instruments to deal with
these places—failed SNGs—was important more for political and social
reasons than for dealing with a macroeconomic problem. Both the
number of places and the size of the drug problem have declined since
the early 2000s. The strengthened regulatory framework, as explained in
the next section, together with the turnaround in the economy and the
broader reform in the intergovernmental fiscal system, have contributed
to the improved subnational fiscal and debt position.
When SNGs have gotten into fiscal distress and excess indebtedness
since the early 2000s, it has usually not resulted primarily from formal
spending and borrowing from banks and capital markets, given the new
fiscal rules on direct borrowing. Rather, much of the excess debt has
arisen from arrears in payments to employees, national tax authorities,
and suppliers, and from contributions to pension funds, from court judgments against the SNGs, and from penalties and interest accrued on these.
Sometimes these arrears occurred when the subnational treasury was
unable or unwilling to pay. In other cases, there was collusion between
the claimant and a local official—and perhaps also with participation of
Colombia: Subnational Insolvency Framework
a lawyer and a judge—who agreed to a delay of payment in order to be
able to sue later for a much larger amount including penalties and interest
for the late payment. A judge could then embargo (sequester) the bank
accounts of the local government in order to seize central government
transfers for payment of these judgments. Such conjunctions of legalism
and collusions have negatively impacted the finances in a few localities,
and the central government has adopted various legal and administrative
measures to try to protect legitimate public finances. Such occurrences are
more frequent in places with a high incidence of poverty and violence—
and with influence by narcotraffickers, guerillas, or the paramilitary.
Legal Framework for Subnational Borrowing
The subnational borrowing framework in Colombia developed in parallel with the political and fiscal decentralization and the fiscal crisis discussed above. The goal of the borrowing framework, developed mostly
during 1997–2003, was to avoid situations of fiscal and debt distress in
SNGs. However, if the ex-ante constraints on borrowing were insufficient, then the ex-post insolvency procedures discussed in the next
section could be applied.
Cross-country experience shows that deficits and debt arise from the
joint decision of governments and their creditors (including suppliers
allowing extended payments). These decisions are made in light not
only of the rules governing issuance of the debt, but also the ex-ante
expectations about what will happen to the debtor and the creditors if
payment difficulties arise—who will lose money or who will be forced
into painful adjustment. The decisions of that lending moment become
a fait accompli conditioning the subsequent decisions. This points
to two important dimensions of control of government borrowing:
(a) their type and timing relative to the initial lending decision, that is,
ex-ante controls or ex-post consequences; and (b) whether the ex-ante
controls and ex-post consequences act on borrowers or on lenders, as
displayed in table 5.6 (Liu and Webb 2011).
The legal framework currently governing Colombia’s subnational
borrowing and insolvency is summarized in table 5.7, which corresponds to the four quadrants shown in table 5.6.
The first major step to increase the ex-ante control of subnational
debt was Law 358/1997, which forbade borrowing to finance current
193
194
Until Debt Do Us Part
Table 5.6 Channels for Control of Deficits and Debt: Lender-Borrower Nexus and Timing of
Controls and Sanctions
Timing relative to
lending decision
For borrowers
For lenders
Ex-ante controls
All governments
• Debt and deficit ceilings
• Restrictions on international borrowing
• Publication of detailed fiscal results
SNGs only
• Regulation of SNG borrowing, based
on fiscal-capacity criteria (regulations
by central government or SNG itself,
central bank, or other institution)
All governments
• No direct central bank financing
• Regulations by central bank or
other financial supervision agency
SNGs only
• Cap on total borrowing by SNGs
• Increased capital requirements for
lending to risky SNGs
Ex-post
consequences
All governments
• Limits on central bank financing
• No bailouts (from central government
or from international community) and
no debt workout without adequate
conditionality
• Publication of detailed fiscal results
SNGs only
• Central government does not accept
SNG debt
• Debt service withheld from transfers
to SNGs
• Insolvency system
All governments
• Strong supervision of banks
SNGs only
• Regulations require capital writeoffs for losses from SNG debt
• No central bank bailouts
• Well-functioning financial market
can increase risk premium for
lending
Source: Adapted from Liu and Webb 2011.
Note: SNG = subnational government.
expenditures and linked subnationals’ issuance of new debt to their
overall payment capacity. This so-called Traffic Light Law (Ley de Semáforo) introduced a rating system for territorial governments based on
a liquidity indicator (interest payment/operational savings17) and a
solvency indicator (debt/current revenue). Both indicators had to be
calculated for each new loan, determining whether the entity had the
capacity to incur further borrowing. Any SNG in the red light category
was prohibited from borrowing without case-by-case permission from
the MFPC.18 Entities in the yellow light category could contract new
debt if the percent increase in debt outstanding was lower than the Consumer Price Index inflation for that year. Otherwise, municipalities and
departments had to obtain permission for new credit from the departmental governor or the MFPC (see table 5.8).19
The MFPC could make the permission conditional on the adoption by the SNG of an adjustment program that included measures to
Colombia: Subnational Insolvency Framework
195
Table 5.7 Ex-Ante and Ex-Post Fiscal Legislation
Borrowers
Lenders
Ex-ante controls
Regulation of borrowing based on fiscal
capacity
• Law 358/1997 and Law 795/2003: Traffic light
system links borrowing to ability to pay
• Law 617/2000: Limits on current spending and
new classification of capacity to pay
• Law 819/2003: Budget management and
transparency rules
Ban on credits to
SNGs in violation of
limits of Laws
358/1997, 617/2000,
795/2003, and
819/2003
Ex-post consequences
(incentives for ex-ante
caution)
Central government does not bail out SNGs
• Law 549/1999: Creation of the Pension Fund
for Subnational Governments
• Ban on financial support of SNGs that are not
in line with Laws 358 and 795
• Debt service withheld from transfers in
restructuring agreements
• Law 550/1999: Restructuring of insolvent
subnational entities
• Decree 28/2007
Write-offs required
for losses from SNG
debt
Law 550/1999
Law 617/2000
Source: Adapted from Liu and Webb 2011.
Note: SNG = subnational government.
cut costs and improve own-revenue collection in order to reestablish
economic and financial stability and guarantee the repayment capacity (Art. 9). The adjustment program would be active until the ratio of
interest payments/operational savings declined below 40 percent. The
SNGs undergoing such a program had to report on a quarterly basis
to the Fiscal Affairs Department (Departamento de Asuntos Fiscales,
DAF), which evaluated the implementation of the program, in coordination with the comptroller general of the republic. Not implementing
the adjustment programs and obtaining new borrowing in violation of
the limits established by the law could lead to sanctions. Also, the Superintendency of Banks could penalize financial institutions that gave credits to subnational entities in violation of Law 358/1997 (Art. 10).
Despite Law 358/1997, some governments with a red-light rating
obtained new financing without MFPC permission by presenting defective financial information, which was only superficially analyzed by the
creditors. In addition, the MFPC gave its authorization in cases where
it should have denied it. As a result, department debt indictors deteriorated from yellow to red instead of improving from yellow to green,
and subnational debt still grew by 15 percent a year, on average, during
196
Until Debt Do Us Part
Table 5.8 Indebtedness Alert Signals
Autonomous
indebtedness
green light
Indicator
Intermediate
indebtedness
yellow light
Critical
indebtedness
red light
Debt interests/
operational savingsa
(liquidity indicator)
< 40%
40% < 60%
> 60%
Debt balance/current
revenue (solvency
indicator)
< 80%
< 80%
> 80%
Effect
Territorial Entity
(Entidad Territorial,
ET) is allowed to
contract new credit
autonomously.
(a) ET can contact
autonomously.
(b) Requires indebtedness
authorization of
the Ministry of
Finance or the
department, which
will be conditioned
to the signing of a
Performance Plan
with the financial
institutions.
Authorization is
required to contract
credit operations,
thus a Performance
Agreement with the
financial entities
should be signed.
Source: MFPC.
a. Operational savings is defined as current revenue – current expenditure (excluding interest payments).
1998–2000. Similarly, in the absence of a fixed ceiling on current expenditures of the core administration, adjustment programs did not always
bring about stronger fiscal discipline at the local level. As a result, the
current expenditures of the core administration continued to rise fast,
and subnational entities remained dependent on transfers (Echavarria,
Renteria, and Steiner 2002).
Law 617/2000 introduced a more systematic framework for ex-ante
measures to avert fiscal crisis, by controlling the growth of operational
expenditures. For this purpose, Law 617 classified departments and
municipalities according to population size and the amount of freely
disposable (nonearmarked) revenues, and set limits on the ratio of
discretionary current expenditure to nonearmarked current revenues.
These limits were established in the law and depend on the size of
the SNG. At the same time, the central administration (including the
central bank) was not allowed to make transfers to SNGs that did not
meet the requirements, and an extensive list of requirements for the
Colombia: Subnational Insolvency Framework
election of governors, mayors, legislators, and their relatives aimed to
increase transparency.
SNGs that exceeded the limits set forth in this law had to execute a
fiscal adjustment program with precise performance targets in order to
regain fiscal viability. The NPD and the MFPC supervised the execution
of the adjustment plan for municipalities and departments, respectively.
The NPD could order a new adjustment program should a municipality
fail to meet the targets in its original program. Should a municipality
again fail to establish viability within two years, it might be required to
merge with another municipality, although this never actually happened.
In addition, Law 617 allowed SNGs under specific conditions to
request a central government guarantee for up to 100 percent of new
credits that were contracted to finance fiscal adjustment plans (for
example, the costs of personnel retrenchment) that were endorsed
before June 30, 2001. To benefit from a guarantee, the territorial entities
(a) had to be in need of a fiscal adjustment program but without own
resources to implement it, (b) had to commit to implement it in line
with the modalities established by the law, and (c) needed to have debt
that had to be restructured to reestablish payment capacity. At the same
time, SNGs’ creditors had to commit to give new credits to finance the
fiscal adjustment program, and the liabilities had to be restructured in a
way that assured payment capacity.
In 2003, Law 795 and Law 819 further strengthened the borrowing
framework for subnational entities. Art. 120 of Law 795 eliminated the
yellow category of the traffic light system. Entities previously categorized as yellow then fell under the red rating, tightening the borrowing
restraints on SNGs. The 2003 Fiscal Transparency and Responsibility
Law, Law 819/2003, strengthened the traffic light system, adding to the
indicators from Law 358/1997 the requirement that the budget has to
be balanced over a 10-year period. The law also increased transparency
and improved fiscal coordination among different levels of government,
since it required both the central administration and local governments
to present each year a consistent 10-year macroeconomic framework.
Expenditure authorizations and revenue collection at all levels of government had to be consistent with this framework, and any deviations
from the framework had to be authorized by the MFPC. Both the central and subnational budgets had to fully comply with the medium-term
197
198
Until Debt Do Us Part
frameworks. Law 819 also introduced market-based incentives, since it
required all subnationals with populations over 100,000 to get a creditrisk rating by a rating agency.
The laws also tightened the regulations on the credit supply side,
prohibiting central government financial support to subnational entities that were not adhering to the fiscal responsibility norms contained
in Laws 617 and 358, or that had debt service arrears for credits to the
national government or for credits that were guaranteed by the nation.
Furthermore, credits from financial and territorial development institutions that did not meet the conditions and limits of Laws 358, 617, 819,
and 795 were invalid, and borrowed funds had be restituted promptly
without interest or any other charges. This would be punishment for
both creditors and borrowers. The MFPC through the DAF monitors
the adherence of the decentralized entities to the different borrowing
and fiscal limits mentioned above.
While the above laws strengthened ex-ante regulations of subnational
debt limitations and fiscal management for both borrowing governments and lenders, Law 550/1999 addressed the bankruptcy proceedings
for SNGs. The details of the content and implementation of Law 550 and
its relationship to Law 617 as part of debt restructuring are discussed in
the next section, “Insolvency Procedures: Implementation and Effects.”
The national government was concerned not only about imprudent
and unsustainable fiscal behavior by SNGs, but also about failures to
deliver essential services, especially those financed by earmarked transfers.
To address the poor service delivery by a few SNGs in sectors financed
with transfers, the Colombian Congress approved in 2007 the Acto Legislativo No. 04, which the executive branch regulated through Decree 28 in
January 2008. Both rules authorized the executive branch to monitor and
control subnationals in the use of earmarked resources financed by the
transfer system. Since entities that get into debt difficulties often, although
not always, have similar problems in service delivery, Decree 28 provides
for an extreme penalty and loss of local control for subnationals that do
not cooperate in a fiscal reform agenda, including insolvency proceedings.
Decree 28 thus added a new element and incentives in the institutional framework that impact the fiscal management of SNGs. The
power of the central government to intervene, when the use of transfers
does not fulfill the expected legal requirements, strengthens the power
Colombia: Subnational Insolvency Framework
of the central government for fiscal oversight and for avoiding situations
where the judicial process impounds their transfers. SNGs with poor
fiscal management face not only the risk of being subject to adjustment
plans defined on the basis of the fiscal insolvency framework; the central
government might also directly take over their financial management.
Insolvency Legal Framework
Subnational entities that surpass the spending and debt limits established in Laws 358/1997, 617/2000, and 819/2003 must undergo fiscal
adjustment plans as described above. In addition, Law 617/2000 also
includes the possibility of restructuring debt and obtaining guarantees
and cofinancing to help subnational entities implement the measures
as outlined in the adjustment plans. A subnational entity that becomes
insolvent, defined as being overdue on payments for at least 90 days,
including court-ordered payments, can also request protection under
Law 550/1999—Colombia’s subnational insolvency or bankruptcy
law—in exchange for a commitment to reduce expenditures and redirect revenues to pay creditors through a restructuring agreement.
Law 550/1999 was introduced as a financial restructuring law for
both private sector companies and subnational entities. The subnational entities covered are departments, municipalities, and districts, as
well as the parts of the decentralized service delivery sector (entidades
decentralizadas del nivel territorial) that are not overseen by any sectoral
superintendency.20 Law 550/1999 was supposed to remain in place for
five years but was subsequently extended and finally made permanent
for subnational entities by Law 1116 (Art. 125 and Art. 126) of 2006.21
Under the jurisdiction of Law 550, subnational entities are protected
from any outstanding or new payment obligations and any court orders
to pay once the insolvency proceedings under Law 550/1999 begin. The
restructuring agreements seek to evaluate, reconcile, and restructure
competing claims on subnational debtors. The insolvency proceeding is
designed to be transparent, with public notices at all stages of the process, including disclosing the restructuring agreement.
Institutional arrangement, process, and triggers. The intervention
under Law 550/1999 is an administrative process, with the SOC acting
as judge. The SOC, a unique institutional arrangement in Colombia,
199
200
Until Debt Do Us Part
Figure 5.4 Restructuring Process under Law 550/1999
SN entity
(mayor or
governor)
Signature of
restructuring
agreement
Solicits
restructuring
DAF verifies if
conditions are
met
Modifications
Vote shares and agreement
are defined
Negotiation
among claimants
and the entity
Supervised by DAF
and Superintendency
of Corporations
Stopped due to
unforseen
circumstances
Implementation
of agreement
Designates
promotor
Promoter
publishes note
Superintendency
of Corporations
Settlement of the
inventory of
claims
Conditions
fulfilled
End
Source: Summary by authors based on consultation with DAF.
Note: SN = Subnational, DAF = Fiscal Affairs Department (Departaments de Asuntos Fiscales).
handles all insolvency cases, due to the recognized weakness of the court
system in the country. The parties involved in the subnational process
are the concerned debtor, the creditors and other claimants, the promoter (to supervise and facilitate the restructuring proceedings), and
the MFPC. Four types of creditors are distinguished: (a) workers and
pensioners, (b) public entities and social security institutions, (c) financial institutions and other entities that are under the supervision of the
Superintendency for Banks, and (d) other claimants (suppliers, contractors, and so forth). In the case of decentralized service providers, creditors can also be internal to the entity, such as shareholders or members
of the service providers. Law 550/1999 outlines in detail the process for
restructuring agreements, as summarized in figure 5.4.
A subnational entity can request debt restructuring under Law 550
if two or more payment obligations are overdue for at least 90 days or
at least two court payment orders have been issued. The accumulated
value of the obligations in arrears must represent at least 5 percent of
the total of obligations that fall due in less than a year. The process is
initiated by a request sent to the DAF by the legal representative of the
concerned subnational entity with the local congress’s authorization.
Colombia: Subnational Insolvency Framework
Next, the DAF verifies that the above-mentioned conditions are
fulfilled and, if they are, accepts the request to start the restructuring
agreement negotiations. The DAF designates a civil servant or contractor to act as a promoter to supervise and facilitate the debt restructuring
negotiations. The promoter checks the accuracy of the financial statements and projections, keeps all involved parties informed, and coordinates the negotiations. The promoter also calculates the voting rights of
the creditors and is part of the committee that supervises the execution
of the agreement (Art. 8, 550). Although the promoter does not vote on
the agreement or any modifications thereafter, she or he plays a pivotal
role in informally advising the subnational debtor and representing the
views of the DAF, whose support is usually needed for the restructuring
to succeed.
Once the negotiations start, public services that were suspended
must be resumed, and all existing or new payment orders, legal claims
(procesos de ejecución), and seizures of assets are suspended and no
new ones can be initiated. The entity is not allowed to make any payments or contract any new obligations that are not strictly necessary to
sustain service provision, except with written previous authorization of
the DAF.22 Administrative costs (mainly wages and utility bills) must be
paid during the negotiations and the life of the restructuring agreement
and have priority over other claims. Creditors whose claims are guaranteed by specific collateral with the entity must decide whether to call the
guarantee (take the collateral) or to include the claim in the restructuring agreement. Other creditors’ claims (including those who got judicial
embargoes) are put on hold until the agreement is signed.
An important responsibility of the promoter is to determine the
voting rights of the individual creditors. The elements included in the
calculation vary among creditors. For instance, the voting-share value
of claims from labor, pension funds, and tax authorities include interest and penalties. Pension fund claimants get an extra 25 percent voting
weight added to the principal of their recognized claim. Voting rights
for financial sector and other claimants, in contrast, are based on only
the principal overdue, excluding interest or penalties that have not been
legally capitalized.
The law provides four months for the promoter to create the inventory of claims and thus set the voting shares of the claimants and then,
201
202
Until Debt Do Us Part
if the claimants accept the inventory, the claimants and the entity have
four additional months to reach and sign an agreement. The process
usually meets these two deadlines. Delays often come in the middle,
when claimants protest because they are not satisfied with the size of
liabilities they are assigned in the inventory. The SOC then resolves
the disputes, which usually takes 6–12 months. The negotiations conclude with the voting on the agreement, which becomes binding if it
is accepted by at least 51 percent of the votes.23 The voting weight of
each claimant in negotiating and settling on the restructuring agreement depends on the amount owed to the claimant (Art. 22); those with
the largest claim (as settled in the middle part of the 550 process, with
the SOC as referee) get the largest weight in determining the terms of
the agreement, although the majority of claimants voting in favor must
include at least one claimant from at least three categories. The agreements establish an 18-month period, starting when the agreement is
signed, to go through the contingent liabilities, except those that require
a judicial pronouncement to be recognized.
The format and content of the restructuring agreement are also regulated in detail: It must be in written form, signed by all creditors who
have voted in its favor, and deposited with the DAF (Art. 31). The agreement must outline detailed rules for the financial and administrative
planning and execution of the entity, and the modalities and conditions
for the payment of pension, labor, social security, and fiscal liabilities
during the restructuring period. The agreement must also outline the
contribution to the pension fund, FONPET, and the conditions and
modalities to repay its creditors. In addition, the agreement must establish and outline the rules of an Agreement Supervisory Committee,
which consists of all the creditors and the promoter, who has a voice
but no voting rights. Other issues addressed in the agreement include
the procedures in case of failure to fulfill the obligations in the agreement and, if necessary, any special labor agreements during the life of
the agreement.
The Superintendent of Banks does not seem to be an important
player in the negotiations, although it plays a background role by
enforcing the regulations for classifying the riskiness of loans and the
corresponding capital risk weighing and provisioning, which helps
shape the incentives of the financial institutions as they participate in
Colombia: Subnational Insolvency Framework
negotiating the restructuring agreements. More important is the sister
agency, the SOC. As the insolvency authority, the SOC plays a key role in
settling disputes over the amounts of claims from various claimants and
in helping to supervise implementation of the agreements.
Priority structure. The agreement also establishes the priority of payments, which must be in accordance with Art. 58, which establishes the
following order:
1.
2.
3.
4.
5.
6.
7.
8.
9.
Pension contributions to individual accounts by subnational entity
Salaries (servicios personales)
Payroll transfers (transferencias de nómina)
General expenditures (gastos generales)
Other transfers
Interest payments
Amortization of debt
Financing of the deficit of previous years
Investment expenditure contracts.
The labor claimants and fiscal (tax) claimants (including the
pension funds) must be paid in full for the claims that are recognized, but the timing of these payments is negotiated and set in the
restructuring agreement.24 So, for example, banks may agree to take
a more reduced share of what is owed them if they get the money
sooner than the others. The restructuring agreement is binding on
all parties involved, including those minority of creditors who voted
against it. The legal consequences of the agreement include the following (Art. 34):
• The disposition of all asset titles to the supervision committee as out-
lined in the restructuring agreement
• The removal of all existing preventive measures except the ones
related to the national tax department (DIAN)
• The suspension of all legal processes against the territorial entity
• The suspension of the eligibility of all charges and guarantees during
the term of the agreement
• Holders of fiduciary guarantees derived from real estate or mortgages
must accept the substitution, if they did not exercise their claims to
real property at the beginning of the process
203
204
Until Debt Do Us Part
• The territorial entity cannot borrow without explicit MFPC authorization, as established in Law 358 (all entities in 550 status would fall
into the red light category under 358, at least initially) and reinforced
in Laws 617 and 319.
Restructuring agreements can be modified by an absolute majority of
voting shares, for instance, if the macroeconomic environment changes
considerably or if liabilities are found that were not addressed in the
original agreement.
Termination of Law 550 agreement. The restructuring agreement
ends (without voting by the creditors) when the date is reached that was
agreed upon or when the obligations in the agreement have been fulfilled (Art. 35). If the fiscal position of the SNG improves more rapidly,
due to, say, an export-led economic boom, the SNG is not obliged to
accelerate payment, but rather may pay only at the minimum required
pace per the agreement and thus prolong its stay in the shelter of Law
550. Other events that can trigger the end are a failure by the concerned
entity to adhere to the agreement or a finding by the supervisory committee that unforeseen circumstances render the agreement impossible.
Nonpayment of liabilities after the initiation of the negotiations or serious misconduct by the subnational entity can also lead to a suspension
of the agreement, if a majority of the creditors approve the suspension.
Suspension of the 550 agreement means, in practice, that claimants can
fall back on judicial orders that allow them to embargo transfers from
the central government, which not only disrupts local service delivery
but would typically alter the shares going to each creditor.
Insolvency proceedings for decentralized service providers. Law
550/1999 covers only the parts of the “decentralized service delivery sector” (entidades descentralizadas del nivel territorial) that are not overseen by any sectoral superintendency.25 The institutions outside Law 550
include private and public health providers, organizations that manage funds from the General System of Social Security for Health, and
providers of other public services. In insolvency cases of such entities,
the concerned superintendency takes possession of the entity to ensure
that the social services continue to be provided. If this is not possible, it
orders the liquidation of the entity. Depending on the superintendency
in charge of the sector, different laws and decrees regulate the details
(see table 5.9), but the proceedings are similar.
Colombia: Subnational Insolvency Framework
205
Table 5.9 Colombian Superintendencies
Responsible superintendent
Sectors
Laws and decrees
Financial Superintendency
(Superindencia Bancaria)
• Banking, insurance, etc.
• Articles 72 and 73 of Decree
4327/2005
National Health
Superintendency
(Superintendencia de Salud)
• General System of Social
Security for Health
• Entities that provide
monopolistic services in the
health sector
• Article 42, 42.8 of Law 715/2001
• Number 5 of article 68 of Law
715/2001
• Number 5 of article 37 of Law
1122/2007
• Number 13 of article 8 of
Decree 1018/2007
Public Services
Superintendency
(Superintendencia
de Servicios Públicos
Domiciliarios)
• Providers of public
services in water, sewerage,
electricity, fuel gas,
basic public telephone
distribution, mobile in rural
areas, etc.26
• Art. 121 of Law 142/1994
Same liquidation proceedings as
for financial institutions
Law 812
Economy Superintendency
(Superintendencia de
Economía)
• Financial Cooperatives
• Saving and Credit
Cooperatives
• Cooperatives that provide
saving and credit services
• Decree 756/2000, Decree
2206/1998, Decree 1401/1999
Source: Based on author’s fieldwork.
Insolvency Procedures: Implementation and Effects
Implementation of Insolvency Procedures
From 1999, when Law 550 was enacted, to 2010, 94 territorial entities—
13 departments, 7 capital cities, and 74 municipalities (not capitals)—
have restructured total claims worth 5.246 billion pesos (approximately
1.23 percent of GDP) (table 5.10).27 Twelve percent of these claims had
been eliminated in the clean-up and reconciliation process (depuración).
Fifty percent of the principal of these claims was reduced through negotiation or has been paid off,28 leaving a balance of 38 percent. Besides
restructuring, SNGs also used the possibility of prepaying domestic debt
using the funds from the Oil Saving and Stabilization Fund (Fondo de
Ahorro y Estabilización Petrolera).29 DAF estimates show that almost
900 billion pesos have been prepaid using these funds.
During the same implementation period, 94 subnational entities
entered Law 550 protection, of which 53 were carrying out debt restructuring agreements; 31 had successfully completed restructuring; 4 had
attempted restructuring but did not reach agreement; and 6 had failed
206
Until Debt Do Us Part
Table 5.10 Total Debt Restructured under Law 550, 1999–2010
thousands of millions of Colombian pesos
Entity
Total claims
% of total
claims
Departments
2,414
Capital municipalities
1,635
Municipalities
1,197
5,246
100
Total
Depuracióna
46
Reduction in
principalb Balance
88
1,139
31
257
23
259
604
% of total
debt
1,187
59
823
555
28
674
264
13
2,636
2,006
100
Source: DAF, MFPC: Informe sobre la Viabilidad Fiscal de los Departamentos, Vigencia 2010.
a. Depuración is the process of refining the list of claims.
b. Combination of payoffs and negotiated capital reductions.
Table 5.11 Entities Restructured under Law 550, 1999–2010
Law 550
Departments
Municipalities
Total
Restructuring under way
7
46
53
Restructuring completedb
5
26
31
Agreement attempted
0
4
4
1
5
6
13
81
94
a
Agreement failedd
Total
c
Source: DAF, MFPC, December 2010.
a. Departments: Cordoba, Narino, Putumayo, Sucre, Bolivar, Magdalena, and San Andres; Municipalities: Achi,
Codazzi, Ambalema, Aracataca, Armero, Ayapel, Barranquilla, Canalete, Casabianca, Cerete, Cienaga, Dagua,
El Molina, Fundacion, Galeras, Guaranda, Herveo, Honda, Isnos, La Jagua, Libano, Lorica, Magangue, Majagual, Monteria, Novita, Nuqui, Patia, Planeta Rica, Pradera, Puerto Libertador, Sabanalarga, Salamina, San
Benito Abad, San Juan de Uraba, San Pelayo, Santa Ana, Santa Marta, Since, Soledad, Tumaco, Tamesis,
Turbo, Valencia, and Zaragoza.
b. Departments: Amazonas, Cauca, Guainia, Tolima, and Vichada; Municipalities: Astrea, Becerril, Buenaventura, Cartago, Belen de los Andaquies, Caucacia, Condoto, Cordoba, Distraccion, El Reten, Fonseca, Inirida,
La Paz, Lerida, Leticia, Maicao, Mariquita, Palmira, Pamplona, Popayan, Providencia, Riohacha, Rovira, Sandona, Tolu, Toluviejo, and Villanueva.
c. Puerto Rico, Corozal, Guapi, and Mocoa.
d. Agreement failed due to noncompliance: Department of Choco; Municipalities: Bahia Solano, Istmina,
Malambo, Plato, and Tado.
due to noncompliance, mainly due to corruption and weak capacity
(table 5.11). Also, the inability to obtain a majority of claimants who
would vote for an agreement led to the failure of 3 agreements and 4
modifications to agreements. These 94 entities comprised 13 departments (with 24 percent of the total population) and 81 municipalities
(with 15 percent of the total population).30
Figure 5.5 shows the frequency per year of insolvency debt restructuring agreements under Law 550 since they started in 2000. Most debt
Colombia: Subnational Insolvency Framework
Figure 5.5 Departments and Municipalities under Debt Restructuring Agreement,
1999–2010
a. Departments
6
Number of entities
5
4
3
2
1
10
20
09
20
08
20
07
20
06
20
20
05
04
20
03
20
02
20
01
20
20
00
0
Year
Agreement signed
First modification
Second modification
Successfully completed
b. Municipalities
25
Number of entities
20
15
10
5
Year
Agreement signed
First modification
Second modification
Successfully completed
Source: DAF, MFPC: Informe sobre la Viabilidad Fiscal de los Departamentos, Vigencia 2010.
10
20
09
20
08
20
07
20
06
20
05
20
04
20
03
20
02
20
01
20
20
00
0
207
208
Until Debt Do Us Part
restructuring took place in the early 2000s. The agreements were modified by 5 departments and 14 municipalities. Three agreements went
through modifications twice.
There are wide differences among the nature of the claimants, differences in their legitimacy, and different legal treatments for each, as
explained above. Outcomes are unpredictable because some claimants
(usually not registered as financial creditors) often appear with inflated
claims, and the outcomes depend on the extent to which these claims
are recognized. The entire array of claims is often not well known in
advance. Some claims are often reduced in the purification process—
such as arrears in wages and pensions—but then the recognized claims
of employees are paid in full, as required by law, and on a schedule over
time. Financial sector claims are typically better documented, so they are
likely to survive intact through the purification process, but then those
creditors usually take a haircut on principal or interest in the restructuring process. The timing of payment is part of the negotiations, and
the financial creditors usually give up some principal or reduce interest
rates in order to clarify their losses and to receive some money sooner.
Implementation of Law 550 in the beginning phase in 2000 was
assisted by Law 617, which facilitated the implementation of Law 550
and provided an alternate route to debt restructuring for insolvent subnational debtors. A subnational debtor would qualify for central government guarantee for debt restructuring, under Law 550, when meeting
certain conditions. If a subnational debtor (a) entered Law 550 during
the first six months of 2000, (b) produced a fiscal program approved
by the DAF, (c) got the creditors to agree to a restructuring plan, and
(d) was on schedule in payment to tax authorities, then MFPC would
guarantee 40 percent of the restructured debt.
Another aspect of Law 617 was available during the same six-month
window: the central government offered a 100 percent guarantee of new
loans to finance adjustments such as retrenchment of workers as part
of the fiscal plan to bring the entity within the spending limits set in
Law 617. Also, the central government offered a 40 percent guarantee to
entities for their adjustment lending from commercial banks or international development banks—that is, restructured debt (see tables 5.10
and 5.11). This support, based on decisions by the DAF, was crucial for
a successful workout under Law 550, because Law 550 only stops the
Colombia: Subnational Insolvency Framework
immediate squeeze from claimants, but it does not give the entity fiscal
space with which to make investments and reforms that could restore
long-term fiscal viability.
Subnational entities have benefited from the assistance of Law 617
with total restructured debt at 1,822 billion pesos. Sixty percent of
the restructured debt had a 40 percent guarantee from the center, and
19 percent had a 100 percent guarantee.
Although Law 550 restructures the debt of insolvent subnational
debtors, it cannot address the root course of insolvency, because Law
550 cannot force subnational debtors to undergo fiscal adjustment. In
this respect, Law 617 supports Law 358 by providing a fiscal adjustment
framework to ensure the effectiveness of ex-ante fiscal rules as defined
by Law 358. Law 617 also supports Law 550 by providing central government assistance in fiscal adjustment so that the debt restructuring
can lead to a realistic and sustainable path. Table 5.12 summarizes the
number of entities with fiscal adjustment programs under Law 617, Law
358, or both, under the supervision of the DAF.31
Effects of Insolvency Procedures
Given the lender-borrower nexus and various channels that would influence government fiscal deficits and indebtedness, it would be difficult to
precisely separate and measure the effects of each individual law that
aims at enforcing fiscal discipline (Liu and Webb 2011).32 The evaluation of the insolvency proceedings must be within the broader context
of fiscal reforms. Law 550 does not operate in a vacuum. As shown, the
insolvency proceedings in Colombia followed a series of reforms including the traffic light law in 1997, and was assisted by parallel or follow-up
legislation, such as Law 617, which helps the fiscal adjustment process.
Other factors have been in play. Colombia has enjoyed solid economic growth and macroeconomic performance from 2003 onward.
Compared with the annual average growth of 1.6 percent from 1997
to 2002, real GDP accelerated to almost 5.9 percent from 2003 to 2007.
The global financial crisis has impacted the Colombian economy, but
the economy still grew at 4.5 percent from 2008 to 2010, on average,
and benefited from higher commodity prices, which increased royalty
income from coal and oil. Inflation, which was in the double digits from
1997 to 2002, was reduced to an annual average of 4.5 percent from 2003
209
210
Until Debt Do Us Part
Table 5.12 Entities under Law 617 and/or Law 358, 1997–2010
Law 617 and Law 358
Restructuring under way
a
Restructuring completed
Departments
Municipalities
Total
3
2
5
1
28
29
Agreement failedc
0
1
1
Total
4
31
35
b
Source: DAF, MFPC, December 2010.
a. Departments: Atlantico, Guajira, and Valle del Cauca; Municipalities: Cali and Corozal.
b. Departament: Santander; Municipalities: Bello, Cajamarca, Caramanta, Cartagena, Charalá, Chinchiná, El
Copey, Espinal, Filandia, Girardota, Guacarí, Guamo, Ibagué, Icononzo, Jamundí, Maria la Baja, Pacho, Palestina, Piedecuesta, Rionegro, San Diego, San Gil, Villa María, and Villarica, under 617; and Manizales, Fresno,
Roldanillo, and Santa Isabel, under 358.
c. Quibdo, capital of Choco, did not complete the performance agreement.
to 2007. Monetary policy has helped maintain an internal balance, since
inflation outcomes remained within the targeted range of 2–4 percent.
Although the fiscal deficit of the combined public sector rose from 2.7 to
3.2 percent of GDP during 2009–10, mainly due to lower oil-related revenues, public debt sustainability is not a major concern in the medium
term.33 The combined public sector debt-to-GDP ratio is projected to
decline from 36.4 percent in 2010 to 31.2 percent in 2016. All three
major rating agencies upgraded Colombia to investment grade in 2011.
Notwithstanding the challenges in analyzing the precise contribution of fiscal legislation and regulatory reforms, to the extent that these
reforms intend to improve government finance and avoid overindebtedness, it is worthwhile ascertaining whether this legislation has been associated with improved fiscal outcomes (Liu and Webb 2011). Overall, the
subnationals’ debt has declined as a share of GDP since the peak of 1999,
their revenues have strengthened, and their credit ratings have improved.
Subnational debt as a share of GDP has declined since its peak—from
almost 8 percent in 2000 to approximately 4.5 percent in 2010—one of
the lowest levels in the last 20 years. Looking more closely, the decline in
debt as a share of GDP has happened for not only the decentralized service entities, as shown in figure 5.3, but also for departments and municipalities, as shown in figure 5.6.
SNG total revenues increased from 7.6 to 10.4 percent of GDP
between 2000 and 2010 as a result of economic growth; improved
revenue collection; and increases in transfers, which accounted for most
of the revenue increase.34 The rating of subnational debt improved,
Colombia: Subnational Insolvency Framework
Figure 5.6 Department and Municipality Debt Balance as Percentage of GDP,
2000–10
3.5
3.0
Percent
2.5
2.0
1.5
1.0
0.5
10
20
09
20
08
20
07
20
06
20
05
20
04
20
03
20
02
20
01
20
20
00
0
Year
Department and municipal debt
Municipalities
Departments
Source: NPD, Desempeño Fiscal de los departamentos y municipios, 2010.
Note: Decentralized service entities are not included in the figure to show only subnational government
debt. GDP = gross domestic product.
because the share of debt that qualified as being served and as posing
little risk (an A rating) increased from 35 percent in 2000 to 90 percent
in 2010 for departments and from 49 to 78 percent for municipalities.35
For those subnational officials in office when the fiscal problems
originated, the political and fiscal consequences are largely moot, since
mayors and governors cannot be reelected and are thus usually out of
office by the time the insolvency crisis must be worked out. Nonetheless,
the governments and parties in power suffer the stigma of bankruptcy
if they go through a 550 process, which may hurt them in the next election. A government’s performance in office correlates somewhat with
election chances of associated candidates. Thus, electoral incentives
exist, albeit weakly, for officials to show good fiscal performance.36
There have been a few clear failures of the insolvency process, mostly
municipalities, and the fiscal problems in those places stem from severe
governance problems—guerillas, narcotraffickers, and so forth—that go
deeper than overborrowing and fiscal irresponsibility. Neither Law 550
nor other parts of the SNG fiscal regulatory regime could have solved
those problems.37
211
212
Until Debt Do Us Part
The most important sanction for subnational officials who do not
comply with the terms of a 550 agreement is that fiscal resources are
withheld, since claimants can again get court-ordered embargoes. A
more complete loss of resources under local control occurs when the
national government authorities use Decree 28, since 2008, to take over
the fiscal and other resources of a sector because the SNG has failed to
deliver critical social services.38
Implementation of Law 550 has succeeded in many ways. The process is transparent and follows the procedures as prescribed by the law.
There have been two unpredicted effects—some agreements take a long
time to finalize, and some SNGs remain in the 550 regime even after
they have become fiscally creditworthy again.
Although many SNGs signed 550 agreements almost a decade ago,
fewer than half have fully paid off their debt since the restructuring agreements were signed. Most of the SNGs are fulfilling their debt
restructuring agreements, although some are struggling financially.
Some subnationals have strengthened their finances sufficiently so that
they prepay their obligations and exit the 550 regime ahead of schedule,
but they prefer to simply pay on schedule and keep the 550 protection
against judicial embargoes.39 Some of the embargoes have legitimate
origins, but others arise from collusions and schemes to obtain public funds for private purposes. They have caused serious disruption
for many subnationals, and some SNGs have used Law 550 as a way to
block the embargoes.40
In judicial embargoes, a judge has no option but to apply the law,
and in individual enforcement, the judge needs to enforce the presented claim, not the other claims. One problem lies in the possibility
that assets belonging to public institutions are available for individual
enforcement of private claims. One potential solution could be to avoid
enforceability, and the judge instead would notify the administrative
authorities and would trigger an “alarm” system, eventually forcing the
public institution to initiate a restructuring proceeding.
Summary and Conclusions
The protection offered by bankruptcy Law 550 enables insolvent SNGs
to reach orderly debt restructuring agreements with creditors, since the
Colombia: Subnational Insolvency Framework
other laws aiming at ex-ante control cannot eliminate the risks of bankruptcy. Courts dealing with individual case-by-case claims are not set up
to solve the problems of competing claims against insolvent subnational
debtors. Corruption in some of Colombia’s local courts exacerbate the
problem, but even without the corruption issue, if a case were not under
a bankruptcy law, courts would have to act on claims in the order the
claimants came to the court, rather than the seniority order of claims.
The factors leading to insolvency changed over time in Colombia,
and therefore the clientele and effects of the law changed. During 2000–
02, most departments and larger municipalities going into 550 had had
access to credit markets in the 1990s and overborrowed. The successful
550 processes for them led to restoration of market access, and new laws
after 2000 regulated SNG borrowing, so that both lenders and SNGs
became more cautious and overborrowing rarely occurred. After 2003,
the new insolvency cases entering the 550 process were mostly municipalities, and mainly resulted from (nondebt service) arrears and the
subsequent penalties and interest.
Law 550 focuses on workouts from bankruptcy and has limited ability
to address the root causes of fiscal stress and debt. Other complementary laws—mainly Laws 358, 617, and 819—work in several ways by
(a) limiting borrowing, (b) promoting fiscal transparency, (c) strengthening the budgetary process, and (d) helping to finance debt restructuring.
Although Law 550 does nothing to address governance issues directly, it
has helped create fiscal space for some newly elected governments with
insolvent situations to have a fresh start and carry through reforms.
A unique feature of the Colombia case is the use of an administrative
apparatus to deal with insolvency. The SOC, a venerable administrative
agency in Colombia, fills the role that bankruptcy courts do in some
countries. This agency was designated in the Bankruptcy Law, rather
than the courts, which have historically been weak and are sometimes
complicit in pushing claims that lead to bankruptcy. The Ministry of
Finance, through its DAF, works closely with the insolvent SNGs and
with the SOC to resolve the insolvency cases. Despite the country-
specific institutional setup, the results of the insolvency process in
Colombia are similar to the results elsewhere.
Macroeconomic fundamentals matter for the success of insolvency
procedures. The implementation of Law 550 and other fiscal legislation
213
214
Until Debt Do Us Part
has taken place in the context of improving the macroeconomic performance of the country since 2003. Colombia’s key macroeconomic indicators, such as GDP and inflation, have remained relatively strong even
after the global financial crisis of 2008–09. This helped the turnaround
of the fiscal balance through higher revenues and moderate interest rates.
The reform in the intergovernmental fiscal transfers system also played a
positive role in stabilizing central government transfers to SNGs.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. In the Colombian system, all insolvency processes are administrative; there is
no intervention of judges in normal insolvency procedures.
2. For a review of these country experiences, see Liu and Waibel (2009) for a
review of cross-country experience, chapter 7 by Jókay (2103) for Hungary and
chapter 8 by De Angelis and Tian (2013) for the United States in this volume.
Not all local governments are allowed to use the federal court for Chapter 9
filing in the United States, since some states do not give authorization for their
local governments for filing in the federal court. See chapter 14 by Liu, Tian,
and Wallis (2013) in this volume for a review of how 50 states handle local government insolvency in the United States.
3. Subnational insolvency was added to the jurisdiction of the SOC through Law
550 in 1999. The SOC was started in the 1930s, and it is an institution with
multiple functions as a company registry, a company supervisor, and an insolvency regulator with quasi-judicial functions.
4. This section draws on chapter 1 of MFPC (2009).
5. Law 33/1968 and 46/1971 implemented the constitutional mandate through
formulas for revenue sharing of funds earmarked for specific sectors, mostly
education and health.
6. Such as water utilities, construction of health and education facilities, agricultural technical assistance, urban development and transport, and other local
infrastructure.
7. The term territorial governments is used in Colombia and means the same thing
as subnational governments.
8. Data: National Statistics Administrative Department (Departamento Administrativo Nacional de Estadística, http://www.dane.gov.co/#twoj_fragment14MFPC: Balance Fiscal Gobierno Nacional Central 1994–2010. The SNG share
in spending excluded spending by decentralized entities (mainly infrastructure
companies owned by SNGs).
Colombia: Subnational Insolvency Framework
9. Education and health account for about 41 percent of spending by departments
and 45 percent by municipalities (World Bank 2009).
10. Including the Situado Fiscal and the Participaciones Municipales.
11. Initially, there was a danger that the specified real growth of transfers would
outpace the (perhaps negative) growth of real GDP and revenues, as happened
in Argentina during 1999–2001, but fortunately Colombia’s real growth turned
increasingly positive (see Gonzalez, Rosenblatt, and Webb 2004).
12. The last revision to the GTS under Law 1176/2007 slightly modified the distribution system and defined the amounts over the medium term.
13. See World Bank (2009), tables 1.1C and 1.2C for details.
14. Subnational Development Financial Entity (Financiera de Desarrollo Territorial, SA, FINDETER), whose responsibilities are rediscount credits from
subnational entities and its decentralized entities, metropolitan areas, municipality entities or associations, among other financial services; http://www
.findeter.gov.co/aymsite/index_en.php. Institutes for Territorial Promotion
and Development (Institutos de Fomento y Desarrollo Territorial, INFIS) are
financial institutions not overseen by the Financial Superintendency. See Duff
and Phelps de Colombia SA, Finanzas Territoriales, Special Report, April 2009.
15. These three companies are TGI International Ltd., Bogotá Electricity Company,
and Medellin Public Enterprises. Source: Bloomberg.
16. MFPC 2009, 28.
17. Operational savings is defined as the difference between current income and
noninterest expenditures.
18. Art. 15 establishes a transition period after the law becomes effective during
which entities with both indicators above the critical indebtedness threshold
would be able to increase their net debt up to 60 and 40 percent of the Consumer Price Index for the first and second year, respectively.
19. Districts and municipalities that are capitals of departments must obtain permission from the MFPC.
20. Law 550/1999, Art. 58 and modifications to Law 617/2000 Art. 69.
21. Insolvency proceedings for private companies are now regulated by Law 1116/
2006 (and modifications introduced by Law 1173/2007 and Law 1380/2010).
22. Art. 17 and Art. 3 of Decree 694/2000. Subnational entities are forbidden
from taking the following actions: (a) incurring new debt or other liabilities;
(b) modifying labor contracts or hiring new staff; (c) modifying the budget
or starting projects that incur substantial expenditures; (d) buying or selling
assets; (e) fulfilling guarantees; and (f) compensating, paying, or clearing any
obligations except the ones that are strictly necessary to avoid paralysis of the
basic service provision or that affect fundamental rights.
23. The votes are weighted by the share of each claimant in the total liability, as
determined by the promoter and the SOC. The pension claimants choose a
spokesperson who has a unified vote for them as a group. Law 549 (1999) provides further details on defining pension liabilities.
215
216
Until Debt Do Us Part
24. Pension funds arrears are often overstated, sometimes unintentionally, because
workers often switch pension funds, but the old fund does not know this and
sees only missed payments to its client accounts (based on mission interview in
October 2010 with a lawyer for pension funds).
25. Law 550/1999, Art 58 (1) and modifications in Law 617/2000 Art. 69.
26. Services related to telecommunications such as telephone distribution and local
mobile service in rural areas are no longer regulated by the Superintendency of
Public Services, but by the Telecommunications and Information Technology
Law 1341 of 2009.
27. The total claims are calculated by adding the claims recognized by subnational
debtors and the additional claims brought by creditors.
28. Payments done by the territorial entities.
29. This was under Law 633/2000, which regulates the Oil Saving and Stabilization
Fund.
30. In addition to departments and municipalities, four entities had also used
Law 550 during the same period. These entities are hospitals under the Cesar
Department (Universidad del Atlántico and ESE Hospital Olaya Herrera de
Gamarra, ESE Hospital Regional Jose Villafañe de Aguachica, and ESE Hospital
Inmaculada Concepción de Chimichagua).
31. A case study of Nariño, a department at the south end of the Pacific coast,
shows that Nariño benefited from MFPC support under both Law 550 and
Law 617. Indeed, it seems to be a classic case of using these two programs, in
the sense that it needed both the bankruptcy protection of Law 550 and the
structural adjustment supported by 617. The downsizing of personnel is especially important to Nariño’s fiscal adjustment. Nariño is a successful example of
implementing debt restructuring under Law 550. The implementation started
without errors and there was apparently no delay or backsliding in implementation. No second debt restructuring was needed. Nariño is on track to exit
from the 550 bankruptcy process as originally scheduled. (Based on authors’
2010 fieldwork.)
32. Corbacho and Schwartz (2007) discuss the problems of determining the direction of causality. Their study compared national fiscal deficits in countries with
and without Fiscal Responsibility Laws, and found that the former had smaller
deficits. Data on subnational deficits for such cross-country comparisons, however, are not readily available.
33. A Debt Sustainability Analysis, prepared by International Monetary Fund staff
(in the context of the 2011 Article IV Consultation), indicates that the trajectory of public debt is declining in the baseline scenario.
34.
Central Bank (Banco de la República), http://www.banrep.gov.co/seriesestadisticas/see_finanzas_publi.htm; IMF - Colombia - Data and Statistics.
35. Source: DAF, MFPC, “Informe sobre la Viabilidad Fiscal de los Departamentos,”
Vigencia 2010.
36. In 2008, a new government in Barranquilla formulated an ambitious plan for
debt restructuring and economic development. The agreement under Law 550
Colombia: Subnational Insolvency Framework
was revised a third time at the end of 2008. It also got new lending under Law
617 to finance administrative reform, with the lending and refinancing backed
by a central government guarantee. In 2010, Barranquilla was able to access the
financial markets without government guarantee. It was the first SNG entity to
get new market financing while still under 550 protection. Combined with the
substantial fiscal adjustment for spending and own revenue, Law 550 protection actually improved Barranquilla’s creditworthiness. (The information on
Barranquilla is based on 2010 fieldwork.)
37. The department of Chocó and other Pacific departments and the Amazonian
region share the common challenge of flourishing illegal armed forces. The
impact of central-government-led efforts in promoting fiscal and debt management in Chocó has been reduced due to the governance-related challenges such as
the absence of the rule of law and of a national political culture (Shepherd 2009).
38. The department of Chocó and some small municipalities with very weak management capacity (and often much corruption and violence) have fallen into
such situations and exist largely only as entities on paper, without substantive roles, since the central government departments manage the social service
spending (Shepherd 2009).
39. Based on discussions with the authorities during October 2010 and July 2011.
For instance, this was the case for Barranquilla; in 2010, it got a good credit rating and returned to the financial markets for fresh lending, although it was still
under 550 protection.
40. Based on field interviews during 2011. Law 550 may not protect SNGs against
all embargos. In the case of St. Marta, several court orders led to an embargo of
up to 25 percent of annual own revenues. Meanwhile, the district did have an
agreement under Law 550. But the 550 order was not respected and the embargoed funds got first priority.
Bibliography
Ahmad, Etisham, and Katherine Baer. 1997. “Colombia.” In Fiscal Federalism: Theory and Practice, ed. Teresa Ter-Minassian, 457–503. Washington, DC: International Monetary Fund.
Bird, Richard M. 1984. “Intergovernmental Finance in Colombia: Final Report of
the Mission on Intergovernmental Finance.” Harvard Law School International
Tax Program, Cambridge, MA.
Central Bank (Banco de la República). http://www.banrep.org/series-estadisticas/
see_precios.htm.
———. http://www.banrep.gov.co/series-estadisticas/see_finanzas_publi.htm.
Corbacho, Ana, and Gerd Schwartz. 2007. “Fiscal Responsibility Laws.” In Promoting
Fiscal Discipline, ed. Manmohan Kumar and Teresa Ter-Minassian. Washington,
DC: International Monetary Fund.
De Angelis, Michael, and Xiaowei Tian. 2013. “United States: Chapter 9 Municipal
Bankruptcy: Utilization, Avoidance, and Impact.” In Until Debt Do Us Part:
217
218
Until Debt Do Us Part
Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and Lili Liu,
311–52. Washington, DC: World Bank.
Dillinger, William, and Steven B. Webb. 1999. “Decentralization and Fiscal Management in Colombia.” Policy Research Working Paper 2122, World Bank,
Washington, DC.
Duff and Phelps de Colombia SA/Fitchratings. 2009. “Finanzas Territoriales.” Special
Report. http://www.fitchratings.com.co/resources/getresource.aspx?ID=534.
Echavarria, Juan José, Carolina Renteria, and Roberto Steiner. 2002. “Decentralization and Bailouts in Colombia.” Research Network Working Paper R-442,
Fedesarrollo, Bogotá, Colombia.
FINDETER. http://www.findeter.gov.co/aymsite/index_en.php.
Gonzalez, Christian, David Rosenblatt, and Steven Webb. 2004. “Stabilizing Intergovernmental Transfers in Latin America: A Complement to National/Subnational
Fiscal Rules?” In Rules-Based Fiscal Policy in Emerging Markets, ed. G. Kopits.
Washington, DC: Palgrave-Macmillan and International Monetary Fund.
Jókay, Charles. 2013. “Hungary: Subnational Insolvency Framework.” In Until Debt
Do Us Part: Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto
and Lili Liu, 261–310. Washington, DC: World Bank.
Liu, Lili, Xiaowei Tian, and John Joseph Wallis. 2013. “Caveat Creditor: State Systems of Local Government Borrowing in the United States.” In Until Debt Do
Us Part: Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and
Lili Liu, 539–90. Washington, DC: World Bank.
Liu, Lili, and Michael Waibel. 2009. “Subnational Insolvency and Governance:
Cross-Country Experiences and Lessons.” In Does Decentralization Enhance
Service Delivery and Poverty Reduction? ed. Ehtisham Ahmad and Giorgio
Brosio, 333–76. Cheltenham, U.K.: Edward Elgar.
Liu, Lili, and Steven B. Webb. 2011. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper 5587,
World Bank, Washington, DC.
MFPC (Ministerio de Hacienda y Crédito Público). 2009. Informe sobre la Viabilidad
Fiscal de los Departamentos, Vigencia 2008. Bogotá, Colombia: MFPC.
———. 2010. Informe sobre la Viabilidad Fiscal de los Departamentos, Vigencia 2009.
Bogotá, Colombia: MFPC.
———. 2011a. Informe sobre la Viabilidad Fiscal de los Departamentos, Vigencia
2010. Bogotá, Colombia: MFPC.
———. 2011b. Marco Fiscal de Mediano Plazo. Bogotá, Colombia.
MFPC Statistics. http://www.minhacienda.gov.co/MinHacienda/haciendapublica/
politicafiscal/informacionfiscal/datoshistoricos.
NPD (National Planning Department). 2010. Desempeño Fiscal de los departamentos
y municipios, 2010. Bogotá, Colombia: NPD.
Rojas, Fernando. 2003. “At the Crossroads of Decentralization: Recentralization,
Federalization.” In Colombia: The Economic Foundation of Peace, ed. Marcelo
Colombia: Subnational Insolvency Framework
Giugale, Olivier Lafourcade, and Connie Luff, 871–96. Washington DC: World
Bank.
Sánchez, Fabio T., and Catalina Gutiérrez. 1995. “Colombia.” In Fiscal Decentralization in Latin America, ed. Ricardo López Murphy. Baltimore: The Johns
Hopkins University Press for the Inter-American Development Bank.
Shepherd, Geoffrey. 2009. “Colombia and the Department of Choco.” Political
Economy Note, World Bank, Washington, DC.
World Bank. 2009. Colombia: Decentralization: Options and Incentives for Efficiency.
Washington, DC: World Bank.
219
6
France’s Subnational Insolvency
Framework
Lili Liu, Norbert Gaillard, and
Michael Waibel
Introduction
France had a long tradition as a centralized state. However, two waves of
decentralization laws—during 1982–83 and 2003–04—contributed to
devolving more powers to the three types of subnational governments
(SNGs): municipalities, departments, and regions. SNGs in France now
have administrative autonomy, own responsibilities, executive powers
and, since 2003, financial autonomy. Because there are a large number
of small municipalities in France, intermunicipal cooperation arrangements are common, covering a range of services such as water supply,
household waste collection, and sewerage. In addition, SNGs may own
50–85 percent of joint public-private partnerships (sociétés d’économie
mixte locales, SEMs).
Before 1982, the state controlled all the loans made to SNGs, usually
at favorable interest rates. With the 1982 Decentralization Act, SNG access to borrowing was no longer submitted for approval of the Prefect
(the representative of the state in each department). The monopoly status of public financial institutions ended in the 1980s, and since then,
the credit market for SNGs has consisted mainly of a few banks. Since
221
222
Until Debt Do Us Part
the 2008–09 global financial crisis, interest in the development of a
subnational bond market has grown.
After some municipalities experienced severe financial distress in the
first half of the 1990s, the central government tightened the regulatory
framework for SNG borrowing and introduced greater disclosure and
transparency. Prudential rules are now used to monitor debt, liquidity, and contingent liabilities. In addition to the borrowing framework,
SNGs must also follow accounting and budget rules. Although SNGs
have considerable fiscal autonomy, the state exercises strong supervision
and monitoring of SNG financial accounts through three institutions:
the Prefects, the Regional Chambers of Accounts (Chambres Régionales
des Comptes), and Public Accountants.
By law, SNGs cannot declare bankruptcy, and public assets cannot be
pledged as collateral. If SNGs become insolvent, the central government
intervenes, enforcing fiscal adjustment and facilitating debt negotiations among creditors and the borrower. The central government does
not guarantee SNG borrowing but may provide exceptional financial assistance. However, the amount of assistance by the central government
is extremely small, and there is no expectation of a state bailout.
SEMs are subject to standard corporate insolvency law. Other government-owned entities, such as établissements publics (public establishments), which are financially autonomous agencies, are as a general rule
subject to administrative law. In such cases, liquidation is not an option.
This chapter reviews how the French system combines decentralized
responsibilities and fiscal decisions with fiscal monitoring from the central government and how central monitoring, supervision, and intervention deals with SNG insolvency. As part of this volume that reviews
and shares country experiences in managing subnational debt and related regulatory reforms, this chapter covers up to 2010.1
The rest of the chapter is organized as follows. Section two reviews
SNG institutional structures and finances. Section three focuses on
three channels of control and monitoring of SNGs by the central government. Section four presents the subnational borrowing framework
and the evolution of subnational credit markets. Section five discusses
contingent fiscal risks arising from various forms of intermunicipal and
public-private arrangements. Section six explains how the French system resolves SNG insolvency. Section seven offers conclusions.
France’s Subnational Insolvency Framework
Institutional Structures and Finances of Subnational
Governments in France
Administrative Structure
French decentralization dates back a thousand years to the emergence
of France as an independent state. Municipalities (communes) were
created relatively recently—in 1789—the year the French Revolution
(Révolution Française) began. The departments (départements) were
created the following year, in 1790, before becoming local authorities
at the beginning of the Third Republic, owing to the first decentralization law of August 10, 1871. That law reorganized the balance of power
within municipalities and provided a check on mayoral authority
through town councils. Regions (régions) were created in 1972 and were
given the status of local authorities in 1982.2
Before 1982, departments and municipalities had limited competences. As representatives of the central government in each department,
the Prefect held the executive power and in that capacity supervised the
laws passed by all SNGs and had the power to approve or cancel subnational decisions. The two waves of decentralization since the early 1980s
have led to a major shift in the history of French institutions. The resulting administrative structure is summarized in figure 6.1.
Under Title XII of the French Constitution,3 SNGs are administrative structures—distinct from the national administration—that
exercise competences within a given territory. France consists of 26
regions, 100 departments, and 36,682 municipalities, which account
for 40 percent of all European SNGs. There is no hierarchy among
the three levels of government, and none may exercise authority over
another territorial entity.4 Four regions and four departments have a
special status.5
Two Waves of Decentralization
In the early 1980s, a national law was passed mandating major reform of
French institutions.6 Reforms included:
•
Replacement of ex-ante prefectoral control with ex-post legal control exercised by the Prefect, the National Court of Accounts (Cour
des Comptes), and the newly created Regional Chambers of Accounts
(RCA), in charge of budgetary control of SNGs.
223
224
Until Debt Do Us Part
Figure 6.1 Subnational Governments in France
State (Etat)
Regional Prefect
(Préfet de Région)
26 Regions (Régions)
Regional Council (Conseil Général)
President of the Regional Council
Departmental Prefect
(Préfet)
100 Departments (Départements)
General Council (Conseil Général)
President of the General Council
36,682 Municipalities (Communes)
Municipal Council (Conseil Municipal)
Mayor
State control
Subnational control
Source: Dexia 2008.
•
Transfer of executive power at the departmental level to the president of the General Council (that is, the executive power in the
department).
The laws of January 7, 1983, and July 22, 1983,7 determined the
respective competences of the central government and SNGs and set
up the rules for transferring financial resources. These laws are now
regarded as “Act I of French decentralization.”
A second wave of decentralization began in 2003. Constitutional Law
No. 2003-276 refers to the “decentralized organization of the Republic”8
and mentions the regions in the Constitution.9 The three local government levels (municipalities, departments, and regions) are considered
“territorial communities” (collectivités territoriales). SNGs have their
own executive powers and, since 2003, the Constitution recognizes their
financial autonomy.10 The Constitution also recognizes various intercommunal structures and allows for the merger of existing SNGs with
larger entities.11 The constitutional reform also enshrines direct democracy at the local level,12 the financial autonomy of territorial entities,13
and the governance of overseas territories.14
Law No. 2004-809 of August 13, 2004, on local liberties and responsibilities, devolved additional responsibilities from the central government to territorial communities starting January 1, 2005 (table 6.1).
France’s Subnational Insolvency Framework
225
Table 6.1 New Powers Devolved to SNGs in 2004, France
Municipalities
Powers before 2004
Economic and
territorial
development
• Local public utilities (waste
collection and treatment, water
distribution, sewerage)
• Urban planning
Social aid and
health
• Social aid (childcare and
kindergartens)
Education and
culture
• Construction and maintenance of
primary schools
Departments
Powers before 2004
Additional powers since 2004
• The law of August 13, 2004, did not
significantly affect the responsibilities
and finances of municipalities.
Additional powers since 2004
Economic and
territorial
development
• Construction and maintenance of
secondary roads
• Interurban public transport
• Management of some formerly
national roads and staff (since 2006)
Social aid,
solidarity, and
health
• Social assistance (children, disabled
adults, and elderly)
• Implement the minimum
subsistence allowance (since 1988)
as set at the national level
• Preventive health care
• Medical prevention
• Financial assistance for the young
• Finance social funds for housing
Education and
culture
• Construction and maintenance of
public junior high schools
• Management of some nonteaching
staff in junior high schools
Regions
Powers before 2004
Additional powers since 2004
Education
• High school construction,
maintenance, and operating costs
• Fund professional training
• Cofund universities
• Establish education requirements
• Fund schools
• Management of nonteaching staff in
high schools (since 2006)
Economic,
territorial
development,
and transport
• Organize and finance regional
passenger rail services
• Responsibility for infrastructure
• Allocation of subsidies (business
tax exemption)
• Transfer of private-sector economic
aid to regions
• Formulate regional plans for economic
distribution
Culture
• Organization and financing of
regional museums
• Responsibility to compile inventory
of cultural patrimony
Health
Source: Fitch Ratings 2008a.
Note: n.a. = not applicable.
n.a.
• Health care education services
(prevention actions, vaccination, etc.)
Possibility of financing sanitary
facilities
226
Until Debt Do Us Part
The two waves of decentralization changed not only the structure
and responsibilities of SNGs, but also central government oversight and
control. Departments and regions received additional responsibilities
and resources, mainly through greater tax-raising power, and administrative autonomy (Fitch Ratings 2008a).
The Importance of Inter-SNG Entities
and SNG-Owned Firms
Compared to other countries, France has a large number of tiny
municipalities, with an average population of just over 1,600. More
than 20,000 municipalities have fewer than 500 inhabitants, and 32,000
municipalities have less than 2,000 inhabitants.15 Intermunicipal cooperation began in the late 19th century to overcome fragmented services
of small SNGs. Since then, many cooperative legal structures (établissements publics de coopération intercommunale) have often been established across municipal boundaries to carry out responsibilities more
efficiently.
There are two major types of intermunicipal cooperation structures:
(a) structures without own-source tax revenue, that is, single-purpose
structures (syndicats de communes à vocation unique) and multipurpose
structures (syndicats de communes à vocation multiple); and (b) large
intermunicipal structures that can levy their own taxes (for example, a
local business tax).
Three types of municipal cooperation with own-source tax revenue
were created in 1999. They are (a) communautés de communes (associations of villages, by small towns), (b) communautés d’agglomérations
(associations of towns) for medium-size urban areas of 50,000–500,000
inhabitants with at least one city with a minimum of 15,000 inhabitants), and (c) communautés urbaines (urban communities with at least
500,000 inhabitants).16
In 2008, 91 percent of French municipalities, accounting for 87 percent
of the population, belonged to at least one of the 2,583 intermunicipal
cooperation structures with own-source tax revenue. These structures are
funded through taxes they levy and grants from the central government.17
Intermunicipal cooperation structures have a deliberative body made up
of elected counselors from the participating municipalities, an executive,
and a committee. The intermunicipal structures delegate a growing share
France’s Subnational Insolvency Framework
of the service delivery functions to the three main types of public entities
they own or over which they exercise control.
The first type of public entity is SEMs—public-
private
partnerships—which are commercial companies subject to the corporate insolvency law18 and commercial laws. The delegation of powers
to SEMs often concerns infrastructure service delivery, such as public
transport. SNGs must own between 50 and 85 percent of the capital. If
an SEM finds itself in financial distress, its “public” assets are typically
transferred to a new entity that continues its public service function, a
procedure that is sometimes used for privately owned companies. SNGs
bear the losses of SEMs as shareholders and may be liable under implicit
or explicit guarantees. Although the SNG is the majority shareholder,
oversight by subnational deliberative bodies is sometimes weak.
The second type of public entity is public establishments (établissements publics)—financially autonomous agencies subject to administrative law. Public establishments are mainly created in the health,
education, and cultural sectors. Despite their relative autonomy, public
establishments are controlled by SNGs. As entities governed by public
law, they cannot be liquidated.19 The largest public establishments are
supervised by the central government.
The third type of public entity is the nonprofit association, which
often receives grants from subnational administrations. Elected officials
generally hold key posts in these associations. This kind of collaboration
could pose fiscal risks for the local government, since the local government may have to help close the fiscal deficits of associations that are
deemed to be essential for social cohesion.
Structure of Subnational Finances
In 2009, SNG expenditures in France amounted to €214 billion, which
represented 21 percent of total public expenditures.20 On average,
municipalities accounted for about 49 percent of SNG expenditures,
departments for 36 percent, and regions for 15 percent.21 Tax revenues
accounted for about 49 percent of SNG total revenues. Financial
transfers from the central government to SNGs accounted for about
21
percent for municipalities and departments and 31 percent for
regions. Borrowing accounted for less than 10 percent of SNG total revenue (figures 6.2, 6.3, and 6.4).
227
228
Until Debt Do Us Part
Figure 6.2 Sources of Municipal Revenues in France
Borrowing
7.3%
Investment
revenues
14.1%
Local taxes
31.2%
Other sources
11.6%
State transfers
21.0%
Nonlocal taxes
14.8%
Source: General Directorate of Subnational Entities (Direction Générale des Collectivités Locales, DGCL) 2011.
Note: The latest data from DGCL 2011 are 2009 figures.
Figure 6.3 Sources of Departmental Revenues in France
Borrowing
7.6%
Investment
revenues 6.9%
Other
sources 9.3%
Local taxes
33.1%
State
transfers 20.1%
Nonlocal taxes
23.0%
Source: DGCL 2011.
Note: The latest data from DGCL 2011 are 2009 figures.
France’s Subnational Insolvency Framework
Figure 6.4 Sources of Regional Revenues in France
Borrowing 13.1%
Local
taxes 18.0%
Investment
revenues 9.6%
Other
sources 2.4%
Nonlocal
taxes 26.2%
State
transfers 30.6%
Source: DGCL 2011.
Note: The latest data from DGCL 2011 are 2009 figures.
SNGs have three main kinds of revenues: state transfers (about
€72 billion in 2008) (Fitch Ratings 2008a), local taxes, and shared taxes
(nonlocal taxes). The four local taxes are as follows:
• The residential tax, based on the rental value of property (taxe
d’habitation)
• The building tax (taxe sur le foncier bâti)
• The land tax (taxe sur le foncier non-bâti)
• The business tax paid by owners (companies and individuals) and
by companies based on fixed-asset rental value and on payroll (taxe
professionnelle) (Rapport Balladur 2009).22
The shared taxes are (a) the oil tax (taxe intérieure sur les produits
pétroliers), (b) the tax on insurance contracts, (c) the tax on real estate transactions, and (d) the vehicle registration tax. The transfers
from the central government are mainly composed of the Dotation
Globale de Fonctionnement (General Public Service Grant), which is
nonearmarked.
229
230
Until Debt Do Us Part
The 2008–09 global financial crisis weakened the financial position
of French SNGs. Although the impact has been mitigated by the central
government’s stimulus package (Fitch Ratings 2009),23 the sustainability
of SNG fiscal positions, in view of the uncertainty in the global recovery,
remains in doubt. In April 2010, the Jamet Report ( Rapport Jamet 2010)
projected that 11 departments might face financial distress.24 These departments have faced dramatic increases in social spending25 and dramatic decreases in revenues.
Control and Monitoring by the Central Government
While there is substantial decentralization in France today, the central
government continues to exercise control over SNGs through balanced
budget rules for SNGs, and through three key institutions—the Prefect,
the RCA, and the Public Accountants—all of which play a large role in
policy making at the subnational level. The Prefect is the representative of the central government in the department and exercises general
oversight of municipal activities. The RCA is responsible for controlling SNG accounts. The Public Accountants, as public servants in the
central government, are responsible for ensuring the regularity of SNG
payments.
Balanced Budget Rules and Processes
The state supervises SNGs through balanced budget rules; that is, the
budget of an SNG must be balanced. This principle is implemented
through the budgetary processes; the assemblies that approve the budgets must vote budgets in which the investment section and the operating (or current) section are each balanced.26 This equilibrium must be
based on reasonable assumptions about the revenue and expenditures
(real equilibrium) and must include all the compulsory expenditures
including debt service.27 Compulsory expenditures must be paid from
revenues and cannot be paid from borrowed funds. Box 6.1 presents the
specific rules regarding the budget processes.
The Prefect and the Regional Chamber of Accounts
The central government also exercises detailed and regular oversight
to ensure that SNGs are able to meet their debt obligations.28 This
France’s Subnational Insolvency Framework
Box 6.1 Subnational Financial Rules: Fundamental Principles, France
Following are the fundamental principles that guide subnational financial rules in
France.
The Principle of Annuality: The budget is voted for the calendar year—that is, for
January 1 through December 31.
The Principle of Anteriority: The budget must be voted before March 31 of the
current year (or April 15 if a new territorial assembly has been elected). Decisions that
modify the budget must be adopted before December 31 for investment credits and
before January 21 of the following fiscal year for operating credits (article L. 1612-11 of
the General Code of Territorial Entities (Code Général des Collectivités Territoriales,
CGCT).
The Principle of Unity: A single document, comprising the whole budget, is
voted on.
The Principle of Universality: All expenditures and revenues must be listed. No
revenue item can be earmarked.
The Principle of Speciality: All revenues and expenditures are classified according
to an accounting system.
The Principle of the Balanced Budget: The budget of municipalities, departments,
and regions must be balanced when it is voted on. Both the operating and investment
sections must be balanced (Article L. 1612-4 of the CGCT). Revenues must be sufficient
to cover annual debt repayments, and the revenues and expenditures figures must be
in good faith (that is, revenues must not be overestimated and expenditures must not
be underestimated). If the SNG refuses to increase revenues to balanceexpenditures,
the Prefect can impose new taxes (this procedure is called the inscription d’office;
see the decision by the Regional Chamber of Accounts of Languedoc-Roussillon,
Commune de Bolquère, January 18, 1989).
versight comes in three forms: legal, budgetary, and management. The
o
focus is on the prevention of subnational financial distress through an
elaborate system of controls, including oversight by Public Accountants
of all SNG disbursements. These ex-ante controls are designed to substantially reduce the probability of default.
Legal controls concern the legal validity of acts, such as compliance,
respect for procedural rules, and mistakes on points of law or fact or
abuse of power or procedure. In particular, assembly votes or the imposition of taxes are subject to a review of their legality. The Prefect or any
affected person may appeal to the administrative judge, who has power
to annul the act in whole or in part.
Budgetary control concerns the budgetary process, as prescribed by
law. The deadlines for the regular (as opposed to extraordinary) budgetary process are presented in table 6.2.
231
232
Until Debt Do Us Part
Table 6.2 Deadlines for the Regular Budget Process, France
Date
Process
January 1
Beginning of the financial year
April 15
Deadline for the adoption of the draft budget
June 1
The current financial account kept by the public treasury must have been
transferred to the local assembly
June 30
The current administrative account kept by the local government must have
been approved by the local assembly
December 31
End of the financial year
Source: Robert 2009.
The Prefect controls the legality of the local budget ex post (Articles
L. 1612-1 to L. 1612-20 of the CGCT). Article 47-2 of the French Constitution states:
The National Court of Accounts assists Parliament in the control of the
action of the government. It assists Parliament and the government in
the control and execution of finance laws and the social security financing laws as well as in the evaluation of public policies. It helps to inform
the electorate by publishing public reports. The public administration
accounts must be regular and in good faith, and give an accurate picture
of the results of their management, of their property and of their financial situation.
SNG budgetary decisions must be sent to the Prefect. The Prefect, or
any other person affected by the decisions (for example, any taxpayer
living in the jurisdiction of the SNG or any debtholder) can contest the
legality of the decisions before the administrative tribunals within two
months of their adoption. An example would be new borrowing to cover
a deficit in the operating budget, which is prohibited by the budget legislation. The administrative judge has the power to nullify the decision.
The Prefect and the RCA are in charge of budgetary controls.
RCAs were created by the law of March 2, 1982. Each region has its
own Chamber, and all members, called magistrates, have life tenure.
These courts have jurisdiction not only over SNGs, but also over public establishments (hospitals, public secondary schools, and so forth),
public-private partnerships, and associations funded by SNGs. Budget
decisions need to be sent to the Prefect when adopted. RCAs have the
power to make adjustments to budget proposals (fiscal adjustments).
France’s Subnational Insolvency Framework
Should these proposed adjustments by the Chamber not be incorporated into the budget, the Prefect may intervene to ensure that SNGs
meet their compulsory expenditures and comply with balanced budget
rules. The Prefect can choose to disregard the RCA recommendations
but must explain the reasons for doing so.
Budgetary control concerns only budgetary instruments in the strict
sense: the draft budget, any supplementary budget, decisions to modify
the existing budget, any annexed budget, and the administrative account. Control is limited to the following: (a) the dates of the vote and
the submission of the initial budget, (b) the balancing of the budget,
(c) the approval of the accounts, and (d) the inclusion and payment of
compulsory expenditures. If a violation occurs, the Prefect refers the
matter to the RCA,29 which may then modify the budget decision.
The draft budget must be sent to the Prefect within 15 days of passage. Otherwise, the Prefect asks the RCA to intervene. In 2007 and
2008, the Chambers rendered 114 decisions on budgets, 49 of which
were for missed budget deadlines (National Court of Accounts 2009c).
Regarding balancing the budget, the Prefect can refer the matter to
the RCA up to one month after receiving the budget. The court has one
month from that date to propose measures to the concerned SNG to
balance the budget. The Chambers made 112 decisions in 2007 and 146
decisions in 2008 concerning unbalanced budgets (National Court of
Accounts 2009c).
If the budget deficit exceeds 5 or 10 percent of receipts, depending
on SNG size,30 the Prefect may call on the RCA to intervene. The Prefect
is required to refer inaccurate or illegal budget decisions to the RCA
within one month. The cases to be referred include, but are not limited to, those in which the operating budget is in deficit, or in which
compulsory expenditures such as debt service are not included in the
budget.
The RCA issues proposals within two months to ask SNGs to balance their budgets, which may include the SNGs cutting expenditures
and raising taxes. The Prefect must send the revised budget for the next
fiscal year to the Chamber. If the municipality has not taken sufficient
deficit-cutting measures, the RCA proposes other measures to the Prefect within one month. The Prefect may seize fiscal control and impose
a budget. Alternatively, the Prefect may reject the proposals made by the
233
234
Until Debt Do Us Part
Chamber, but must explain why. There were 119 such RCA interventions in 2006 for these controls.31
Compulsory expenditures must be included in the budget and must
be paid. The Prefect, or any other person adversely affected by their
exclusion, can initiate proceedings before the competent Chamber. This
is the most common reason the RCAs are called on (Bouvier 2008). For
instance, debt service, including both principal and interest, constitutes
a compulsory expenditure. Debt service must, therefore, be entered as
an expenditure item in the annual budget. If debt that is due is unpaid,
a lender can ask an administrative tribunal to include the debt service in
the SNG’s budget and enforce payment (mandatement d’office) (Articles
L. 1612-15 to L. 1612-17 of the CGCT). The Prefect has the right to
refuse the RCA recommendations, but must explain the reasons for the
decision. If the Prefect accepts the decision of the Chamber, and unless
the SNG can defend against such action on the grounds that the sum is
not in fact due, the tribunal decides in favor of the lender. The payment
order is issued, and the Public Accountant pays the debt.
Funds, of course, need to be available in the SNG account to execute
a payment order. This is essential for budget control. The measures may
not be successful if the financial distress is discovered late. Herein lies
the essential role of RCAs. In the case of Pont-Saint-Esprit, the National
Court of Accounts recommended several measures to ensure the enforcement of recovery proceedings, such as the compulsory publication
of prefectoral decisions before town council deliberations, and the recognition of personal liability if the measures are not executed (National
Court of Accounts 2009a).32
Tax receipts regularly flow into SNG accounts. If the available funds
are insufficient for payment, the representative of the central government in the department issues an overdue notice and calls on the entity
in question to create the necessary resources for payment. If the assembly of the SNG or the public entity does not create the necessary resources, the representative of the central government in the department
or the supervisor takes over and, when necessary, orders payment (mandatement d’office) (figure 6.5).
Oversight of SNG budgets by the central governments requires internal controls of SNGs. Indeed, facing an increase of SNG expenditures
over the last 25 years, SNGs have strengthened internal management
France’s Subnational Insolvency Framework
235
Figure 6.5 Budgetary Control and Appeal Proceedings, France
Possible Appeal
Two months
Agreement on
the budget
Local Entity
Vote on the
budget
Prefect
Budget
inspection
30 days
CRC Seisin
Budget
inspection
30 days
Local Entity
Adjustment (if
necessary)
One month
No Seisin of
the CRC
Agreement
on the
budget
Possible
Appeal
Two months
CRC
Inspection of
the modified
budget
15 days
Disagreement
Prefect
Budget
Settlement
No deadline
In accordance
with CRC
instructions
Beyond CRC
instructions
(justification
needed)
Binding budget
Possible Appeal
Two months
Source: Robert 2009.
Note: CRC = Chambre Régionale des Comptes (Regional Chamber of Accounts).
control systems not only to ensure the legal validity of their budgetary
accounts but also to achieve their goals at lower costs. More recently,
a growing number of municipalities have collaborated with respect
to their financial control systems through intermunicipal cooperation arrangements, for example, as regards the (costly) training of staff
specialists. The Rapport Balladur (2009), the Rapport Mauroy (2000),
and the Rapport Richard (2006) encouraged such cooperation.
Management control by the central government refers to the decisions made ex post by the courts, most often in the form of reports.
236
Until Debt Do Us Part
The role played by courts has increased over time, though some SNG
politicians have been critical of their growing role. Indeed, even if the
courts only pass on whether the funds have been used for their intended
purposes, local councillors often consider this a means of questioning
their management and policy priorities, which is why these proceedings
are suspended during elections. The law of December 21, 2001, forbids
value judgments (see Article L. 211-8 of the Code des juridictions financières [Code of Financial Tribunals]).
The Chamber has special powers to enable it to perform its duties
(for example, it has access to all necessary documents). At the end of
the audit, the Chamber sends an interim report to the controlled local
government, which can reply or request a hearing. The Chamber analyzes the audit and publishes a final report. The local government can
respond a second time. Its replies will be published with the report and
will be accessible to the public.
The Public Accountants
France distinguishes between the expenditure function and the treasury function. Ordonnateurs (ministers, mayors, and presidents of local
councils), who order expenditures, are separate from comptables (Public
Accountants), who make payments and manage the funds.33 All Public
Accountants are central government civil servants. They are appointed
by the Finance Minister to perform treasury and accounting functions.34
They record SNG receipts and order payment of SNG expenditures on
behalf of SNGs. Public Accountants ensure compliance with budgetary
rules; they do not decide budgetary priorities. The decision by a Public
Accountant not to pay an expenditure item may be overridden by the
ordonnateur pursuant to an ordre de requisition, to make the payment
out of the SNG funds. However, the ordonnateur is liable with his personal assets if the expenditure violates the budgetary rules.
As regards receipts, the control by Public Accountant is more limited. Tax revenues are controlled by the fiscal administration, and Public
Accountants simply record and verify receipts.
With respect to expenses, Public Accountants are personally
responsible for failing to perform their duties. They are potentially
liable for all financial transactions. They verify the budgetary allocation, expenditure items, debt and its servicing, revenue receipts, the
France’s Subnational Insolvency Framework
a vailability of funds, and the clauses of the financial transactions. In the
event of noncompliance with these requirements, Public Accountants
are obliged to suspend the financial transaction.
Public Accountants, who exercise ongoing controls over SNG finances, are themselves subject to two types of control: administrative
and jurisdictional. With respect to administrative control, Public Accountants are supervised by the Tresorier-Payeur General (State Treasurer), the highest-ranking public accountant.35 Internal audit within
the Ministry of Finance is performed by an internal audit unit—the
Mission d’audit, d’évaluation et de contrôle (Audit, Evaluation and Control Group)—and the Inspection Generale Des Finances (General Inspection of Finance)—a group of top civil servants acting as a senior auditor
and advisory service for the Ministry of Finance. This is to ensure that
Public Accountants comply with laws and regulations and carry out
antifraud enforcement. Members of the General Inspection of Finance
have special powers (access to all necessary documents, particularly to
the records of all Public Accountants, and the right to suspend accountants temporarily in cases of fraud).
With respect to jurisdictional control, though not in charge of administrative control of Public Accountants, the RCA can issue judgments against a Public Accountant, as a follow-up of the compliance
audits they perform on local authorities. If a Public Accountant fails to
perform his or her duties, he or she is legally liable and must pay with
his or her own personal funds, if necessary. If the accounts are judged as
regular, Public Accountants are exonerated from liability.
France has a unique system for managing SNG cash. All state and
SNG funds are centralized daily in a single state treasury account. Public accountants accept all receipts and authorize all expenditures that are
in compliance with budgetary rules. The general rule is that SNGs must
deposit all cash with the State Treasurer. For payments, the SNG requests
the Treasurer to pay the debtor, the so-called payment order (mandatement). Revenues, transfers from the central government, and subnational
tax receipts—subnational taxes are set by the SNG but are collected by
the state’s tax collection agents—are also transferred to the SNG’s noninterest-bearing cash account held by the Treasurer. In return, the Treasury
does not charge SNGs for managing their cash and can advance money
against future tax revenues when SNGs do not have sufficient cash.
237
238
Until Debt Do Us Part
New procedures have been developed since 2006 to facilitate the
control exercised by Public Accountants. These controls operate at
several levels. New instruments, such as the computer system Helios,
ensure that the control of recurrent expenditures is systematized and
separated from the control of unusual expenditures. In the same way,
the system of Contrôle hiérarchisé de la dépense (Multi-layered Control
of Expenditures) establishes a set of controls. This system helps Public
Accountants redefine their priorities.
Subnational Debt
Subnational Borrowing Framework
Before the decentralization laws of 1982, the Prefect exercised ex-ante
control (contrôle a priori), preventing a municipality, department, or
region from borrowing if the loan was determined to be unfavorable to
the interest of the SNG (that is, if the loan would result in overborrowing). Moreover, only two state-owned financial institutions could lend
money to subnationals.36
In 1982, SNGs became responsible for their own borrowing and
could decide whether and how much to borrow without ex-ante control. In 1986, the central government stopped offering loans with privileged interest rates to SNGs. The same year, the requirement to borrow
from government lenders was abolished. Currently, SNGs mainly rely
on bank loans, and compared to the subnational bond market in the
United States, the French subnational bond market is tiny: more than
20,000 U.S. SNGs and federal and local government-owned enterprises
are rated compared to about 40 in France.37
In the early 1990s, some municipalities in France experienced severe
financial distress. In the second half of the 1990s, the central government
tightened the regulatory framework for SNG borrowing, introduced
greater disclosure and transparency requirements, and implemented an
early warning system.
Key elements of prudential rules regulating debt, liquidity, and contingent liabilities include the following:
•
•
New long-term borrowing must fund capital investments only.
Debt payments are compulsory expenditures and must be fully
budgeted.
France’s Subnational Insolvency Framework
•
•
•
Annual debt service, including that paid on guaranteed loans, must
be less than 50 percent of operating revenue.38
No single borrower may benefit from a guarantee exceeding 5 percent
of the SNG’s operating revenue.39
Guarantees may not exceed 50 percent of the principal.
SNGs are required to deposit cash with the central government,
which carries out all payments following the control framework discussed in the previous section. In addition to the borrowing framework,
SNGs must also follow certain accounting and budget rules. These rules
include balanced budget rules requiring that both operating and capital
accounts be balanced and annual debt service be covered by SNG own
revenues.
Subnational borrowing can take the form of a loan, a bond issue, or
other instruments including structured products. There must be a positive balance in the operating budget so that it can cover principal payments.40 Article L. 2122-22 of the CGCT states:
The mayor may, in addition, on delegated powers of the city council, be
authorized, in whole or in part, and for the duration of his mandate:
To borrow, within the limit set by the city council, for the purpose of
financing investments as foreseen in the budget, and for financial operations that are useful for managing its lending, including hedging for
interest and exchange rate risk.
According to articles L. 200-3, L. 335, and L. 4333-1 of the CGCT,
municipalities, regions, and departments may borrow. According to
Article L. 2331-8 of the CGCT, the proceeds from borrowing constitute
nonfiscal income in the investment section of the SNG budget. Borrowing is defined as the aggregate of all debts contracted with a maturity
exceeding one year during the legislative period and allowed for capital investment only. Short-term borrowing with a maturity shorter
than one year is treated differently.41 This short-term debt finances the
liquidity of SNGs and is allowed for all types of expenditures.
Subnational borrowing is not subject to the compulsory submission
of bids from several lenders,42 although SNGs are required to hold a tender for other banking and investment services if they exceed a certain
threshold.43
239
240
Until Debt Do Us Part
Nevertheless, even if SNGs became responsible for their own
borrowing, their freedom of management is still limited by the control
of RCAs, which can declare certain SNG acts void. This control does
not contravene the progressive autonomy granted to SNGs, but is in
accordance with the law. Thus, subnational borrowing must conform
to several rules, such as the prohibition against financing day-to-day
management by resorting to borrowing, the prohibition on speculation, and the requirement of publishing borrowing activities as part of
budget documents. In this respect, contracts must fulfill the conditions
defined by the Conseil national de la comptabilité (National Council of
the Comptroller) in its ruling of July 10, 1987, to be considered as financial risk management instruments. The publication of information on
borrowing activities attached to budgetary documents is necessary to
inform local councillors and citizens about the financial commitments
of the subnational entity (with a list of the establishments the entity has
contracted with) (Robert 2009).
Evolution of Subnational Debt in France
Following passage of the 1982 decentralization laws, the subnational debtto-expenditure ratio increased. SNG debt accounted for 11.1 percent of
GDP in 2007, up from 8.5 percent in 1982 (Rapport Pébereau 2005). This
increased borrowing financed new social, economic, and infrastructure
programs resulting from the devolution of new responsibilities to SNGs.
In 2009, municipal borrowing reached €6.75 billion compared to
€5.1 billion for departments and €3.6 billion for regions. Until 2002,
the three levels of SNGs managed to have positive net repayments
(meaning that their debt repayment was higher than their borrowing).
Since then, the trend has reversed. There has been a significant increase
in the borrowing of regions, departments, and municipalities—up 153,
64, and 10 percent, respectively, between 2003 and 2009 (figure 6.6).
This trend is partially due to the Law on Local Rights and Responsibilities (Law 2004-809), which deepened the devolution process. New
responsibilities were transferred to SNGs during 2005–07, particularly
to regions and departments. For instance, regions are responsible for
social and health care and education services, economic development,
vocational training, and grants to private high schools. Departments are
responsible for assistance for the young, social funds for housing, grants
provided to disabled people, management of some nonteaching staff in
France’s Subnational Insolvency Framework
241
Figure 6.6 SNG Borrowing and Debt Repayment in France
9
8
7
Euros, billion
6
5
4
3
2
1
09
20
08
20
07
20
06
20
05
20
04
20
03
20
02
20
01
20
00
20
99
19
98
19
19
97
0
Year
Municipalities—borrowing
Departments—borrowing
Regions—borrowing
Municipalities—dept repayment
Departments—dept repayment
Regions—dept repayment
Source: DGCL 2011.
junior high schools, management of some formerly national roads, and
grants to private junior high schools (Fitch Ratings 2008a).
The Lenders
Until 1987, the Crédit Local de France44 had a monopoly on municipal
lending. Since then, the French subnational credit market has been, in
essence, an oligopoly composed of four banks—Dexia (by far the largest bank in this market), Caisse d’Epargne, Crédit Agricole, and Société
Générale—which together control 80 percent of the market. Dexia, in
particular, played a leading role in the aggressive selling of structured
products, and received heavy blame following the global financial crisis.
With the crisis, Dexia entered a period of turbulence and survived due to
central government support. The bank shed a number of its subsidiaries
abroad. The support by the government created particular controversy
in light of widespread allegations that Dexia aggressively overloaned to
SNGs, both in volume and in terms of risky products.
From 2000 onward, few foreign banks have made loans to SNGs.
The Swiss bank UBS tried unsuccessfully to enter the market in the
early 2000s. In contrast, the Royal Bank of Scotland managed to lend
242
Until Debt Do Us Part
to some SNGs, such as the city of Aubagne.45 Most SNGs get financing
through bank loans. There are exceptions, such as the City of Paris: in
2008, i 331 million was raised, i251 million of which was in the form
of bonds and only i80 million of which was in the form of bank loans
(Mairie de Paris 2009). Since the early 2000s, SNGs have had the opportunity to borrow through the following channels: bank loans (fixedinterest loans, variable-interest loans, structured loans, and revolving
loans denominated in euros or in foreign currency), or bond issues
(short-term commercial paper, medium- or long-term securities, and
derivative bonds).
Creditworthiness of SNGs
SNGs willing to issue bonds need a rating. As of December 26, 2011,
Moody’s, Fitch, and Standard & Poor’s (S&P) rated 3, 9, and 13 SNGs,
respectively. Unlike sovereign and corporate borrowers, most SNGs are
rated by a single agency. All rated SNGs are in the investment grade category
with the exception of the Overseas Territory of French Polynesia, which was
rated BB+ by S&P.46, 47 Eighty-eight percent of SNGs are rated AAA/Aaa or
AA/Aa, which reveals strong creditworthiness overall (figure 6.7).
Figure 6.7 Distribution of Ratings Assigned to French SNGs
BB 4%
A 8%
AAA 28%
AA 60%
Source: Authors’ computations from Fitch, Moody’s, and S&P.
Note: SNG = subnational government. Ratings are Fitch, Moody’s, and Standard & Poor’s as of December 26,
2011. There are 25 SNG ratings.
France’s Subnational Insolvency Framework
243
Managing Subnational Contingent Fiscal Risks
There are two types of SNG contingent fiscal risks.
The first fiscal risk comes from potential liabilities of companies
owned by SNGs.48 In addition, SNGs sometimes guarantee liabilities of
other entities. Rating agencies typically count the debt incurred by such
companies as the indirect debt of the local government that owns the
SEM. In Fitch’s ratings of the nine French SNGs in December 2011, the
net indirect debt plus guarantees range from 0.1 to 74.7 percent of net
overall debt, with an average of 31.1 percent (table 6.3). Regulations on
SNG companies have been tightened, including the compulsory publication of guarantees, ownership, and subsidies. No single borrower may
benefit from a guarantee exceeding 5 percent of operating revenue, and
guarantees may not exceed 50 percent of the principal. Table 3 provides
a snapshot of the scope of indirect debt of selected French SNGs.
Table 6.3 Debt Data for French SNGs Rated by Fitch
Fitch data
Net
Net direct guaranteed +
debt
indirect
(million
debt (million
euros)
euros)
Net
overall debt
(million
euros)
Net
guaranteed +
indirect
debt)/net
overall
debt (%)
Year
Rating
(as of
December 26,
2011)
Essonne
(department)
780.3
220.9
1,001.2
22.1
2009
AA
Guadeloupe
(department)
110.6
215.8
326.4
66.1
2010
AA-
Île-de-France
(region)
3,506.8
11.3
3,518.1
0.3
2010
AAA
Paris (city)
2,695.0
7,957.0
1,0652.0
74.7
2010
AAA
Picardie (region)
471.5
0.4
471.9
0.1
2009
AA-
Provence-AlpesCôte d’Azur
(region)
1,664.1
150.0
1,814.1
8.3
2010
AA
Rhône-Alpes
(region)
1,357.2
48.2
1,405.4
3.4
2009
AAA
Seine-et-Marne
(department)
892.2
585.3
1,477.5
39.6
2010
AA
Strasbourg (city)
151.9
285.1
437.0
65.2
2010
AA+
Source: Compilation of various reports from http://www.fitchratings.com.
Note: SNG = subnational government.
244
Until Debt Do Us Part
The second fiscal risk results from the strong growth of complex
structured products since the 1990s. Structured products with increasingly risky instruments combine various derivative instruments with
a loan structure. At end-2007, the volume of structured products was
i30 billion to i35 billion in total borrowings of i137.5 billion (Fitch
Ratings 2008a).
The first step toward structured products took place in the 1990s,
when SNGs tried to renegotiate their debts with fixed interest rates
(Paris 2009). Since interest rates were higher in the 1980s and declined
sharply in the 1990s, such renegotiation enabled the SNGs to reduce
their debt. Banks even offered such products to small SNGs. The National Court of Accounts had already published a report, in 1991, about
debt management, focusing on the increasing importance of structured
products in SNG debt.
Four types of structured products were used: (a) borrowing with
variable interest rates (with options to minimize risk), (b) products
with barriers (the interest rate is guaranteed until an index reaches a
threshold), (c) snowballing products (discussed below), and (d) products indexed on an exchange rate or a difference in inflation levels.
For products with barriers, the interest rate is fixed until a threshold is crossed. For most of the 2000s, the threshold was not reached.
When the global financial crisis hit, interests rates often came close to
the threshold. Once the threshold is crossed, the interest rate increases
quickly.
Snowballing products are those for which the interest rate depends
on the difference between a long-term and a short-term interest rate.
Traditionally, the long-term rates are higher than short-term rates,
which explains the attractiveness of this product and reflects the risk
linked to the length of the maturities. However, when financial conditions deteriorate, the yield curve often reverses, with short-term interest
rates being higher. In this kind of product, the interest rate is fixed as
long as the requisite difference between the long-term and the shortterm interest rate persists. If the difference narrows by too much, the
fixed rate turns into a variable rate. Since maturities are typically long,
foreseeing the full budgetary impact of such products is difficult, especially for SNGs without sufficient in-house financial expertise.
France’s Subnational Insolvency Framework
The same principle applies to other products: the initial fixed interest rate (which is linked to an index) is attractive to the borrower, but
beyond a given threshold of this index, this interest rate increases. Considering that the maturities are long (several decades for the majority
of these structured products), the central problem is that interest rates
cannot be reliably projected in the long run.
The transition from a fixed interest rate to a variable interest rate can
gravely impact SNG financial accounts. In this case, two solutions are
possible: to settle borrowers’ accounts or to renegotiate the loans.
In addition, several surveys conducted by RCAs established that
SNGs having financial difficulties are frequently those that have exposure to these structured products.49 The recent evolution of the SNGs’
financial position confirms this. Since 2003, SNG borrowings have had
longer maturities, with an increased use of structured products. The
increased availability of structured products with combined fixed and
variable rate terms has proved to be particularly risky.
Public accounting standards have lagged behind in reflecting the new
challenges that these structured products pose for transparency and
debt profile. Risks and costs of these products are often not adequately
evaluated and understood. In December 2011, a report released by an
ad-hoc parliamentary commission estimated that i13.6 billion of SNG
structured debt was “toxic,” thereby likely to weaken the financial position of French local governments.50 Efforts are ongoing to better regulate and restrict the use of structured products.51
Resolving Subnational Financial Distress
French law distinguishes between private and public law. Private law is
the law of coordination and voluntary cooperation, and in private law,
the actors are in a horizontal legal relationship. Public law is characterized by vertical legal relationships, with the central government exercising sovereign authority and commanding the legal subjects to act as
prescribed. Public law is fundamentally asymmetric. The mechanisms
for coping with subnational financial distress are found in administrative law (public law), which emphasizes prevention and oversight
by the central government. However, the financial distress of SEMs
245
246
Until Debt Do Us Part
(with public ownership between 50 and 85 percent) is subject to private
law (including corporate insolvency law).
Before decentralization, subnational defaults in France were
extremely rare.52 Because the decentralized system of subnational fi
nance
is less than three decades old, case studies are few. In the early 1990s, one
estimate put the number of municipalities experiencing overindebtedness at about 2,000 (out of more than 36,000 municipalities), of which
40 were experiencing severe financial difficulties (Moody’s Investors
Service 2002). Other tiers of SNGs (departments and regions) have also
experienced subnational financial distress, although there have been few
cases.53 Actually, both domestic and foreign lenders rushed to lend to
municipalities, in the mistaken belief that there was little risk because
public entities were involved.
After several cases of subnational financial distress in the early 1990s,
both the subnational borrowing framework and accounting systems
were strengthened, as discussed above. (The risks of structured products materialized only in the 2000s [these products were not previously
available]). An early warning system (réseau d’alerte) was launched, and
the training of local civil servants was intensified. The overall goal was
to achieve greater transparency and disclosure, and thereby reduce municipal financial risk. Recourse to speculative financial instruments was
tightly regulated, but these standards were loosened after 2000.
Starting in 2008, the global financial crisis put the finances of French
cities and regions under strain. Most of them did not use structured
products, even if such borrowing looked attractive. Nevertheless, a
growing number of SNGs experienced difficulties when credit markets dried up in the second half of 2008 (Cossardeaux 2008). The subnational borrowing market shrank in volume. The crisis also led to a
decrease in tax revenues and tax allocations from the central government. About 1,595 SNGs used structured products that turned out to
be toxic.54 Other estimates are that up to 25 percent of SNG debt used
structured products. The French central government committed to help
local authorities face their credit needs and pressured banks to accept
subsovereign debt restructurings.
Facing a slowdown of their tax revenues and losses related to the
slowdown in the real estate market (for example, the droits de m
utation
[property transaction tax]), SNGs were experiencing a severe credit
France’s Subnational Insolvency Framework
crunch in late 2008. The risks of structured products added to these
financial difficulties. The crisis in SNG finances may continue for
years. Some SNGs may have to choose between cutting investments
or b
orrowing more to fulfil their enlarged responsibilities, thereby
jeopardizing their creditworthiness. Ex-ante rules, such as the balanced
budget rule or the debt-service-to-revenue ratio, combined with the
early warning system, may have reduced the risk of systemic SNG financial distress in the future.
To help SNGs cope with the impact of the global financial crisis, the
central government reduced the delay in refunding the value-added tax
to SNGs, enabling them to maintain capital expenditures in 2009 above
the 2004–07 average. The expected value-added tax refund payments
in 2009 totaled more than i4 billion, equivalent to 8 percent of SNG
capital expenditures that year. Overall net borrowing by French
SNGs increased by i5.1 billion in 2009, while total outstanding debt
as a share of GDP increased by 4 percent over 2008. The growing
financial d
ifficulties of French SNGs (particularly departments and
municipalities) have prompted the central government to establish a
new bank that would fund local governments and make an exceptional
transfer (of i3 billion to i5 billion) to SNGs in 2012.55
The ex-ante regulations and the oversight by the central government,
as discussed in previous sections, have substantially reduced the risks
of defaults and, thus, uncertainties facing lenders. Nonetheless, fiscal
and default risks exist despite ex-ante rules. If an SNG becomes insolvent, the negotiated nature of debt workout and fiscal adjustment cause
uncertainty to lenders due to a lack of a priority structure that would
have existed with a formal insolvency system. As mentioned, there is no
insolvency law that applies to SNGs. Although the procedures for dealing with SNGs that are breaching fiscal rules are clear, the procedure
for dealing with SNG defaults is not detailed in legislation, such as the
CGCT, but is shaped by administrative practice.
These mechanisms are largely informal, unbounded by legal constraints and with less reliance on precedents than civil law. Even though
the payment order mechanism theoretically offers extremely strong
security to lenders, it requires sufficient income for the payment of
all compulsory expenses. Moreover, lenders have few remedies under
civil law, such as seizing SNG assets. Liens on assets that are e ssential
247
248
Until Debt Do Us Part
for the performance of public services are illegal based on the premise
that interrupting public services would be counter to the public interest. Similar restrictions exist in other countries, including the United
States.
Even though the Prefect and the RCAs have the authority to enforce
a recovery plan and to release additional revenue in the case of financial difficulties, these powers have proved insufficient to prevent some
municipalities from defaulting. The view that the central government
always guarantees subnational debt is incorrect. Nevertheless, under the
provisions of articles L. 2335-2 and L. 1524-4 of the CGCT, there is a
procedure for exceptional state aid for SNGs in financial distress. Article
L. 2335-2 states:
Under the provisions of article 1524-4, exceptional subventions can
be granted by Ministerial order … to municipalities in which unusual
circumstances have led to specific financial difficulties.
In reality, the amount of extraordinary aid has been small. In 2006,
only 12 municipalities benefited from these subsidies, for a total of
i1,593,682 (Robert 2009). The allocation of these subsidies is subject
to three conditions: (a) a municipality can receive exceptional subsidies only if the budget is unbalanced (as defined in article L. 1612-4
of the CGCT), (b) this imbalance led to a case before the RCAs (indeed, the subsidies can be granted only after the audit of the budget by
the RCAs), and (c) the recovery measures taken by the Chamber must
have been inefficient (they cannot absorb the deficit of the operating
budget).
The Ministry of Home Affairs and the Ministry of Finance have compiled a list of subnational entities in financial distress. However, there
is no power of coercion against an entity that refuses to cooperate with
a restructuring plan. A subnational entity cannot be placed under supervision a priori. The current framework reaches its limits in cases of
extreme financial distress; that is, there is no established procedure for
restructuring debt with lenders that are unwilling to cooperate. In crisis,
there is a three-party negotiation system: the central government, SNGs,
and lenders (only a limited number of banks lend to SNGs). It is a predictable negotiation system, even though the outcome of the negotiation
depends on the balance of power among three parties. Notwithstanding
France’s Subnational Insolvency Framework
249
the above, there is no rules-based framework for dealing with subnational financial distress, and the nature of the policy response may thus
differ from case to case.
The implementation of an early warning system began in 1993 and
only applied to municipalities of more than 10,000 inhabitants. The
Direction Générale des Collectivités Locales (General Directorate of Subnational Entities, DGCL) and the Direction Générale de la Comptabilité
Publique (General Directorate of Public Accounting, DGCP) had each
developed a system to analyze the accounts of municipalities to detect
financial distress. The two systems were based on the analyses of certain
ratios, which are presented in table 6.4.
Under the DGCP system, the financial situation of a municipality is
considered to be critical if it exceeds three thresholds, and extreme if it
exceeds four thresholds, as explained below.
The DGCP and the DGCL later realized the efficiency of their system
was limited, for two reasons: the system was centralized nationwide, and
the number of ratios was excessive. In this respect, the revision of the
warning system in 2001 was based on implementation of this system
in all municipalities, the decentralization of the system at the departmental level, and the simplification of the ratios system, with four ratios
inspired by the initial DGCP system. Each ratio has a critical threshold and rates the municipality on a scale of zero to 100. The situation
is considered critical when the ratio falls below 30, and extreme when
it falls below 20. The list of municipalities in financial distress is sent
to the Prefect, who has the option of intervening, of providing such
information to the RCA, or both. The results are for the mayor’s private
Table 6.4 Main Ratios Used by the French Central Government to Detect
Financial Distress
DGCL
DGCP
Ratios
Indebtedness and financing = 4 ratios
Budget structure and rigidity = 2 ratios
Fiscal pressure = 1 ratio
Cash flow = 1 ratio
Rigidity of structural expenses = 1 ratio
Fiscal leeway = 1 ratio
Excessive debt = 1 ratio
Method
Qualitative evaluation
Implementation of quantitative thresholds
Source: Robert 2009.
Note: DGCL = Direction Générale des Collectivités Locales (General Directorate of Subnational Entities), DGCP =
Direction Générale de la Comptabilité Publique (General Directorate of Public Accounting).
250
Until Debt Do Us Part
information (publication is at his or her discretion), and the list of the
concerned critical municipalities is centralized through the DGCP and
the DGCL. In the most extreme cases, the departmental treasury can
conduct an additional audit.
Generally, caution must be used when assessing the creditworthiness of SNGs. Ratios taking into account the local population are not
necessarily accurate, because most local revenues come from firms and
not from householders. The most reliable indicators are discretionary
own-source revenues to operating revenues, the total debt to operating
revenues, and the debt-service-to-operating-revenues ratios. Also worth
considering is the taxing capacity and assessing to what extent the SNG
is able to increase taxes.56
There is a question as to whether SNGs use structured products to
engage in speculation57 or to actively manage debt. In practice, distinguishing between speculation and optimization of debt is difficult.
Risky borrowing just after elections provides a measure of budgetary
freedom to the new executive. At the same time, it creates an electoral
cycle in SNG investments, and thus substantial fiscal risk. According to
one view, banks and SNGs share the blame. Banks object, saying that
SNGs took advantage of the lower interest rates provided by structured
products in the past and must live up to their responsibilities when circumstances change. They also claim that the current situation is exceptional due to the global financial crisis.
The recent growth in structured products used by a growing number of French SNGs poses additional challenges to the resolution of financial distress. The growth of structured products combined with the
financial crisis has increased the risks of financial distress. The government has begun to address the challenges of structured products. In its
report, “Les risques pris par les collectivités locales et les établissements
publics locaux en matière d’emprunt,” (National Court of Accounts
2009b), the National Court of Accounts stresses three points on how to
improve SNG debt management.
First, the role of elected assemblies needs to increase. Currently, the
CGCT grants the regions, departments, and municipalities c onsiderable
freedom regarding the operations they consider necessary, and there
are no specific measures regarding structured products. According to
France’s Subnational Insolvency Framework
the Court, it would be advisable to structure and redefine precisely the
delegation of power to these assemblies.
Second, improved accounting and transparency is needed. Indeed,
the accounting norms enforceable to SNGs do not at present include
any specific accounting framework for structured products. Moreover,
the lack of accounting of the risks linked to the possession of structured
products seems to be in contradiction to the general principle of caution, as defined in the General Accounting Plan (Article 120-3).58
Third, since decentralization abolished direct control by the central
government over SNG actions, and especially over their borrowing,
there is a need to improve the openness of SNG policies. Indeed, since
SNG goals reflect the intentions of the party in power, communication
between citizens and deliberative assemblies must be improved. In this
regard, the SNGs could make available to citizens reports that could inform their choices.
In 2008, a tripartite meeting gathered representatives of banks,
SNGs, and the Minister of Home Affairs. They decided to treat the difficult cases individually. The minister insisted on better information by
comparing the financial situation of all SNGs. Another emphasis was
on transparency: banks are subject to stress tests and must keep politicians abreast of the financial situation.59 After the meeting among the
representatives of the central government, the SNGs, and the banks,
agreement was reached in November 2008 on two propositions:
(a) dealing with subnational financial distress is the responsibility of
the concerned SNG and its banks, and (b) a code of good conduct is to
be developed to provide a framework for the relationship between the
banks and the SNGs.
In May 2009, the Ministry of Finance proposed a code of good
conduct. Under it, banks may not sell structured products to SNGs that
risk a loss of capital, or risky products indexed to volatile variables such
as exchange rates. Banks must indicate the position of their products
on a predetermined risk scale. These measures preserve much of the
autonomy of SNGs.60 The fact that they are not binding illustrates the
ad-hoc nature of the policy response to relatively widespread subnational financial distress, and the reluctance of the central government to
develop a more structured framework.
251
252
Until Debt Do Us Part
Both banks and SNGs agreed on several points: (a) SNGs should
actively manage their debts without a priori control by the central government; (b) competition among banks should be maintained; and
(c) a balance should be reached between financial innovations and
constraints specific to public administration. The agreement took e ffect
on September 1, 2009, and the elements, called commitments, contained therein are as follows:
Commitment 1:
Banks promise not to sell products whose principal payment is risky or linked to risky indexes. Risky
indexes include an exchange rate in a currency not
held by the SNG, prices of raw materials, and equity
investments. It is also forbidden to sell products linked
to indexes not related to the SNG investments or those
from a non-Organisation for Economic Co-operation
and Development country.
Commitment 2: Banks commit not to sell products with cumulative
products that are particularly sensitive to interest rate
shocks.
Commitment 3: Banks accept the requirement to provide a transparent index of riskiness when proposing products. Each
product is rated on two common scales of five levels
each. One scale refers to the riskiness of the index the
product is linked to, and the other refers to the structure of the product. This should increase the transparency in comparing structured products.
Commitment 4: Banks acknowledge that SNG executives do not have
financial expertise and that the executives should be
well informed at all times. All the documents should
be in French. The drawbacks and the risks of each
product should be clearly shown. Past behavior of the
indexes should be analyzed. Stress tests should be conducted showing how the product behaves in case of
sharp corrections of the index.
Commitment 5:
SNGs commit to improving the transparency of
the decisions taken regarding borrowings and debt.
SNG executives should present the current financial
France’s Subnational Insolvency Framework
s ituation to the entire municipal council. The council
may authorize the executive to buy products of a certain risk level on the two scales.
Commitment 6: SNGs commit to make public the structured products
they subscribe to, the indexes they are linked to, and
the structure of the products. This increased transparency should improve the decisions taken in the
budgetary meetings. The executives should receive
authorization from the municipal council before committing to a new product.
Former French president Nicolas Sarkozy created a Comité pour la
réforme des collectivités territoriales (Committee for the Reform of Territorial Entities) at the end of 2008 to “determine measures to simplify
SNG structures, to clarify the distribution of their competences and to
allow a better allocation of their financial means, and to formulate useful recommendations.”61 Twenty propositions were presented by the
committee to the president on March 5, 2009, (Rapport Balladur 2009).
Regarding the simplification of SNG structures, the committee suggests
reducing the number of metropolitan regions from 22 to 15. Abolition
of the departments is not planned, but the report insists on the necessity
of redefining the competencies of existing departments. At the municipal level, the prerogatives of the mayors are not modified, but the committee intends to strengthen the intermunicipal cooperation structures.
Between 2009 and 2014, a new land allocation and new voting system
could be implemented.
Conclusions
During 1982–83 and 2003–04, two waves of decentralization in France
devolved more powers to the three levels of SNGs: the municipalities,
the departments, and the regions. This new institutional framework has
enabled SNGs to enjoy a greater degree of autonomous expenditures,
to raise their own taxes, and to borrow from financial markets, within
ex-ante rules established by the central government. However, SNGs are
subject to ex-post controls by the Prefect and the Regional Chambers of
Accounts, and to ongoing controls by the Public Accountants.
253
254
Until Debt Do Us Part
The ex-ante fiscal rules and the regulatory framework for managing
SNG fiscal risks were established after a period of unregulated borrowing by SNGs following the decentralization and subsequent debt stress
experienced by some SNGs. The regulatory framework combines the
laws and regulations with three sets of institutions, while preserving
considerable SNG fiscal autonomy. The laws and prudential rules regulate debt, liquidity, and contingent liabilities. The state exercises strong
supervision and monitoring of SNG financial accounts through the Prefect, the Regional Chamber of Accounts, and Public Accountants.
The French system, which combines decentralized responsibilities
and fiscal decisions with fiscal monitoring by the central government,
offers valuable experience for countries undergoing decentralization.
State supervision has resulted in the avoidance of major SNG defaults—
although several debt restructurings occurred in recent decades, which
partly explains the high credit ratings—“AA,” on average—assigned by
rating agencies. Nonetheless, the lack of a clear, established legal structure for priority payments creates uncertainties. Off-budget entities,
such as SEMs, pose contingent fiscal risks, a common challenge across
countries.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Standard & Poor’s downgrading of France’s sovereign credit rating in
January 2012 (Standard and Poor’s 2012) will have implications for subnational
finance in France; the assessment of such implications is beyond the scope of
this chapter.
2. In this chapter, the term local authorities and the term subnational governments
are used interchangeably.
3. Titre XII: Des Collectivités Territoriales (Title XII: Territorial Entities).
4. Article 72 of the French Constitution, October 4, 1958.
5. Article 74 of the French Constitution. The four regions are French Polynesia,
Guyana, New Caledonia, and Wallis and Futuna. The four departments are
Corsica, Guadeloupe, Martinique, and Reunion Islands.
6. The Law of March 2, 1982, known as the “Defferre Law,” because Gaston
Defferre was then Minister of Home Affairs.
France’s Subnational Insolvency Framework
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Laws are referred to by date in this chapter.
Article 1 of the Constitution.
Article 72 of the Constitution.
Article 72-2 of the Constitution. The terms are defined under the constitutional
law of July 29, 2004.
Article 72 of the Constitution.
Articles 72 and 72-1 of the Constitution.
Article 72-2 of the Constitution.
Article 72-3 of the Constitution.
DGCL (Direction Générale des Collectivités Locales) 2011.
Law on Inter-Municipal Bodies of July 12, 1999 (also called the Chevènement
Law).
Dexia 2008.
Law of July 7, 1983.
Because their assets are immune from attachment, public establishments are
not subject to the liquidation law of January 25, 1985.
DGCL 2011.
DGCL 2011.
The business tax was abolished in 2010 and was replaced by the Territorial Economic Contribution (Contribution Economique Territoriale).
The central government reduced the delay in refunding the value-added tax
to SNGs, enabling them to maintain capital expenditures in 2009 above the
2004–07 average. The expected value-added tax refund payments in 2009 was
more than i4 billion, equivalent to 8 percent of SNG capital expenditures in
2009. Overall net borrowing by French SNGs increased by i5.1 billion in 2009,
while total outstanding debt as a share of GDP increased by 4 percent over 2008.
However, debt accounts for only 4.2 years of overall SNG current balance.
These 11 departments are Ardennes, Cher, Corrèze, Creuse, Haute-Loire,
Haute-Saône, Indre, and Meuse (which are rural and poor departments), and
Pas-de-Calais, Seine Saint-Denis, and Val d’Oise (which are urban and poor
departments).
For instance, the number of beneficiaries of the Active Solidarity Income
(Revenue de Solidarité Active) in the department of Seine-Saint Denis increased
41 percent from 2008 to 2009.
An example of an SNG with an unbalanced budget is Seine-Saint Denis which,
in April 2010, voted an unbalanced budget to protest the abolition of the professional tax and the reduction of transfers from the central government.
Compulsory expenditures include wages of local civil servants; SNG financial
contributions to local-interest services; maintenance of the city hall, roads, and
cemeteries; and debt services.
Acts No. 82-213 of March 2, 1982; No. 82-623 of July 22, 1982; No. 83-1186 of
December 29, 1983; No. 85-97 of January 25, 1985; No. 86-972 of August 19,
1986; No. 88-13 of January 5, 1988; No. 90-55 of January 15, 1990; No. 92-125
255
256
Until Debt Do Us Part
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
of February 6, 1992; No. 93-122 of January 29, 1993; and No. 94-504 June 22,
1994. For an overview, see Bertucci (1995).
The National Court of Accounts and the Regional Courts of Accounts constitute a separate judicial branch in administrative-financial matters.
L. 1612-14 CGCT provides that if the projected deficit exceeds 10 percent of
operating expenses for a municipality with less than 20,000 inhabitants, or
5 percent for other municipalities, the Regional Chamber of Accounts recommends measures, on request of the Prefect, to reestablish budgetary equilibrium within a month.
Fabrice Robert, Les Finances Locales, 2009, 158.
Pont-Saint-Esprit is a municipality of 9,523 inhabitants. The Regional Chamber of Accounts of Languedoc-Roussillon in 2006 began implementing several
actions to cope with Pont-Saint-Esprit’s financial distress, such as an audit of
Pont-Saint-Esprit financial management during 1999–2005.
The 1962 Public Accounting Regulations specify this principle of separation.
In the area of education, an important public function in France, Public
Accountants are appointed by the Minister of Education. Otherwise, there are
Public Accountants at the three SNG levels (municipalities, departments, and
regions); they are, respectively, the receveurs municipaux (municipal public
accountant), the payeurs départementaux (departmental public accountant),
and the payeurs régionaux (regional public accountant).
The State Treasurer services also audit local public finances management of
only municipalities with a population under 3,500 inhabitants whose operating
revenues are under i750,000.
These two public financial institutions are la Caisse des Dépôts et Consignations
and la Caisse d’aide à l’équipement des collectivités locales.
This figure includes SNGs and public-private enterprises.
Article D. 1511-32 of the Code Général des Collectivités Territoriales.
Article D. 1511-34 of the Code Général des Collectivités Territoriales.
Article L. 1614-4 of the CGCT.
Law of May 15, 2001, Nouvelles régulations économiques (New Economic Regulations), authorizes SNGs to issue short-term negotiable debt instruments with
a minimal nominal value of i150,000 and a maximum duration of one year.
These instruments are not considered to be long-term debt.
Decree No. 2005-601 of May 27, 2005 modifying Decree No. 2004-15, relying
on Directive No. 92/50 of June 18, 1992.
Circulars of September 6, 1999, and May 15, 2000.
The successor company is Dexia.
Based on interviews with Valérie Montmaur, S&P; and Gilbert Payan, municipality of Aubagne.
Most French SNGs do not access the capital market; hence, they are not rated.
Since the majority of French SNGs do not access the capital market, relying
mainly on bank loans, the rated SNGs may represent the most creditworthy
borrowers.
France’s Subnational Insolvency Framework
48. An example of this fiscal risk can be seen in the municipality of Levallois-Perret.
In the late 2000s, the city faced acute financial woes. The costs of running the
city increased much more rapidly than receipts, due mostly to increased investment and personnel expenses. Though there was room for raising taxes, the
city’s self-financing capacity in 2006 remained unchanged relative to 2002. The
municipality relied increasingly on borrowing. The biggest source of c ontingent
liability stemmed from urban development operations. The municipality owns,
de facto, 80 percent of three public limited companies and 15–80 percent of
another 40 commercial companies. Some SEMARLEP activities, such as real
estate sales, are outside the competences of SNGs and their government-owned
corporations. Yearly reports appear to contain insufficient information for the
municipal council to understand either the true financial stakes of the group’s
activities or the indirect commitments and the risks flowing from them. Source:
National Court of Accounts Report (2009b).
49. The National Court of Accounts 2009b.
50. This amount accounts for 58.4 percent of total SNG structured debt (see
Crouzel 2011).
51. National Court of Accounts 2009b.
52. Statement of financial executive Wallace O. Sellers before the Subcommittee
on Economic Stabilization of the U.S. House Committee on Banking, Currency and Housing, October 22, 1975, 1418. (“In England or France, it would
be unthinkable for a major city to go bankrupt. In France, with a highly centralized system developed under Napoleon, the state exercises substantial control
over the finances of SNG units.”)
53. One example is the municipality of Saint-Étienne, whose heavy exposure to
financial derivatives was discovered in 2008. Facing a skyrocketing debt in the
1990s, the municipality tried to smooth the payoff by actively managing the
debt through structured products (“Saint Étienne: dégonflement programmé
de la dette,” Les Echos, July 7, 1994). The central government did not want to
intervene because this could create a precedent of a moral hazard (“Pas de
dérogation pour Saint-Étienne,” Les Echos, April 15, 2009a). There were calls for
a more structured legal framework and the creation of a “bad bank” in order to
get rid of all the toxic products (“Saint-Étienne veut une régulation plus stricte
qu’une charte,” Les Echos, May 29, 2009b).
54. Cécile Crouzel, “La difficile évaluation des prêts toxiques,” Le Figaro, December
15, 2011.
55. Garabedian, A., “Le puzzle des collectivités locales se met en place,” L’Agefi, February 9, 2012.
56. This view is advocated by Michel Klopfer, an independent consultant who specializes in local finances (see Klopfer 1996).
57. Circulaire No. 92-260, Contrats de couverture du risque de taux d’intérêt
offerts aux collectivités locales et aux établissements publics locaux, September 15, 1992. See also Fitch Ratings, French Local Government’s Structured Debt:
Active Management or Speculation?, July 15, 2008b.
257
258
Until Debt Do Us Part
58. Article 120-3 of the General Accounting Plan.
59. Les Echos, “Emprunts des collectivités locales: le gouvernement cherche à rassurer,” November 3, 2008.
60. Joël Cossardeaux, “Emprunts toxiques: élus locaux et banques font le ménage
pour l’avenir,” Les Echos May 29, 2009.
61. Executive Order No. 2008-1078 of October 22, 2008.
Bibliography
Bertucci, Jean-Yves. 1995. “Le Contrôle des Budgets des Collectivites Locales Par les
Chambres Régionales des Comptes.” In Le financement du développement local.
Paris: Presses Universitaires de France.
Bouvier, Michel. 2008. Les Finances Locales. Paris: L.G.D.J.
Circulaire No. 92-260. 1992. “Contrats de couverture du risque de taux d’intérêt
offerts aux collectivités locales et aux établissements publics locaux.” La Documentation française, Paris, September 15.
Cossardeaux, Joël. 2008. “Les élus locaux redoutent désormais les effets d’un argent
cher,” Les Echos, October 20.
———. 2009. “Emprunts toxiques: élus locaux et banques font le ménage pour
l’avenir,” Les Echos, May 29.
Crouzel, Cécile. 2011. “La difficile évaluation des prêts toxiques.” Le Figaro, December 15.
Dexia. 2008. EU Sub-national Governments, 2007 Key Figures. 2008. Brussels: Dexia.
DGCL (Direction Générale des Collectivités Locales). 2011. “Les collectivités locales
en chiffres 2011.” La Documentation française, Paris.
Fitch Ratings. http://www.fitchratings.com.
———. 2008a. “Institutional Framework for French Subnationals.” February 28.
———. 2008b. “French Local Government’s Structured Debt: Active Management
or Speculation?” July 15.
———. 2009. “European Local and Regional Government Outlook 2010.”
December 16.
Garabedian, A. 2012. “Le puzzle des collectivités locales se met en place.” L‘Agefi,
February 9.
Klopfer, M. 1996. “Apprécier le risque financier d’une collectivité locale.” Banque,
No. 572.
Les Echos. 1994. “Saint Étienne: dégonflement programmé de la dette.” July 7.
———. 2008. “Emprunts des collectivités locales: le gouvernement cherche à rassurer.” November 3.
———. 2009a. “Pas de dérogation pour Saint-Étienne.” April 15.
———. 2009b. “Saint-Étienne veut une régulation plus stricte qu’une charte.” May 29.
Mairie de Paris. 2009. “Annual Report for the Financial Year 2008.” Mairie de Paris,
France.
France’s Subnational Insolvency Framework
Moody’s Investors Service. http://www.moodys.com.
———. 2002. “Subsovereign Defaults in France.” December.
National Court of Accounts. 2009a. “Les limites des procédures de contrôle budgétaire des collectivités territoriales: le cas de la commune de Pont-Saint-Esprit.”
In Annual Public Report. Paris: National Court of Accounts.
———. 2009b. “Les risques pris par les collectivités territoriales et les établissements publics locaux en matière d’emprunt.” In Annual Public Report. Paris:
National Court of Accounts.
———. 2009c. Rapport d’activité des jurisdictions financiéres 2008. La Documentation francaise, Paris.
Paris, Antoine. 2009. “La dette structurée des collectivités locales: gestion active ou
spéculation.” Mémoire de fin d’études, Jouy-en-Josas, École des Hautes Études
Commerciales, Paris.
Rapport Balladur. 2009. “Comité pour la réforme des collectivités locales.” La Documentation française, Paris, March.
Rapport Jamet. 2010. “Rapport à Monsieur le Premier Ministre sur les finances
départementales.” La Documentation française, Paris, April.
Rapport Mauroy. 2000. “Refonder l’action publique locale.” La Documentation
française, Paris, January.
Rapport Pébereau. 2005. “Rompre avec la facilité de la dette publique.” La Documentation française, Paris.
Rapport Richard. 2006. “Solidarité et performance, Les enjeux de la maîtrise des
dépenses publiques locales.” La Documentation française, Paris, December.
Regional Chamber of Accounts of Languedoc-Roussillon. 1989. “Commune de
Bolquère.” Regional Chamber of Accounts of Languedoc-Roussillon, Montpellier, January 18.
Robert, Fabrice. 2009. “Les Finances Locales.” La Documentation française, Paris.
Standard & Poor’s. http://www.standardandpoors.com.
———. 2012. “Eurozone Sovereign Rating and Country T&C Assessment Histories.” January 13.
259
7
Hungary: Subnational Insolvency
Framework
Charles Jókay
Introduction
Hungary’s experience with municipal insolvency is unique. It has one
of few insolvency systems in the world where municipal insolvency can
lead to a court-supervised “bankruptcy and reorganization” process
that is led by an independent receiver or trustee.1 The Municipal Debt
Adjustment Law (Law XXV, 1996) was enacted a few years after the start
of economic and political transformation in Hungary, and studies have
examined its results (for example, Jókay, Szepesi, and Szmetana 2000).
Hungary’s experience has influenced similar legislation in Estonia,
Latvia, and Romania in the first half of the 2000s, although the courtsupervised system has not been emulated entirely.
A review of the Hungarian experience will add to the understanding of a subnational insolvency system, particularly with respect to the
potential role of an independent judiciary, importance of properly allocating risk, ensuring the viability of local government in delivering vital
services, and balancing the needs of creditors and debtor. Hungary’s
case emphasizes the importance of drawing a distinction between the
municipality itself, and the private or publicly owned enterprises contracted to perform vital public services. A further distinction involves
261
262
Until Debt Do Us Part
whether or not the municipality provides an explicit guarantee to
private or publicly owned enterprises and whether or not these enterprises perform any legally mandated public function. This distinction
affects the potential recourse of lenders to the municipality.
Recent examples in Hungary demonstrate implications for both the
owners—that is, municipalities and enterprises—that have mutual and
often contradictory effects on each other. According to the Hungarian
experience, only municipalities are covered by municipal bankruptcy
legislation, and all corporate entities regardless of ownership and type
(limited liability, share companies, for profit, or nonprofit) fall under the
jurisdiction of corporate bankruptcy law.2 The provision of guarantees
by municipalities, and the obligation of municipalities to have certain
services delivered regardless of the form of delivery complicate these relationships. The Municipal Debt Adjustment Law extends the obligations
of providing public services to the municipality itself if the company
(private or public) that provides a particular municipal service such as
water or public transportation goes through bankruptcy proceedings
under the commercial code.3
This chapter analyzes the Hungarian municipal insolvency experience since 1996. The chapter also analyzes issues that have emerged,
such as treatment of municipal enterprises that may perform a mandatory function but operate under the commercial code, and others such
as guarantees and hidden contingent liabilities that have been identified
in insolvency cases.4 The chapter presents the historical context for the
adoption of the Municipal Debt Adjustment Law and the implementation experience, describes the fiscal functioning of local governments
and their responsibilities, and identifies issues to be dealt with going
forward. A review of the Hungarian experience could offer lessons for
other countries with a relatively decentralized local government system.
The scope of this chapter does not extend to enterprises in which
the national government is the sole or dominant shareholder. For this
reason, companies such as the national railways (MAV) and the power
grid (MVM) will not be discussed. Municipally owned enterprises are
not governed by any special legislation of their own and belong under
the corporate regulatory regime regarding taxes, accounting, and bankruptcy. These entities are treated as corporate borrowers by lenders, and
in most cases, a municipal guarantee is also sought. Such entities do not
Hungary: Subnational Insolvency Framework
fall under the Municipal Debt Adjustment Law but can have a direct
impact on the financial viability of the municipality through called
guarantees (the creditor submits a claim to the guarantor), and invoked
service provision obligations. A countervailing tendency is that municipal enterprises are consolidated into holding companies that amass
profits and assets, all of which are off the balance sheet of the owner,
except through the nominal value of shares that appear as “share assets”
on the balance sheet of the municipality.5
On corporate bankruptcy vs. municipal debt adjustment, the rules on
the effect of both insolvency laws are clear. For cases of insolvency of a
municipal government as an entity, the Municipal Debt Adjustment Law
applies with all its practical consequences: the debts should be partly or
entirely settled, and the entity shall be able to provide services. Should
any of the business associations that are rendering public utility services
become insolvent, the corporate insolvency law (Law XLIX) applies.
There is an option for pursuing the business activity for a definite time
provided that the company in insolvency proceedings is able to pursue
its activity without any losses, and the creditors agree.
A country’s macroeconomic framework impacts the financial health
of municipal governments, primarily through the cost of borrowing, fiscal
transfers, and economic growth (Canuto and Liu 2010). This chapter discusses how the 2008–09 global financial crisis impacted local government
finance in Hungary, although the focus of the chapter is on the subnational
insolvency system, which has its own distinct issues, such as the design of
the framework and the priority structure for resolving competing claims.
A more detailed analysis of the macroeconomic challenges facing the
country, though important, is outside the scope of the chapter.
The rest of the chapter is organized as follows. Section two
reviews the macroeconomic and institutional context for the development of the Municipal Debt Adjustment Law. Section three presents
the structure of subnational governments and their finance. Section four
details the subnational borrowing framework with a focus on ex-ante
regulations. Section five examines subnational insolvency procedures
and how the law deals with different types of debt instruments and debtors. Section six reviews insolvency implementation experience. Section
seven draws lessons and summarizes ongoing discussions in Hungary
on potential reforms to strengthen subnational insolvency procedures,
263
264
Until Debt Do Us Part
in the context of broader institutional reforms. Section eight presents
conclusions.
Macroeconomic and Institutional Context for the
Development of the Municipal Debt Adjustment Law
Roots in “reform communism.” Hungary’s economic and political transformation that culminated in free elections in April 1990 had roots
in the “reform communism” of the 1980s. Hungary’s first significant
transformational laws were passed in the late 1980s before the systemic
change that shook Central Europe during 1989–90. The banking system
became multitiered in 1987, when the central banking functions were
separated from the commercial and retail banks, all still owned by the
state. The first international banks to appear in Hungary after nationalization in the late 1940s were Citibank and Unicbank, the precursor to
Raiffeisen, both of which appeared in the 1980s. A form of Party cadrelead self-privatization6 began in 1988–89, as well. Direct investment and
limited currency liberalization for both industry and households were
in place by 1990. Following Hungary’s free elections in early 1990, all
of the major sectoral and framework laws were quickly passed. These
included the Law on Local Government, and laws on the stock market,
property transfer, banking, privatization, pensions, enterprises, elections, media, accounting, and the state budget system.7 Although some
of these laws have been amended in minor ways, they remain in effect
(see box 7.1).
Box 7.1 Major Acts Regulating the Municipal Sector, Hungary, 1990–2011
1990: Law on Local Government created municipalities out of the existing local
council system
1991: Property Transfer Act assigned and returned most state property on municipal
land to municipalities, transferred other state property to the local level, transferred “council” property to newly formed municipalities
1995: Modifications to the Law on Local Government imposed borrowing limits
based on a share of net own revenues
1996: Municipal Debt Adjustment Law
1996: Debt service limit added to the Law on Local Government
2011: Law on Local Government (effective January 2012)
Hungary: Subnational Insolvency Framework
True self-government. The 1990 Municipal Act and various constitutional provisions guaranteed that Hungarian municipalities were
considered an equal and unsubordinated branch of government. Hungarian local self-governments were granted important freedoms to
engage in business activities, impose local taxes, plan their own development, and borrow in an entirely unregulated manner until 1995. The
3,194 self-governments, which range from villages of a few dozen people to the capital city of Budapest, have essentially equal status, rights,
powers, and duties. The national government operated public administration offices at the county level until 2012 to check compliance with
procedures and administrative requirements, but they are not allowed
to comment on the substance of municipal decisions and budgets provided to them. The State Audit Office, independently of the government
and reporting directly to Parliament, does engage in performance, compliance, and financial audits on an as-needed and random basis; however, it can only recommend changes to local leadership and file formal
charges with the prosecutor’s office. The State Audit Office has no investigative powers.
In this environment, Hungarian local governments were, until 2012,
not required to seek permission of a ministry to engage in borrowing
and had the right to freely choose a bank. Attempts to have an integrated
single treasury account do not work politically in Hungary, given the
resistance of commercial banks, the municipal sector, and the persistence of the liberal tradition underlying the Law on Local Government.
Notwithstanding that, each municipality has a treasury account to net
out transfers and shared taxes received from, and payroll taxes and the
value-added tax (VAT) owed to, the central government. These treasury
accounts are only for clearing, while all municipalities keep their excess
cash in commercial banks and savings cooperatives.
Macroeconomic context. In the context of the rapid collapse of the
Soviet markets and of trade among the former Council for Mutual
Economic Assistance states, by 1994 Hungary suffered a collapse of its
industrial sector, resulting in high unemployment, energy price shocks,
and other adjustment costs. In 1994, inflation was 15–20
percent.
Municipalities, without any restriction on their investment, cash management, and borrowing activities, began to behave irresponsibly, and
engaged in for-profit businesses such as hotels and industrial parks.
265
266
Until Debt Do Us Part
unicipalities engaged in an investment boom, borrowing at high
M
interest rates from the handful of state-owned banks. Most of this
investment was stimulated by grants from the central government for
vital environmental investments in safe drinking water, sanitation, and
wastewater treatment.
Threat of soft budget constraints. In a strange role reversal, in 1993
and 1994, representatives of state-owned banks lobbied for special bailouts for their clients, while the Ministry of Interior faced the first major
policy challenge to Hungary’s liberal Municipal Act: bailouts would lead
to more bailouts, and economic freedom also meant facing the consequences of bad decisions on the part of both lender and borrower.
Unfortunately, at that time, there was no framework in place to deal with
defaults by municipalities, and, notwithstanding pressure from those
few in trouble and their banks, policy makers and bureaucrats faced a
serious dilemma. Solving the problem of irresponsible lending and borrowing has two possible solutions: restricting municipal borrowing by
requiring higher-level approval and a stricter debt limit formula, or,
alternatively, a no-bailout policy combined with no higher-level ex-ante
or ex-post oversight.8 The second approach would not restrict debt
issuance beyond the existing debt service formula, but in addition to a
no bailout policy would create a legal framework for debt adjustment
and restructuring similar to that in the corporate sector. In 1994, design
objectives for a debt adjustment procedure included the following:
•
•
•
•
•
To maintain municipal autonomy from the central government by
not instituting a permitting process
To assure that mandatory tasks would not be endangered by the consequences of failed projects or irresponsible borrowing
To offer assurance to lenders and other creditors that their claims
would be adjudicated fairly9
To protect municipal assets classified as being essential for providing
public services (schools, the town hall, parks, and so forth)
To encourage voluntary agreements among the creditors and borrowers, while supporting structural changes at the local level needed
for long-term fiscal viability.
Given the unrestricted freedom of local governments to manage
their assets and budgets (within the constraints of shared taxes, transfer
Hungary: Subnational Insolvency Framework
ayments, and local tax capacity), the central government faced the posp
sibility of potential contingent liabilities in the form of “moral obligations” and political pressure to fund local mandatory tasks. Or worse,
the central government faced the possibility of directly carrying out
mandatory local tasks if local governments failed to. By the end of 1995,
several local governments lobbied for and received one-time grants from
the central government’s general reserves to resolve insolvency caused
by mismanagement and excessive debt.10 This precedent created a moral
hazard and gave no incentive to the local governments to make more
reasonable investment decisions. The major lenders began to lobby the
central government for “one time” bailout assistance to their clients.
Toward institutionalized debt adjustments. By 1995, Hungary’s
deteriorating macroeconomic situation and adjustment resulted
in a currency devaluation and real cutbacks in spending and other
restrictions. The problem of indebted local governments became
more serious. Municipalities began to borrow to finance short-term
operating deficits and nonmandatory infrastructure, creating substantial refinancing risks given that the short-term interest rates on Treasury securities were between 25 and 30 percent during 1994–95. The
restrictions on bailouts and the use of transfers cited above were not
adequate to withstand the political pressure on the central government
to provide emergency funds to assist potentially insolvent local governments whose poor investment decisions were endangering the provision of mandatory public services. Without the political will or ability
to tightly control local government borrowing and business practices by
constitutional and legislative fiat, the Hungarian government decided to
propose a municipal debt adjustment law that could be invoked if prudence and other preemptive measures failed.
Structure of Subnational Governments and Their Finance
Structure of Subnational Governments
Hungary is a unitary (centralized) state, with three levels of elected government and appointed administration: central, counties, and municipalities. Local governments that are in full harmony with the European
Charter of Local Government11 can be found only at the level of municipalities. Each village, small city, large city, Budapest district, and the city
267
268
Until Debt Do Us Part
of Budapest itself is called a municipality, regardless of whether it has
a legal status of village, town, or city. For the purposes of this chapter,
a Hungarian municipality is one of 3,196 settlements that has its own
mayor, council, budget, and assets, as well as mandatory duties and full
authority to pass local ordinances allowed by law.12 No municipality is
subordinate to any other municipality, and only an act of parliament
can dissolve an assembly or involuntarily combine several jurisdictions
into one.
Only 10 percent of municipalities have a population over 5,000,
and only 5 percent have a population over 10,000. By international
comparison, the average size of a Hungarian local government is relatively small. Notwithstanding the small size of local governments, they
are responsible for a significant number of tasks. The budget of all local
governments (including counties) amounts to 12–13 percent of gross
domestic product (GDP), which is average for Europe.13
The development of the Hungarian system complied with the
requirements of the European Charter of Local Government. The legal
foundations of the present framework are the Constitution (1989) and
the 1990 Law on Local Government. They define the economic basis of
the independence of local governments. These include municipal ownsource revenues, their assets, and grants from the central government.
Local governments have considerable autonomy in local decision-
making processes. The central government has limited authority over
them. The Fiscal Stability Act (No. CXCIV, Law of 2011), effective
January 1, 2012, generally binds the borrowing of the municipalities to
the permission of the central government.
The mandatory duties of Hungarian municipalities numbered in the
hundreds until the end of 2011. The Law on Local Government, and
dozens of acts regulating sectors such as education, health, social welfare, and so forth, add to the duties.14 These specific functions, such
as primary and secondary education and many public administration
tasks, will be centralized effective 2013. It is too early to forecast the fiscal impact on the municipal sector.
This list of mandatory tasks, regardless of the size and economic
potential of municipalities, is a source of many fiscal stresses that are outside
the scope of this study. In most cases, the delivery method and standards
of performance for these tasks are not specified. Therefore, m
unicipalities
Hungary: Subnational Insolvency Framework
are free to contract out, hire private providers, form their own firms, and
seek forms of cooperation with the nonprofit sector and with neighboring
municipalities in associations authorized by specific legislation. Municipalities have great autonomy by law in local decision making.
Municipal services are delivered by many types of legal entities, budgetary or subject to the commercial code. Municipal services of all types are
delivered by the municipalities’ own departments, by public bodies that
are subjects of the municipal budget, by commercial code enterprises
wholly owned by the municipality and established for one specific service, by enterprises with mixed municipal and private ownership, by
contractors, by public-private partnership (PPP) entities, by concession
companies (water, wastewater, solid waste), and by nonprofit corporations that may or may not be in municipal ownership, among others.
The relationship of these entities to the municipality is governed by
founding acts, by concession and procurement law, by contracts, and by
the type of service being performed on behalf of the municipality using
public funds and user charges.
Sources of Subnational Finances
Capital investment funding. Since European Union (EU) accession in
2004, Hungarian municipalities have relied almost exclusively on various grants that originate from EU sources, since domestic capital grants
have all but been eliminated. These grant structures typically require
cofinancing; municipalities need to cofinance 20–50 percent of project costs. The source of financing may be from asset sales, accumulated
savings, and from bond issues and bank loans. The central government budget provides aid for municipalities to supplement their own
resources, such as local taxes and fees. The “EU Own Resource Subsidy,”
which is a part of central government appropriations, lends assistance to
municipalities in disadvantage for the purpose of cost sharing required
by investments aided by the EU.
There are four types of municipal revenues15:
•
Own current revenues, such as local taxes, user charges, and other
nontax revenues such as income of municipal institutions (catering
fees), rental income, and interest and dividends, which account for
about one-third of municipal revenue.
269
270
Until Debt Do Us Part
•
•
•
Own capital revenues, such as sales of real estate and other tangible
assets, sales of shares in enterprises, and dividend earnings, which
account for about 10 percent of municipal revenue.
Intergovernmental grants that are either earmarked or freely usable,
both for mandatory services, which account for about 15 percent of
municipal revenue.
Minor tax sharing of the personal income tax and the full amount
of the motor vehicle tax (the capital city and the towns with county
status also share the duty revenue), which account for about one-
quarter of municipal revenue. Both the motor vehicle tax and the
personal income tax share will become a fully central source of revenue as of 2013.
The formulas for tax sharing and grant design have shown great
volatility during the last two decades. As a result, it is difficult for local
governments to exercise strategic financial management, considering
the risk that the annual central government budget will alter normative
grant calculations to their disadvantage.
The local business tax (essentially a tax on gross turnover set at
2 percent) accounts for 84 percent of local taxes collected, followed
by a building tax at 11.2 percent.16 The turnover tax is vital in that it
is the primary source of cash to finance debt service. All other taxes,
including the sojourn tax (hotel tax) and minor taxes on the area of land
and buildings, are not substantial sources of revenue. Budapest collects
more than half of the total business tax for all municipalities, and most
small municipalities impose only minor communal and property taxes,
since they have little economic activity. In the larger municipalities that
are categorized as cities, local taxes account for no more than 20 percent
of total revenue. Villages collect virtually no local tax, and only less
than 5 percent of their budgets are funded from own-source revenues,
including all fees, local taxes, and business income.
De jure financial freedom vs. de facto central control. Hungarian local
governments have considerable de jure financial freedom. In the euphoria of the transition from a planned economy to a market economy, a
highly decentralized and undersupervised local government system was
created. De facto fiscal independence of local governments is contradictory, at a minimum. First, 60–70 percent of their funds actually depend
Hungary: Subnational Insolvency Framework
on the annual decision on the central budget. Second, the government
grants paid to finance their compulsory functions cover a decreasing
portion of the costs of these services. An increasingly higher proportion
of the local governments’ own revenues finances the centrally mandated
functions. As a result, the most effective instrument for the central government to exercise influence on the functioning of local governments
is to regulate tax sharing and grant design.
Restrictions on Debt Collateral
Municipalities in Hungary own significant portfolios of assets, such as
buildings, land, shares in enterprises, equipment, and financial investments. These properties are required by the Municipal Act to be categorized by the local administration applying the law, as being (a) core, that
is, essential for the delivery of a public service and hence unavailable
for sale or use as collateral; (b) essential but negotiable; or (c) nonessential and fully negotiable. In most cases, it is not the type of property
that defines its negotiable or essential status, but rather its function in
delivering mandatory services. In essence, with a few exceptions such
as historical monuments, archives, museums, parks, and so forth, the
local council may choose to reclassify property as being negotiable if its
underlying function changes. This distinction of core vs. noncore property is critical in determining whether an asset is available as collateral,
and whether it can be sold during a municipal insolvency proceeding.
Besides “core property,” certain sources of revenue are not available for
debt payment. These are normative state transfers, the personal income
tax, and other shared taxes. The turnover tax (84 percent of local taxes
collected) is the primary source of financing debt service.
Impact of the 2008–09 Global Financial Crisis
After a 6.8 percent drop in GDP in 2009, Hungary’s GDP grew modestly in 2010 and 2011. Inflation declined from 7.9 percent in 2007 to
3.9 percent in 2011. The ratio of fiscal deficit to GDP improved during
the period.17 With high exposure to foreign currency debt at the state,
household, and enterprise levels, Hungary’s gross external debt ratio to
GDP rose from 105 percent in 2007 to 140 percent by 2011.18 These
macro tendencies impacted local governments indirectly through rapid
devaluation19 and higher risk premia for borrowing in foreign currency,
271
272
Until Debt Do Us Part
since domestic currency borrowing by local governments was not
significant during this period; prevailing domestic interest rates continued to be a multiple of euro or Swiss franc rates.20
Hungary’s local government debt stock21 (in bonds) jumped from
212 billion forint (US$1.23 billion) in 2007 to over 555 billion forint
(US$2.3 billion) in 2011. Bond issuances increased on a net transaction basis annually during 2007–10. However, net bond issues were over
325 billion forint (US$1.35 billion) from 2007 to 2011, with an over
200 billion forint (US$830 million) increase as the result of revaluation due to the strong Swiss franc. On the loan side, loan debt stock
increased by only 200 billion forint (US$830 million) during this period,
with actual declines in 2008 and 2011 due to prepayment. In some
cases, revaluation of existing debt stock was large enough to counteract
changes in net transactions (for example, from 2008 to 2011, revaluation overwhelmed repayment of loans and increased debt stock).22
Hungarian municipalities face three risks from the banking and fiscal crisis in Europe. The first risk is currency risk and rapid devaluation
of the forint, a risk they cannot hedge since all of their revenues are in
domestic currency. Over 80 percent of municipal bonds are denominated in Swiss francs and half of bank loans are in euros. The second risk
stems from a decline in fiscal transfers, which has affected their revenues
used to deliver mandatory services and make payments. From 2007 to
2011, operational transfers dropped 23 percent in nominal terms.23 The
third risk is the impact of slower economic growth on own revenues,
which are the only funds available for debt service.
Subnational Borrowing Framework:
Ex-Ante Regulations
Subnational Debt Market
Ex-ante rules on borrowing. Hungary’s debt regulations operated under
several important rules and laws that regulate the economic and budgetary functions of local government. These controls emerged during
1990–2002, but most of them are in basic laws and were in effect by
1996, when the debt adjustment law was passed. Prior to 1995, there
were no restrictions on municipal borrowing. In 1995, a debt service
limit was introduced that was based on a restricted definition of funds
Hungary: Subnational Insolvency Framework
that were available for debt service. Only own revenues (local taxes,
local fees, business income, earned interest, rental income, and so forth)
would be available for debt service, only after all mandatory services
were paid for, and only 70 percent of such surplus was available calculated on the last completed budget year.24 The formula was difficult
to enforce since guarantees, contingent liabilities, PPP payments, and
grace periods could not be taken into consideration, and it was built on
the assumption that full operating costs were actually known, and the
surplus would be consistent over time. There is no early warning system, since the cash accounting system is delayed, inaccessible, and can
hide deferred payments and contingent liabilities with ease.
The Law on Local Government (2011), which went into effect in
January 2012, stipulates a new kind of limit calculation that essentially restricts debt service to half of annual own revenues (excluding
all transfers and shared taxes and capital revenues). There is no stock
limit, and, more significantly, each borrowing or bond issue must be
approved by government decision, a practice that did not exist between
1990 and 2011.
General characteristics of municipal banking and lending. OTP, the
former monopoly state-owned savings bank, privatized in 1995, and the
network of savings cooperatives (owned by depositors) dominate about
70 percent of accounts management for municipalities in Hungary.
The rest of the commercial banks, all of them foreign-owned subsidiaries or representation offices, are more successful in lending and bond
investing, since OTP’s dominance in lending volume is much lower, at
about 50–55 percent of long-term lending. There are no public issues
of municipal debt. Almost all bonds are held by the arranger bank, the
same banks that dominate term lending and accounts management, and
only a handful of bonds have been sold to investors based outside of
Hungary. Even though almost all bonds (some 80 percent) are denominated in Swiss francs and euros, they are held by Hungarian-registered
banks. Loans are also heavily in euros and Swiss francs, but mainly from
Hungarian-registered banks, since Hungary’s legislation allows public
bodies to borrow in any currency.25
Bank management of local governments’ accounts. Prior to its privatization in 1995 and public offering in 1997, OTP, known as the National
Savings Bank, had a monopoly on managing municipal bank accounts
273
274
Until Debt Do Us Part
and was responsible for nearly all lending and other financial services
offered to the sector. Municipalities may have as many banking relationships as they wish, as long as they maintain one primary account
for their transactions.26 The OTP’s primary banking relationship with
the municipal sector until 1995 earned benefits for the bank in terms
of being familiar with municipal management, having a monopoly on
information on the finances of the client, and essentially having the
right of first refusal on any new loan, overdraft product, or even bond
issuance. Savings cooperatives, with extensive but localized networks,
took business from OTP in the smaller towns and larger villages.27
The separation of primary banking from other financing activity is very
significant, and the resulting competition among banks can explain, in
part, the favorable, almost irrational terms offered to Hungarian municipalities in the lending boom between 2006 and 2008, which came to
an abrupt ending during the financial crisis as foreign headquarters
restricted lending. EU accession in 2004 meant full adoption of EU
procurement directives on financial services. The EU exempts certain
services, such as the issuance and sale of securities and other financial instruments, from procurement, and this provision was inserted
directly into Hungarian procurement rules. Since no formal procurement is needed, bond issuance enjoys a significant advantage over bank
loans, and a series of intense competitive negotiations and offers can be
executed in a short time.
Municipal Debt Composition
Changes in debt stock. Hungarian municipalities’ long-term debt
remained modest during the decade 1990–2000, not exceeding 100 billion
forint (about US$600 million at prevailing exchange rates) until 1998.28
This is in contrast with indirect liabilities such as vendor loans and
long-term service contracts. Between 2000 and 2009, bank borrowing increased by a factor of 6. Bond issuance went from zero in 2004
to a stock total of nearly 500 billion forint (US$2.5 billion) in 2009.
Significantly, short-term debt amounted to a quarter of total debt by
2009, a sign that serious arrears and loans from vendors had accumulated in the sector, with short-term cash flow loans essentially financing
hidden operational deficits. The only restriction on short-term debt is
that it has to be zero by the last day of each year, a practice that does not
Hungary: Subnational Insolvency Framework
eliminate the rollover. As shown in table 7.1, the total long-term debt
of Hungarian municipalities reached 4.5 billion euros by March 2011,
but the share of bonds fell to 48 percent. Table 7.1 also reflects changes
in the Hungarian forint exchange rate compared to the Swiss franc and
euro as well, so the relative share of loans and bond issues also depends
on the cross rate between euros and Swiss francs—since 80 percent of
bonds are issued in Swiss francs, while only half of loans are in euros.
The aggregate data shown in table 7.1, however, do not reflect the
extent of the liabilities of the municipalities, since given guarantees do
not show up in financial statistics reported to the National Bank, and do
not appear in balance sheets unless those guarantees are actually paid.
Debt stock could exceed total annual revenues by several factors, when
the most common debt stock limitation in use in European transition
countries is a debt stock limit of 50–100 percent of the annual budget.
This budget-related debt stock limit is a rule of thumb, since property
taxes, property valuations, and other measures related to assessed and
taxed value are not informative in this context.
Figure 7.1 shows neither guarantees given by municipalities nor
other contingent liabilities. Furthermore, enterprises owned by municipalities are also borrowing as commercial entities, and their liabilities
and assets do not appear on municipal balance sheets, except in the
Table 7.1 Debt of Hungarian Municipalities Calculated at Prevailing
Euro Exchange Rates
National bank financial accounts
Municipal debt expressed in euros
Long-term bonds
(in thousands)
Long-term loan
(in thousands)
Total long-term
debt (in thousands)
% of long-term
debt in bonds
2003
22,800
730,000
752,800
3.03
2006
100,000
2,150,000
2,250,000
4.44
2007
83,399
1,707,000
1,790,399
4.66
2008
1,735,000
1,660,000
3,395,000
51.10
2009
1,775,000
1,660,000
3,435,000
51.67
2010
2,071,890
2,177,900
4,249,790
48.75
2011 1Q
2,188,000
2,310,000
4,498,000
48.64
Source: Author’s calculations based on National Bank of Hungary “Financial Accounts,” http://english
.mnb.hu/Statisztika/data-and-information/mnben_statisztikai_idosorok/mnben_elv_net_lending/
mnben_0601_osszefoglalo_informaciok), published quarterly; and official exchange rates published daily.
275
Until Debt Do Us Part
Figure 7.1 Bonds, Bank Loans, and Other Debt, Hungary, 2005–11
700
600
500
Forint, billions
400
300
200
100
11
20
10
20
09
20
08
20
07
20
06
20
05
0
20
276
Year
Bank loans
Other debt
Bonds issued (stock)
Source: National Bank financial statistics published quarterly; http://english.mnb.hu/Statisztika/data-andinformation/mnben_statisztikai_idosorok/mnben_elv_net_lending/mnben_0601_osszefoglalo_ informaciok.
form of subscribed or base capital, a figure that often represents only a
small fraction of the balance sheet of these enterprises. Various studies
by the State Audit Office have estimated that the guarantees given by
municipalities not recorded as explicit debt anywhere have an estimated
value equal to 7 percent of total liabilities, or about 70 billion forint at
the end of 2009.29 Thus, total municipal debt is larger when including
guarantees of enterprise debt that do not show up in the financial and
budget reports filed with the State Treasury.
The borrowing began to increase in 2000 in anticipation of EU
accession in 2004. That borrowing was used to finance environmental
and other infrastructure projects, while the bond issuance boom that
started in 2006 was a function of the availability of structural funds,
inexpensive lending in Swiss francs and euros, and a strong forint, combined with the realization that all bond issuances would be exempt from
public tendering.30
Hungary: Subnational Insolvency Framework
Exposure to currency risk. Figure 7.1 shows a nearly 19 percent
increase in bond debt stock in the first nine months of 2010. Over
86 percent of that approximately 90-billion-forint (US$50-million)
increase in bond debt stock was due to “revaluation.” During that
period, the Swiss franc gained 19 percent and the euro appreciated
23 percent against the forint. Since 80 percent of bonds issued are
denominated in Swiss francs and euros, the bond debt stock of
Hungarian municipalities increased by 19 percent in nine months.
This highlights the exposure of the municipal sector to currency risks
in debt service payments in late 2010 and early 2011, as the bonds
with three-to-five-year grace periods issued during the 2007–08 boom
started to incur principal payments. This annual capital repayment
was estimated at 50 billion forint (US$250 million) on an aggregate
basis. Again, the microlevel impact of capital repayment at Swiss franc
exchange rates that were nearly 50 percent higher in 2010 than they
were in 2007 could be devastating.31
A common explanation for the boom in bonds from 2006 to 2008
is that interest rates offered on these Swiss-franc-denominated bonds
were about a third of the forint rate available at the time. The forint
was appreciating during 2005–08 for the most part against the euro
and the franc; thus, municipalities played arbitrage with higher forint
deposit rates and low Swiss franc interest rates on their debt. Another
factor was the ease with which competitive tendering of bonds could
be avoided, leading to fierce negotiations over basis points. Banks could
not offer ever lower Swiss-franc interest rates, so they began to negotiate
the terms of the bonds, for example, extending the maturities to 20–25
years, and adding three-, four-, and five-year grace periods on principal payments. The market rumors that the government would move to
restrict municipal borrowing caused a rush to accumulate borrowed
cash and to invest it at temporarily favorable rates, that is, arbitrage.32
During this same period, municipalities accumulated cash and
deposits that were equal to 40–60 percent of their total long-term debt.
Bond proceeds undergo less routine monitoring than do bank loans.
The cash earned interest in foreign currencies, while the grace period
on the bonds gave a false sense of liquidity. Restrictions on municipal borrowing were too risky politically to be enacted in the annual
budget.
277
Until Debt Do Us Part
The accumulation of cash, which almost doubled between the end
of 2005 and 2009, is not justified by other uses such as payment of
short-term loans, vendor loans, or other accounts payable, as figure 7.2
indicates. This means borrowed money was in part deposited in the
banking sector for use in later periods. Taking into account changes in
fixed assets, due to both sales and the completion of new projects and
the revaluation of existing assets, a portion of borrowing has certainly
been “consumed” for operational expenses, and not invested. The figures in this section are all aggregate, and include the debt of Budapest
the capital and its 23 separate districts; thus, these trends will apply in
differing degrees to each type and size of municipality. Figure 7.2 shows
liquid financial assets of about 600 billion forint in June 2010. Figure
7.1 showed all forms of debt, including short-term debt, at over 1,400
billion forint. When liquid financial assets are dropping, the amount
Figure 7.2 Cash and Liquid Financial Assets, 2004–11
600
500
400
Forint, billions
300
200
100
11
20
10
20
20
09
20
08
07
20
20
06
05
0
20
278
Year
Other financial assets
Bonds held
Cash and deposits
Source: National Bank financial statistics published quarterly, http://english.mnb.hu/Statisztika/data-andinformation/mnben_statisztikai_idosorok/mnben_elv_net_lending/mnben_0601_osszefoglalo_informaciok.
Hungary: Subnational Insolvency Framework
available for anticipated rises in annual debt service becomes scarcer,
perhaps eventually inducing an insolvency situation at the macro level.
These systemic risks, however, have so far not led to a bankruptcy
procedure. Cash on deposit may be used to fund current deficits, to pay
down debt, to make investments in fixed assets, or a combination of
these. Thus, cash on deposit is a stock of money that may or may not
be used. The operational deficit or surplus thus is entered only when
the accounts are closed. During a budget year, cash on hand from previous years can be used for either current or capital spending. That cash
stock may change due to operational deficits, surpluses, or new borrowing that is deposited and not used within a current budget year. Cash on
hand can thus cushion crisis years and finance current deficits, and thus
can have both positive and negative effects.
Subnational Insolvency Framework
Why a municipal insolvency act? During the 1994–95 debate on the
creation of the debt adjustment law, policy makers in the Finance and
Interior Ministries and experts concluded that if the state is not willing
to administratively or legislatively control municipal borrowing, then a
legal framework would need to serve the following policy goals:
•
•
•
•
•
•
•
Prevent and preempt municipal defaults with respect to any lender,
bondholder, or vendor
Provide clear administrative and legal procedures for affected
creditors
Provide reorganization and workout procedures
Make clear that the national government will not guarantee local
borrowings or provide bailouts33
Maintain essential public services
Allow for expansion of borrowing as local taxes and revenues
increase
Discourage irresponsible borrowing and lending though municipal
debt adjustment, combined with preemptive reorganization, budget
cutbacks, and some emergency funding from the state.
The Municipal Debt Adjustment Law passed with 82 percent support
in the Parliament, and went into effect in April 1996.
279
280
Until Debt Do Us Part
Subnational Insolvency: Procedural Steps
Procedural phases in the Municipal Debt Adjustment Law. The Hungarian Municipal Debt Adjustment Law specifies a procedure consisting of
seven major phases:
1.
2.
3.
4.
5.
Initiation of debt adjustment procedure
Court review of the petition
Creating a debt adjustment committee
Adoption of budget developed for financial crisis
Formulating the financial reorganization plan and the proposed
agreement
6. Debt agreement negotiations
7. Asset liquidation if no agreement is reached.
Initiation (1). If a municipality does not pay an acknowledged obligation to a creditor, vendor, or other party, documented either as an
invoice or a court order to pay, within 60 days of the due date, the mayor
is obligated to notify the city council and to petition the County Court34
within eight days. If the court agrees that the municipality is truly in
a crisis situation and cannot meet its obligations, it declares the debt
adjustment process’s initiation, and certain obligations are then imposed
upon the mayor and the city council, and a separate set of actions is
required of the creditors. A creditor may also petition the court if a municipality is in default, that is, more than 60 days late in paying an obligation.35
The court may reject a debt adjustment petition if it determines that the
obligation can easily be met with existing a ssets and cash flow. The mayor
faces strict financial sanctions as a private person if a debt adjustment process is not initiated due to delays on his or her part (see figure 7.3).36
Court review of the petition (2). If the court finds that the conditions
of insolvency are met—that is, any financial obligation is late 60 or
more days—it decrees the commencement of the debt adjustment procedure or can terminate the procedure. The declaration by the court to
start the debt adjustment process is posted in the Enterprise Registry,
an official document of the court, and public notices are placed in the
appropriate newspapers and published electronically.
Creating a debt adjustment committee (3). Simultaneously with
the public notice, the court appoints a trustee (called a “receiver” in the
international context), and the municipality has eight days from the
Hungary: Subnational Insolvency Framework
Figure 7.3 The Debt Adjustment Process, Hungary
Creditor/Municipality
Petition
Court
Review corrections
Examination of
petition
Correction made
Rejection
No correction made
Appeal
Process halted
Appeal
Publication, notice of creditors,
debt adjustment starts
Debt adjustment
begins; Trustee
appointed
Appeal
Workout negotiations
No agreement
Agreement
Court orders continuation,
study asset liquidation
Procedure ends
Appeal
Publication
Court reviews plan
Orders new plan
Court approves
Appeal
Trustee paid
procedure ends
Assets liquidated
Appeal
Source: Jókay, Szepesi, and Szmetana 2000.
Appeal
Publication
281
282
Until Debt Do Us Part
ublic notification to form a crisis committee. The membership includes
p
the mayor, the notary, the chair of the financial committee (in the absence
of a financial committee, a council member), and a council member. The
trustee serves as the committee’s chair. The court appoints a trustee from
a list of qualified bankruptcy trustees and liquidators maintained by the
Interior Ministry.37 These are all corporate bankruptcy experts with public sector training and budget sector qualifications. Some work independently, but most are affiliated with liquidation firms (see box 7.2).
Box 7.2 point “h” is significant in that the trustee may void contracts
and transactions it deems to be “grossly disadvantageous” to the municipality. In other words, if closed-door negotiations between a creditor
and the debtor involve the transfer of assets, the granting of additional
mortgages, or other securities that hinder the rights of all the other
creditors in a later workout negotiation, the trustee may ask the court to
nullify transactions that took place up to a year before the formal filing.
Adoption of emergency budget (4). The municipality has 30 days
from the decree to prepare an emergency budget that services only
Box 7.2 Powers of the Trustee: Article 14 (2) of Hungary’s Municipal Debt
Adjustment Law
Under the law, the powers of the trustee are as follows:
a. Review the financial management of the local government and make a finding as
to the reasons for the insolvency.
b. Inspect all documents pertaining to local government assets.
c. Attend public and closed sessions of the local government and the committees
that have bearing on local government assets.
d. Make motions regarding debt settlement, which are to be deliberated by the representative body or the debt adjustment (also called crisis) committee as a first
priority on the agenda.
e. Initiate collection of the local government’s accounts receivable.
f. Inform the creditors, if requested, about the local government assets and the debt
settlement procedure.
g. Inform the head of the county/Budapest public administration office if the representative body does not meet its obligations as stipulated by the law.
h. Within 90 days of the decree, the trustee may, by filing a claim on behalf of the
local government at the court exercising general jurisdiction, contest contracts and
legal statements made by the local government or its budgetary organ executed
within one year before the commencement date of the debt adjustment procedure if they are grossly to the disadvantage of the municipality.
Hungary: Subnational Insolvency Framework
mandatory tasks allowed by the law. These mandatory services, defined
in sectoral laws as well, do not necessarily have to be performed by
the municipality directly through one of its budgetary agencies. The
trustee examines the legality and legitimacy of all decisions leading to
the financial crisis, and makes recommendations to the court for criminal and civil prosecution, if needed. The trustee must cosign all payments made by the municipality during this period, and all creditors
are notified in the Enterprise Gazette to file their claims. The notary
(chief administrator) prepares the draft crisis budget by law within
30 days of the commencement date of the debt settlement procedure
(box 7.3).
Box 7.3 Restrictions on a Municipality Undergoing Debt Adjustment
in Hungary
From the date of the commencement of the debt settlement onward, the local
government may not:
a. Make decisions through which it incurs additional pecuniary liabilities
b. Establish businesses
c. Acquire ownership in businesses
d. Service debts or other obligations assumed prior to the decree.
Within 30 days of the date of the commencement of the debt settlement, the mayor
shall provide to the trustee:
a. A report on financing and the proposed method of implementing mandatory or
optionally assumed local government duties and excising mandatory or optionally
assumed local government powers
b. An inventory of, and an annual report on, the local government assets prepared as
of one day before the date of the commencement of the debt settlement which
includes, with adequate justification, the following categories: registered assets, assets necessary for performing and exercising duties and powers required by law,
and assets that can be used to meet creditors’ claims
c. A draft crisis budget
d. A detailed summary of proceedings in progress at court and before state authorities and a detailed summary of execution proceedings in progress
e. Contracts regarding local government assets that were concluded within a year
before the date of the commencement of the debt settlement procedure
f. Detailed information on business organizations operating with the involvement of
the local government
g. Detailed information on the financial situation, debts, and accounts receivable of
local government institutions
h. Other requested documents that the trustee determines are needed to perform
its responsibilities.
283
284
Until Debt Do Us Part
Financial reorganization plan and draft agreement (5). The council
shall approve the crisis budget, and the debt adjustment committee
draws up a reorganization plan and drafts an agreement for a compromise with the creditors. In addition to a detailed description of the
financial situation of the local government, the reorganization plan
includes proposals regarding the use of the assets that may be involved
in the debt settlement and proposals for taking various measures to
expedite the debt settlement effort, also indicating the level of income
each of these measures yields to the local government.
Debt agreement negotiations (6). The trustee sends the reorganization
program and the compromise proposal approved by the representative
body to all the creditors, and invites them to negotiate the compromise.38
A compromise may be concluded if more than half of the creditors having extant claims at the time of the decree date consent to it, provided that
the total aggregate claims of these creditors amount to at least two-thirds
of all the reported and uncontested creditors’ claims. If the creditors were
divided into groups in the compromise proposal, then at least a majority of
the creditors in each group must consent to the compromise in bankruptcy.
Asset liquidation (7). If the debt adjustment negotiations fail to reach
agreement or the city council fails to develop a crisis budget within
60 days, the court shall continue the debt settlement procedure according to the rules of the partition of assets by the court. No appeal of such
an order is possible.39
The worst case scenario. If the debtor and creditors cannot come to an
agreement 210 days after publication of the debt adjustment decree by
the court, or more than 270 days have lapsed after nonpayment, taking
into account time needed for public auctions of assets, the court can
request the Parliament to dissolve the city council and call for new elections. At this time, criminal and civil prosecution can also take place.
The entire process, including appeals, and if needed, public auctions of
municipal assets, is set to take place within 270–300 days of the decree,
depending on whether each actor—the municipality, the court, and the
trustee—uses their full allotment of processing time.
Types of Debt and Their Treatment during Debt Adjustment
Three types of municipal debt are significant: loans (secured and
unsecured), bonds, and arrears (preferential arrears such as taxes and
Hungary: Subnational Insolvency Framework
s alaries and nonpreferential arrears). What matters in the debt adjustment process is whether the particular type of debt is preferential or
not. Although the implicit (service-obligation-based) guarantees are
not covered by the debt, once the guarantees are called, the law does
cover them.
Typically, loans taken by local governments are secured by real estate
mortgages. The pledged property may even be a movable asset, or, less
frequently, securities or business shares. The guarantee may also be
granted as a security either on a cash-flow basis or based on offering
other forms of collateral. Issuing bond securities provides an opportunity to raise funds that is simpler and easier than borrowing from a
bank. The hierarchy of preference rules governing secured claims does
not apply to the enforcement of claims based on bonds. The enforcement of claims arising from this and the analysis of legal relationships
on which they are based may also be an important issue in relation
with the debt adjustment procedure.40 On arrears, all claims in the debt
adjustment procedure will be classified as to the order of satisfaction
depending on the nature of the person making the claims and whether the
specific claim is a preferential one. Arrears such as wages and government
revenues have higher priority than other suppliers. Claims for banks
and bonds are considered equal to the claims of suppliers; the difference
may be based only on how these claims are secured previously.
Priority of claims during debt adjustment. The assets must be divided
among the creditors in the following way:
a. Regular wages and salaries, including severance
b. Claims secured with a lien, mortgage, or caution money up to the
pledged value, provided that the security was stipulated at least six
months prior to the commencement date of the debt settlement
procedure
c. The state’s claims arising from interest, payment, support p
rovided
for a compromise in bankruptcy concluded in the course of a
previous debt settlement procedure, and the amounts of reimbursable targeted support and further reimbursable central budget support. (These are claims for grants, subsidies, transfers, and so forth
that must be reimbursed for some reason by the municipality. This
also includes repayment of the loan the state gives during a settlement
285
286
Until Debt Do Us Part
of a bankruptcy that happened before. So these are amounts owed by
the municipality to the state)
d. Social security debts, taxes, and public debts that may be collected
like taxes
e. Other claims (loans, bonds, and arrears to suppliers)
f. Interest, also default penalties and fees on claims listed under “d”.
The claims of lower-priority-group creditors shall be settled after the
claims of higher-priority-group creditors are fully settled. The amount
available for settling the claims of creditors shall be divided among the
creditors in proportion to their claims.
Municipally Owned Utilities and Corporations
How are services delivered? A municipality may render mandatory and
optional services through a multitude of legal entities, such as (a) a budget agency of its own or that of another local government, including
municipal associations; (b) a municipally owned enterprise that may be
for profit or nonprofit; and (c) a private enterprise, including nonprofit
enterprises and foundations, based on a service contract (typically concessions, PPPs). Local governments may participate only in business
enterprises with limited liability. Thus, there are basically two types of
municipal companies that are suitable for managing municipal assets
or providing utility or other public services: a limited liability company
(LLC) and a company limited by shares (Ltd).
Which insolvency act applies? The 1996 Municipal Debt Adjustment
Law applies only to local governments, defined as (a) municipalities
(both villages and cities), (b) Budapest (capital) and its districts, and
(c) the local government of the counties (administrative units of the
county). A budgetary agency or administrative unit of a municipality
does not borrow on its own, and neither do municipal associations. In
these cases, there is full recourse to the municipality as the legal entity
engaged in borrowing. The insolvency cases of all business entities
(limited companies or unlimited, any kind of business association) fall
under the scope of Law XLIV of 1991 on Insolvency and Compulsory
Liquidation, regardless of whether the actual company in question is
owned partially or wholly by a subnational government or any other
private shareholders.
Hungary: Subnational Insolvency Framework
Obligation to continue providing services. In contrast with corporate
restructuring, municipal debt adjustment requires that certain municipal competences be carried out during the debt adjustment process.
Among other things, utility services (water, solid waste, wastewater)
shall also be rendered by the municipalities, even during bankruptcy of
the municipality or an enterprise providing the services. The corporate
bankruptcy act does not prescribe any kind of obligation for the liquidator to continue the business activity of the insolvent company. The
obligation to provide, or to make arrangements to provide, services falls
on the municipality, regardless of whether its own budgetary unit, its
enterprise, or a private enterprise is actually contracted to do so. Thus,
a challenge emerges: the bankruptcy of a utility, regardless of ownership,
falls under corporate law. But the service provision obligation that has been
outsourced to this separate entity now falls back on the local government.41
The obligation to provide, or provide for, a service cannot be subordinated to any other municipal obligation. In other words, the potential bankruptcy of public service provision enterprises, such as water
companies, hospitals, and district heating enterprises, does not entail
financial recourse of creditors to the budget, but, rather, means service
delivery recourse to the municipal budget.42
Other Issues
Assets created with EU Cohesion and Structural Funds, and cofunded
with matching grants from the municipality and/or central government,
face the contractual obligation of being in use for a minimum period of
time. In other words, even nonessential assets that are funded with these
EU monies must observe these “usage” rules, or else the grant recipient
must refund 120 percent of the grant amount. In the case of a municipal
insolvency, these assets may not change their use, let alone be a part of
a debt settlement. If a municipality intentionally or mistakenly changes
the use of an EU-funded asset subject to usage restrictions, then the
national government, obliged to refund the EU, could place a lien on
the municipality for 120 percent of the amount involved. The option
to enforce or not enforce this kind of lien could be a source of a soft
budget constraint and political bargaining, since the member state will
refund the EU, and further steps are at its own option.
287
288
Until Debt Do Us Part
Subnational Insolvency Framework: Implementation
Experience, 1996–2010
Database of Events43
There are 38 known formal cases of municipal debt adjustment in Hungary from 1996 to 2010 (table 7.2).44 In the table, amount of debt means
how much debt was recognized by the bankruptcy trustee and was actually a part of the voluntary settlement or forced liquidation. Of the 38
known cases, 30 have been settled. Of the 30 settled cases, 18 resulted
in a voluntary workout agreements, and in 12 cases the court had to
force liquidation of assets and debts. The cases’ debt adjustment process
reflects the distribution of municipal size and economic potential in
that there were only 5 cases filed in municipalities with populations
above 3,000, and only 2 cases filed in municipalities with populations
above 5,000, that is, Szigetvár, with a population of over 11,000 and,
recently, Esztergom, with a population of over 30,000.
Note that the categories in table 7.2 overlap, and that it is unclear
in two cases what led to the debt adjustment filing. Viewed differently,
10 cases apply to excessively expensive and underused utility projects,
mostly in the natural gas and wastewater sectors.45
Causes of Municipal Insolvency
Municipal insolvency cases often reflect several years of financial difficulties. Years before the filing, the municipality, its bank, and its suppliers attempted to work out informal arrangements to avoid the publicity
and potentially unpredictable results of a formal debt adjustment procedure. By the early 2000s, it became known based on multiple liquidations that banks could expect to lose the entire interest claims, and
on average up to 95 percent of its claimed unpaid principal in liquidation. Therefore, lenders were reluctant to file debt adjustment claims
upon reaching the 60-day deadline. In addition, the creditors (all types,
including suppliers) reportedly colluded to avoid a required municipal
filing after 90 days. The State Audit Office has identified several hundred
“latent” bankruptcies, where according to aggregate numbers and other
indicators, the municipalities would have a legal obligation to file, but
they do not, while lenders and suppliers tend to avoid filing, because
it is not in their interest to involve all others who have claims against
the debtor. “Latent” insolvency caused by systemic budgetary shortfalls,
Table 7.2 Municipal Debt Adjustment Filings, Hungary, 1996–2010
Population
Debt
(million
forint)
Date of filing
petition
Date of
exiting
bankruptcy
Reason
Result
289
1
Atkár
1,685
98
10/25/2001
8/1/2002
Debt from utility project
Workout agreement
2
Bakonszeg (I.)
1,278
152
8/22/1996
7/23/1998
Imprudent profit-seeking project,
illegalities
Liquidation
3
Bakonszeg (II.)
1,278
60
8/3/2000
9/26/2001
New claim by creditors dissatisfied
during first procedure
Liquidation
4
Bátorliget
783
79
8/22/1996
3/26/1997
Debt from utility project
Workout agreement
5
Biri
1,398
60
2/24/2009
12/10/2009
Mismanagement of school and public
catering facilities
Workout agreement
6
Boba
822
40
1/16/2008
5/22/2008
Environmental fines due to
incomplete sewerage project
Workout agreement
7
Csány
2,298
46
8/15/1996
4/3/1997
Debt from utility project
Workout agreement
8
Csepreg
3,333
89
4/15/1999
4/27/2000
Illegal VAT refund
Liquidation
9
Domaháza
1,082
22
11/20/1997
6/1998
Illegalities
Workout agreement
10
Dunafalva
1,185
69
3/13/2003
12/29/2005
Illegal public works contracts, arrears
to suppliers
Liquidation
11
Esztergom
30,928
22,600
11/25/2010
In process
Divided government, political strife,
unpaid invoices
In process
12
Egerszólát
1,107
24
8/25/1996
4/3/1997
Debt from utility project
Workout agreement
13
Felsőmocsolád
559
10
8/11/2005
8/3/2007
Miscalculation of operation and
management costs
Liquidation
14
Forró
2,547
163
4/7/2005
12/15/2005
No information available
Workout agreement
(continued next page)
Table 7.2 (continued)
15
Gilvánfa
341
26
9/21/2000
2003
16
Kács
654
32
12/12/1996
7/24/1997
17
Jásztelek
1,775
10
12/2008
–
18
Kajászó
986
85
4/3/2008
19
Magyardombegyház
259
20
Nágocs (I.)
856
21
Nágocs (II.)
856
22
Nemesgulács
23
Nemesvid
24
Neszmély
25
Nick
26
Ópályi
290
Population
Debt
(million
forint)
Date of filing
petition
Date of
exiting
bankruptcy
Reason
Result
Debt owed to utilities
Liquidation
Debt from utility project
Workout agreement
Labor rights lawsuit won by
fired teacher; court ordered
compensation payment caused
insolvency
Unknown
12/22/2008
Excessive investment in new and
upgaraded health, cultural, burial,
sports facilities
Workout agreement
9/9/2010
In process
Excessive pay for mayor; kindergarten
with only 10 children
In process
123
9/5/1996
7/23/1998
Criminal and imprudent business
activities
Workout agreement
46
9/21/2000
5/9/2002
Unmet claims resubmitted
Liquidation
1,100
118
6/21/2007
1/28/2008
Disputed final payment to contractor
Workout agreement
830
750
6/15/2010
In process
Sw F debt from sewerage system
project, Sw F currency rate rising
In process
1,444
140
7/23/2008
11/11/2009
Environmental fines for illegal waste
dump; illegal guarantee offered to
municipal nonprofit enterprises for
their borrowing
Liquidation
563
91
11/22/2007
In process
Illegal VAT refund
In process
2,983
64
11/7/2008
7/17/2009
Excessive operational expenses related
to elementary school and its
refurbishment
Workout agreement
27
Páty
28
Pilisjászfalu
4,998
400
8/15/1996
3/4/1999
Debt from utility project
Liquidation
993
300
1/7/2008
11/18/2008
Succession from larger community;
inability to finance any basic
functions
Workout agreement
29
Ráckeresztúr
3,300
1,000
3/2010
–
VAT fraud, vendor loans, criminal
investigations, Audit Office
investigaton, countersuits in play
Under appeal
30
Sáta
1,391
55
2/25/1999
8/1/2002
Debt from utility project
Liquidation
31
Sáta (II.)
1,391
90
4/14/2010
In process
Unpaid invoices due to loss of school
funding/unwilling to consolidate
with neighboring villages
In process
32
Somogyfajsz
553
86
7/29/1999
9/13/2001
Criminal activity
Liquidation
33
Somogyudvarhely
1,208
31
34
Somosköújfalu
2,234
35
Sorokpolány
825
36
Sóstófalva
3,509
37
Szigetvár
38
Tiszaderzs
3/5/1998
11/19/1998
Debt from utility project
Workout agreement
10/6/2010
In process
Utility construction
In process
11
4/1/1999
12/30/1999
Illegal VAT refund
Workout agreement
6
1/21/1999
12/30/1999
Default on loan to remodel
community center
Workout agreement
11,353
3,500
2/26/2010
10/29/2010
Complex reasons, overinvestment,
lack of funds for operations, faulty
planning, PPP projects
Workout agreement
1,357
71
1/7/2008
In process
Default on loans to build local roads
and sports arena; mayor under
criminal investigation
Liquidation process
started
Source: Author.
Note: There are five more cases with the date of filing and the date of process ending indicated in parantheses: Kisnamény (1/8/2009, 1/4/2010), Nagydobos (12/9/2010, 9/6/2011), Selyeb
(10/12/2009, 10/14/2010), Ősi (12/28/2010, in process), and Tiszavalk (3/25/2009, 2/5/2010). Sw F = Swiss franc, VAT = value-added tax, – = not available.
291
292
Until Debt Do Us Part
illustrated by the large number of deficit grant applicants, reflects structural fiscal deterioration.46 It is difficult to enforce the sanction (Article
5 of the Municipal Debt Adjustment Law), given the current state of
accounting and budgeting practices.
The compliance and enforcement aspects of the Municipal Debt
Adjustment Law have been its weakest links. The cash-based accounting system makes it difficult to detect insolvency and impose sanctions
on an involuntary basis. Based on the State Audit Office report (Vigvári
2009), the rollover of payment arrears, hidden and explicit short-term
borrowing, and contingent liabilities suggest a multitude of latent
defaults that are never uncovered, nor do they trigger events as per the
Municipal Debt Adjustment Law. There is little consequence for ignoring the mandatory filing threshold.47
Table 7.3 summarizes the causes of municipal insolvency for these
formally filed cases. These cases represent different origins of financial
problems. The three “older cases” during the five years since the 1996
Municipal Debt Adjustment Law reflect different causes of financial problems: inappropriate and unauthorized borrowing, VAT disputes, and
imprudent utility projects and guarantees given by municipalities to
private ventures. About 61 percent of cases are project-related fiscal
Table 7.3 Main Causes of Bankruptcy in Hungary
Cause of financial distress in filing documents
Number of filings
(of 38)
External hits
VAT refund clawed back by tax office
4
Environmental fines
2
Criminal cases
Illegal and criminal activities, fraudulent contracts
3
Project related
Overinvestment, imprudent borrowing, miscalculated operating
expenses and revenue
16
General liquidity and management
Procedural causes, contract disputes, labor problems, political strife,
and so on.
5
Arrears in operational expenses (schools, etc.), structural
6
Source: Based on table 7.2.
Note: Two cases have unknown causes. VAT = value-added tax.
Hungary: Subnational Insolvency Framework
risks including overinvestment, ill-planned borrowing, miscalculated
operating expenses and revenues, and arrears and contingent liabilities.
External shocks such as environmental fines and taxation disagreements
with the central government account for 17 percent. Managementrelated issues such as labor and contract issues account for 14 percent.
Four more recent cases reinforce the general conclusions drawn
from cases and highlight additional elements that lead to debt adjustment.48 The environmental fine applied in the case of Boba was massive,
equaling 25 percent of the village’s annual budget; the Environmental
Inspectorate did not apply similar fines to many other villages in Boba’s
region. Ráckeresztúr added a new element in that a vendor attempted to
exercise its rights under the Municipal Debt Adjustment Law but was
foiled through an appeals process for two years. This is the first example
of a vendor attempting to exercise its options but having to wait until
the village was forced to declare bankruptcy for other reasons.
Tiszaderzs is an example of new borrowing from a secondary financial institution once the village’s primary bank refused further loans.
The village also overcommitted itself in a sports facility project that was
built using a PPP scheme subsidized by the state-owned development
bank. The anticipated cash flow for debt service based on the planned
student enrollment did not materialize, because students were later
consolidated with another village under the government incentive for
consolidation. Tiszaderzs’ experience indicates that the budget forecast
for debt service can be impacted by central government fiscal incentives
designed for different objectives.
Finally, Szigetvár is the largest city so far to undergo bankruptcy in
Hungary. In this case, there was an interrelated web of guarantees provided by the city for its hospital, PPP contracts with its own water company,
and direct borrowing by the city that was beyond its ability to pay, combined
with political events that worked to the disadvantage of the city.
Lessons Learned and Strengthening
the Legal Framework
The legal procedure in the bankruptcy law is transparent and explicit.
Each step is described in detail so that each participant knows what
comes next and what his or her responsibility is. There were no disputes
293
294
Until Debt Do Us Part
concerning procedures in any of the cases. The bankruptcy trustee
(receiver) and court have authority and are respected. The quality
and performance of the county courts and judges have not been a
problem.49
No moral hazard. The law has clearly established that the state will
not guarantee the debt of a municipality (unless through an act of Parliament and after a fee has been paid to the National Bank at the request
of a multilateral lender), nor does the state assume responsibility for a
municipality’s borrowing after it has defaulted. Commercial guarantees
are available in the market from financial institutions, and from a guarantee company operating under the commercial code that is co-owned
by the state and all commercial banks.50
The local assemblies cooperated with the court and the trustee in each
bankruptcy procedure. No assembly was threatened with new elections
or dissolution. One source of difficulty was that some municipal assemblies hoped that the bankruptcy trustee would provide them with ideas
for financial and organizational reforms, and also make difficult decisions on their behalf. The trustees can only suggest the details of viable
reorganization plans and supervise the negotiation of workout agreements. If the parties involved could not or did not want to agree, then
the trustees have legal power to suggest workout and liquidation plans
to the court, which does impose settlements in about 40 percent of the
cases.
Vital public services were maintained in each case. Successful reorganization plans came about with the full involvement of the assembly and
the management of municipal institutions. In these cases the trustee
simply reviewed the suggestions made by the reorganization committee.
In all of the cases known to date, vital services were maintained, which
is one of the purposes of the Municipal Debt Adjustment Law.
The debt adjustment procedures gave participating municipalities a
clean slate to move forward, enabling them, in theory, to continue to
borrow for development purposes. But some municipalities that have
undergone debt adjustment may continue to operate in difficult economic conditions, and are likely to remain uncreditworthy for reasons
other than an earlier debt adjustment. Furthermore, external events
such as tax disputes could put pressure on municipal finance, triggering
insolvency procedures.
Hungary: Subnational Insolvency Framework
In a significant portion of the cases, insolvency was due to accounting
and internal regulatory shortfalls. In these cases, there were no countersignatures, no internal controls, receipts were missing, and assembly decisions were a result of incomplete or misleading information.
Nor did the local assembly decisions address all of the issues related to
borrowing.51 As a consequence, unpaid bills accumulated, if they were
recorded at all.
Informal preemptive arrangements head off formal filings. There has
been a relatively low number of formal debt adjustments. Although a
municipality may not file for bankruptcy, creditors and debtors often
negotiate in the shadow of the law. From a bank regulatory perspective,
this may be a problem if insolvency becomes more frequent, because
nonperforming assets such as loans to municipalities and bonds in
default must be accounted for under Basel II,52 with additional reserves
set aside. Transparency can be compromised if a municipality is in
default but keeps the information from bank shareholders, regulators,
and the public. From the perspective of suppliers and vendors, municipalities tend ultimately to pay their bills, although often after the payment deadline. Vendors often include a “late fee” to price in the risks of
arrears and nonpayment.
Lenders, suppliers, and other vendors have little incentive to initiate a
bankruptcy proceeding (which is their right),53 since they stand to lose
a significant portion of their claim unless they are directly involved in
operating a mandatory service. The Municipal Debt Adjustment Law
was implemented only when basic municipal functions were endangered by a lien on funds, court-ordered collection, or the execution of
a judgment against the municipality (e.g., an environmental fine or a
labor lawsuit) used as a last resort. Those vendors and suppliers that
provide mandatory services will continue to be paid during an adjustment procedure and to enjoy priority in the emergency budget. Lenders,
however, often fund nonmandatory services and have individual claims
that are larger than the individual claims of a group of vendors.54 Since
all debt (including debt still under a grace period) becomes due in a
debt adjustment procedure, the creditor side has a great incentive to
make a deal in private, to the potential detriment of others.
Once debt adjustment negotiations start, the larger creditors stand
to lose the most, since proportional reductions are nominally larger,
295
296
Until Debt Do Us Part
and penalty interest and late fees are assessed in full, while suppliers
have limited ability to add such charges, and they seem to be the ones
that hold out, bringing the liquidation process upon all the rest, and in
the end, all creditors take a hit. Larger creditors such as banks tend to
argue intensely for full payment of interest, interest penalties, and capital, knowing that their claim alone exceeds funds available to all other
classes of creditors. But the tendency is for these large creditors to insist
on their claims at the risk of moving into a court-ordered liquidation,
which in relative terms will hurt them the most.55
In most cases under debt adjustment, municipalities were already
late in payments to creditors and vendors for several years, but it was
the significant operational deficits during the period immediately
before a bankruptcy that led to the filing by municipalities. Thus, in
the months before insolvency, the amount of unpaid bills for operational expenses could substantially increase. In these situations,
the municipalities could not even stay current on their invoices for
operational expenses. The suppliers’ patience eventually runs out,
and court-ordered payment liens arrive, forcing the municipality
to ask for bankruptcy protection. Creditors may wait four to five
years before taking action, and then endure protracted lawsuits and
appeals, when they could simply exercise their rights 60 days after a
valid invoice becomes due. Yet, they do not use their powers granted
in the Municipal Debt Adjustment Law, based on the fear that in an
adjustment scenario, all claims will be reduced, especially if they pertain to “nonessential” services. No business wants to be seen as being
tough on municipalities, because that could generate bad publicity in
a competitive business environment.
There are no credit bureaus, public sources of budget data, or early
warning systems. A potential supplier or lender to a municipality has no
access to a public database of payment histories. The bank that handles the transaction account has a significant advantage in knowing the
financial management skills and payment discipline of its client. The
account management bank is hence more willing to extend overdraft
loans and renegotiate long-term loans, since it has the power to seize
funds from the municipal account directly. The rest of the banks and all
of the vendors and suppliers lack this security mechanism and instead
rely on “relationship” banking.
Hungary: Subnational Insolvency Framework
Recent Changes in the Legal Framework
Except for adding the provision for the possibility of appeals to the
law in 2001, the Municipal Debt Adjustment Law itself has not been
amended since it came into force in 1996. Recently, decrees announcing
decisions by judges regarding bankruptcy procedures have been published electronically, so the time to reconsider and perhaps withdraw a
petition has been drastically reduced to a few days.
Two changes in legislation (in 2008 and 2010) would impact the municipal fiscal accounts. The first change is the Act on the Status of Budget
Subjects (Law CV, 2008). This law states that municipal enterprises operating under the Enterprise Law that receive more than two-thirds of their
revenue from the municipal budget in any form (subsidies, contracts)
must be dissolved and their functions transferred to an existing or new
budget agency. If a municipal enterprise performs a mandatory service
delivery function but is financially dependent on municipal support, the
2008 law will make transparent the contingent liabilities of the enterprises to the municipal budget. The off-budget entities that are financially self-reliant are not likely to be compelled to perform mandatory
municipal functions. Revenues from water fees collected by a municipal
water enterprise, for example, may flow directly into a budget agency,
and the municipal water enterprise cannot use the water fees for other
nonmandatory, often profit-seeking activities. But the apparent mixing
of mandatory and optional functions in a municipal enterprise56 may
make it difficult to show that two-thirds of revenues come from the budget, because those flows will be diluted with other enterprise revenues.
The second change was in 2010 with the enactment of a new Treasury rule that intends to reduce the potential negative impact of the
central government decisions (with respect to funds allocated to the
local governments) on local government finance. The change could
be in direct reaction to the five bankruptcy cases. In these cases, the
municipal finance deteriorated after the central government seized
VAT refunds, overpaid grants, and other revenues, or after the central
government imposed fines (such as in the case of Boba). According to
the 2010 change in treasury rules, when the Treasury places a lien on or
intercepts a municipal fund, it could allow installment payments by the
municipality to avoid “adversely affecting” the performance of mandatory tasks.57
297
298
Until Debt Do Us Part
Strengthening the Insolvency Framework
Consensus is emerging among stakeholders (such as representatives of
the central government, lenders, and service providers to municipalities) on the need for strengthening monitoring, audit, oversight, disclosure, accounting standards, and budget procedures for municipal
governments. This section summarizes a common set of observations
derived from discussions in Hungary on potential reforms to strengthen
the municipal insolvency system.
There are several hundred latent cases of insolvency among municipalities, but these are not formalized into bankruptcy cases, due to a
shared interest among lenders, municipalities, and their suppliers to
settle matters informally and out of public view. In all documented
cases of debt adjustment, the debtor and some of the creditors engaged
in informal negotiations, to reschedule debt or to gain access to new
forms of security. This kind of pre-adjustment negotiation essentially is
expanded during the first formal phases of bankruptcy, when the emergency budget and workout agreement are negotiated, and the trustee
prepares an examination of the finances of the municipality.
Bilateral negotiations are an integral part of all insolvency regimes
(and are substantial instruments in corporate insolvency cases, as well).
The priorities as prescribed by insolvency laws provide the backstop for
voluntary restructuring negotiations, shaping the bargaining power of
creditors and debtor even outside bankruptcy (Liu and Waibel 2008).
However, a lack of transparency may benefit one class of creditors at
the expense of other classes. A pre-bankruptcy negotiated restructuring will need to be made transparent, with fair access to information
by all affected stakeholders. Information transparency helps the public
to access municipal financial and budget data. In the United States, the
voiding of preferred arrangements and recovery of preferred payments,
coupled with transparency, public access, and “sunshine,” have substantially reduced the problem of nontransparent prefiling negotiations (De
Angelis and Tian forthcoming).
The bankruptcy proceeding will need to clarify the treatment of the
PPP contracts: Are they debt? Do they provide mandatory services?
And what happens if they provide blended services? PPP payments are
treated as service contracts. Under a debt adjustment scenario, a type of
service must be classified as either mandatory or nonmandatory. If part
Hungary: Subnational Insolvency Framework
of the service provided by a PPP is classified as nonmandatory, then
part of the fee relating to the “nonmandatory” portion may be withheld
from payments. This may cause further insolvency for the PPP provider
as the legal entity.58 Cutting off nonmandatory payments to an operator that also conducts mandatory operations may impact the financial
health of the operator. International experience shows that PPP contracts are complex and their treatment as debt or long-term service
contracts needs to be clarified in budgeting, accounting, debt limitation,
and debt adjustment legislation (Irwin 2007). Long-term PPP contracts
affect not only the sharing of risks, including performance risk over the
long run, but funds paid to PPP contractors impinge upon cash available for other expenditure items, including debt service.
Another reform relates to the ex-ante regulation of borrowing
in the Municipal Act, in force until 2012. Restrictions on the use of
long-term borrowing to cover operational deficits (such as requiring
short-term loans to be cleared perhaps 30 days in advance of the end
of the budget year) would help prevent the rolling over of overdraft
and short-term loans into the next budget year. Better management
of refinancing risks such as currency risks would help municipalities
improve their debt portfolio. The risks of guarantees would need to
be accounted in the borrowing rules; cross-country experiences offer
valuable references.
Reforms in intergovernmental fiscal systems and other areas complement the debt restructuring framework. The insolvency law alone
cannot address the root causes of fiscal imbalance. Basic mandatory
functions are underfinanced for many municipalities, with an average
level of financing deficit ranging from 30 to 40 percent. Strengthening municipal own revenues will help improve their creditworthiness.
Reform of the budgeting and reporting system is also needed; the
current reporting formats are difficult to use and often miss critical
information. Better recording, release, and monitoring of financial
information could help prevent the accumulation of arrears and operating deficits. An early warning system via better reporting would be
beneficial. A reform change that went into effect in 2011 requires a
municipality considering borrowing or issuing a bond to conduct an
independent audit, the results of which must be disclosed to all concerned parties and the public.
299
300
Until Debt Do Us Part
An important budgetary reform has taken place—the separation of
operational and capital budgets. Starting with the 2010 budget, the State
Budget Act requires the separate listing of operational and capital revenues and expenditures, showing the balance in each account. Deficits
and the sources of financing both capital and operational deficits would
have to be shown explicitly: from cash accumulations, long-term borrowing, or an operational surplus. This is a major improvement in budget presentation. Properly implemented, it could reduce risk through
better information. It has been too easy to hide operational deficits
by simply showing a planned asset sale that covers the planned deficit,
bringing the budget plan into balance.
A more drastic improvement would involve the calculation of operational and capital balances separately, and limiting current and future
debt service by a coverage ratio. This would involve significant changes
to the budget and accounting structure, but in the long run is a more
predictive measure than a simple debt service ratio that allows debt service even if there is an operational deficit. A final step would be to add
an element of time, that is, to require the stock and flow limits to be
estimated into the future, taking projected revenues and debt payments
into account on a multiyear basis.
Conclusions
The 1990 Law on Local Government granted Hungary’s local governments independence in financial management. Municipalities had
unfettered freedom to manage their finances and started to borrow for
commercial activities, thus increasing the risks of insolvency. The macroeconomic deterioration in 1995 exposed the seriousness of subnational
financial distress. Furthermore, municipalities began to borrow long
term to finance short-term operating deficits. Several local governments
successfully lobbied for one-time grants from the central government.
This threatened to set a bailout precedent, raising concerns of adverse
incentives for both local governments and creditors.
Hungary’s Municipal Debt Adjustment Law, enacted in 1996, works
to reduce the uncertainty faced by creditors and debtors in the case
of municipal default on loans or vendor payments. The implementation experience has exceeded the original expectations of the framers
Hungary: Subnational Insolvency Framework
of the law. There have been few complaints against the law, the courts,
the trustees, or the implementation process. The legal procedure per the
law is transparent and explicit. The moral hazard of bailouts has been
minimized, essential services have been maintained, and local assemblies have cooperated with the court and the trustee in each bankruptcy
procedure. The debt adjustment procedures have given participating
municipalities a clean slate to move forward.
Its main implementation challenge—noncompliance—cannot be
directly attributed to the legislation itself. Many of the cases reviewed
here have debt and vendor payments that are over 90 days late, when
the law requires a filing to take place. Filings do not take place for many
reasons, including the lack of monitoring and sanctions, and informal
rescheduling and negotiations that could negatively affect the interests
of less-informed or smaller creditors.
Consensus is emerging among stakeholders on the direction of
strengthening the municipal insolvency system. There is a need to make
the pre-bankruptcy negotiated restructuring more transparent so that
the pre-negotiations are fairly and transparently conducted among all
creditors and the municipal debtor. An early warning fiscal monitoring system would help prevent the accumulation of operating deficits, arrears, and contingent liabilities. The treatment of PPP contracts
would have implications for the performance of mandatory and nonmandatory functions. Reforms in the municipal budget, accounting,
reporting, disclosure, and oversight systems will complement the insolvency law. The Municipal Debt Adjustment Law works, but it cannot be
used to correct the structural challenges in Hungary’s local government
system.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Inspired in part by the U.S. experience with Chapter 9 municipal bankruptcy
and with various forms of financial control and reorganization boards, Hungary’s law was written with significant input from U.S.-based consultants
301
302
Until Debt Do Us Part
supported by the U.S. Agency for International Development and private
foundations.
2. See the 1991 Law on Bankruptcy and Liquidation (Law XLIX), revised version,
in force since March 2012.
3.
The bankruptcy of municipally owned enterprises such as water utilities
or public transport companies falls under corporate bankruptcy rules. The
obligation to deliver such services is always municipal. If the firm providing
such services is wound up (i.e., forced to close), the service obligation, and
all assets and liabilities associated with it, are passed on to the legal successor,
most likely a municipal department. Assets and liabilities not directly affiliated
with the mandatory public service remain part of the corporate bankruptcy.
Contracting out of mandatory services in any form (PPPs, concessions, and
so forth) does not absolve the municipality of its obligation to deliver a list of
mandatory services that are defined specifically in several pieces of legislation.
4. These cases occurred after the last major study, which was made by Jókay et al.
(2004), and Jókay, Szepesi, and Szmetana (2004).
5. The National Bank of Hungary reports on the financial accounts of the public
sector on a quarterly basis. Municipal financial assets, such as cash and deposits,
are reported along with ownership shares in municipal companies at face value,
and these shares are not revalued in line with changes in total balance sheets or
owners’ equity.
6. Communist-era managers were able to purchase state assets at fire-sale prices,
then sold the assets to foreign investors.
7. See Kopányi et al. 2004, 15–75.
8.
During 1994–95, when the Municipal Debt Adjustment Law was being
designed, a working group that included bankers tried to lobby against the
act, since the bankers believed they were doing the public a service by making risky loans that, upon default, should naturally be paid by the central
government.
9. The existence of the Debt Adjustment Law reduced the uncertainty that troubled banks in the early part of the 1990s: what happens if there is a default?
A clear answer was given: the risk of making imprudent loans would have to
be borne by the lender. This notwithstanding, imprudent loans were made in
many cases that led to debt adjustment during 1996–2010, but fear of the debt
adjustment process did, as many bankers indicated informally, restrain them
from making too many marketing-based loans that were too risky.
10. Government Decision 1092/1995 (IX.28) transferred funds to cover a portion of the expenses of municipalities in distress due to their own fault. These
communities included Bakonszeg, Bátorliget, Nágocs, Páty, and Szerencs, all of
which (except Szerencs) eventually underwent a debt adjustment process.
11. See http://conventions.coe.int/Treaty/en/Treaties/html/122.htm.
12. Counties, the middle tier, do not have these rights, and regions, a creation for
the sole purpose of gathering statistics for EU projects and planning, have none
Hungary: Subnational Insolvency Framework
of these characteristics and are not recognized in the Constitution). At the end
of 2011, there were 3,196 municipalities in Hungary: Budapest capital, 23 districts of the capital city, 19 counties, 23 towns with a county status, 304 towns
(cities), 120 villages, and 2,706 parishes (small towns).
13. Starting in 2013, education and many other human services will be centralized,
signalling a significant drop in the share of GDP accounted for by municipal
spending.
14. The Municipal Debt Adjustment Law defined the tasks of municipalities. This
part of the law has changed more than once since, with some tasks removed
and others added.
15. Data for the share of each type of revenue in total municipal revenue are from
the “Annual Budget Report” of the Ministry of Finance.
16. Data are for 2008. Source: “Annual Budget Report,” 2008, Ministry of Finance.
17. Data for this and the following two paragraphs come from the National Bank of
Hungary, annual budget reports of the Ministry of National Economy (previously Finance), the State Debt Management Agency, Eurostat, the International
Monetary Fund’s Fiscal Outlook, and Article IV reports on Hungary.
18. Hungary’s public-debt-to-GDP ratio increased from 67 percent in 2007 to
81.4 percent in 2010, with a significant portion of the increase from a rapid
devaluation of the forint compared to the euro and the Swiss franc.
19. Between 2007 and 2011, the forint dropped by 66 percent against the Swiss
franc, by 23 percent against the euro, and by 40 percent against the dollar.
20. Hungary’s prime rate was 7.5 percent in 2007, peaked at 11.5 percent in October
2008, was lowered to 5.75 percent in 2010, and was 7 percent in June 2012. The
10-year bond rate vacillated between 7.08 and 9.75 percent, ending 2011 at
8.48 percent.
21. At prevailing year-end exchange rates.
22. Based entirely on National Bank of Hungary financial accounts statistics; http://
www.mnb.hu.
23. For 2013, these transfers are to be cut by 40 percent to reflect the centralization
of most municipal functions to central state organizations.
24. This is different from the usual 15 percent of total revenues debt service formula used in the region.
25. Private placement criteria in Hungary are derived from the 2004 EU Markets in
Financial Investments Directive (MiFid). This means that no permit for bond
placement is needed from the securities regulator, and only qualified investors
may purchase bonds (to a numerical limit of 100 qualified investors). The face
value of each bond should be no less than 50,000 euros. The offering statement
must be filed with the regulator. Regulation prior to accession in 2004 restricted
private placements to 50 “professional” investors. There are no specific municipal disclosure standards except for public offerings.
26. They may change banks for this primary account once a year, reporting the
change by October 31. Thus, September and October are intensive bank lobbying
303
304
Until Debt Do Us Part
months, when financial institutions work hard to land accounts currently managed by their competitors.
27. Author’s estimates based on reports from the Treasury and press accounts.
28. Based on the debt stock figures from the National Bank of Hungary.
29. See, for example, Vigvári 2009.
30. These reasons were common in professsional discourse, but were summarized
in the National Bank Review (September 2008), as cited at http://www.mnb
.hu/Root/Dokumentumtar/MNB/Kiadvanyok/mnbhu_mnbszemle/mnbhu_
szemle_cikkek/homolya_daniel_szigel_gabor.pdf.
31. One small village, Nemesvid, faced serious currency risks.
32. See National Bank Review, September 2008 cited in footnote 30.
33. Guarantees are provided only if required by the lender, such as by the international financial institutions, but the National Bank charges a guarantee fee, and
sovereign guarantees of municipal borrowing in such multilateral frameworks
are rare. In onlending situations, the onlending institution, a domestic bank,
lends at its own risk, and has no recourse to the sovereign, even if the funds
originally enjoy a sovereign guarantee in favor of the original source of funds.
34. In Hungary, County Courts are the court of first instance for most civil and
criminal cases, except for courts in the larger cities, such as the Budapest City
Court. Judges are assigned based on their experience and qualifications. Corporate bankruptcy and liquidation began before the system change in Hungary
in the late 1980s. By the time the Municipal Debt Adjustment Law was passed,
county judges amassed both bankruptcy and public administration experience.
There has been no documented or reported failure on the part of any court to
follow procedures or to adjudicate liquidations (where necessary) at the request
of the trustee. Apart from the usual capacity and infrastructure problems,
these courts have performed well. Since the inception of electronic publishing
of decisions, the initial steps in a debt adjustment filing have been reduced to
48–60 hours, as seen in the case of Szigetvar.
35. See the complete English translation of the Municipal Debt Adjustment Law for
a more detailed description of the procedure (see Jókay 2012).
36. Article 5 of the law states that the court may impose fines of 100,000 forint
(about US$500) on the mayor or a person acting on his or her behalf for violations of any section of the law. Not filing for debt adjustment after 90 days
or not cooperating with the crisis budget committee, the court, or the trustee
could lead to multiple penalties.
37. The trustees are selected through an open competition held every five years
(the latest occurred in 2009). The Interior Ministry requires that those who
apply already be certified corporate liquidators, or represent firms with at
least three years of liquidation experience. In addition, trustees need to have
at least two years of municipal finance or budgeting experience and to have
liability insurance of at least US$150,000. For this reason, only 23 firms qualified based on their experience as firms, not individuals. These rules are laid
Hungary: Subnational Insolvency Framework
out in Government Decree 95/1996. The latest list of firms eligible for appointment was published in the Interior Ministry’s Gazette on July 16, 2009.
38. The receiver shall forward the invitation complete with its attachments to the
creditors at least eight days prior to the meeting. Depending on the number of
creditors, creditors may be invited in separate groups to negotiate a compromise in bankruptcy.
39. The law was modified in 2001 to include more possibilities for appeals. As a
result, several cases have taken much longer than envisioned in the original law.
Forced liquidations have taken place in these cases, since there was no incentive
to reach an agreement, and assets available to cover debts were insufficient. An
impasse could only be overcome by an agreement imposed by a judge.
40. Municipal finance procurement advisor Dr. Gábor Szepesi has indicated that
about 80 percent of bond issues are not secured by any collateral at all, since
interest rates (10–20 basis points above the benchmark) and grace periods
(four to five years on a 20-year bond) became so competitive that arrangers had to agree to no collateral investing to stay attractive in competitive
negotiations.
41. There are examples of the opposite. A district heating plant operated by a
chemical factory that went bankrupt also served a residential area with hot
water and heat. The company operating the heating plant was liquidated, and
since district heating is not a mandatory function, many households went without service. The municipality of Fuzfo tried to take control of the heating assets,
but the liquidators were not obliged to hand them over and did not (based on
author’s field interview).
42. If an enterprise has majority municipal ownership, defined as more than a
50 percent share, and it is performing public service functions, special mandatory liquidation rules apply, according to the Act on the Status of Budget
Subjects (Law CV, 2008). This means that if a majority-owned municipal
enterprise that is carrying out municipal functions becomes insolvent and is
subject to corporate bankruptcy, the founding public entity is fully responsible
for providing uninterrupted and consistent service (water, solid waste, wastewater, public sanitation, and so forth) instead of the bankrupt enterprise. The
cost of providing a service will be paid for from the municipal budget), and
those assets, liabilities, and contractual obligations that are related to that mandatory service pass directly from the bankrupt entity to the owner, that is, the
municipality.
43. This database was assembled using paper and electronic court records and press
reports, since no central listing of bankruptcies of municipalities is available
from the government or private credit bureaus. Only Budapest is rated by several agencies, so this constitutes original research by Laszlo Osvath and Charles
Jókay using previous data until 2004, then new data since then.
44. There may be incomplete information on some, where the exact date and text
of the court decrees were not available. This study examined Szigetvár in detail,
305
306
Until Debt Do Us Part
including field interviews with all stakeholders, and examined the 5 new cases
using desk research and interviews when appropriate. The rest of the database,
or the first 28 to 30 cases, was examined in detail by previous studies (Jókay and
Szepesi 2003; Jókay et al. 2004; Jókay, Szepesi, and Szmetana 2004; Jókay and
Veres-Bocskay 2009), but there is no extensive literature on municipal bankruptcy or the Municipal Debt Adjustment Law itself. The electronic database of
newspaper and court documents extends back to about 2000; earlier materials
exist only in paper form. Almost all of the material used to describe the cases
and to formulate generalizations is available only in Hungarian. The general
observations and lessons apply to a composite of the cases known to date. Boba,
Ráckeresztúr, Szigetvár, and Tiszaderzs produce many of the same conclusions;
however, each new case introduced elements that were not that relevant earlier.
45. Several newer cases that are not described in detail in this study nevertheless
deserve to be briefly mentioned. Sáta, a village of 1,400, is repeating a debt
adjustment procedure eight years after having ended the previous one with
forced liquidation. The newer case was caused by a deeply divided council that
refused to consolidate the village’s primary school with that of a neighbor. Persistent unpaid bills in the school budget led the mayor to file for debt adjustment
to show their inability to finance their school under the current fiscal system.
Neszmély, a village of wineries on the Danube, turned to debt adjustment when
it could not pay a large environmental fine, and even tried to convince a local
business to lend it money in lieu of paying the local business tax. The village
management turned to unauthorized borrowing, while the initial problem was
compounded by another extension of an environmental penalty. Biri, a village
of 1,200, amassed 17 million forint in unpaid utility bills related to its school
out of a total budget of 190 million forint. The unpaid bills eventually totaled
50 million forint in unpaid current obligations. Felsó́mocsolád, a village of 600,
decided to build a sports facility but failed to complete it. It did not qualify for
the deficit grant since its problem was caused by an overextended investment.
Its annual budget was 270 million forint, and the sports facility cost 170 million forint. Although it claims that not getting the deficit grant is the reason for
the bankruptcy, in reality, the sports facility was oversized, and the operations
and maintenance expenses of the unfinished facility could not be covered by its
budget. (These cases are based on field interviews and press reports).
46. Vigvari (2009) introduced the concept of latent bankruptcies.
47. The amended Municipal Debt Adjustment Law in 2011 relieves mayors of the
obligation to file for insolvency upon reaching certain thresholds. A mayor may
only petition the court if the local assembly has approved the filing.
48. For a detailed analysis of the select cases, see Annex 2 to Jókay (2012).
49. When the Municipal Debt Adjustment Law was being designed in 1994–95,
there were concerns about whether corporate bankruptcy experts would be
competent to serve in municipal situations. This concern was overcome by
implementation experience. The chief judge assigns cases to the bankruptcy
Hungary: Subnational Insolvency Framework
trustees based on experience or qualifications. Trustees must be on a list of qualified firms with reorganization and liquidation skills in the corporate sector.
In addition, the firms must demonstrate competence in public sector finance,
accounting, and municipal governance. Twenty-three firms were recertified by
the Ministry of Finance in 2009. There is no evidence that the county courts
and their judges have faced difficulties adjudicating and supervising these cases.
The county judges assigned to these 38 cases already had 20 years of corporate
bankruptcy experience or experience in the general court system handling civil,
criminal, and administrative cases. There are no specialized bankruptcy courts
or public administration courts in Hungary.
50. Garantiqa Plc (see http://garantiqa.hu/en/local-governments) reported a perfect payment history on its 140 guaranteed municipal bonds and loans. This is
due to active management of their clients, constant monitoring, and a product
that “buffers” the debtor from the lender in a “pre-insolvency” period that is
backed by the guarantee. By the time the lender is not paid, the problem has
been solved by the guarantee company.
51. In one case, the mayor exceeded his legal authority in signing contracts and in
making verbal commitments to vendors that were not documented.
52. Hungarian banks, like all banks domiciled in the EU, are subject to the Basel
II Revised International Capital Framework on prudency, reserves, and capital
requirements (see http://www.bis.org/publ/bcbsca.htm).
53. With two exceptions, creditors and suppliers did not initiate debt adjustment
procedures against a municipality.
54. The low recovery rate is also explained by a combination of factors including overestimation by lenders of cash flow and proper collateral, financial shocks caused
by regulatory and tax events, or fiscal mismanagement by borrowers. The rapid
depreciation of Hungarian currency also contributed to the difficulty of making
debt service payments on foreign debt (see discussion on the impact of the global
financial crisis in “Structure of Subnational Governments and Their Finance” section). Assets available as collateral are used to cover the claims of all creditors in an
insolvency case. Book value, market value and potential liquidation value at distress sales are often a part. Experience suggests that banks do not make a windfall
by gaining access to book-value land; their overall claims are not being met in full.
55. Based on previous research by the author (Jókay, Szepesi, and Szmetana 2000,
2004).
56. As in the case of the Szigetvar water utility.
57. The Treasury has the authority to extend the one-year installment option to
three years. But the law does not invalidate the 60- and 90-day trigger events in
the Municipal Debt Adjustment Law.
58. In the case of Szigetvar, for example, the water company providing a mandatory service is also the PPP contractor. Withholding PPP payments for operating an instructional pool may endanger the built-in cross-subsidies the water
company needs to survive.
307
308
Until Debt Do Us Part
Bibliography
Canuto, Otaviano, and Lili Liu. 2010. “Subnational Debt Finance: Make It Sustainable.” In The Day After Tomorrow: A Handbook on the Future of Economic Policy
in the Developing World, ed. Otaviano Canuto and Marcelo Giugal, 219–38.
Washington, DC: World Bank.
Dafflon, Bernard, ed. 2002. Local Public Finance in Europe: Balancing the Budget and
Controlling Debt. Cheltenham, U.K.: Edward Elgar.
De Angelis, Michael, and Xiaowei Tian. 2013. “United States: Chapter 9 Municipal
Bankruptcy: Utilization, Avoidance, and Impact.” In Until Debt Do Us Part:
Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and Lili Liu,
311–52. Washington, DC: World Bank.
Homolya, Dániel, and Gábor Szigel. 2008. “Önkormányzati hitelezés – kockázatok
és banki viselkedés.” National Bank of Hungary Monthly Review 3 (2): 20–29.
Irwin, Timothy. 2007. Government Guarantees. Washington, DC: World Bank.
Jókay, Károly (Charles). 2012. “Hungary: Subnational Insolvency.” Economic Policy
and Debt Department, World Bank, Washington, DC.
Jókay, Károly (Charles), and Gábor Szepesi. 2003. “Municipal Debt Management
and Bankruptcy Intervention in Hungary 1995–2002: Policy Suggestions for
Russian Federation Draft Legislation.” Manuscript prepared for Institute for
Urban Economics, IGE Consulting Ltd., Moscow. http://www.ige.hu.
Jókay, Károly (Charles), Gábor Szepesi, and György Szmetana. 2000. Municipal Debt
Management and Bankruptcy Intervention in Hungary: Case Studies and Lessons Learned (1996–2000). IGE Consulting Ltd. and World Bank Subnational
Development Program, Budapest. http://www.ige.hu.
———. 2004. “Municipal Bankruptcy Framework and Debt Management Experiences 1996–2000.” In Intergovernmental Finance in Hungary: A Decade of Experience 1990–2000, ed. Kopányi Mihály, Deborah Wetzel, and Samir El Daher.
Washington, DC: World Bank Local Government and Public Service Reform
Initiative; Budapest: Open Society Institute.
Jókay, Károly (Charles), László Osváth, Gyula Sóvágó, and György Szmetana. 2004.
Az önkormányzati adósságrendezések oknyomozása 1996–2004. Unpublished.
IGE Consulting Ltd.–State Audit Office, Budapest.
Jókay, Károly (Charles), and Katalin Veres-Bocskay. 2009. “Only in Hungary: Experiences with Municipal Debt Adjustment and Suggested Regulatory Changes.”
Public Finance Quarterly 54 (1): 111–29.
Kopányi, Mihály, Samir El Daher, Deborah Wetzel, Michel Noel, and Anita Papp.
2004. “Modernizing the Subnational Government System.” In Intergovernmental Finance in Hungary: A Decade of Experience 1990–2000, ed. Kopányi Mihály,
Deborah Wetzel, and Samir El Daher, 15–75. Washington, DC: World Bank
Local Government and Public Service Reform Initiative; Budapest: Open Society Institute.
Hungary: Subnational Insolvency Framework
Liu, Lili, and Michael Waibel. 2008. “Subnational Insolvency: Cross-Country Experiences and Lessons.” Policy Research Working Paper 4496, World Bank, Washington, DC.
Vigvári, András. 2009. “Financial Risk in the Municipal Sector.” Report (in Hungarian)
published by the State Audit Office, January, http://www.asz.hu/tanulmanyok/
2009/penzugyi-kockazatok-az-onkormanyzati-rendszerben/t272.pdf 9.
———. 2010. “Is the Conflict Container Full? Problems of Fiscal Sustainability at
the Local Government Level in Hungary.” ACTA Oeconomica 60 (1): 49–77.
309
8
United States: Chapter 9 Municipal
Bankruptcy—Utilization,
Avoidance, and Impact
Michael De Angelis and
Xiaowei Tian
Introduction
As a result of the global trend of decentralization and increased subnational1 fiscal autonomy, the restructuring and discharge of subnational
debt has emerged as a critical issue. Due to the 2008–09 economic crisis and the declining fiscal conditions facing U.S. municipalities,2 and
to certain municipal financial practices related to funding of pension
obligations,3 municipal bankruptcy has become a relevant and much
discussed issue in the arena of municipal finance. The importance of the
issues and the risks associated with municipal insolvency are increasingly recognized.4
In the United States, Chapter 9 of the U.S. Bankruptcy Code5 for
municipalities6 has long been established although rarely used. However, past experience may not be an accurate predictor of the future.7
Observers speculate that future Chapter 9 filings may be driven by
municipal pension and health care liabilities. Press accounts indicate
that these liabilities may reach crisis levels for many municipalities.8
During the past two years, there been considerable media attention and
market concern about the prospect of municipalities filing for protection under the provisions of Chapter 9 of the U.S. Bankruptcy Code.9
311
312
Until Debt Do Us Part
The increased interest in Chapter 9 and its recent use or consideration in
several high-profile cases has resurrected interest in its provisions.10 This
chapter assesses the current state of Chapter 9 municipal b
ankruptcy
and its use and impact on municipalities facing severe fiscal distress.
Prior to the enactment of Chapter 9 in 1937, the only remedies available to creditors when a municipality was unable to pay the creditors
were for the creditors to pursue an action of mandamus11 and compel
the municipality to raise taxes or to seize its accounts. The general rule
is that “public property” dedicated to a public use is not subject to debt
foreclosure. In practice, very little property falls into the “proprietary”
category. This argument may also apply to funds in the public treasury
accounts to be applied to public purposes.12 These remedies were largely
ineffective and, in particular, created an environment that induced
individual creditors to race to the courthouse to file separate mandamus
suits. Creditors that might be disposed to negotiate a settlement were
dissuaded if any creditor refused to agree to a settlement and held out for
full payment, called the holdout problem. During the Great Depression,
these remedies proved ineffective.
The fundamental objective underlying the enactment of Chapter 9
is to provide a distressed municipality court protection from creditors,
while it develops and negotiates a plan for adjusting its debts in a manner that enables it to continue to provide essential services.13 A municipality, unlike a private corporation, is not created to generate profits but
to provide public services to its residents, and it has an obligation to continue to provide these services even when facing economic difficulties.
Approximately 600 municipal bankruptcy petitions have been filed
through 2011.14 Most of these filings were by small, special-purpose districts such as water and sewer districts or small rural municipalities.15
There were only 252 municipal bankruptcy filings between 1980 and
2011.16 This compares to 51,259 business filings under Chapters 7 and
11 in 2010 alone.17 A Chapter 9 filing for municipal bankruptcy by a
general purpose municipality is a relatively rare event.18 Default on debt
appears to be equally rare. A study by Moody’s Investors Services found
that only three general purpose governments rated by Moody’s had
defaulted on long-term bonds in 30 years.19,20
Municipal bankruptcies are less frequent. As a result of the lack of
judicial precedents interpreting the provisions of Chapter 9, many
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
issues relating to the application of Chapter 9 are also not fully developed. Notwithstanding the shortage of case experience, in recent years
there have been several significant cases that have enhanced the ability
to assess the potential impact of Chapter 9,21 the impact of municipalities seeking to avoid a Chapter 9 filing by negotiating with their creditors, and the impact of using the threat of a Chapter 9 filing as leverage
in such negotiations.22
This chapter is organized as follows. Section two presents an overview
of Chapter 9, with a focus on key elements of Chapter 9 that are shaped
by the unique federal structure of the United States. Section three reviews
the use of Chapter 9 and focuses on selected cases. Section four analyzes the impact of Chapter 9 and assesses the benefits and limitations of
Chapter 9. Section five concludes and points to future areas of research.
Chapter 9: An Overview
Much of the structure of Chapter 9 is shaped by two federal constitutional constraints: the Contracts Clause23 and the Tenth Amendment to
the U.S. Constitution.
The Contracts Clause prohibits the states from passing laws that
impair, that is, interfere with, existing contracts. Therefore, states cannot
pass laws that would adjust a municipality’s debt obligations, in effect
impairing the creditors’ interests in the debt obligation contracts. This
constitutional restriction does not apply to the federal government.24
The Tenth Amendment to the U.S. Constitution reserves certain
powers to the states regarding the management of their internal affairs.
Chapter 9 must balance a bankruptcy court’s power to restructure
municipal debts with the sovereignty of a state and its municipal e ntities’
ability to control their own affairs. As a result, the bankruptcy court plays
a much more limited role in Chapter 9 than in the bankruptcy proceedings of private entities.25 Although Chapter 9 contains many provisions
similar to other chapters of the Bankruptcy Code applying to private
entities, Chapter 9 is significantly different. For example:
•
Creditors cannot force an involuntary filing, submit their own Plan
for the Adjustment of Debts, move for the appointment of a trustee,
or contest the decisions of the municipality regarding its property
and revenues.26
313
314
Until Debt Do Us Part
•
•
•
•
There is no provision in the law for liquidation of the assets of a
municipality and distribution of the proceeds to creditors.
The bankruptcy court cannot impose taxes.27
The bankruptcy court generally is not as active in managing a
municipal bankruptcy case as it is in corporate reorganizations under
Chapter 11.28
A municipality must be specifically authorized by the state to file for
Chapter 9 Bankruptcy.29
In addition, state laws governing the activities and finances of
municipalities cannot be interfered with. Chapter 9 is respectful of not
interfering with a state’s control over its municipalities by reserving
to the state the power to control its municipalities and limiting the
jurisdiction and powers of the Bankruptcy Court.30
Eligibility
A municipality may only use Chapter 9 of the Bankruptcy Code31 and
only a municipality may file for relief under Chapter 9.32 In addition,
Chapter 9 requires that the municipality must33:
•
•
•
•
Be specifically authorized by state law to be a debtor
Be insolvent34
Desire to effect a plan to adjust its debts
Engage in certain prefiling efforts to work out its financial difficulties. The debtor must have reached agreement toward a plan or must
have failed to do so despite good faith negotiations, or such negotiation must be impracticable.35
The threshold for seeking bankruptcy protection is higher for a
municipality than for a private business entity filing a Chapter 11 petition. In addition, a municipal debtor is subject to fewer constraints in
its operations, and the court’s role and powers are far more limited. The
bankruptcy court cannot take over the governance of the debtor. Nor
can the court interfere with the municipality’s political or governmental powers or with its properties or revenues. The court cannot order a
reduction in expenditures, an increase in taxes, or sales of property.
Due to these limitations on the court’s jurisdiction over a municipality, some have argued that Chapter 9 may be used too easily by
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
unicipalities since they receive protection from creditors and the credm
itors are subject to debt adjustment pursuant to a Plan of Adjustment
proposed by the municipal debtor, while at the same time the court cannot substantially interfere with municipal affairs, thus creating a moral
hazard of abusing the Chapter 9 process. To counter this possibility,
Chapter 9 provides for the dismissal of any petition not filed in good
faith. This good faith requirement has been interpreted to mean that
the municipal debtor must be attempting to effect a speedy, efficient
reorganization on a feasible basis and to prevent the municipal debtor
from attempting to unreasonably deter and harass its creditors.36 Such
good faith negotiations must be wary of preferring certain creditors
over others, as in the event of bankruptcy such preferred arrangements
may be voided. The voiding of preferred arrangements and the recovery
of preferred payments, coupled with transparency, public access, and
“sunshine,” have substantially reduced the problem of nontransparent
prefiling negotiations.
The intention to counter moral hazard, or abuse of the protection,
also lies behind many other provisions: the insolvency test, for example, is designed to protect creditors and avoid abuse when less drastic
remedies are available. The potential moral hazard of a debt adjustment procedure that is too easily available by not inflicting significant
penalties on municipal affairs seems to be effectively countered by
the stringent eligibility requirements and evidenced by the low use of
Chapter 9 by municipalities.
Definition of municipality. The term municipality is defined as “a politi-
cal subdivision or public agency or instrumentality of a State.”37 The
definition is broad enough to include cities, counties, townships, school
districts, and public improvement districts. It also includes revenueproducing bodies that provide services that are paid for by users rather
than by general taxes, such as bridge authorities, highway authorities,
and water and sewer authorities.38
Although this is a broad definition that clearly includes general
purpose municipalities and special service districts, it is not without
limitation. In the Orange County bankruptcy, the court held that the
Orange County Investment Pool (OCIP) was an instrumentality of
Orange County and not of the state; therefore the Investment Pool was
315
316
Until Debt Do Us Part
not eligible to file under Chapter 9 as a municipality.39 In addition, a
recent case involving the Las Vegas Monorail Company’s40 filing for
reorganization under Chapter 11 was challenged by Ambac Assurance
Corporation, which had issued a guarantee of the Las Vegas Monorail’s
outstanding bonds.41 Ambac argued that the Las Vegas Monorail was a
“public instrumentality” of the state of Nevada and as such could only
file pursuant to Chapter 9.42 Although the interest on the Las Vegas
Monorail’s bonds was exempt from federal income taxation as a public instrumentality of the state, Ambac’s motion was denied by the
Bankruptcy Court of the District of Nevada.43 The judge argued that
although the Las Vegas Monorail Company had expressly acknowledged
itself as an “instrumentality of the state of Nevada” for obtaining the
tax exemption on its debt and that it was a company controlled by the
Governor of the state of Nevada, the term “public instrumentality” of
Chapter 9 was vague and that the Las Vegas Monorail Company did not
have sufficient municipal qualities and characteristics to be considered a
municipality within the meaning of Chapter 9.44
This case demonstrates that the determination of eligibility is not
a simple matter and may vary among states.45 For example, in a case
involving New York City’s Off-Track Betting Corporation (OTB) filing
under Chapter 9, the court found that OTB was a municipality since it
is a public benefit corporation “created by the State for the general purpose of performing functions essentially governmental in nature.”46
The eligibility determination is critical because it may be more beneficial for a municipality to have one of its special purpose entities to
proceed under Chapter 9 than it would be to proceed under Chapter 7
or Chapter 11. This is because Chapter 9 is more restrictive of creditor
rights, reflecting the need to preserve essential public services.
State authorization. A municipality must be specifically authorized by
the state to file for Chapter 9 bankruptcy.47 This requirement of state
authorization derives from the Tenth Amendment principle that the
federal government may not interfere with states’ internal governance.
Chapter 9 must respect states’ sovereignty over their political subdivisions. While Chapter 9 offers a municipal bankruptcy process, the state
authorization requirement leaves to each state the final say over whether
and which of its political subdivisions may have access to this process.48
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
A state has significant interests related to its municipalities’ filing
pursuant to Chapter 9. For example, a state may be concerned that,
among other things, the impact of such a filing would limit the access of
other municipalities in the state to the credit markets by lowering credit
ratings in the state and increasing borrowing costs of all municipalities
within the state.49 However, such state interests do not necessarily coincide with the interests of the municipality. In addition, the state may be
a creditor of the municipality. The requirement of state authorization
may not be in the best interests of a financially distressed municipality.
States have approached the authorization requirement in several
ways. In some states, there is a broad statute that grants filing authority to all municipalities. However, many states—including California,
which until recently had such broad authorization50—limit which
entities can file and under what circumstances, or require special
approval of state authorities to permit a filing.51 Twenty-three states
prohibit their municipalities from filing pursuant to Chapter 9.52 (See
table 8.1.)
Table 8.1 State Authorization of Chapter 9 Bankruptcy
Chapter 9 eligible
Chapter 9 ineligible
Alabama
Missouri
Alaska
New Mexico
Arizona
Montana
Delaware
North Dakota
Arkansas
Nebraska
Georgia
Oregon
California
New Jerseya
Hawaii
South Dakota
Colorado
New York
Indiana
Tennessee
Connecticuta
North Carolinaa
Kansas
Utah
Florida
Ohio
Maine
Vermont
Idaho
Oklahoma
Maryland
Virginia
Illinois
a
Pennsylvania
Massachusetts
West Virginia
Iowa
Rhode Islanda
Mississippi
Wisconsin
Kentucky
South Carolina
Nevada
Wyoming
Louisianaa
Texas
New Hampshire
Michigana
Washington
a
a
Minnesota
Sources: Laughlin 2005; Spiotto 2008.53
a. States that conditionally authorize municipal bankruptcy.
317
318
Until Debt Do Us Part
Insolvency. A municipality must be insolvent.54 Only municipalities
filing in Chapter 9 face a statutory requirement of a determination of
insolvency. However, because municipal assets are not subject to seizure
or liquidation, insolvency of a municipality is not determined by examining its balance sheet but rather is based on cash flow. A municipality
either must not be paying its debts when due or must be unable to pay
such debts when they become due in the future.55
Determination of a municipality’s insolvency requires a comprehensive cash flow analysis of factors including multiyear cash flows, available reserves, ability to reduce expenditures or borrow, and legal options
to postpone debt payments. The municipality is expected to continue to
operate and provide at least a minimal level of services.
A municipality’s taxing capacity also enters into the analysis of insolvency. Although a municipality need not exercise its taxing authority to
the fullest extent to be insolvent, a failure to consider any reasonable tax
increase may lead a court to conclude that the good faith requirement
(discussed under Eligibility) has not been met. In a case involving
Bridgeport, Connecticut, the court held that the city, which had chronic
financial problems, a US$16 million annual deficit, and the h
ighest
effective tax rates in the state, was not insolvent because it had not
exhausted its financing power and, therefore, could not demonstrate
that it would run out of funds in the next fiscal year.56
Commencement of Chapter 9: Automatic Stay and
Revenue Bond Preference
One of the most important and immediate advantages of a C
hapter 9
filing is the protection from legal actions that might be taken by creditors.57 The automatic stay prohibits the continuation of creditors’
lawsuits and the exercise of remedies against a debtor until the creditor
obtains relief from the stay.58 This protection provides the municipality with a period of time to deal with its financial crisis and to conduct
negotiations without having to deal with legal claims of creditors.
Different types of bonds receive different treatment in municipal
bankruptcy cases. General obligation bonds are treated as general debt
in Chapter 9 cases. During the period of the automatic stay the municipality is not required to make payments on general obligations bonds.
The obligations created by general obligation bonds are subject to negotiation and possible restructuring under the Plan of Adjustment.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
Special revenue bonds,59 by contrast, will continue to be secured and
serviced to the extent that special revenues are available after the payment of the operating expenses of the project or system from which the
revenue is derived.60,61 Such revenues may be applied to payments coming due on special revenue bonds without violating the automatic stay.62
Although general obligation debt constituting the full faith and credit
of a municipality may be generally viewed as the best credit a municipality can offer a creditor, in a Chapter 9 proceeding, debt secured by a
single, limited, special revenue, having a protected status from impairment, may have a preferred credit status.
Plan of Debt Adjustment
Chapter 9 provides the municipal debtor with a means to refinance or
reduce its debt and to obtain relief from burdensome contractual obligations, such as collective bargaining agreements. At the time a municipal
debtor files for Chapter 9,63 it must file a disclosure statement and a plan
for the adjustment of its debts. The disclosure statement and the Plan of
Adjustment are sent to the creditors for a vote. The Plan of Adjustment is
proposed by the municipal debtor and submitted to the court and must
be fair and equitable and in the best interests of the creditors.64
Executory contracts. The Plan of Adjustment may include, and the
court may approve, the assumption or rejection of executory contracts.65 The municipal debtor can assume unexpired leases and executory contracts that are beneficial and reject those that are burdensome.66
For many municipalities the financial obligations associated with labor
agreements and pensions are a substantial source of the financial distress. (For example, see the discussion of Vallejo, California, later.)
Labor agreements and pension obligations are subject to an assumption
or rejection in Chapter 9.67,68
Debt adjustment. In addition to the automatic stay, a significant benefit
of Chapter 9 is that the bankruptcy court has the power to approve the
Plan of Adjustment over the objection of creditors so long as the requisite majorities of creditors holding similar claims have approved the Plan
and so long as the Plan does not discriminate among holders of similar claims.69 In order to be confirmed, the Plan of Adjustment must be
accepted by one-half in number and two-thirds in amount of each class of
319
320
Until Debt Do Us Part
claims that is impaired under the Plan of Adjustment.70 This provision was
one of the primary motivations behind the enactment of Chapter 9.71,72
The Plan of Adjustment can impair the rights of holders of secured
and unsecured debt. A vote of a majority of each class of debtor will bind
dissenting creditors in that class. Notwithstanding a rejection by a class,
if at least one impaired class approves the plan, the court may confirm
the plan, forcing creditors to go along with a plan they have not accepted.
The bankruptcy court’s role is limited to the acceptance or rejection
of the plan. However, the court must still determine that the plan is fair
and equitable, feasible, and in the best interests of the creditors.73 Feasibility of a plan would be based on the expectation that the municipality is
capable of carrying out the plan.74 The best interests of the creditors is a
more ambiguous standard. The test has been interpreted to mean that
the plan must be better than other alternatives available to the creditors. In a Chapter 9 case, the alternative would be dismissal of the case,
leaving a chaotic situation in which every creditor must fend for itself.
An issue of some ambiguity is the extent to which the best interests test
requires a municipality to raise taxes in order to meet debt obligations.
The Supreme Court has held that the fairness of a plan cannot be evaluated without specific findings on a district’s ability to pay bonds with
tax revenues.75 Determining the point to which taxes can be effectively
raised is difficult. At some point tax increases will result in a decreasing collection rate, causing a decline in tax revenues.76 In addition, the
Plan of Adjustment must comply with state law that may be different in
each state. For example, in the recent Chapter 9 filing by Central Falls,
Rhode Island, a provision of a recent Rhode Island law providing that
bondholders are to be paid first became a contentious issue with other
creditors such as the pension funds and labor unions.77
Bankruptcy Courts: Restricted Powers
Chapter 9 is designed to recognize state sovereignty and the court’s limited power over operations of the municipal debtor78 by restricting the
power of the bankruptcy court to interfere with:
•
•
Any of the political or governmental powers of the municipality
Any of the property or revenues of the municipality
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
•
The municipality’s use or enjoyment of any income-producing property unless the municipality consents or the plan so provides.
These provisions clearly provide that the municipality’s day-to-day
activities are not subject to court approval and that the debtor may borrow money without the court’s approval.79 In addition, the court cannot
appoint a trustee (except for limited purposes80) and cannot convert the
case to a liquidation proceeding.81, 82
If the Chapter 9 proceeding fails to produce a Plan of Adjustment
acceptable to the bankruptcy court, the case will be dismissed and the
relationship between the municipality and its creditors will continue
as before the Chapter 9 filing, with whatever remedies are available to
the municipality and its creditors under state law. Dismissal of the case
without the approval of a plan puts the municipality in a difficult situation, because the municipality remains unable to pay its debts and
is now without the protection of the automatic stay. The power of the
bankruptcy court to reject the plan and force the municipal debtor and
creditors into the maelstrom and unpredictability of litigation is the
only, although substantial, leverage that the bankruptcy court has in
Chapter 9.
Use of Chapter 9
Statistics on Chapter 9 Use
There were approximately 600 municipal bankruptcy filings from 1937
to 2011.83 For bankruptcy practitioners, this number is small, and the
use of the law is often described as “rare.” For example, in 2010 there
were only 6 Chapter 9 filings compared to 56,282 business bankruptcy
filings.84
As shown in figure 8.1, from 1980 to 2011, there were 252 Chapter
9 petitions filed, or about eight filings annually. The annual number of
filings peaked in 1990 at 18, while there was only one filing in 1980, the
lowest number.
One crucial feature of Chapter 9 use is that most filings are not by
general purpose municipalities, but by municipal utilities, special purpose districts, and other types of municipalities. From 1980 to 2007,
321
Until Debt Do Us Part
Median population
Figure 8.1 Annual Chapter 9 Filings, 1980–2011
20
18
16
14
12
10
8
6
4
2
0
19
8
19 0
19 81
8
19 2
8
19 3
8
19 4
8
19 5
86
19
8
19 7
8
19 8
8
19 9
90
19
19 91
9
19 2
19 93
9
19 4
9
19 5
96
19
9
19 7
9
19 8
2099
0
20 0
20 01
2002
20 03
2004
2005
2006
20 07
2008
0
20 9
1
200
11
322
Year
Sources: 1980–2010 data are from the American Bankruptcy Institute, http://www.abiworld.org; 2011 data are
based on cases recorded at http://www.pacer.gov.
only 17.5 percent of Chapter 9 filings were from general purpose
municipalities—cities, villages, or counties—while 61.8 percent were
from utilities and special purpose districts.85 Other Chapter 9 filers
were mainly schools, public hospitals, and transportation authorities.86
Only four of the 13 Chapter 9 filings in 2011, for instance, are from general purpose municipalities, including Boise County, Idaho; the city of
Central Falls, Rhode Island; the city of Harrisburg, Pennsylvania; and
Jefferson County, Alabama.87
From 1980 to 2007, more than 60 percent of filings were concentrated
in four states: California, Colorado, Nebraska, and Texas. Nebraska had
39 Chapter 9 filings from 1980 to 2007, the highest number, followed by
Texas with 33 filings, and California and Colorado, with 22 filings each.88
Close scrutiny reveals that most general municipalities that filed for
Chapter 9 tend to be small entities. Based on cases recorded in Public
Access to Court Electronic Records (PACER), the population median is
1,305 for those that filed, and more than 75 percent had a population of
less than 10,000.89 This fact, coupled with the frequently observed state
involvement in the fiscal distress of large municipalities, may support
the hypothesis that states tend to aid big municipalities and would not
allow them to go broke.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
Five general purpose municipalities filed under Chapter 9 twice, but
refiling is rare for other types of municipalities.90 Three of the five refilings occurred in 2009, highlighting the impact of the recession on local
government finances and the potential for revisiting Chapter 9 by previous filers (see figure 8.2).91
From 1937 to April 2012, 162 of 636 of the Chapter 9 filings, or
approximately 26 percent, have been closed or dismissed without a plan
of adjustment being filed. Since 1980, 81, or approximately 31 percent,
have been dismissed or closed without a plan of reorganization of the
filings (Spiotto 2012).
Selected Chapter 9 Cases
Below are brief descriptions of important Chapter 9 cases, of recent
municipal experience with Chapter 9, and of municipalities considering
Chapter 9. They illustrate different origins of Chapter 9 filing and reflect
applications of the Chapter 9 framework.
Figure 8.2 Chapter 9 Filings by Type of Municipality, 1980–2007
Transportation,
4.4%
Other, 0.5%
Special municipal
district, 20.8%
School, education,
3.3%
Municipal
utilities, 41.0%
Hospital, health
care, 12.6%
City, village, or
county, 17.5%
Source: Spiotto 2008.
323
324
Until Debt Do Us Part
Until 2010, Orange County, California, was the largest municipal
bankruptcy in the United States. However, Jefferson County, Alabama,
which filed for bankruptcy in 2011, is now the largest to file a petition under Chapter 9. Both municipalities experienced fiscal distress
as a result of the use of certain derivative debt instruments. Vallejo,
California, is an example of a municipality financially burdened by
labor agreements and pension obligations in the face of continuing
economic decline. Harrisburg, Pennsylvania, the state capital, is experiencing financial distress as a result of a guaranty that it issued on
the debt of a local authority used to build an incinerator. Westfall,
Pennsylvania’s financial distress was the result of a one-time liability
judgment, and its Chapter 9 experience appears to have been efficient
and effective.
Orange County, California. Orange County, California, was one of the
fastest growing, richest counties in the United States and, as mentioned,
was, at the time, the largest municipality in U.S. history to file for Chapter 9 bankruptcy, in 1994.92
At the time of its bankruptcy, the county was the fifth-most-populous
county in the United States, with 2.5 million residents, had a budget that
exceeded US$3.7 billion, and employed about 18,000 people.93
As a result of the restrictions imposed by the California Constitution94 on the ability of local governments to raise local tax revenues, and
the increasing demand for high-quality public services, public officials
have been tempted to search for creative solutions to these challenges.95
The County Treasurer was in charge of the OCIP, which invested funds
of Orange County and of more than 200 other local public agencies
including 31 cities, regional transportation agencies, local school districts, local water agencies, sanitation districts, and many small local
agencies. The OCIP had assets of US$7.6 billion in 1994 that were
invested in derivative instruments and high-yield long-term bonds. In
addition, the OCIP borrowed US$2 for every US$1 on deposit, creating total liabilities of US$20.6 billion. As a result of market conditions
that devalued OCIP investments, by November 1994, the OCIP had lost
US$1.64 billion.
Awareness of the situation caused many Wall Street firms to commence legal actions to seize collateral, that is, the remaining assets of
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
the OCIP. Orange County filed for Chapter 9 in December 1994 to
avail itself of the automatic stay protection of Chapter 9. The filing did
not stop the creditors’ legal proceedings against the OCIP assets held
as collateral by banking institutions, but it froze OCIP funds, preventing withdrawals and causing severe distress for Orange County and the
local agencies that had invested their funds with it.
Both Orange County and the OCIP filed for Chapter 9 bankruptcy.
The OCIP filing was rejected by the bankruptcy court based on a determination that it was not a municipality pursuant to Chapter 9 (see the
discussion on Eligibility). Orange County initially submitted a Plan of
Adjustment (Plan A) that called for an increase of one-half percent in
the sales tax. Such an increase was subject to voter approval pursuant
to California law, in effect requiring voter approval of Plan A. Voters
overwhelmingly rejected the increase. Orange County then developed
Plan B, which was substantially based on forbearance by the local public agencies that had invested in the OCIP and their willingness to seek
reimbursement of their investment losses from the results of litigation
by Orange County against the banking institutions and other professionals involved with the OCIP.96
Plan B also provided for refinancing the outstanding county debt.
This was accomplished in June 1996 while Orange County was still in
bankruptcy, through the issuance of US$880 million in 30-year bonds
that were insured by a municipal bond insurer. This refinancing permitted Orange County to exit Chapter 9 by the end of June 1996.
Much of the impact of Plan B was felt after Orange County exited
from bankruptcy.97 This huge amount of debt for Orange County
prevented the county from borrowing for other purposes, and the transfer of certain revenue sources to the payment of the debt put substantial stress on the Orange County budget. The Orange County budget
constraints, together with an US$850 million shortage for local public
agencies that had invested in the OCIP, resulted in severe budget cutbacks by Orange County and the investor local public agencies. Many
of the local public agencies98 that were OCIP investors were deliverers
of public services, such as school districts, utilities, and health care and
other social services. Most of the resulting budget cuts were in public
protection, general government services, and community and social services. The impact fell disproportionately on the poor99 since they are
325
326
Until Debt Do Us Part
more dependent on county government programs. There were large
budget cuts in social service agencies that serve the poor and cuts in
basic infrastructure and transportation programs, and user fees for services were increased.100
The Orange County bankruptcy was precipitated by a risky investment strategy rather than a shortage of tax revenues and increased
spending. The county emerged from Chapter 9 18 months later, in
June 1996, and at that time sold US$880 million of insured bonds
needed to refinance its debts. From the perspective of the current
county treasurer, bankruptcy was beneficial; Orange County was
insolvent and bankruptcy allowed it to reduce its debt to an affordable level and begin a path to sound fiscal health. Just two years after
filing, it had access to the lending markets, and seven years after
filing it had an AA bond rating. The downside was the risk to its
reputation.101
The Orange County bankruptcy was both orderly and quick. Within
18 months, a Plan of Adjustment had been adopted that called for full
repayment of creditors’ claims (excluding lost interest and the forbearance of the shortfall to the local public agencies, which would be paid
to the extent of amounts recovered as a result of litigation against the
banking institutions and other professionals involved in the OCIP).
This probably would not have been possible without the automatic stay
on litigation and the financial relief provided by the suspension of payments to creditors during the stay.102 Chapter 9 appears to have been
sufficiently flexible to accommodate the operational needs of the county
and the interests of its creditors.103
City of Vallejo, California. Vallejo, a community of 120,000, is the larg-
est California city, by population, ever to file for Chapter 9 bankruptcy,
and the only general purpose municipality to do so in California since
2001. Vallejo’s finances have long been dominated by the costs of its
labor agreements, and its distress was caused not by a debt issue but
by a budget issue, that is, a long-term structural imbalance that was
the result of a declining economic base, decreased revenues from property and sales taxes, cuts in funds from the state, and labor contracts
that were out of line with the city’s budget realities.104 This trend was
exacerbated by the recent economic slowdown. A large part of Vallejo’s
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
fiscal problems had to do with diminishing revenue; city tax collections
plummeted from US$83 million during 2007–08 to US$65 million
during 2010–11, a result of the recession and the housing bust. Housing
values have fallen an astonishing 67 percent.105
Pension liabilities and financial obligations per labor contracts are by
far Vallejo’s largest debt. Prior to filing for Chapter 9, Vallejo had negotiated with several of its labor unions but was unable to reach an agreement. Vallejo filed for Chapter 9 bankruptcy in May 2008.106
Vallejo submitted a Plan of Adjustment it deemed feasible at the time
and sought to adjust its labor contracts. The labor unions challenged
the right of the bankruptcy court to approve a plan that abrogated their
collective bargaining agreements. The court ruled that such executory
labor contracts can be voided in a Chapter 9 proceeding.107 Since the
court decision, Vallejo has negotiated contracts with three of its four
labor unions.108
During the bankruptcy proceedings, Vallejo continued to make all
payments on its non-General Fund obligations (including water revenue bonds, tax allocation bonds, and assessment and improvement
district bonds) on time and in full. The majority of this debt, approximately US$62 million, consisted of water revenue bonds, which were
paid from the net revenues of the city’s water enterprise. Payments on
General Fund debt service, however, were paid at less than contractual
rates.
During the Chapter 9 proceedings, the city’s finances continued to
deteriorate.109 The feasibility of the original Plan of Adjustment diminished over time and municipal officials had to renegotiate further concessions from its unions.110
After spending three years and five months in Chapter 9 proceedings, the bankruptcy judge approved Vallejo’s revised five-year Plan of
Adjustment and its exit from Chapter 9 in November 2011.
Vallejo has closed fire stations; cut funding to senior centers, libraries and public works; eliminated minimum staffing requirements for
the fire department; and sought new sources of revenue. Among other
changes, city workers now contribute more to their health insurance,
pension benefits are reduced for new employees, and pension contributions by current workers are increased. Pension benefits for current
retirees were not changed.111 The Plan does not adjust debt that is
327
328
Until Debt Do Us Part
secured by designated revenue sources, such as water revenue bonds,
and it restructures the debt owed to unsecured creditors, which will
receive between 5 and 20 percent of their claims over two years.112
Unlike Orange County, Vallejo’s bankruptcy process has not been
quick, and unlike Orange County, where the distress was precipitated
by a one-time event, the financial distress of Vallejo is based on structural fiscal imbalance, which was exacerbated by the economic decline.
The Chapter 9 process does not seem to be as effective at resolving this
type of fiscal distress. The process took more than three years at a cost
of approximately US$9.5 million in legal fees.113 Despite its limited
effectiveness, bankruptcy has enabled the control of wage cost and pension liabilities, which account for more than three-quarters of Vallejo’s
General Fund spending.114 However, Vallejo continues to face fiscal
challenges.
Jefferson County, Alabama. Jefferson County, Alabama’s most populous
county, which includes Birmingham,115 filed for Chapter 9 bankruptcy
in November 2011 and has become the largest municipal bankruptcy in
U.S. history. The filing is to resolve the overindebtedness of the county’s
sewer system—a special purpose vehicle. The sewer system, since inception in 1994, has suffered a structural imbalance in revenue and expenditure. The city resorted to structured financial products to reduce debt
service obligations. However, the 2008–09 global financial crisis destabilized the market for such debt instruments.
The county began a sewer restoration and rehabilitation program
in 1994. That effort, initially estimated to cost US$1 billion, grew into
a US$3.2 billion project to rebuild and expand the system.116 Jefferson
County issued US$3.2 billion in bonds to finance the project.
The county’s bankruptcy filing represents that sewer rates in Jefferson County increased 400 percent. In an attempt to reduce debt service costs while limiting increases in tariffs, the county swapped its
long-term fixed higher interest rate into a short-term variable rate by
entering into interest rate swap agreements. The 2008–09 financial crisis
destabilized the market for such debt instruments, resulting in increased
debt service largely as a result of financial market illiquidity.117 In 2008,
Jefferson County defaulted on its sewer debt payments, which resulted
in an acceleration of the debt.118
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
The county had also been hurt by the loss of an occupational tax that
brought in 44 percent of its discretionary revenue. The state Supreme
Court ruled the tax unconstitutional in 2011, and the county has laid off
hundreds of employees as a result.119
Unlike the city of Harrisburg, Pennsylvania (see below), where the
state of Pennsylvania moved swiftly to intervene in the city’s financial
situation, the state of Alabama has resisted providing any assistance to
Jefferson County.120
Jefferson County had been considering filing for bankruptcy pursuant to Chapter 9 for several years.121 In lieu of such filing, it reached a
forbearance agreement122 with creditors in 2009 while it negotiated
with creditors.123 The governor and a majority of council members
supported the negotiation of the debt in lieu of Chapter 9 filing because
they wanted to avoid the “stigma” of bankruptcy. However, the possibility of a Chapter 9 filing and the desire of both the county and creditors to avoid Chapter 9 was part of the dynamic of these negotiations,124
which revolved around125:
•
•
•
Writing down a significant portion of the sewer debt
Restructuring the remaining debt at fixed rates
Limiting sewer rate increases to the rate of inflation.126
However, in November 2011, the negotiations were suspended and
the county filed for Chapter 9 bankruptcy. This bankruptcy proceeding will raise several legal issues relating to Chapter 9, including but not
limited to the issue of application of special revenues, pledges, and statutory liens, which have real significance to the municipal market.
Harrisburg, Pennsylvania. Debt issued by a special purpose vehicle
for an incinerator plant was guaranteed by the city of Harrisburg,
Pennsylvania, the capital of the state. Projections for the construction
and operation of the plant were not met and forecasts of the revenues
that would be generated were overly optimistic. As a result, the special
purpose vehicle defaulted on its debt, and the guaranty of Harrisburg
was activated. In 2010, Harrisburg owed US$68 million in interest payments, US$3 million more than its entire annual budget.127
Harrisburg sought forbearance by its principal creditor128 for time
to negotiate a settlement.129 The mayor resisted filing for Chapter 9;
329
330
Until Debt Do Us Part
owever, the governor has vowed that the state will not bail out the city,
h
and the city controller considered Chapter 9 bankruptcy the city’s best
option.130 The option of Chapter 9 bankruptcy was part of the dynamic
of the negotiations with creditors.
Notwithstanding the negotiation efforts, on October 11, 2011, the
Harrisburg city council authorized the filing for Chapter 9 bankruptcy
amidst discord among state officials, the city council, and the mayor. In
November 2011, the bankruptcy filing was dismissed by the court as not
having been properly authorized by Harrisburg.131 The dismissal leaves
the state to move forward on its takeover of the city’s finances. The state
governor has asked a state judge to appoint a receiver for the city pursuant to state intervention procedures for municipalities in fiscal distress.132
Westfall, Pennsylvania. Westfall, Pennsylvania, with a population of
2,400 and an annual budget of US$1.5 million, faced an unusual
US$20 million expense from a legal judgment obtained by a property
developer whose civil rights were violated. Westfall tried to negotiate
with the developer, who was willing to reduce the debt some, but not
enough for the township to be able to pay.133 In April 2009, the township learned that the developer planned to file a mandamus order to
force Westfall to make the payments. On April 10, 2009, Westfall filed
for bankruptcy.134
The Plan of Adjustment submitted by Westfall and approved by the
bankruptcy court reduced the claim to US$6 million to be paid over 20
years with no interest. To pay the US$6 million legal settlement owed to
the housing developer, township officials increased property tax rates
on the community’s residents by 48 percent—a rate that will drop gradually over the 20-year repayment period. Westfall’s attorney believes
that the developer agreed with the plan because the judge might have
crammed down a less favorable plan if there was a fight in bankruptcy
court. The judge cannot cram down a plan unless at least one class of
creditors agrees to the plan. In Westfall’s case, even though it was only
one developer who was owed money, Westfall owed three other parties
smaller sums. They all agreed to the plan even though the developer initially did not. Before Westfall filed for bankruptcy, it was known that at
least one class of creditor would likely go along with the plan.135
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
The financial distress precipitated by a one-time event was effectively
dealt with by the Chapter 9 proceeding.
Prichard, Alabama. Prichard, which is located outside of Mobile, has a
population of 25,000—half the population it had 50 years ago. It is a classic case of a dying city, owing to, among other things, the closure of a
military base, the shift in business and commerce to Mobile suburbs, and
declining property values. Only the poorest citizens in the Mobile area
live in Prichard, which has created challenging social problems. Housing
infrastructure and law enforcement became serious problems.136
In October 1999, Prichard filed for Chapter 9 bankruptcy when
it was unable to pay US$3.9 million in delinquent bills. In addition,
Prichard admitted that it had not made payments into its employees’
pension fund for years and had withheld taxes from employees’ paychecks, but had not submitted the withholdings to the state and federal
governments.
In the years following the bankruptcy filing, Prichard made some
progress enhancing social, financial, and technological growth, as
well as economic development. Its 2001 budget predicted a 4 percent
increase in revenue over its 2000 budget, and the city exited from bankruptcy in 2001.
Although Prichard had some success in revising its budget, so that it
no longer operated at a deficit, it was not able to meet its pension obligations. Prichard filed for Chapter 9 bankruptcy for the second time on
October 27, 2009, eight years after exiting the previous Chapter 9 filing. In
its filing, Prichard claimed a US$600,000 deficit in the prior fiscal year’s
US$10.7 million budget. In addition, it owed a US$16.5 million payment
to its pension fund under the earlier Chapter 9 settlement. Prichard was
being sued by its pensioners for failure to make pension payments for six
months, and filed for Chapter 9 to “stay” those proceedings.137
On August 31, 2010, the bankruptcy court rejected Prichard’s filing
for Chapter 9 protection on a technical interpretation of the requirement for Alabama’s consent for municipalities to file for Chapter 9. The
court ruled that only municipalities with bonded debt may file. Prichard
does not have any outstanding bonds. Prichard has filed an appeal of
this decision.138
331
332
Until Debt Do Us Part
Impact of Chapter 9
Although Chapter 9 continues to be rarely used by municipalities,139
there has been a marked increase in both interest in Chapter 9 by financially distressed municipalities and concern by creditors and rating
agencies about municipalities filing for Chapter 9.140 A number of
factors may contribute to the scarcity of cases. They include:
•
•
•
•
•
•
The threshold requirements for Chapter 9 eligibility are substantial,
including the prohibition and limitations by states for a municipality
to file for Chapter 9.141
Municipalities are not exposed to some risks that lead private creditors to seek bankruptcy protection; for example, their assets are not
subject to seizure.
Municipalities are concerned about the stigma effect of bankruptcy
on their ability to borrow and the cost of such borrowing, and the
public perception of the municipality.
Municipal officials may be wary of the political stigma of a bankruptcy filing, that is, constituents may link the bankruptcy to officials’ policies and behaviors.142
State intervention programs exist in some states, which could be
effective in the sense that states could force tough fiscal adjustmenttax increases and service cuts that cannot be imposed by the court in
a Chapter 9 proceeding.143
The process is expensive.
In addition, it is apparent that the availability of Chapter 9 to municipal debtors has an impact on the dynamic of forbearance by, and negotiations with, creditors.144 Chapter 9 may have a substantial impact in its
avoidance, even if rarely used. Even when used, it is clear that Chapter 9
is perceived as a last resort to deal with a municipality’s financial distress after all other options have been explored, including available state
remediation.
An analysis of Chapter 9 must recognize the following basic, unique
principles not common to other chapters of the Bankruptcy Code that
put the municipal debtor in an advantageous position:
•
•
Municipalities are not subject to liquidation or strict judicial control.
The Plan of Adjustment is proposed by the municipal authority.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
•
The municipal authority does not need judicial permission to exercise governmental functions.
Pros and Cons of Chapter 9 Bankruptcy
Fiscally distressed municipalities may turn to a number of options short
of default or bankruptcy to put their fiscal house in order. These include
(a) cutting expenditures, (b) raising taxes, (c) postponing payment of
obligations, (d) drawing down reserves, (e) renegotiating debt obligations to reduce or defer payments, and (f) borrowing from government
entities or commercial lenders.145
However, Chapter 9 bankruptcy may benefit a municipal debtor in
several ways:
•
•
•
•
•
•
•
It provides immediate relief by “staying” the municipality’s obligation to make payments on debt other than special revenue bonds;
that is, it stops the run on municipal funds.
It provides immediate relief from legal actions being pursued by
creditors.
It provides a means of obtaining long-term relief, including reduction in debt and other obligations, which will bind a dissenting
minority if a majority of creditors consent.
It may protect a municipality and its residents from untenable levels
of taxation by blocking creditor lawsuits from seeking to force officials to raise taxes to support debt service.
Since postfiling borrowing to support a municipality’s operations
is given a higher priority than prefiling borrowings, it may in some
cases facilitate new borrowing.
It provides the ability to renegotiate contract agreements and pension plans.
It provides a municipal debtor with a single forum in which to consolidate and address each of its various issues under the expert supervision of a bankruptcy judge.
A Chapter 9 filing also comes with potentially significant costs including costs associated with retaining legal and financial professionals to
administer the case, complying with court requirements, and negotiating
with creditors. Moreover, any municipality engaged in a Chapter 9
proceeding faces the unpredictability innate in legal proceedings. This
333
334
Until Debt Do Us Part
unpredictability may be a substantial factor in Chapter 9, a result of the
uncertainty owing to limited case law relating to the interpretation of its
provisions.
One of the most cited reasons to avoid Chapter 9 has been the alleged
“stigma” of bankruptcy and the need of a municipality to have access
to the credit markets that would arguably be limited, or available at an
increased cost, by the stigma of bankruptcy.146 Access to credit is a serious issue for a municipality faced with major infrastructure needs. It
affects not just creditworthiness but the perception of life in the city and
the economic vitality of the city for years to come.147
However, distressed municipalities have been able to gradually return
to the credit markets. For example, New York City returned to the credit
markets six years after its fiscal crisis, and Cleveland returned five years
after its 1978 default.148 Orange County was able to access the credit
markets almost simultaneously with its exit from Chapter 9, 18 months
after filing for Chapter 9.149
This experience raises the question of whether the stigma of
bankruptcy is exaggerated by creditor interests fearing debt adjustment or loss of control over the debt adjustment process. Is it the
bankruptcy procedure more than the fiscal distress that may increase
future borrowing costs? That is, is the impact of Chapter 9 worse
than the impact of default? If and when Jefferson County determines
to return to the credit markets, will it be treated less favorably as a
result of a Chapter 9 filing than as a result of its default and negotiated debt adjustment with its creditors? A Chapter 9 filing is not the
cause of the fiscal problem but the result of not being able to resolve
them any other way. Orange County’s experience may indicate that a
municipality’s putting its financial house in order is more important
to accessing credit markets than the process used to achieve financial
well-being.150
Can Chapter 9 Save Fiscally Stressed Municipalities?
If the primary objective of a financial distress mechanism is to provide
a process to develop a solution to the financial difficulties of a municipality that can be sustained over time, the effectiveness of Chapter 9 may
depend on the underlying causes of the financial distress. The cases seem
to indicate that many of the Chapter 9 filings are by municipalities that
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
have experienced one-time events, for example, Orange County’s use of
strategic investments, and Westfall Township’s liability for a legal judgment to a property developer. The Chapter 9 process seems to have been
effective in these cases by providing a mechanism for debt adjustment
and protection from legal proceedings. These municipalities have exited
from Chapter 9. Orange County has since accessed the credit markets
and currently enjoys an AA credit rating.151
In contrast, Vallejo’s financial distress is the result of systemic budget
distress, and notwithstanding concessions made by some of its creditors,
it remained in Chapter 9 for more than three years as its fiscal condition
continued to deteriorate and it incurred substantial administrative and
legal costs.152
In addition, there is some evidence that the municipalities that have
filed more than once for Chapter 9 did so as a result of systemic budget problems. For example, the city of Mack’s Creek, Missouri, filed for
Chapter 9 in 1998, then for a second time in 2000, and contemplated
bankruptcy again in 2004.153 The city of Prichard, Alabama, filed for
Chapter 9 in 1999, exited from Chapter 9 in 2007, and filed for Chapter 9
again in 2009 (see section on Prichard). Without addressing the cities’
core problems, the Chapter 9 process seems to have little impact on
reversing the structural fiscal decline without debtors undertaking sustained fiscal consolidation.
Many of the potential remedies for systemic fiscal distress relate to the
political and governmental management of municipalities that a court in
Chapter 9 procedures is restricted from interfering with. Chapter 9 procedures do not operate in such a manner as to be able to force reform,
facilitate reorganization, impose taxes, cut expenditures, or enable other
interventions that may interfere with state sovereignty. The role of state
intervention procedures and the active participation of market players
may have more authority to impose such changes than Chapter 9.154
Fiscal stress related to a one-time problem appears to be more susceptible to resolution through the debt adjustment procedures of Chapter 9.
Fiscal stress related to ongoing structural deficits is more difficult since
Chapter 9 has limited impact on solving the underlying structural problems. Although Chapter 9 can facilitate fiscal adjustment, it lacks the
authority to compel budgetary decisions that are under the purview of
the executive and legislature.
335
336
Until Debt Do Us Part
Conclusion
The design of the Chapter 9 legal structure is specific to the U.S. legal
system and is largely determined by the need to comply with the Tenth
Amendment to the U.S. Constitution. However, the issues and objectives of a legal framework to resolve financial distress are common
across many countries, that is, the public nature of municipalities, the
interest in the functioning of local government autonomy, safeguarding
essential public services and the assets that provide such services, transparent procedures, the interests of creditors, and functioning subsovereign capital markets. Strategic default by municipalities is a potential
risk. The effective design of the insolvency procedure can deter strategic
default but also allow a debt adjustment with less risk for moral hazard.
The issues of maintaining essential services and assets and limited interference with the authority of democratically elected local officials must
be dealt with in any public entity insolvency procedure. This represents
a delicate balance of interests. The economic reality is that if creditors
are not treated fairly in an insolvency proceeding, they may severely
limit their lending to the municipal sector.
In addition, the U.S. Chapter 9 system is based on a respected, independent, and competent judiciary that has the authority to reject a
municipality’s Plan of Adjustment. This role of the judiciary in many
countries may not be appropriate given the development of a country’s
judiciary. Other jurisdictions have relied on more administrative procedures or a combination of administrative and judicial procedures.155
In the municipal sector, bankruptcy is considered a remedy of last
resort. However, when all other options have been exercised and have
failed, it is useful to have access to this process. Municipal bankruptcy
is not a perfect solution for a municipality’s fiscal problems, but it can
provide breathing room while other long-term options are pursued, and
can provide the important element of debt adjustment. Municipalities
must continue functioning, and temporary or partial relief from debt
obligations can make a difference, particularly when the cause of the
financial distress is a one-time event.
Chapter 9 appears to be less effective in providing a solution to
municipalities facing long-term, endemic problems involving erosion of
the tax base, loss of manufacturing jobs, and a decaying infrastructure,
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
all of which will require substantial funding and significant structural
changes that go beyond the scope of Chapter 9.
Notwithstanding this limitation, insolvency proceedings and debt
adjustment are legitimate tools in a regulatory framework of subnational debt management and should be considered by municipalities
experiencing financial distress. Limitations and implications must be
carefully evaluated, notwithstanding the advantages of suspending legal
actions by creditors, debt adjustment, reducing the holdout problem,
and access to new financing. An insolvency system such as Chapter 9 is
an important part of a regulatory framework of subnational financial
management that strengthens ex-ante borrowing regulation. As shown
by Liu and Waibel (2009), ex-ante rules for debt procedures are not sufficient without an ex-post insolvency mechanism that manages efficient
debt workout and facilitates fiscal adjustment.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank,
its Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Subnational refers to all tiers of governments and public entities below the federal government or central government. This chapter focuses on insolvency of
municipalities in the United States. In broader terms, municipal bonds in the
United States include bonds issued by states, local governments, and special
purpose vehicles of states or local governments. For the purpose of Chapter 9,
subnational governments applies only to local governments that are political
subdivisions of states. Therefore, this chapter refers solely to municipalities,
that is, local governments.
2. Moody’s U.S. Public Finance, Moody’s Assigns Negative Outlook to U.S. Local
Government Sector, 2009.
3. A substantial problem facing municipalities today is the shortfall in public
pension funds—estimated to be between US$1 trillion and nearly US$4 trillion nationwide. In California alone, the shortfall could be as high as US$500
billion. Howard Bornstein, Stan Markuze, Cameron Percy, Lisha Wang, and
Moritz Zander, “Going for Broke: Reforming California’s Public Employee
Pension Systems,” SIEPR Policy Brief, April 2010, at 2.
4. Council of Europe Recommendation No. 96 (3) “the consequences of financial difficulties among local authorities should be made clear, for example in a
municipal bankruptcy code.”
337
338
Until Debt Do Us Part
5. Municipal Bankruptcy is covered by Chapter 9 of the United States Bankruptcy
Code (11 U.S.C. Sec. 901 et seq.).
6. See discussion below as to the definition of municipalities that are eligible to
file pursuant to Chapter 9.
7. Robin Jeweler, Municipal Reorganization: Chapter 9 of the U.S. Bankruptcy
Code, Congressional Research Service, March 8, 2007.
8. See, for example, Gary Kaplan and Joel Moss, “Distressed Cities See No Clear
Path: Health, Pension Obligations Threaten Fiscal Crisis,” National Law Journal,
March 6, 2006, at S1; Mary Williams Walsh, “Once Safe, Public Pensions Are Now
Facing Cuts,” New York Times, Nov. 6, 2006 at A1; and “Paying Health Care From
Pensions Proves Costly,” New York Times, Dec. 19, 2006 at A1. Estimates of public
pension liabilities for states and local governments range from US$1 trillion to
US$3 trillion. “The $2 Trillion Hole,” by Jonathan Laing, Barrons, March 15, 2010.
9.
Municipal Bankruptcy—A Story in Search of a Trend?, by Chris Hoene, March
13, 2010; The Official Blog of the National League of Cities; “The Perils of Considering Municipal Bankruptcy, Special Report,” January 27, 2010, Fitch Ratings;
“Muni Threat: Cities Weigh Chapter 9,” Wall Street Journal, February 18, 2010.
10. Recent filings by Central Falls, Rhode Island; Harrisburg, Pennsylvania; and
Jefferson County, Alabama and have raised the profile of Chapter 9 as an
option for municipalities in fiscal distress. See Paul Maco et al. (2011). International legal scholars have suggested that the principles of Chapter 9 may be
appropriate for sovereign bankruptcies. See Raffer, “Internationalizing U.S.
Municipal Insolvency; A Fair, Equitable and Efficient Way to Overcome Debt
Overhang,” 6 Chicago Journal of International Law 361 (2005).
11. A mandamus is a court order obliging public officials to take a certain course of
action. For a description of mandamus, see McConnell, Michael W. and Randall C. Picker, “When Cities Go Broke: A Conceptual Introduction to Municipal Bankruptcy,” University of Chicago Law Review Vol. 60, No. 2 (Spring, 1993)
pp. 425–495.
12. The general rule is that “public property” dedicated to a public use is not subject to debt foreclosure. In practice, very little property falls into the “proprietary” category. This argument may also apply to funds in the public treasury
accounts to be applied to public purposes. See McConnell, supra, note 11
p. 431, 433, 444.
13. House Report No. 95-595, 95th Congress, 1st Session 263 (1977), U.S. Code
Cong. & Admin. News 1978 pp. 5787, 6221.
14. Authors’ estimation based on “Bankruptcy Basics” (2006) by Administrative
Office of the United States Courts and the American Bankruptcy Institute.
2007 U.S. Census. http://www.abiworld.org/; and Public Access to Court Electronic Records, http://www.pacer.gov. 1980–2010 data are from the American
Bankruptcy Institute, http://www.abiworld.org; 2011 data are based on cases
recorded at http://www.pacer.gov. For statistics covering up to April 2012,
see Spiotto (2012). According to the 2007 U.S. Census, there are 89,476 local
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
governments, which include those that are prohibited by state law from filing
under Chapter 9.
15. See figure 8.2. Special Purpose Districts are independent governmental units
that exist separately from general purpose local governments such as county
and municipal governments for a specified purpose such as airports, cemeteries, conservation, electric power, fire protection, gas utility, highways, hospitals, irrigation, libraries, mass transit, parking facilities, parks, sewerage, solid
waste, stadiums, water ports, and water supply.
16. Data for 1980–2010 are from the American Bankruptcy Institute; data for 2011
are based on cases recorded at http://www.pacer.gov.
17. United States Courts, http://www.uscourts.gov.
18. “Muni Threat: Cities Weigh Chapter 9; Wall Street Journal, February 18, 2010.
See McConnell and Picker, “When Cities Go Broke: A Conceptual Introduction
to Municipal Bankruptcy”, University of Chicago Law Review 60(2)(Spring):
425–95 (1993). A general purpose municipality is an administrative subdivision governing general municipal functions, as opposed to a special-purpose
district which has a defined and limited purpose.
19. “Municipal Bankruptcy in Perspective, A Joint Report from the Bureau of
Governmental Research and the Public Affairs Research Council of Louisiana,”
April 2006.
20. In 1988, a study by Enhance Reinsurance Co. looked at historical patterns of
municipal defaults from the 1800s to the 1980s and concluded that municipal
defaults usually follow downswings in business cycles and are also more likely
to occur in high-growth areas that borrow heavily. Following the 1873 Depression, when more than 24 percent of the outstanding municipal debt was in
default, the greatest number of defaults occurred in the South, the fastest
growing region at the time. Factors that caused defaults included fluctuating
regional land values, commodity booms and busts, cost overruns and financial
mismanagement, unrealistic projections of the future, and private-purpose
borrowing. The report also said that since World War II, revenue bonds have
been a new source of default, largely because actual revenues were less than
projected revenues.
21. For example, Orange County, California, which at the time of its Chapter 9 filing was the largest municipal bankruptcy in U.S. history.
22. “Alabama County Brainstorms,” Wall Street Journal, July 2, 2010.
23. U.S. Constitution Article 1, 8, 10.
24. State programs that deal with municipalities experiencing financial distress,
while having many tools to affect municipal financial affairs, may not impair
the outstanding obligations to creditors without a substantial g overnmental
interest (for a review of state programs dealing with financially distressed
municipalities, see chapter 14 by Liu, Tian, and Wallis in this volume). The
Supreme Court laid out a three-part test for whether a law violates the Contracts Clause in Energy Reserves Group v. Kansas Power & Light 459 U.S. 400
339
340
Until Debt Do Us Part
(1983). First, the state regulation must substantially impair a contractual relationship. Second, the State “must have a significant and legitimate purpose
behind the regulation, such as the remedying of a broad and general social or
economic problem.” 459 U.S. at 411–13. Third, the law must be reasonable and
appropriate for its intended purpose. Only once has the alteration of a municipal bond contract been sustained by the Supreme Court. In Faitoute Co. v.
Asbury Park, 316 U.S. 502 (1942), the Court sustained a New Jersey statute
authorizing state control over insolvent municipalities. The plan involved an
exchange of securities for new bonds with an extended maturity and a lower
interest rate. In response to this decision, however, Congress amended the
bankruptcy law to proscribe state laws addressing composition of indebtedness from becoming binding on nonconsenting creditors. See 11 U.S.C. § 903.
Allowing each state to enact its own version of Chapter 9 of the Bankruptcy
Code would frustrate the constitutional mandate of uniform bankruptcy laws.
See H. Rept. 686, 94th Cong., 2d Sess. 19, reprinted in 1976 U.S. CODE CONG.
& ADM. NEWS 539, 557.
25. Section 904 of the Bankruptcy Code provides that absent the consent of the
municipality, the bankruptcy court may not interfere with (a) any political
or government power of the municipality, (b) any property or revenue of the
municipality, or (c) any income-producing property of the municipality.
26. For a thorough discussion of such limitations see David L. Dubrow, “Chapter 9
of the Bankruptcy Code: A Viable Option For Municipalities in Fiscal Crisis?,”
24 The Urban Lawyer 3, 548 (Summer 1992), p. 552.
27. “The levying of taxes is not a judicial act … it is an act of sovereignty to be
performed only by the legislature,” Merriweather, 102 U.S. at 515.
28. The functions of the bankruptcy court in Chapter 9 cases are generally limited
to approving the petition (if the debtor is eligible), confirming a plan of debt
adjustment, and ensuring implementation of the plan.
29. 11 U.S.C. § 109(c).
30. It is a result of such Tenth Amendment considerations that a Chapter 9 filing of an insolvent municipality may only be accomplished as a “voluntary”
act of the municipality, and, unlike private entities, a municipality’s creditors
may not force it into a Chapter 9 filing. In the 2011 Chapter 9 filing by Central
Falls, Rhode Island, a provision of a recent Rhode Island law providing that
bondholders are to be paid first (limiting the bankruptcy court’s authority)
became a contentious issue with other creditors such as pension funds and
labor unions. See “Pensions Chopped but Investors Paid,” Wall Street Journal,
December 20, 2011, p. C1.
31. Unlike the traditional individual, corporate, or partnership debtor that has a
largely unfettered right to choose from a variety of chapters of the Bankruptcy
Code (that is, chapters 7, 11, and 13).
32. Id.
33. Id.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
34. As defined in 11 U.S.C. § 101(32)(C).
35. 11 U.S.C. § 109(c).
36. Frederick Tung, After Orange County: Reforming California Municipal Bankruptcy Law, 53 HASTINGS L.J. 885, 907 (2002).
37. 11 U.S.C. § 101(40).
38. 11 U.S.C. Sec 101 (4). A state is not a municipality for purposes of Chapter 9.
39. In re County of Orange; Orange County Investment Pools, 183 B.R. 605
(Bankr. Ct. C.D. Cal. 1995). On May 10, 2010, Ambac appealed the decision.
40. Created under the nonprofit corporation law of the state of Nevada.
41. Ambac’s liability for such bonds was estimated to be US$1.16 billion; Reuters,
January 14, 2010. Ambac believed its position as a creditor would be stronger
in a proceeding pursuant to Chapter 9 rather than Chapter 11.
42.
Municipalities as defined in Chapter 9 are ineligible from filing under
Chapter 11.
43. Decision of the District of Nevada Bankruptcy Court, April 27, 2010.
44. Id.
45. In addition, instrumentality will be determined by each state’s laws and can
produce varying results.
46. In re New York Off-Track Betting Corp. (NYC OTB). March 22, 2010, the
Bankruptcy Court for the Southern District of New York.
47. 11 U.S.C. § 109(c).
48. See Tung, supra note 36.
49. The argument is that allowing one municipality to file signals that the state
will not bail out other municipalities if they get into financial distress, and this
in turn raises municipal borrowing costs within the state. Michelle J. White,
“Sovereigns in Distress: Do They Need Bankruptcy?”, Brookings Papers on Economic Activity, I: 2002. In addition, the rating of the state could be negatively
impacted, as illustrated in the case of Bridgeport, Connecticut.
50. California Government Code S. 53760 (1995).
51. For example, in Connecticut, the Governor must approve a Chapter 9 filing.
In Louisiana, a Chapter 9 filing must have the prior consent of the Governor,
the Attorney General, and the State Bond Commission. Pennsylvania liberally
grants authorization to file Chapter 9, but the effect of filing Chapter 9 automatically triggers the appointment of a state plan coordinator and subjects the
municipality to state procedures that act concurrently with federal bankruptcy
law. 53 P.S. 11701.261-1170.263 (1995).
52. Source: See note 53. Twenty-two states are silent on the issue, and this silence
cannot meet the requirement of “specifically authorize.” Georgia specifically
prohibits the Chapter 9 filings. In these states, a Chapter 9 filing needs special
legislation to be authorized.
53. While the table is based on Laughlin (2005) and Spiotto (2008), we incorporate the new development in which Rhode Island adopted a new law, An Act
Relating to Cities and Towns-Providing Financial Stability, on June 11, 2010,
341
342
Until Debt Do Us Part
which gives the state-appointed receiver the authority to file under Chapter 9.
It is worth noting that the issue of state authorization is changing rather than
static. The information, accurate as of August 2010, is subject to rapid change.
California, for example, recently modified the procedure of its Chapter 9 filing.
For recent development in states’ authorization, see Spiotto (2012).
54. As defined in 11 U.S.C. § 101(32)(C).
55. Id.
56. In re City Of Bridgeport, 129 B.R. 332, 335 (Bankr. D. Conn. 1991). The court
found that the city had access to a US$27 million bond fund and cited this as an
additional reason for not meeting the “insolvency” test. See Rachael E. Schwarz,
“This Way To Egress: Should Chapter 9 Filing Have Been Dismissed?,” 66 American Bankruptcy Law Journal 103 (1992); Dorothy A. Brown, “Fiscal Distress and
Politics: The Bankruptcy Filing of Bridgeport as a Case Study in Reclaiming
Local Sovereignty,” 11 Bankruptcy Developments Law Journal 626 (1994–95).
57. 11 U.S.C. Sec. 901.
58. The stay prohibits a creditor from bringing a mandamus action against an
officer of a municipality on account of a prepetition debt. It also prohibits a
creditor from bringing an action against an inhabitant of the debtor to enforce
a lien on or arising out of taxes or assessments owed to the debtor. Additional
automatic stay provisions are applicable in Chapter 9 that prohibit actions
against officers and inhabitants of the debtor if the action seeks to enforce a
claim against the debtor. 11 U.S.C. § 922(a).
59. Many municipalities have separate enterprises that are owned or operated
by the municipality but are not separate legal entities. Such systems are typically treated as separate accounting units and are paid in the form of fees and
charges for services. Such systems are typically financed by debt obligations
payable from the system revenue, and in many cases this is the sole source of
payment. Such revenues are treated as “special revenues.” Special revenues are
defined in Section 902(c) as (a) receipts derived from projects or systems primarily used for transportation, utility, or other services; (b) special excise taxes
imposed on particular activities or transactions; (c) incremental tax receipts in
a tax increment financing; (d) other revenues or receipts derived from particular functions of the debtor; and (e) taxes specifically levied to finance projects
or systems (excludes general property, sales, or income taxes levied to finance
the general purposes of the debtor.
60. Section 922(d) of title 11 limits the applicability of the stay.
61. 11 U.S.C. § 928.
62. Bondholders have been recognized as having the right to receive those revenues and to block diversion of those revenues to other purposes including
general obligation bonds. Matter of Sanitary and Improvement District No. 7,
98 Bankr 970, 974 (D Neb 1989).
63. Or at a later time agreed to by the court.
64. 11 U.S.C. § 941.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
65. Contracts that are yet to be performed.
66. 11 U.S.C. § 365.
67. Sections 1113 and 1114 of Chapter 11 restricting the ability to reject collective
bargaining agreements and to restructure pension obligations do not apply to
Chapter 9.
68. In re City of Vallejo, 403 B.R. 72 (51 Bankr. Ct. Dec 2009).
69. See 11 U.S.C. §1126.
70. Often referred to as cramdown, 11 U.S.C. 901 (a).
71. See McConnell, supra, note 11.
72. The U.S. Supreme Court upheld the constitutionality of Chapter 9’s debt
adjustment authority in U.S. v. Bekins, 304 U.S. 27 (1938).
73. 11 U.S.C. 941 (b).
74. See below for a discussion of how the feasibility of Vallejo’s plan changed
due to increasing deterioration of its financial condition during the two-year
period it has been in Chapter 9 bankruptcy.
75. In an early irrigation district case, the Ninth Circuit Court of Appeals required
a showing that the taxing power was inadequate to raise taxes to pay debt. See
Fano v. Newport Heights Irrigation District, 114 F2nd 563 (9th Cir. 1940).
76. In addition, increased rates can dampen economic activity.
77. Supra, note 31.
78. Sections 903 and 904 of the Bankruptcy Code.
79. 11 U.S.C. § 903 states that “chapter [9] does not limit or impair the power of a
State to control, by legislation or otherwise, a municipality of or in such State
in the exercise of the political or governmental powers of the municipality.”
80. Specified in 11 U.S.C. § 926(a).
81.
Moreover, a Chapter 9 debtor may employ professionals without court
approval, and the only court review of fees is in the context of plan confirmation, when the court determines the reasonableness of the fees.
82. The restrictions imposed by 11 U.S.C. § 904 are necessary to ensure the constitutionality of Chapter 9 and to avoid the possibility that the court might
substitute its control over the political or governmental affairs or property of
the debtor for that of the state and the elected officials of the municipality.
83. Authors’ estimation based on “Bankruptcy Basics” (2006) by the Administrative Office of the United States Courts and the American Bankruptcy Institute;
and http://www.pacer.gov. The U.S. Congress amended Chapter 9 in 1937, and
the amended law was upheld by the Supreme Court in 1938.
84. United States Courts, http://www.uscourts.gov. For statistics covering up to
April 2012, see Spiotto (2012).
85. See Spiotto, James E. (2008) “Chapter 9: “The Last Resort for Financially Distressed Municipalities,” in Handbook of Municipal Bonds, ed. by Sylvan G. Feldstein and Frank J. Fabozzi. Since 1980, 49 of the 265 Chapter 9 filings have
been traditional local governments, towns, cities, villages, and counties (Spiotto 2012).
343
344
Until Debt Do Us Part
86. Id.
87. See cases recorded at http://www.pacer.gov.
88. See Spiotto 2008.
89. The authors collected 217 Chapter 9 filings recorded at http://www.pacer.gov,
which cover the period starting from 1981. Forty-two filings are by general
purpose municipalities. Some filings may be missing since some filings may
not be recorded in PACER, and some recorded filings do not give information
on the filer. The population data are from the 2000 U.S. Census. We counted
those entities that filed twice only once to calculate the median population.
90. Id. See cases recorded at http://www.pacer.gov. The 42 Chapter 9 filings of general purpose municipalities comprise 37 entities and 5 entities that filed twice.
These 5 entities include (a) City of Macks Creek, Missouri, filed in 1998 and
2000; (b) City of Prichard, Alabama, filed in 1999 and 2009; (c) City of Westminster, Texas, filed in 2001 and 2004; (d) Town of Moffett, Oklahoma, filed in 2006
and 2009; and (e) Village of Washington Park, Illinois, filed in 2004 and 2009.
91. Supra, note 89. See cases recorded at http://www.pacer.gov.
92. A petition for Chapter 9 protection was filed on December 6, 1994.
93. The rest of this section draws mainly from Baldassare (1998).
94. The California Constitution limits local and city control over most tax and
many fee revenue sources. Proposition 13 sets property tax rates and caps on
the annual growth of parcel assessed valuations. Sales tax rates are also controlled by the state Bradley-Burns Act, with the exception that the local electorate can vote to self-assess at a greater rate for specific or general programs.
Fees, assessments, and any new or increased taxes are subject to the constraints
of Proposition 218. Fees can be assessed and used only to recover the actual
cost of service, and assessments and taxes require property owner approval,
voter approval, or both.
95. See Baldassare 1998.
96. See Baldassare 1998, p. 4. The county sued a dozen or more securities companies, advisors, and accountants. Merrill Lynch settled with Orange County,
California, for US$400 million to settle accusations that it sold inappropriate
and risky investments to Orange County. The county lost US$1.69 billion.
The county was able to recover about US$600 million in total, including the
US$400 million from Merrill Lynch.
97. The plan provided for the refinancing of outstanding debt, forbearance by
the investor local public agencies, and diversion of certain revenue sources to
secure the refinancing debt. The impact on the county budget was not a result
of the Plan of Adjustment; rather, the impact was felt after Orange County
exited from bankruptcy with a substantial debt burden and fewer revenue
sources available for the budget.
98. Thirty-one cities, regional transportation agencies, local school districts, local
water agencies, sanitation districts, and many small local agencies. See Baldassare (1998), p. 9.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
99. See Baldassare (1998), p. 131.
100. Id. p. 179. The social program cuts included (a) a 50 percent reduction in child
abuse prevention programs, (b) elimination of a program for the homeless,
and (c) closure of a prenatal clinic and 15 clinics for children. The basic infrastructure cuts included cuts in funds for beaches, flood control, harbors, parks,
and redevelopment projects.
101. http://www.PennLive.com, May 16, 2010, interview with Chris Street, Treasurer of Orange County, California.
102. White, supra, note 49. Orange County raised funds for the plan by laying
off workers, selling some assets and cutting expenditures, and issuing new
bonds. It attempted to raise its local sales tax, but voters rejected the proposed
increase. See “Orange County Adopts Plans To Get Out of Bankruptcy,” New
York Times, December 22, 1995, p. D2.
103. “In the Orange County financial crisis, the bankruptcy forum appears to have
provided an appropriate and efficient judicial mechanism for its resolution.
The County qua municipality remained in control of its ‘political’ affairs, that
is, the operation of government and the provision of public services, while the
County qua debtor was free to pursue both litigation and negotiated settlement with its creditors. The uniquely binding effect of a Chapter 9, federally
confirmed reorganization plan coupled with the inherent limitations creditors
face in dealing with a municipal debtor may promote consensus towards an
achievable composition of debt.” See Jeweler, supra note 7.
104. For the years preceding 2008, the City of Vallejo had difficulty balancing
its contractual commitments in its General Fund with its General Fund
revenues. For fiscal years 2005–06, 2006–07, and 2007–08, General Fund
expenditures exceeded revenues by US$3 million to US$4 million per year,
resulting in a reduction of General Fund reserves. At the time of the bankruptcy filing, projections were that the city’s General Fund reserves would
be depleted by June 30, 2008 and that in fiscal year 2008–09, General Fund
expenses could exceed General Fund revenues by US$16 million, meaning that the city could not meet its obligations and was technically insolvent. The city was unable to reach agreements with its primary creditors
(employee labor associations) that would ensure ongoing General Fund
solvency. See official website http://www.ci.vallejo.ca.us/GovSite/default
.asp?serviceID1=712&Frame=L1.
105. “For Vallejo, Bankruptcy isn’t exactly a fresh start,” The Bay Citizen, January
23, 2011.
106. “Tough Budget Arithmetic Puts Vallejo in Bind,” Wall Street Journal, July 15, 2010.
http://www.ci.vallejo.ca.us/GovSite/default.asp?serviceID1=712&Frame=L1.
107. In re City of Vallejo, 403 B.R. 72 (51 Bankr. Ct. Dec 2009).
108. The agreements saved the city over US$6 million in General Funds through
June 30, 2010. After attempts to facilitate an agreement between the city and
the International Brotherhood of Electrical Workers failed, the United States
345
346
Until Debt Do Us Part
Bankruptcy Court upheld the city’s motion to reject their labor contract.
http://www.ci.vallejo.ca.us/GovSite/default.asp?serviceID1=712&Frame=L1.
Much of the savings in the renegotiated union contracts come from severe
workforce reductions: the police department is down to 90 sworn officers from
155 in 2003, and the fire department was slashed from 122 people and 8 firehouses to 70 people and 5 firehouses. See “For Vallejo, Bankruptcy isn’t exactly
a fresh start,” The Bay Citizen, January 23, 2011.
109. Wall Street Journal, July 15, 2010. “As Vallejo slogs through its third year of
bankruptcy, city officials are giving police a blunt choice; forgo a pay raise
agreed to in January 2009” (agreed to in bankruptcy eight months after they
filed for bankruptcy protection in May 2008).
110. Id. “What was feasible in January 2009 does not seem feasible 18 months later.”
111. Pension plans for retirees and current city employees, including one that allows
police officers to retire at 50 with as much as 90 percent of their pay, remain
untouched. The city chose not to test whether the attempt to change the existing pensions would be allowed even in bankruptcy, and so remains responsible
for some US$195 million in unfinanced pension liabilities. “For Vallejo, Bankruptcy isn’t exactly a fresh start,” The Bay Citizen, January 23, 2011.
112. http://www.huffingtonpost.com/2011/11/02/vallejo-bankruptcy-ends-after-three-years
_n_1072;usactionnews.com/2011/01/Vallejo-bankruptcy-plan-offers-unsecuredcreditors-5-20jpm.
113. blog.al.com/birmingham-news-stories/2011/11/jefferson_county_among_
several.html.
114. May 25, 2008, www.cbs13.com news report, “Vallejo Facing Uncertain Road
After Bankruptcy.”
115. 665,000 residents.
116. http://www.al.com, published Monday, September 27, 2010. The county entered
a consent decree in 1996, agreeing to fix the sewer system after the Cahaba
River Society and individuals successfully sued in federal court to show that the
county was illegally polluting area creeks and rivers with untreated waste.
117. www.CNNMoney.com May 28, 2010.
118. Shelley Sigo, “JeffCo Has 1st Missed Payment; Defaults on $46 million of
Accelerated Principal,” The Bond Buyer, July 9, 2009, at 1.
119. “Largest Municipal Bankruptcy Filed,” Wall Street Journal, November 10, 2011.
120. www.articles.businessinsider.com/2011-10-23/wall-street/30312613_1_jeffersoncounty-sewer.
121. “The Chapter 9 Filing Would Be for All of Jefferson County, Not Just the Sewer
System.” www.al.com, published Monday, September 27, 2010.
122.
Pursuant to which the creditors will not pursue legal remedies during
negotiations.
123. JPMorgan bankers were among the financial advisers who persuaded county
officials in 2002 to replace traditional fixed-rate bonds with notes having
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
floating interest rates, including auction-rate securities whose terms are set
through periodic bidding.
124. “Largest U.S. Municipal Bankruptcy Looms in Alabama,” Joe Mysak, http://
www.Bloomberg.com, April 11, 2010.
125. “Alabama County Brainstorms,” Wall Street Journal, July 2, 2010.
126. According to attorney Jeffrey Cohen, “Raising the rates [that] are already the
second or third highest in the country will scare away new business. The businesses that can leave, will leave and frankly the people who are going to have
the biggest burden are the homeowners.” www.al.com, published, Monday,
September 27, 2010.
127. http://www.CNNMoney.com, May 28, 2010.
128. Assured Guaranty, guarantor of the defaulted bonds.
129. “Harrisburg Seeks ‘Least Worst’ Path,” Wall Street Journal, April 28, 2010.
130. http://www.CNNMoney.com, May 28, 2010; http://www.PennLive.com, May
16, 2010.
131. “Harrisburg Bankruptcy Filing Voided,” Wall Street Journal, November 25, 2011.
132. “Judge Rejects Harrisburg Bankruptcy Move,” Wall Street Journal, November
25, 2011.
133. www.PennLive.com, May 19, 2010.
134. Id.
135. Comments of Westfall Attorney J. Gregg Miller, www.PennLive.com, May 19,
2010.
136. Financial Distress and Municipal Bankruptcy: The Case of Prichard Alabama, by
Douglas J. Watson, et al., July 1, 2005.
137. The Deal Magazine, March 19, 2010.
138. This ruling has no impact on Chapter 9 itself and would be limited to the
requirements for municipalities in Alabama to use Chapter 9.
139. And almost never used by large cities such as New York, Cleveland, and
Detroit. Large cities have persuaded their states to intervene and provide
financial relief. The states of New York and Ohio were heavily involved in their
cities’ resolution of financial distress, and Michigan has shown no hesitation in
its assistance to Detroit.
140. Supra, note 11.
141. See State Authorization above.
142. In the case of Bridgeport, the mayor who filed for Chapter 9 lost reelection,
defeated by the new mayor whose position was against the filing, supra, note 57.
143. See chapter 14 by Liu, Tian, and Wallis in this volume.
144. For example, like Vallejo, Los Angeles is suffering from weak revenue at the
same time that the cost of its pensions and other retirement benefits are rising.
Former mayor Richard Riordan has said, “The threat of bankruptcy is really
the only way you’re going to get them to make major changes.” Former mayor
Richard Riordan said those factors put the government of the second-largest
347
348
Until Debt Do Us Part
U.S. city on track to declare bankruptcy between now and 2014. Riordan sees
bankruptcy as a necessary tactic for squeezing concessions from the city’s public employee unions. It could also pave the way for 401(k) retirement accounts
for new city workers instead of defined pension benefit plans with escalating
costs, he said. http://articles.latimes.com/2010/may/08/opinion/la-oe-morri
son-20100508, May 8, 2010.
145. See supra, note 19, “Municipal Bankruptcy in Perspective,” A Joint Report from
the Bureau of Governmental Research and the Public Affairs Research Council
of Louisiana, April 2006.
146. Raphael, Richard J., Friedland, Eric, Laskey, Amy R., and Doppelt, Amy S. “The
Perils of Considering Municipal Bankruptcy.” Fitch Ratings, Public Finance,
January 27, 2010, indicating even the discussion of the possibility of filing is a
negative credit factor.
147. Mark Baldassare, president and CEO of the Public Policy Institute of California and author of When Government Fails: The Orange County Bankruptcy,
University of California Press, Berkeley, CA, a joint publication with the Public
Policy Institute of California.
148. www.Ohiohistorycentral.org.
149. Supra, note 95.
150. For example, within seven years of Orange County’s Chapter 9 bankruptcy,
its bond rating has improved from junk status to “Aaa”—the highest rating
offered by Moody’s Investor Services.
151. Fitch Ratings.
152. Supra, note 111.
153. Wes Johnson, “Should Mack’s Creek Exist?”, Springfield News Leader, October
16, 2004 at 1A.; and data from the federal judiciary’s case management files
(http://www.pacer.gov).
154. For a review of state intervention, see chapter 14 by Liu, Tian, and Wallis in
this volume.
155. See Liu and Waibel 2010.
Bibliography
Administrative Office of the United States Courts and the American Bankruptcy
Institute. 2006. “Bankruptcy Basics.” Washington, DC.
Baldassare, Mark. 1998. When Governments Fail: The Orange County Bankruptcy.
Berkeley, CA: University of California Press and the Public Policy Institute of
California.
Bay Citizen. 2011. “For Vallejo, Bankruptcy Isn’t Exactly a Fresh Start,” January 23.
BGR/PARCL (Bureau of Governmental Research [New Orleans, LA] and the Public
Affairs Research Council of Louisiana [Baton Rouge, LA]). 2006. “Municipal
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
Bankruptcy in Perspective, A Joint Report from the Bureau of Governmental
Research and the Public Affairs Research Council of Louisiana,” April.
Bornstein, Howard, Stan Markuze, Cameron Percy, Lisha Wang, and Moritz Zander.
2010. “Going for Broke: Reforming California’s Public Employee Pension Systems.”
Stanford Institute for Economic Policy Research Policy Brief, Stanford, CA, April.
Breckinridge Capital Advisors. 2009. “White Paper: Bankruptcy Basics for Municipal Bondholders.” Special Report, Boston, MA, October.
Brown, Dorothy A. 1994–95. “Fiscal Distress and Politics: The Bankruptcy Filing of
Bridgeport as a Case Study in Reclaiming Local Sovereignty.” 11 Bankruptcy
Developments Law Journal 626.
Ciccarone, Richard A. 2008. “Jefferson County and Municipal Finance, In the Wake
of the Bubble,” Washington, DC, September 9.
Dubrow, David L. 1992. “Chapter 9 of the Bankruptcy Code: A Viable Option for
Municipalities in Fiscal Crisis?” 24 The Urban Lawyer 3, 548 (Summer).
Fitch Ratings. 2010. “The Perils of Considering Municipal Bankruptcy.” Special
Report, January 27.
House Report No. 95-595, 95th Congress, 1st Session 263 (1977), U.S. Code Cong.
and Admin. News 1978, pp. 5787, 6221.
Johnson, Wes. 2003. “Should Mack’s Creek Exist?,” Springfield News Leader, October
16, p. 1A.
Kannell, William W., Richard H. Moche, and Mintz Levin. 2009. “A Guide to Municipal Bankruptcy.” National Federation of Municipal Analysts, Advanced Conference on Bankruptcy and Workouts, Cambridge, MA, October 15–16.
Kaplan, Gary, and Joel Moss. 2006. “Distressed Cities See No Clear Path: Health, Pension Obligations Threaten Fiscal Crisis.” National Law Journal (March 6): S1.
Kimhi, Omer. 2008. “Reviving Cities: Legal Remedies to Municipal Financial Crises.”
88 Boston University Law Review, 633.
Knox, John, and Marc Levinson. 2009. “Municipal Bankruptcy: Avoiding and Using
Chapter 9 in Times of Fiscal Stress.” Orrick, Herrington and Sutcliffe, LLP, San
Francisco.
Kotok, David. 2008. “The Largest Municipal Bankruptcy Ever? Jefferson County and
Municipal Finance in the Wake of the Bubble.” Cumberland Advisors, Washington, DC, September 9.
Jeweler, Robin. 2007. “Municipal Reorganization: Chapter 9 of the U.S. Bankruptcy
Code.” Congressional Research Service, Washington, DC, March 8.
Laing, Jonathan. 2010. “The $2 Trillion Hole,” Barrons, March 15.
Laughlin, Alexander. 2005. “Municipal Insolvencies: A Primer on the Treatment of
Municipalities under Chapter 9 of the US Bankruptcy Code.” Wiley Rein and
Fielding LLP, Washington, DC.
Liu, Lili, and Michael Waibel. 2009. “Subnational Insolvency and Governance:
Cross-Country Experiences and Lessons.” In Does Decentralization Enhance Service Delivery and Poverty Reduction?, ed. Ehtisham Ahmad and Giorgio Brosio,
333–76. Cheltenham, UK: Edward Elgar Publishing Limited.
349
350
Until Debt Do Us Part
———. 2010. “Managing Subnational Default Risk.” In Sovereign Debt and the
Financial Crisis: Will This Time be Different?, ed. Carlos Braga and Gallina
Vincelette, 273–93. Washington, DC: World Bank.
Liu, Lili, Xiaowei Tian, and John Joseph Wallis. 2013. “Caveat Creditor: State Systems of Local Government Borrowing in the United States.” In Until Debt Do
Us Part–Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and
Lili Liu. Washington, DC: World Bank.
Maco, Paul S., Jane L. Vris, and William L. Wallender. 2011. “Public Finance Challenges and Opportunities for Resolution.” Civic Research Institute, Kingston, NJ.
McConnell, Michael W., and Randal C. Picker. 1993. “When Cities Go Broke: A
Conceptual Introduction to Municipal Bankruptcy.” University of Chicago Law
Review 60 (2) (Spring): 425–95.
Miller, Matt. 2010. “Taboo: Chapter 9,” Deal Magazine, March 19; http://www
.thedeal.com/newsweekly/features/coverstories/taboo:-chapter-9.php.
Moody’s U.S. Public Finance. 2009. “Moody’s Assigns Negative Outlook to U.S.
Local Government Sector,” April.
Mysak, Joe. 2010. “Largest U.S. Municipal Bankruptcy Looms in Alabama,” http://
www.Bloomberg.com, April 11.
New York Times. 1995. “Orange County Adopts Plans To Get Out of Bankruptcy,”
December 22, p. D2.
———. 2006. “Paying Health Care from Pensions Proves Costly,” New York Times,
December 19, p. A1.
Raffer, Kunibert. 2005. “Internationalizing U.S. Municipal Insolvency: A Fair, Equitable and Efficient Way to Overcome Debt Overhang.” 6 Chicago Journal of
International Law 361.
Raphael, Richard J., Eric Friedland, Amy R. Laskey, and Amy S. Doppelt. 2010. “The
Perils of Considering Municipal Bankruptcy.” Fitch Ratings, Public Finance,
January 27.
Reuters. 2010. “Sovereign Defaults Top 2010 risk Hit List for WEF,” January 14.
Schwarz, Rachael E. 1992. “This Way To Egress: Should Chapter 9 Filing Have Been
Dismissed?” 66 American Bankruptcy Law Journal 103.
Schwarcz, Steven L. 2002. “Global Decentralization and the Subnational Debt Problem.” 51 Duke Law Journal: 1179, 1190.
Spiotto, James E. 2008. “Chapter 9: The Last Resort for Financially Distressed
Municipalities.” In Handbook of Municipal Bonds, ed. Sylvan G. Feldstein and
Frank J. Fabozzi. Hoboken, NJ: Wiley.
———. 2012. “Primer on Municipal Debt Adjustment - Chapter 9: The Last Resort
for Financially Distressed Municipalities.” Chapman and Cutler LLP, Chicago.
Spitz, Jonathan. 1992. “Federalism, States and the Power to Regulate Municipal
Bankruptcies: Who May Be a Debtor under Section 109(c).” 9 Bankruptcy
Development Journal 621.
Standard and Poor’s. 2009. “What Credit Concerns Does Talk of Municipal Bankruptcy Raise?” Standard and Poor’s Global Credit Portal, December 15.
United States: Chapter 9 Municipal Bankruptcy—Utilization, Avoidance, and Impact
Tung, Frederick. 2002. “After Orange County: Reforming California Municipal
Bankruptcy Law.” 53 HASTINGS L.J. 885, 907.
Wall Street Journal. 2010a. “Muni Threat: Cities Weigh Chapter 9.” February 18.
———. 2010b. “Alabama County Brainstorms,” July 2.
———. 2010c. “Tough Budget Arithmetic Puts Vallejo in Bind,” July 15.
———. 2010d. “Harrisburg Seeks ‘Least Worst’ Path,” April 28.
———. 2011a. “Pensions Chopped but Investors Paid,” December 20, p. C1.
———. 2011b. “Largest Municipal Bankruptcy Filed,” November 10.
———. 2011c. “Harrisburg Bankruptcy Filing Voided,” November 25.
———. 2011d. “Judge Rejects Harrisburg Bankruptcy Move,” November 25.
Watson, Douglas J., Donna Milam Handley, and Wendy L. Hassett. 2005. “Financial
Distress and Municipal Bankruptcy: The Case of Pritchard.” 17 Journal of Public Budgeting, Accounting & Financial Management 129.
White, Michelle J. 2002. “Sovereigns in Distress: Do They Need Bankruptcy?” Brookings Papers on Economic Activity Issue I: 287–319.
Legal Decisions
In re City of Bridgeport, 129 B.R. 332, 335 (Bankr. D. Conn. 1991).
In Re City of Vallejo 403 B.R. 72 (51 Bankr. Ct. Dec. 2009)
Websites
http://www.thedeal.com/newsweekly/features/cover-stories/taboo:
chapter-9.php.
http://www.ci.vallejo.ca.us/GovSite/default.asp?serviceID1=712&
Frame=L.
http://www.uscourts.gov/FederalCourts/Bankruptcy/BankruptcyBasics/
Chapter9.aspx.
http://govinfo.library.unt.edu/nbrc/report/22chapte.html#2415.
http://www.orrick.com/fileupload/1736.pdf.
351
9
When Subnational Debt Issuers
Default: The Case of the
Washington Public Power
Supply System
James Leigland and Lili Liu1
Introduction
The Washington Public Power Supply System (WPPSS) was the largest
municipal bond default in United States modern history. WPPSS was a
municipal corporation of the state of Washington. It expanded r apidly
in the 1970s since power demand in the region was expected to keep
doubling every 10 years, as it had in the recent past. By the end of the
1970s, it had become the largest issuer of municipal revenue bonds in
the United States. In the summer of 1983, after years of deepening problems, WPPSS defaulted on US$2.25 billion in outstanding bonds. Without recovery of the US$2.25 billion debt principal, bondholders also
stood to lose the US$5 billion in interest owed over the lifetime of the
bonds. Of five WPPSS nuclear plant projects for which over US$8 billion
was borrowed, only one eventually became operational but generated
only a fraction of the revenues needed to repay bondholders.
Municipal defaults since the 1950s in the United States have been
very rare (for example, average credit loss rates on Moody’s-rated
municipal bonds have been extremely low). The 1-, 5-, and 10-year
353
354
Until Debt Do Us Part
cumulative default rates for all Moody’s-rated municipal bond issuers
have been 0.0043, 0.0233, and 0.0420 percent, respectively, compared to
0.0000, 0.1237, and 0.6750 percent for triple-A-rated corporate bonds
during 1970–2000 (Moody’s 2002).
The WPPSS debacle has been well documented over the years, beginning with a series of consulting studies commissioned by WPPSS itself
and including hundreds of credit reports by Wall Street firms, analysis by academicians and journalists, and an extensive investigation by
the United States Securities and Exchange Commission (SEC). All of
this information allows the WPPSS story to serve as a useful case study
of what happens in the United States when a subnational government
entity defaults on municipal bond obligations. The story provides particularly valuable lessons for governments in emerging markets undertaking efforts to accelerate subnational capital market development.
This chapter is organized as follows. Section two describes how
WPPSS went into default, section three discusses major factors contributing to the default, section four asks why WPPSS was not bailed out by
either the federal or state government, section five discusses the regulatory reforms in the aftermath of the default, and section six draws lessons that may be relevant for developing countries.
WPPSS: From Creation to Default
The Public Authority Concept
WPPSS was created as a public authority or municipal corporation of
Washington State, located in the northwestern corner of the United States.
Public authorities are a widely used form of American government. These
entities build and run bridges, tunnels, parkways, dams, ports, airports,
public buildings, railroads, and industrial and recreational parks.2 They
provide essential services, including water, gas, e lectric power, and transportation. By the late 1970s, at least 6,000 local or regional public authorities and 1,000 state and interstate public authorities were in operation in
the United States (Walsh and Mammen 1983).
Public authorities such as WPPSS are authorized by legislative
action to function outside of the regular executive structure of state
or municipal government.3 They are independent legal entities with
purposes and powers defined in a statute or charter granted under law.
They generally do not have the power to tax, but because of their legal
When Subnational Debt Issuers Default
identity they can borrow and own assets. They also can sue or be sued,
enter into contracts in their own names, and have liability distinct from
that of the government entities that chartered them. Public authorities
are designed to have the independence and flexibility needed for them
to function as business entities.
This separate legal nature provides two kinds of flexibility, both of
which were enjoyed by WPPSS. First, authorities are usually permitted
a great deal of administrative flexibility by being exempted from many
of the procedures and regulations that apply to executive line agencies,
including civil service and personnel rules, procurement procedures,
and internal operating rules. Second, public authorities are capable of
independent borrowing. In many states, constitutions or legislatures
have instituted strict limits on the amount of general obligation debt
the state can issue (such debt is backed primarily by the taxing powers
of the state). Many states and local governments also require popular
referendums in advance of state borrowing and prohibit executive line
agencies from selling revenue bonds. The revenue bonds secured by the
revenue of a project issued by a public authority are typically not subject to state or local government debt limitations of these kinds.4
Public authorities, however, typically operate within regulatory
frameworks that combine rules set by the federal government, state,
or other “parent” governments, bondholders, and financial accounting
and reporting associations. Regulation usually includes the following
(Liu 2010): federal government regulations require that borrowing in
the municipal bond market must be for a public (not private) purpose
and must finance capital projects related to the stated purpose of the
public authority. Parent governments often require that borrowings
be repayable from a special fund, such as a fund into which revenues
of a water system or special tax are deposited. This helps avoid the
diversion of revenues to the general budget or commingling in such
manner that the funds lose their separate identity. Bondholders usually require that operating revenue be maintained at levels that reflect
specific ratios of debt service, and levels of rates/tariffs for services may
be required to maintain such ratios (in agreements with bondholders,
these kinds of tariff requirements are referred to as rate covenants).
Bondholders may also require compliance with historical and projected
debt s ervice coverage ratios as a condition for the issuance of additional
debt. Accounting standards require adherence to generally accepted
355
356
Until Debt Do Us Part
a ccounting principles for government entities, and the regular public
disclosure of independently audited financial statements. Legal opinions need to be obtained to confirm that each municipal bond issue is
fully compliant with applicable government regulations.
WPPSS was created under Washington state law as a “joint action
agency,” a type of public authority that enjoyed all of the aforementioned administrative and financial flexibility, but was also subject to
few oversight powers from the state. Based on state legislation enacted
in 1953, any two or more public utility districts or municipalities were
allowed to form a municipal corporation for the purposes of purchasing, building, owning, and operating electrical generation and transmission facilities. Washington was only the second state to pass such
legislation (California was the first in 1949), and it was not until 1972
that more states followed suit.
WPPSS faced minimal reporting and auditing requirements—
because it was a municipal corporation, its board was required to
appoint an independent financial auditor whose report was to be filed
with the state auditor. Also, approval for the initiation of certain projects required permits, licenses, and approval from state agencies. But
any kind of ongoing, institutionalized oversight activities by the state
legislature were essentially nonexistent.
It would be inaccurate to suggest that many more controls would
have been available to legislators had WPPSS not been a joint a ction
agency. Municipal corporations in Washington State, like public
authorities in most other states, had control over their management and
operations. But it appears that because WPPSS, as a joint action agency,
was an offspring of the highly respected Washington State Public Utility
Districts (PUDs),5 state legislators considered WPPSS and its problems
to be the responsibility of WPPSS and its members. The state legislature
finally began to take notice of WPPSS problems in the late 1970s, but by
the time an investigation by state officials was concluded in 1981, it was
already too late to head off the WPPSS default.
Background to Default
WPPSS was formed in 1957 as a supply arm of its original owner/
members—19 PUDs and four cities. All of these members were represented on the WPPSS board of directors, which made all key business
When Subnational Debt Issuers Default
ecisions for the organization. To build the nuclear power plants WPPSS
d
recruited dozens of project “participants” in several Pacific Northwest
states, including municipal utilities, other PUDs, irrigation districts, and
rural electric cooperatives. The participants agreed to buy the generating capacity of the WPPSS nuclear plants. But the nuclear plants represented a massive increase in the size and complexity of WPPSS projects.
For the first 14 years of its existence—from 1957 to 1971—WPPSS
constructed and operated only two small power projects with a combined operating capacity of 890 megawatts. In 1971, WPPSS began
work on the first of five nuclear power plants, because power demand in
the region was expected to keep doubling every 10 years as it had in the
recent past. The first three plants received backing from the B
onneville
Power Administration (BPA), a federal agency still active in the region,
which markets and transmits power generated at federal hydroelectric projects in the region. The backing took the form of a “net billing”
arrangement whereby BPA undertook to purchase all of the power from
the three plants of WPPSS, then sell it to its customers, including over
130 utilities and industrial customers in the region, far more than the
WPPSS members or project participants. If individual BPA customers
could not or would not pay, the cost would be shared across the rest
of BPA’s customer base. The arrangement meant that a federal agency
was the principal offtaker of power and that the bonds sold to finance
the first three plants were considered low-risk investments by rating
agencies and bond investors, many of whom assumed that the BPA role
constituted a federal guarantee of debt service payment. What many
investors did not understand was that BPA was not authorized to access
federal treasury funds or borrow on its own account to make good on
such commitments.
By 1972, BPA planning had determined that even more powergenerating capacity would be needed by the early 1980s. The agency
encouraged WPPSS to undertake construction of two more nuclear
power plants to meet regional power needs, and it was the eventual
failure to pay interest on the bonds issued to pay for these additional
two plants that triggered the default in 1983. By the time borrowing was
needed for these two new plants, BPA was no longer able to use the net
billing arrangement to back WPPSS borrowing because of changes in
federal regulations. To provide the security necessary to attract investors
357
358
Until Debt Do Us Part
to the new bond issues needed to finance the new plants, WPPSS signed
“take-or-pay” contracts with 88 project participants, mostly municipal
and regional utilities, which obligated these participants to pay their
shares of project costs, including debt service on the bonds, whether
or not Projects 4 and 5 were ever completed or capable of generating
power.
At the time, the investment community had great faith in take-or-pay
contracts, which had long been used in electric revenue bond financing.
Most experts believed that bonds could not be sold for nuclear power
projects without take-or-pay backing, or some other method almost
as secure, due to the substantial risks of construction delay. The takeor-pay contracts put the project and construction risks on the members.
At the time of the WPPSS default, over a dozen joint action agencies
across the country were financing major power project construction
using take-or-pay that were virtually identical to the ones used by
WPPSS—in several cases state supreme courts had already upheld their
use (Tamietti 1984). Because participating utilities were not subject to
regulation by federal or state utility commissions, they had unlimited
authority to set rates for existing customers. In theory this meant that
they were capable of complying with the take-or-pay contract provisions by raising rates as high as necessary on existing customers to pay
off the WPPSS debt even if the nuclear plants were not completed and
no new customers could be added to their systems. The take-or-pay
contracts were almost as helpful as the BPA net-billing a rrangements
in convincing rating agencies, underwriters, and investors that WPPSS
bonds were highly secure investments. WPPSS therefore had no trouble borrowing from the securities markets. WPPSS Projects 1, 2, and
3 bonds were rated triple-A by Standard & Poor’s and Moody’s until
January 1983 and Project 4 and 5 bonds were rated A-plus by Standard
& Poor’s until June 1981.
But by the end of the 1970s, WPPSS was facing serious problems.
The 1970s energy crisis dramatically slowed growth in demand for electricity and demonstrated to energy planners that simplistic straight-line
demand growth projections were unable to fully capture the potential
impacts of energy conservation.6 At the same time the WPPSS nuclear
projects were plagued with construction delays and cost overruns. The
plants were initially estimated to cost US$4.1 billion, but by 1981 the
When Subnational Debt Issuers Default
estimate had grown to US$23.8 billion. In 1982, WPPSS terminated
construction on Projects 4 and 5, acknowledging that the plants would
never be completed and never generate the revenues needed to pay back
the US$2.25 billion in affected bonds. Pursuant to the take-or-pay contracts, the default signaled that the 88 participating utilities, and ultimately their customers, were obligated to pay back the borrowed money
from whatever other sources were available, even if it meant dramatically raising electricity rates to their customers. But the region (and the
country) was inflicted by a deep recession in the late 1970s to the early
1980s. In some small northwest towns already affected by unemployment, the cost amounted to more than US$12,000 per customer.
When 28 WPPSS participants (municipalities and PUDs), accounting for two-thirds of the Projects 4 and 5 shares, refused to pay for their
share of the projects, a lawsuit was filed by the trustee for the bondholders of the Projects 4 and 5. The Washington State Supreme Court ruled
in July 1983 that the utilities had lacked the legal authority to enter into
take-or-pay contracts (promise to pay for something that they would
not receive, that is, power plant output that might never exist), thus the
contracts were void and unenforceable. The legal sources of the judgment included existing statues, constitutional provisions governing the
authority of home rule cities, and case law.7 Prior to this decision, the
Washington State Supreme Court had never ruled on the legal issue
relating to the legality of take-or-pay contracts. The ruling demonstrates
the autonomy of state law and the risks of assuming that the legal decisions of one state can be used to predict the outcomes of court cases
in other states. This ruling illustrates that debt obligations and their
underlying security arrangements are based on obligations to pay that
are lawfully entered into and can be legally enforced.
The 80,000 affected bondholders began a series of legal actions
against WPPSS and other major actors in the disaster, charging fraud
and misrepresentation in the sale of the bonds.8 The major lawsuits
continued for the next 13 years. In 1988, a settlement was reached for
US$753 million. The last settlement was reached in 1995. Of the five
plants, only one was completed—one of the three backed by the BPA.
Revenues from that plant helped BPA continue to pay debt service on
bonds issued for the first three plants, so there never has been a default
on any of those BPA-secured bonds.
359
360
Until Debt Do Us Part
By August 1989, WPPSS was back in the municipal bond market,
issuing US$450 million in bonds to refinance some of the outstanding debt used to finance the two closed plants backed by BPA. Even
though the projects were no longer being built, and clearly would never
be finished, the refinancing bonds received investment-grade ratings
from Moody’s and Standard & Poor’s because of the BPA support.
The remaining structures of the other four plants were demolished
in 1995. In 1998, WPPSS renamed itself Energy Northwest and began
to focus on wind and solar energy projects. In 2002, 19 years after the
default, Energy Northwest once again sold bonds for the construction
of a new energy-generating facility and US$70 million was raised for a
48-megawatt wind energy project.
Factors Contributing to the Default
The major contributing factor to the default was the failings of
WPPSS management. When the Washington State Senate finally began
investigating WPPSS problems in 1981, they concluded that “WPPSS
mismanagement has been the most significant cause of cost overruns
and schedule delays on the WPPSS projects” (Washington State Senate
Energy and Utilities Committee 1981). An administrative auditor
engaged by the legislature discovered that 11 substantial management
consulting reports incorporating over 400 specific recommendations
had been commissioned and received by WPPSS from 1976 to 1980.
But most of the recommendations were not implemented. The auditor
noted that during the crucial years from 1971 to 1979, “the type of staff
required to manage the growing giant of a program was simply ‘not in
place’” (Washington Public Power Supply System 1980).
The specific management problems are well documented in the consulting studies done during the 1970s (Leigland 1988). At the end of the
1970s, WPPSS was still being managed in many respects as if its size and
responsibilities remained at 1971 levels. The huge growth in organizational size, complexity, and responsibility had not been accommodated
by management and organizational changes. The size of the WPPSS
staff increased from 81 in 1971 to over 550 in 1976, with thousands
of construction personnel working on each of the individual projects.
Administrative expenses increased from just over US$1 million in 1972
When Subnational Debt Issuers Default
to over US$33 million in 1976. With Projects 4 and 5 fully under construction, it was clear that both administrative costs and the number of
personnel would continue to grow rapidly. By 1981, the number of staff
personnel had grown to over 2,000, with an additional 14,000 construction personnel working under 400 separate contracts.
WPPSS quite literally grew out of control because its management
style did not mature. In the late 1970s, the top management of WPPSS
was still exercising a management style appropriate for a small business,
characterized by concentration of authority and accountability at the
top levels. WPPSS directors tended to be successful small businessmen
whose public power management experience was limited to approving
budgets for local utilities and promoting local power use. A few were
utility managers, but none had experience with large-scale construction, nuclear power, or complex financing. All of these shortcomings of
WPPSS governance were documented in the management consulting
reports completed during the 1970s, but the findings were apparently
never fully appreciated in WPPSS until after a new managing director
was recruited from outside the organization for the first time in 1980.
Members of the investment community—Wall Street underwriters,
dealers, institutional investors, rating firms, and bond attorneys—also
played roles in the WPPSS drama. Both ratepayers and many bondholders
argued that the pursuit of short-term profit caused the investment community to lose sight of the true investment quality of the WPPSS bonds.
Credit analysis—the assessment of the ability and willingness of
a debt issuer to pay back those debts in a timely fashion—is considered by nearly all the members of the investment community to be an
important foundation of the work that they do. However, the WPPSS
debacle caught Wall Street by surprise. Credit analyses prepared by
major Wall Street firms routinely misunderstood key legal and economic factors a ffecting WPPSS (see Leigland and Lamb 1986). Authors
of credit r eports took for granted the opinions of bond attorneys who
assumed the legal validity of take-or-pay contracts without considering
how elected state judges might view those contracts under the politically charged conditions of WPPSS project failure. Analysts also failed
to correctly estimate the chances of delayed debt service payments (and
short-term or technical defaults) resulting from court challenges to
those contracts.
361
362
Until Debt Do Us Part
Typically, analysts expected that new federal legislation would allow
BPA to purchase the output of Projects 4 and 5 (as it had for the first
three projects) even after the bill was passed and made such a purchase
extremely unlikely. Straight-line growth projections were taken at face
value, and the conflicting demand growth studies of other groups or
public agencies were almost never mentioned. The many management
consulting studies that painted such damning portraits of management incompetence were never mentioned much less examined by
most analysts, even though those studies were available to the general
public.
Until January 1983, both Standard & Poor’s and Moody’s gave bonds
for WPPSS Projects 1, 2, and 3 triple-A ratings, the firms’ highest rating
(these were the BPA secured bonds with the quasi-indirect federal
guarantee). Roughly US$6 billion bonds were sold with the benefit
of that rating. In June 1981, Standard & Poor’s lowered its rating for
Project 4 and 5 bonds from A-plus to A (these are the bonds secured
by the take-or-pay contract later defaulted). Shortly after, Moody’s also
downgraded these bonds. WPPSS 4 and 5 bonds were never sold again,
but US$2.25 billion had been sold with the benefit of Moody’s A-1 and
Standard & Poor’s A-plus ratings. The downgrades came too late to
benefit the bond-buying public. By June 1983, Moody’s had suspended
ratings for all WPPSS project bonds, and Standard & Poor’s had suspended ratings for Projects 1, 2, and 3, with Project 4 and 5 bonds given
a highly speculative, CC rating.
The financial advice provided to WPPSS was more concerned with
marketing of WPPSS bonds than with strengthening creditworthiness.
In 1980, the principal problem occupying WPPSS financial advisors9 was
the fact that the portfolios of institutional investors, the largest purchasers of WPPSS bonds, were saturated with those securities. The spaces
typically reserved in those portfolios for securities of the type issued
by WPPSS were largely filled. By 1980, WPPSS, as the largest issuer of
revenue bonds, had been marketing major new long-term bond issues
about every six or seven weeks. Other similar kinds of securities were
also competing for the same space. WPPSS thus faced the problem of
raising an additional US$5.34 billion before 1985. WPPSS proposed
that WPPSS tailor its offerings to occupy some new “space” in investor
portfolios.10
When Subnational Debt Issuers Default
WPPSS underwriters proposed a different strategy for effectively
marketing new WPPSS debt, one which WPPSS executives decided to
implement (Marion and Quinn 1980). Instead of selling short-term
debt, WPPSS was advised to shift from competitive to negotiated
underwriting. The number of underwriters competing for WPPSS business was declining. As this happened, bidders were asking for higher
remuneration. By switching to noncompetitive negotiated underwriting, WPPSS could establish a relationship with a single underwriter/
arranger who would not jeopardize the relationship and future business
by trying to maximize current returns on a single bond sale. WPPSS was
convinced of the soundness of this approach and launched its negotiated bond program with a US$750 million issue in 1981.
Unfortunately, the negotiated offering procedure did nothing to
improve WPPSS underlying creditworthiness. The risks associated
with the bonds were clearly increasing and were already apparent to
most s ophisticated investors. Knowledgeable institutional investors had
begun to shy away from WPPSS, so bonds had to be sold in smallerthan-usual lots to attract individuals. The new negotiated underwriter
selected by WPPSS used extensive presale surveys and public relations
efforts that were successful in appealing to individual investors. Since
the underwriter was preparing a WPPSS issue, the underwriter’s own
research department (a separate unit in the company) issued one of the
most thorough critiques of WPPSS bonds prepared to date, citing cost
overruns, management inadequacies, and the possibility of plant shutdowns (Sitzer and Karvelis 1981).
No Bailout
After the default, WPPSS bondholders hoped to benefit from a bailout.
A series of bailout plans had been proposed including (a) a proposal by
WPPSS participants to use a federal subsidized loan to purchase new
bonds issued by WPPSS and to use the bond proceeds to pay WPPSS
debt service,11 (b) a regionalization plan by mostly WPPSS participants
to spread Project 4 and 5 debt through BPA billings to its regional customers,12 and (c) a proposal to create a federally chartered regional
financing agency to sell bonds to pay off WPPSS debt and finance final
construction costs.13
363
364
Until Debt Do Us Part
But none of the many bailout proposals developed by and for WPPSS
were seriously considered by Congress. The tradition of federal unwillingness to become involved in state and local bailouts went back more
than a century, with the federal position crystallized in a series of refusals to assume state debts after several state defaults occurred in the mid19th century (Wallis 2005). The default justification was that the federal
government did not want to signal that fiscal irresponsibility would not
be penalized by default or bankruptcy, but instead would be rewarded
with bailouts paid for by taxpayers. The federal refusal to bail out an
entity like WPPSS was also related to the concept of state sovereignty
in the U.S. system of federalism—if states were going to jealously protect their right to self-determination from federal interference, including their right to sell tax-exempt bonds, then they should be prepared
to handle their own problems without the expectation of extraordinary
federal help.
There were also several more practical reasons for the federal position. By the early 1980s, the failings of WPPSS management were well
documented and widely believed to be the primary cause of WPPSS
problems. Project failure (if not the long-term default itself) did not
appear to have been beyond the control of WPPSS management, and
therefore special outside assistance was not easily justified. Even among
local and regional politicians there was little support for a bailout, perhaps largely because justifying outside assistance would mean admitting
that WPPSS was not solely responsible for its own problems.
From 1972 to 1983, over 100 nuclear-power-generating plants were
cancelled in the United States, twice the number of coal plants. In 1982
dollars, those plants represented almost US$10 billion in investments.
With some nuclear utilities experiencing some problems similar to
those of WPPSS, Congress was wary of putting itself in a position of
obligation to all by recognizing an obligation to WPPSS.
The WPPSS default lacked an injured party, other than bondholders. No massive unemployment or disruption of crucial municipal services would result from the WPPSS default—the five plants were not
yet remotely close to operational. Nor did it appear that the economic
interests of the region would be significantly harmed by the default. Various Wall Street executives made vague threats of penalizing Washington
State with higher interest rates on new debt because of the state’s failure
When Subnational Debt Issuers Default
to come to the aid of WPPSS, but knowledgeable observers knew that
competitive pressures within the financial community would lead to
business as usual when state bonds next came to market, as was indeed
the case. The 80,000 affected WPPSS bondholders certainly constituted
an injured party. A sample survey of 10,000 WPPSS bondholders conducted in 1985 revealed that two-thirds were over 60 years old, and more
than half were retired and seeking a modest supplement to their social
security income (Lehmann 1986). The votes of disgruntled bondholders
were dispersed across the country, with ratepayers affected by WPPSS
focused in two states (Washington and Oregon).
Regulatory Reforms of the Municipal Bond Market in
the Aftermath of Default
Participants in the municipal bond market include issuers, bondholders, underwriters, fund managers, bond attorneys, and advisory
services. Total municipal debt outstanding at the end of 2011 was
US$3.05 trillion.14 The United States has the largest municipal bond
market in the world.
Beginning in the 1960s, the use of revenue bonds began to grow rapidly, accounting for 50 percent of all tax-exempt bonds by 1975 and
83 percent by 1983 (Bond Buyer 1985). These kinds of securities are
usually issued by public enterprises or public authorities, and the
security behind them is the explicit stream of revenues identified in
the bond resolution and offering. Revenue bond repayment usually
depends heavily on the contemplated operating revenues of the project to be built—tolls from roads and bridges; mortgage payments from
housing developments; or, in the case of WPPSS, revenues from the
sale of electricity when and if the plants are completed, fully tested,
and o
perating. With the shift from general obligation bonds to revenue bonds, real credit quality depends on the economic and financial
feasibility of the issuer and each individual project rather than on the
full faith, credit, and tax base of governments.
Each bond resolution and prospectus designates a trustee, usually
a bank, to represent the bondholders over the terms of the bonds for
purposes of collecting and distributing interest payments and pursuing
legal protection of bondholders’ rights. Bond attorneys and financial
365
366
Until Debt Do Us Part
advisors often devise special arrangements to strengthen the security
behind revenue bonds and thereby make them more attractive to the
investors to whom the bonds will be sold by underwriters. A syndicate
of underwriting firms usually buys the entire bond issue from the public enterprise at a negotiated price or by competitive bid. The members
of the syndicate then attempt to sell the bonds to the public at a higher
price, and the difference, or spread, supplies profits to the underwriting firms. WPPSS and the utilities participating in its projects, like other
public agencies, depend heavily on the advice of a relatively limited
number of national bond attorney firms and investment advisory and
underwriting firms both to understand their own credit situation and to
design the most appropriate security arrangements behind their bond
issues. The ultimate buyers of the bonds also depend on the opinions
and representations of these firms.
Until the mid-1970s, the municipal bond market was essentially
unregulated by the federal government. Although municipal securities are subject to the antifraud provisions of the Securities Act of 1933
and the Securities and Exchange Act of 1934, the two acts exempted
municipal securities, and almost all parties associated with the municipal securities market, from most of their regulations affecting corporate underwriting, buying, selling, and trading. Congress was also wary
of violating the sovereignty of state governments by imposing federal
rules on states or local government borrowers. The municipal market
was relatively small at that time and considered free from the problems
that plagued the corporate securities market. As a result, the municipal
market was run on a self-regulated basis by the issuers themselves and
their service providers from the investment community.
The near default of New York City in 1975 prompted Congress
to introduce changes in municipal bond regulations.15 But largely
because of the state sovereignty issue, Congress chose not to eliminate
the exemption that municipal bonds were granted by the Securities
Act of 1933. Instead, Congress compromised by creating an industry
self-regulating body known as the Municipal Securities Rulemaking
Board (MSRB), made up of members from the investment community and general public. The MSRB was to establish fair practices for
underwriting and trading municipal bonds. As part of the system,
brokers and dealers had to register with the SEC, but government
When Subnational Debt Issuers Default
issuers were not required to follow any pre-issuance or post-issuance
filing requirements. The system was based on good-faith voluntary
disclosure.
After the WPPSS default in 1983, the SEC noted the many parallels
between the New York City crisis and the WPPSS default: offering documents did not disclose key facts about the borrowing, like accurate cost
estimates or demand projections; underwriters did not attempt to verify
disclosures but instead used the vague or incorrect information to market the bonds; Unit Investment Trusts purchased the bonds on the basis
of ratings that were widely known to be unrealistically high; and bond
counsel did not fully disclose potential legal problems. The SEC noted
that in both the New York City and WPPSS cases “neither the underwriters nor the rating agencies accepted responsibility for reviewing the
offering documents” (Ruder 1988).
In 1988, following a five-year investigation, the SEC decided not to
propose new legislation or authorize any enforcement action relating to
the WPPSS case, citing the “massive private damage litigation” already
underway (U.S. Securities and Exchange Commission 1988). The SEC
did adopt a new rule requiring underwriters to obtain an offering prospectus from public issuers for bond sales of US$1 million or more,
assess the document (known in the municipal market as an Official
Statement) in terms of truthfulness and completeness, distribute these
documents to potential purchasers, and file them with private entities
designated as repositories of municipal securities information. In other
words, again the SEC refrained from directly regulating the activities of
government issuers, but instead tried to effect this regulation indirectly
by creating responsibilities for underwriters. In addition to be being
indirect, the content and form of the disclosure document was not specified and therefore not regulated (except that the antifraud provisions of
the Security Act did prohibit materially misleading statements, even in
municipal offering documents).
In 1994, the SEC turned its attention to secondary market disclosure
issues, another problem that had arisen with the WPPSS securities flooding the market in the 1970s and early 1980s. As the problems of WPPSS
became known to Wall Street insiders, sophisticated investors looked for
ways to sell their holdings in the secondary market, where many investors knew little about the current financial situation of WPPSS, or any
367
368
Until Debt Do Us Part
other government entity whose securities they were buying. In 1994, the
SEC imposed another rule indirectly on government issuers by prohibiting underwriters from handling securities from issuers that did not
agree to provide annual reports with updated financial information
and timely reports of material events. These reports were to be sent to
municipal information repositories, where they would be available to
the general public.
The SEC’s use of indirect action illustrates the challenges of balancing the need to regulate the subnational debt market and the cost of
enforcing regulations. There was criticism of the SEC’s indirect securities regulation.16 The costs to the federal government of closely regulating the municipal market, however, would be exorbitant. It also raises
the issue of the federal and state relation.
The dependency on voluntary compliance has not affected the
growth of the municipal bond market. From the US$26 billion in new
municipal bonds sold in 1975 when WPPSS started planning Projects
4 and 5, the new issue market had grown sixfold by the time the SEC’s
report on WPPSS was issued in 1989. Today, the new issue market is
over US$400 billion, with 50,000 issuers and 40,000 daily transactions.
The corporate securities market has twice as much debt outstanding,
but only 8,000 issuers (Municipal Securities Rulemaking Board 2012).
The default rate in the municipal market remains low.
Lessons Learned
The WPPSS default offers lessons for governments in emerging
economies attempting to create subnational bond markets or to
accelerate the issuance of such debt for infrastructure investment purposes. The U.S. municipal market is often used as a model for such
efforts. It has powerfully attractive features, particularly in terms of
its ability to allocate huge amounts of capital for government projects, especially in infrastructure sectors. Three-quarters of this debt is sold
to individual investors, either directly or indirectly via mutual funds or
other investment vehicles. A municipal bond market is not just another
mechanism to intermediate savings to help finance much needed largescale infrastructure; it is also a way to effectively access the domestic savings of individuals and provide individuals with fixed-income
When Subnational Debt Issuers Default
investment returns. Finally, the fact that less than one-quarter of the
U.S. m
unicipal market in 2010 involved general obligation debt backed
solely by taxing power means that it is possible to back borrowing from
the revenues of the facilities to be built with the bond proceeds. Subnational governments can indeed raise money for investment in a commercial manner.
The WPPSS default, a rare default event in terms of scale and frequency in the modern U.S. subnational debt market, offers a window
into the interactive roles of the market, the courts, the regulators, the
debtor, the creditors, and taxpayers. Even in a developed functioning
market like the United States, the issuance of debt for infrastructure
has endemic risks that must be dealt with by the legal and regulatory
structure. WPPSS illustrates the risks of the lack of transparency and
disclosure, the risks of project analysis and feasibility projections, and
the risks of poor management and construction delays. These risks exist
in all infrastructure projects whether in developed or developing countries. Since debt is a legal obligation to pay, it must be lawfully authorized and enforceable. The role of the legal and regulatory environment
is to attempt to minimize such risks and to respond to risks that will
be identified from time to time, as has happened over the course of the
subnational debt market development in the United States.
The lessons revolve more around the fact that the U.S. municipal
market is a unique product of 200 years of experimentation. As shown
in chapter 14 in this volume, the U.S. municipal securities market has
developed gradually over a long period of time, with gradual and incremental reforms. For short stretches within that time span, for example
during the mid-19th century, the market was plagued by defaults and
other scandals. The market has had a considerable amount of time to
evolve and mature into the mechanism that it is today. Nothing illustrates this better than the fact that the U.S. municipal market showed
very little evidence of damage resulting from the WPPSS default. The
WPPSS default briefly pushed up municipal bond interest rates, but
only slightly. Not only did the market quickly return to normal after
the WPPSS default, but the period during which the WPPSS drama
unfolded, from 1975 to 1985, was one in which total annual municipal
bond issuance grew tenfold—the most dramatic decade of growth in
the history of the modern market.
369
370
Until Debt Do Us Part
Already by the 1980s, the municipal market had the size and
durability to survive a US$2.25 billion default,17 and also fully take
advantage of such a default by benefiting from the sort of disciplinary message sent to market participants that the financial and legal
consequences of mismanagement by and poor oversight of municipal
borrowers are real. Rescues and bailouts of such borrowers and their
bondholders are unlikely under most circumstances in the United
States. And even though the SEC took minimal enforcement action
against actors in the WPPSS drama, the principle of self-regulation did
not mean that these actors could avoid responsibility for their actions.
The amount of p
rivate damage litigation that followed for years in the
wake of the d
efault was unprecedented and resulted in many failed
careers and business collapses. Very few individual market participants
gained from the WPPSS disaster, but the market more than weathered
the storm. It b
ecame stronger as a result.
Most developing countries do not have the luxury of benefiting from
defaults in quite the same way. The risks of default on new subnational
issues, and the resulting risks of long-standing damage to nascent bond
market activity, are dangerously high. Many of these economies simply lack experience with debt issuance of any kind. Private companies
often raise capital by issuing equity rather than borrowing, because of
macroeconomic instability, lax bankruptcy laws, burdensome tax laws,
and government policy promoting share ownership. The sovereign debt
market may be in an early stage of development. The absence of longterm corporate or treasury debt means that benchmark yield curves are
not available for pricing long-term municipal bonds. Investment intermediaries are usually not familiar enough with municipal issuers to be
able to distinguish creditworthy borrowers from uncreditworthy borrowers. In such weak market environments, investor confidence in longterm municipal issues tends to be low. Subnational markets in many of
these economies are at risk of being shut down by early defaults.
As noted, revenue bonds issued by municipal corporations account
for over two-thirds of total subnational debt outstanding in the United
States. Outside the debt limitation imposed by state constitutions or
legal frameworks, municipal corporations have taken on major responsibilities in infrastructure investments. As developing countries take on
When Subnational Debt Issuers Default
infrastructure investments to meet the growth demand, WPPSS offers
a lesson in managing infrastructure projects with care and prudent
governance and management structure. It is important to highlight that
although municipal corporations have borrowing and management flexibility, they are subject to regulatory rules and market discipline. The
municipal corporations should not be used as a way of circumventing
borrowing rules. While there was no federal or state bailout in the case of
the WPPSS default, the desire to bail out in a developing country could
be strong, particularly if the project entails significant liabilities and
affects a large population.
Policy makers in developing countries need to identify a number of
basic ways in which issuers and investors are attracted to the municipal
bond marketplace. Many of these policy makers have been innovative
in trying to find proxies to achieve the same results, using the key characteristics of the U.S. market as targets to be achieved in some fashion if municipal bond market development is to be facilitated in their
countries. For example, careful prescreening of issuers helps to substitute for the ability possessed by many market players in the United
States to spot obvious problems with creditworthiness before issuance
(see Leigland 1997 for examples of this from Indonesia, Poland, and
South Africa).
One lesson of the WPPSS experience is that it is necessary to develop
subnational bond markets with care, and that it should not be assumed
that features of the U.S. market can simply be transplanted to developing countries. The WPPSS case happened in a country with a developed
and sophisticated system of legal jurisprudence, which should serve as
a caution to developing environments where the rule of law and legal
principles are much less developed and clear. Legal risk is a substantial
factor in debt markets. Although the regulatory framework for the U.S.
municipal markets cannot be directly replicated in other counties, some
of its most important features are generally applicable—creating clear
interest among creditors to support strengthening both the rule of law
and incentives for private market development, enhancing transparency
and disclosure of the credit risks of all issuers, and a no-bailout policy
to reduce moral hazard and enforce market discipline on debtors and
creditors.
371
372
Until Debt Do Us Part
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. This chapter draws mainly from the book WPP$$: Who is to Blame for the
WPP$$ Disaster by James Leigland and Robert B. Lamb (1986).
2. Public authorities are also called corporations, authorities, agencies, commissions, and so forth.
3. They can also be created by local governments.
4. For a history of revenue bonds and their relation to debt limitation set by states,
see chapter 14 by Liu, Tian, and Wallis (2013) in this volume.
5. When direct federal sponsorship of public hydropower development began to
taper off in the mid-1950s, the PUDs inherited a highly successful legacy of
hydropower expansion and became the focus of regional public power development. In Washington State the PUDs combined the legal autonomy and
revenue-debt-issuing capacity of public authorities with the taxing powers of
special districts, and had democratically elected boards. But because many of
them were small, they considered legislation enabling the formation of joint
ventures a necessary part of their role in developing the area’s power resources.
The desire to help PUDs develop the hydroelectric potential for the region was
the motive behind the state law authorizing the creation of joint action agencies. The locally based democratic control over PUDs was to be used to control
joint action agencies as well.
6. A study commissioned by the Washington State Legislature predicted in late
1981 a 1.5 percent annual growth rate, much lower than the 7.5 percent that
had justified the huge Hydro-Thermal Power Program.
7. For a detailed account of the legal decision and the legal sources for the court
decision, including the appeal to the U.S. Supreme Court (which declined to
review the Supreme Court rulings without comment), see Leigland and Lamb
(1986, 167–77). See chapter 14 by Liu, Tian, and Wallis in this volume for a discussion of these factors in American history more generally.
8. In the case of WPPSS Project 4 and 5 bonds, the trustee (Chemical Bank) in
August 1983 initiated the first in a long list of WPPSS-related lawsuits by suing
WPPSS, the participating utilities, and the BPA on behalf of the bondholders.
The charge was fraud in the sale and subsequent failure to pay interest on the
bonds.
9. Blyth Eastman Paine Webber Public Power Finance Group 1980.
10. One way of doing this was to encourage investors to view WPPSS bonds as backed
more by hydropower revenues from the first three plants supported by BPA than
by revenues from the nuclear power plants (BPA had no legal connection to the
bonds sold for the nuclear power plants). A second way was to issue short-term
When Subnational Debt Issuers Default
debt instruments rather than long-term bonds, which entails refinancing risks.
See Blyth Eastman Paine Webber Public Power Finance Group 1980.
11. In 1982, American Express Company, at the request of the WPPSS participants,
devised this direct federal rescue plan.
12. In March 1983, just prior to the Washington State Supreme Court decision
invalidating the take-or-pay contracts, members of the Washington Public
Utility Districts Association (most of whom were WPPSS participants) suggested a number of support options. One possibility offered was a regionalization plan, similar to the original BPA net-billing arrangements, which would
spread Project 4 and 5 debt through BPA billings to its regional customers,
including non-WPPSS participants and other service providers.
13. A special commission appointed by the governor of Washington State recommended the plan.
14. Federal Reserve Board, Flow of Funds, Accounts of the United States, 2005–2011,
June 7, 2012b, table L.104, p. 63. The term municipal bond market in the United
States includes bonds issued by states, cities, special purpose vehicles, and public authorities.
15. At the time, New York City and its corporate entities had US$14 billion in debt
outstanding, of which US$6 billion was short-term, requiring constant refinancing. The city also had an operating deficit of over US$2 billion. Underwriters
began to resist city efforts to sell more debt, particularly when a New York State
public authority, the Urban Development Corporation (UDC), defaulted on its
short-term debt. Although separate from the city, most UDC projects involved
city housing. When the state legislature moved to pay suppliers and contractors
but not bondholders, investors began to question what would happen to city
bondholders if the city defaulted, or worse, declared bankruptcy. The state government eventually stepped in and created the Municipal Assistance Corporation as
an independent corporate entity of the state empowered to control New York City
budgets and financing activities. New York City only narrowly avoided default.
16. See, for example, Seligman 1989.
17. Total state and local government securities outstanding at the end of 1980
were US$337 billion, so the WPPSS bonds account for less than 1 percent of
the market (Federal Reserve Board, Flow of Funds Accounts of the United States,
1975–1984, June 7, 2012a. Table L.104, p. 63).
Bibliography
Blyth Eastman Paine Webber Public Power Finance Group. 1980. Presentation to
Washington Public Power Supply System: A Balanced Financing Program. New
York: Blyth Eastman Paine Webber Inc.
Bond Buyer. 1985. The Bond Buyer’s Municipal Statbook 1984. New York: The Bond
Buyer.
373
374
Until Debt Do Us Part
Federal Reserve Board. 2012a. Flow of Funds Accounts of the United States, 1975–
1984. Washington, DC, June 7.
———. 2012b. Flow of Funds, Accounts of the United States, 2005–2011. Washington,
DC, June 7.
Lehmann, C. Richard. 1986. Interview conducted with the Treasurer of the National
WPPSS 4 and 5 Bondholders Committee. February 13.
Leigland, James. 1985. “Reforming the State’s Public Authorities.” New York Affairs
9 (2) (December): 50–57.
———. 1988. “Causes and Consequences of Management Failure in Public Enterprise: The Case of the Washington Public Power Supply System.” In Public
Enterprise: The Management Challenge, ed. Colm O. Nuallain and Roger Wettenhall, 15–32. Brussels: International Institute of Administrative Sciences.
———. 1992. “External Controls over Public Enterprise.” In Public Authorities and
Public Policy: The Business of Government, ed. Jerry Mitchell, 31–47. Westport,
CT: Greenwood Press.
———. 1994. “Public Authorities and the Determinants of Their Use by State and
Local Governments.” Journal of Public Administration Research and Theory
4 (4) (October): 521–44.
———. 1997. “Accelerating Municipal Bond Market Development in Emerging
Economies: An Assessment of Strategies and Progress.” Public Budgeting &
Finance 17 (2): 57–79.
Leigland, James, and Robert B. Lamb. 1986. WPP$$: Who Is to Blame for the WPPSS
Disaster. Boston, MA: Ballinger Publishing.
Liu, Lili. 2010. “Strengthening Subnational Debt Financing and Managing Risks.”
Economic Policy and Debt Department, World Bank, Washington, DC.
Liu, Lili, Xiaowei Tian, and John Joseph Wallis. 2013. “Caveat Creditor: State Systems
of Local Government Borrowing in the United States.” In Until Debt Do Us Part:
Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and Lili Liu,
539–90. Washington, DC: World Bank.
Marion, Joseph, and Francis J. Quinn. 1980. Competitive and Negotiated Offerings: The
Relative Merits of the Negotiated Method. A report prepared for WPPSS by Merrill Lynch White Weld Capital Markets Group, Merrill Lynch Pierce Fenner &
Smith Inc., November 24.
Moody’s. 2002. “Moody’s U.S. Municipal Bond Rating Scale.” November.
Municipal Securities Rulemaking Board. 2012. MSRB 2011 Fact Book. http://www
.msrb.org/msrb1/pdfs/MSRB2011FactBook.pdf.
Ruder, David S. 1988. “Underwriter Responsibilities in Municipal Bond Offerings after
WPPSS.” Remarks of the SEC Chairman before the 37th Annual dinner of the
Investment Association of New York, Waldorf Astoria, New York, September 22.
Seligman, Joel. 1989. “The Washington Public Power Supply System Debacle.”
Journal of Corporate Law 14 (4): 889–905.
Sitzer, Howard, and Leon J. Karvelis. 1981. “Washington Public Power Supply System: At the Crossroads.” Merrill Lynch Pierce Fenner & Smith Inc., Fixed
Income Research Department, New York, July 24.
When Subnational Debt Issuers Default
Tamietti, Robert L. 1984. “Chemical Bank v. WPPSS: A Case of Judicial Meltdown.”
Natural Resources Lawyer 17 (3): 373–96.
U.S. Securities and Exchange Commission. 1988. “Staff Report on the Investigation in the Matter of Transactions in Washington Public Power Supply System
Securities.” Washington, DC, September.
Wallis, John Joseph. 2005. “Constitutions, Corporations, and Corruption: American
States and Constitutional Change, 1842 to 1852.” Journal of Economic History
65 (1): 211–56.
Walsh, Annmarie H., and David Mammen. 1983. State Public Corporations: A Guide
for Decision Making. New York: Institute of Public Administration.
Washington Public Power Supply System. 1980. “Report on Implementation Status
of Consultant Recommendations—Parts I, II, and III.” WPPSS, Richland, WA,
August 8.
Washington State Senate Energy and Utilities Committee. 1981. “Causes of Cost
Overruns and Schedule Delays on the Five WPPSS Nuclear Power Plants.”
Vol. 1: Report to the Washington State Senate and the 47th Legislature.
Olympia, WA, January 12.
375
Part 3
Developing Subnational Debt
Markets
10
Transition from Direct Central
Government Onlending to
Subnational Market Access
in China
Lili Liu and Baoyun Qiao
Introduction
China has been investing about 10 percent of its gross domestic product
(GDP) annually in infrastructure since the 1990s, a much higher rate
than in many other developing economies. Under China’s decentralized fiscal structure, subnational governments (SNGs)1 have taken on
a large share of infrastructure investments, particularly in urban areas.
A substantial part of subnational urban infrastructure investments has
been financed by debt instruments. China’s large national savings, at
about 50 percent of GDP, has made the debt financing feasible.
The legal and institutional frameworks for subnational debt financing in China have undergone significant reforms gradually, in recognition of the fact that institutions, capital markets, and market access take
time to develop. The debt instruments have been evolving, from simple
to more sophisticated instruments. The development of regulatory and
institutional frameworks and of market debt instruments is characterized by learning by doing. The reform is still unfolding.
379
380
Until Debt Do Us Part
Prior to 2009, SNGs relied on central government onlending and
their own off-budget vehicles—Urban Development and Investment
Corporation (UDIC) borrowing directly from the financial markets—to
finance infrastructure investments. The UDIC started in the mid-1990s
in tandem with significant fiscal decentralization in China, and with the
1994 national taxation reform that delineated the spending and revenue
assignments between the national government and SNGs. In addition
to debt financing, UDICs have also extensively used land asset-based
finance to build urban infrastructure.
The central government’s onlending and UDICs’ direct borrowing
from the financial markets, together with land asset-based financing,
have helped transform the urban infrastructure landscape in China,
including the rapid development of transportation systems, power
systems, and water and sanitation systems. By the mid-2000s, the limitations of these financing instruments had become evident to policy
makers.
First, direct central government onlending to SNGs separates the
borrowing power (the central government) and the payment obligations
(SNGs). When the central government issues the debt and then onlends
the proceeds to an SNG, the SNG has no market interaction with creditors, and the market assigns the responsibility to the central government
and assumes an explicit guarantee by the central government. Second,
UDICs’ implicit off-budget debt and liabilities are nontransparent and
difficult to monitor. They create contingent liabilities for SNGs and
may also implicate the central budget. By 2010, the total subnational
debt outstanding in China was estimated at RMB 10.7 trillion, about
27 percent of GDP.2 More important, information on the subnational
debt was asymmetric between the central government and SNGs due
to the off-budget practices of UDIC borrowing.3 Third, financing infrastructure through land lease is not sustainable in the long run, because
of its one-time nature.4
Since 2009, China has undertaken substantial reforms to address the
above challenges. The reforms have consisted of three critical elements:
(a) providing a more direct link between borrowing and debt-service
responsibility by allowing the issuance of provincial bonds, and later
by piloting municipal bonds; (b) conducting a comprehensive audit of
UDIC off-budget debt and moving toward greater fiscal transparency
Transition from Direct Central Government Onlending to Subnational Market Access in China
by bringing off-budget contingent liability onto the subnational budget;
and (c) developing an institutional framework for regulating and managing subnational debt. These reforms have been augmented by other
reforms in fiscal management and better management of land assetbased financing.
The 2008–09 global financial crisis provided a broad context, and
an opportunity, for the transition toward market access for SNGs. To
cushion the impact of the crisis, China implemented a proactive fiscal policy—the key components of which included an RMB 4 trillion
economic stimulus package to promote public investments and a tax
cut to promote private investment and consumption, such as raising
the e xport tax rebate and reducing the tax burden on businesses and
residents. Most of the public investment programs under the stimulus
package were to be undertaken by SNGs. The implementation of the
fiscal stimulus brought fiscal challenges to the SNGs, because of the projected large financing gap between SNG expenditure and revenue.
There are two ways to close the financing gap through borrowing:
the central government can onlend to SNGs, or SNGs can access market
financing. A basic problem with onlending is the absence of a link
between debt issuance and the responsibility for debt service. China thus
took a decisive step to begin the transition from onlending to SNG market access. The reform had two objectives: (a) ensuring fast access to market financing by SNGs when dealing with the global financial crisis, and
(b) developing institutions for prudent management of subnational debt.
The State Council authorized the issuance of RMB 200 billion (US$30
billion) in provincial bonds in 2009.5 However, policy makers recognized
that important preconditions for the issuance of provincial bonds by
provinces did not exist. It would take time to develop credit rating systems; SNGs would need to learn market access, including auctions; and
the reform of the legal framework would need to carefully review the
1994 Budget Law, which restricts the access of subnationals to market
borrowing. Thus, the issuance of provincial bonds needed to proceed in
parallel with developing institutional, legal, and market infrastructure.
The reform thus took a gradual, learn-by-doing approach, with the central government as the issuing agency and SNGs participating in the
auctions to take advantage of the sovereign bond market experience and
lower-cost financing. An additional RMB 200 billion (US$30 billion) of
381
382
Until Debt Do Us Part
provincial bonds was authorized and issued in 2010 and again in 2011.
In 2011, reform took a further step: the State Council approved piloting of direct bond issuance by four cities (RMB 23 billion of municipal bonds issued) without the central government acting as the issuing
agency.6
The consequences of the reform are significant. The reform provided
the instrument for the SNGs to finance capital investment and interact
with market creditors; improved the intergovernmental fiscal relationship, particularly on capital expenditure; enabled the accumulation of
lessons from experiments; and facilitated the process of establishing a
regulatory framework for subnational debt management.
The transition from central government onlending to market access
has been a challenge for many countries. Market development requires
certain preconditions and coordinated reforms such as developing
credit rating systems for subnational borrowers and establishing a regulatory framework for subnational debt management. Equally important, it takes time for SNGs to accumulate experience in, and develop
capacity for, market access.
This chapter focuses on China’s experience in developing market
access through a series of coordinated reforms. The remainder of the
chapter is organized as follows. Section two discusses the framework
for subnational borrowing in China prior to 2009. Section three presents the transition from direct central government onlending to market a ccess in China since 2009. Section four reviews the experiences and
implications of the transition. Section five provides concluding remarks.
Frameworks for Subnational Borrowing
in China Prior to 2009
Over the past two decades, China has been investing about 10 percent of
GDP in infrastructure annually, a much higher rate than that of many
developing economies. Large-scale investments in urban infrastructure—
power, roads, railways, bridges and tunnels, water systems, and sanitation
facilities—have facilitated rapid urbanization. The fiscal decentralization,
started in the early 1980s and formalized in 1994 with the Tax Sharing
System reform, granted major responsibility to SNGs in capital investments and operations of urban infrastructure.
Transition from Direct Central Government Onlending to Subnational Market Access in China
Taxation and fiscal transfers were not sufficient to finance the
large-scale urban infrastructure transformation required to accelerate
economic growth in China. In addition to own revenues and fiscal
transfers, debt financing has been important to SNGs. Prior to 2009,
various financing channels were utilized, including central government
onlending and off-budget vehicles.
Central Government Onlending
According to Article 28 of China’s 1994 Budget Law, subnational budgets at various levels should be balanced; that is, expenditures shall not
exceed revenues, and budget deficits shall be restricted. Consequently,
SNGs should not finance expenditures through borrowing, such as
from banks or by issuing bonds, except as otherwise prescribed by law
or approved by the State Council.
Various formal financing channels had existed for SNGs. Since the early
1980s, the central government could borrow externally (for example, from
international financial institutions and bilateral sources) and onlend to
SNGs.7 The Road Law of 1998 allows SNGs to raise funds for the construction of toll and nontoll roads and other transportation infrastructure.8
From 1998 to 2004, the Chinese Ministry of Finance (MOF) onlent
funds to provincial governments by issuing treasury bonds as part of the
countercyclical fiscal policy to mitigate the impact of the Asian financial crisis.9 The fiscal policy was designed to expand domestic demand
and stimulate economic growth. The onlent proceeds were required
to be used mainly for investment in infrastructure such as large-scale
construction projects as defined by the central government.10 The MOF
and the subnational financial departments were creditor and debtors,
respectively. SNGs had no market interaction with creditors.
UDIC Borrowing
The demand for infrastructure due to rapid industrialization and
urbanization increased an imbalance between subnational expenditure
responsibilities and their revenue assignment, which motivated SNGs
to seek different sources of financing. The UDICs—investment and
financing platforms of SNGs—have since the mid-1990s become the
main instrument for financing subnational infrastructure investments
and construction.11 Their rapid growth has helped close the widening
383
Until Debt Do Us Part
financing gap between SNG expenditure responsibilities and revenue
sources (figure 10.1).
In general, the UDICs have satisfied subnational financing demand
without violating the Budget Law. The UDICs, controlled and financed
by different levels of subnational government, are mainly responsible
for capital investment and maintenance of infrastructure. Their financing sources come from land asset-based revenues, government equity,
user charges, onlending from Treasury bonds, the central government
subsidy, and government guarantees. The UDICs have borrowed from
the financial markets (mainly the banking system) and issued bonds to
raise funds. Bank loans have been the predominant debt instrument,
but bond issuances have also grown, particularly since the late 2000s.
The State Council in 2004 granted the UDICs access to more
financing channels and further encouraged expanding UDICs.12 To put
subnational bonds in China in international perspective, figure 10.2
represents subnational bond issuance from 2000 to 2009 by the top five
countries outside the United States.13 Subnational bond issuance by
UDICs in China grew rapidly—from less than US$10 billion in 2002
to over US$60 billion in 2009. These bond issuances excluded the RMB
200 billion (US$30 billion) of provincial bonds issued by the central
government on behalf of provinces in 2009.
Figure 10.1 Revenue and Expenditure of Subnational Governments in China, as
Percentage of Total Government Revenue and Expenditure, 1985–2010
90
80
70
60
50
40
30
20
10
0
19
8
19 5
86
19
8
19 7
88
19
8
19 9
90
19
9
19 1
92
19
9
19 3
94
19
9
19 5
96
19
9
19 7
98
19
9
20 9
00
20
0
20 1
02
20
0
20 3
0
20 4
0
20 5
06
20
0
20 7
08
20
0
20 9
10
Percent
384
Year
Subnational revenue (%)
Source: National Bureau of Statistics 1993, 2011.
Subnational expenditure (%)
Transition from Direct Central Government Onlending to Subnational Market Access in China
385
60
09
20
20
20
20
20
20
20
20
20
20
Percent
65
0
08
70
10
07
75
20
06
80
30
05
85
40
04
90
50
03
95
60
02
100
70
01
80
00
US$, billion
Figure 10.2 Top Five Countries Issuing Subnational Bonds, 2000–09
(Excluding the United States)
Year
Spain
Canada
China
Share of top 5 issuers
Japan
Federal Republic of Germany
Source: Canuto and Liu 2010, with data source from DCM Analytics.
Note: In the United States, annual average issuance of subnational bonds from 2007 to 2011 was about US$450 billion
(data source: Thomson Reuters).
To mitigate the impact of the 2008–09 global financial crisis, China
launched an RMB 4 trillion fiscal stimulus package in late 2008. SNGs
provided matching funds as part of the package, which spurred the
rapid expansion of the UDICs. By 2010, total subnational debt (loans
and bonds) outstanding in China amounted to RMB 10.7 trillion, about
27 percent of GDP.14
Although UDICs significantly improved the urban infrastructure,
their rapid growth caused both a significant increase in local debt and
new challenges, such as discretionary operations and risky guarantees
from the SNGs. To control the increasing debt risk, the central government required stricter regulation of SNG liabilities.15 It required that
different levels of governments clean up UDIC liabilities and apply different treatment strategies to different types of debt to ensure credit
safety.
Land Asset-Based Financing
SNGs have also utilized land assets as an important source of financing
infrastructure. Land asset-based financing is an important ingredient of
386
Until Debt Do Us Part
SNG finance in many countries. Land frequently is the most valuable
asset on the asset side of subnational balance sheets. There are various
instruments for converting public land rights to cash or infrastructure,
including “capital” land financing via direct sales of land and using
land as collateral for borrowing (a practice that has a long history of
financing urban investment). Land is often the most important public contribution to public-private partnerships that build transit lines,
airports, or other large infrastructure projects. Beyond physical land,
rights to more intensive land development—a higher Floor Space Index
or higher Floor Area Ratio—may also be sold by public development
agencies. These “excess density rights” in effect represent the publicly
controlled share of privately owned land. The development rights have
economic value that can be sold by public authorities.16
SNGs in China have actively used land asset-based financing since
the mid-1990s.17 There are two important aspects to such financing:
(a) land asset-based financing and UDICs are linked—the UDICs use
various instruments of land asset-based financing to convert land value
into infrastructure assets;18 and (b) UDICs have used publicly owned
land as collateral for borrowing to finance investment.19 (See section
four for a discussion on land financing.)
Informal Practices of Subnational Borrowing
Given the constraints placed on SNG formal borrowing, SNG informal
borrowing has grown, and it can take the form of arrears on wages and
on what is owed to suppliers. A uniform statistical framework for subnational debt is lacking. Apart from the technical issues, the inaccuracy
of statistical data may also be attributed to different motivations on the
part of local governments, such as overstating the size of liabilities to
win potential assistance from the higher-level government, or understating it to demonstrate governing performance. And the measurement
becomes more problematic when taking into account implicit and contingent liabilities.
One example of such informal borrowing is the accumulation of
rural education debt. China started promoting nine-year compulsory
education in rural areas throughout the country in the mid-1980s. For
the next 10 years, local governments (towns and villages), with limited fiscal resources, resorted to borrowing to build or improve school
Transition from Direct Central Government Onlending to Subnational Market Access in China
387
facilitates to help meet the minimum facility standards for all schools.
In 2000, nine-year compulsory education became universal and China
achieved a historic improvement in education. Meanwhile, the accumulation of rural compulsory education debt had become a significant fiscal burden on local governments. The aggregate rural education debt
was about RMB 110 billion20 at the end of 2007 (3.9 percent of aggregate
subnational own fiscal revenue, excluding transfers), of which about
RMB 80 billion was used to finance capital expenditure, and about RMB
30 billion was used to finance operational expenditures. The central
government later restructured the rural education debt with innovative
reforms (see chapter 2 by Liu and Qiao in this volume).21
Toward a New System of Financing
The financing channels prior to 2009 contributed to the rapid urbanization and transformation of the urban infrastructure landscape in China.
Table 10.1 shows the rapid growth of urban infrastructure in China.
China also has the world’s second-largest highway network, the world’s
most densely trafficked railway network, the largest high-speed rail
network, the world’s three longest sea bridges, and three of the world’s
four-largest container ports.22
Table 10.1 Urban Infrastructure Development in China: Selected Indicators, 1990–2009
Indicators
1990
1995
2000
2009
Population density of city districts
(persons/square kilometer)
279.0
322.0
442.0
2147.0
Water consumption for residential use
(100 million cubic meters)
100.1
158.1
200.0
233.4
Consumption of natural gas for residential use
(100 million cubic meters)
11.6
16.4
24.8
91.3
Coverage rate of urban population with access to gas (%)
19.1
34.3
45.4
91.4
Area of centralized heating (100 million square meters)
2.1
6.5
11.1
38.0
Length of city sewage pipes (10,000 kilometers)
5.8
11.0
14.2
34.4
Number of public vehicles under operation at year end
(buses and trolley buses, etc.) (10,000 units)
6.2
13.7
22.6
37.1
Per capita area of parks and green land (square meters)
1.8
2.5
3.7
10.7
6,767.0
10,671.0
11,818.9
15,733.7
Volume of garbage disposal (10,000 tons)
Source: National Bureau of Statistics 2010.
388
Until Debt Do Us Part
The limitations of the financing models prior to 2009 became e vident
in the mid-2000s, as did the conditions facilitating market access.
First, it was recognized that the direct central government onlending
to SNGs separates the borrowing power (the central government) and
the payment obligations (SNGs). Basically, debt issued by the central
government and onlent to SNGs is difficult to classify as either central
or subnational debt. In practice, the onlending funds were included in
neither the central nor the subnational budget. SNGs have no market
interaction with creditors.
When the security markets, particularly the sovereign debt market,
were in their nascent stage of development, the central government’s
targeted onlending instruments would help meet the urgent funding
needs without waiting for the eventual development of capital markets.
As the sovereign debt market was developing and maturing, it helped
lay a solid foundation for the development of the subsovereign debt
market. In the 2000s, China’s sovereign debt market was developing fast.
Second, UDICs’ off-budget debt and liabilities are nontransparent
and difficult to monitor. This type of debt not only creates contingent
liabilities for SNGs, but it may also implicate the central budget. More
important, the information on subnational debt was asymmetric
between the central government and the SNGs due to SNG off-budget
borrowing.
By 2009, the Chinese government was ready to address the contingent liability issue. The fiscal system after the tax-sharing reform in
1994 delineates the intergovernmental fiscal relation between the c entral
government and SNGs. During the 15 years of development since 1994,
the division of fiscal revenue and responsibility between the c entral government and SNGs had become clearer. In particular, SNGs take the
responsibility for local economic development, which makes it possible
for the hard budget constraint between the central government and
SNGs. This implies that SNGs have become more independent in dealing with risks.
At the same time, a series of reforms in fiscal budget management
(for example, department budget reform and the national treasury single account system reform launched in 2001;23 and budget expenditure,
particularly project expenditure reforms in 2005) improved the
transparency of budget management and facilitated subnational bond
Transition from Direct Central Government Onlending to Subnational Market Access in China
financing in the capital markets. Moreover, reforms in 2003 related
to the state-owned asset management system affirm governments’
property rights on state-owned enterprises. Meanwhile, the reforms
also affirm that the ownership of a public enterprise would be separated
from the management of the public enterprise, so that the management
can follow commercial and market principles.24 The regulations on
state-owned property further clarify the rights and obligations of stateowned assets. And these policies could help prevent public institutions
from transferring debt risks to government agencies.25
Third, financing infrastructure through land lease is not sustainable
in the long run, because of its one-time nature.26 The practice of UDIC
borrowing through securitization of land values magnifies the borrowing risks, because land values decline during periods of economic stress,
when it is more difficult to finance loan repayments. There is potential
for heightened systemic risk when the entire subnational sector relies
heavily on land and land values to provide security for subnational
borrowing.27
There are tremendous benefits from granting SNGs access to
financial markets, and debt financing would be a more efficient and
equitable approach to finance infrastructure (Liu 2008). For the central
government, the benefits include (a) strengthening fiscal transparency
of SNGs, (b) helping the central government assess the fiscal capacity of
SNGs and supervise subnational budgets, (c) providing information for
the determination of intergovernmental transfers, and (d) providing a
self-sustaining mechanism that clearly delineates the rights and obligations of borrowed funds and decreases the soft budget constraint.
For the SNGs, the benefits of accessing capital markets would
include (a) providing SNGs with more financing sources for infrastructure investments, (b) creating profit- and risk-sharing mechanisms
between creditors and debtors, and (c) exposing SNGs to market discipline and reporting requirements, which helps improve government
accountability.28
The capital market might also benefit from subnational bond
financing. Compared to corporate bonds or stocks, investments in
subnational bonds can be relatively less risky, rewarding investors with
stable and attractive returns. Subnational bond financing provides
additional financial products in the securities market, diversifying the
389
390
Until Debt Do Us Part
securities structure. Subnational bonds can better fit the asset and liability structure of such investors as life insurance companies, mutual
funds, and pension funds. In addition, issuing subnational bonds
might not only increase securities volume, but also add new elements
to diversify the securities markets.29
Expanding Market Access since 2009
Since 2009, China has moved forward with developing subnational
capital markets in three significant ways: (a) allowing the issuance of
provincial bonds in 2009, 2010, and 2011 totaling RMB 600 billion
(approximately US$90 billion), (b) piloting the direct issuance of
municipal bonds by four cities in 2011, and (c) developing an institutional framework for managing subnational debt.
Issuance of Provincial Bonds
The 2008–09 global financial crisis provided a broad context, and an
opportunity, for the transition toward market access. The State Council authorized the issuance of provincial bonds in 200930 as part of the
large-scale stimulus package. SNGs were to provide RMB 2.82 trillion in
investments, accounting for 71 percent of the RMB 4 trillion fiscal package, to match the RMB 1.18 trillion in central government investments.31
The issuance of provincial bonds in 2009 differs from the period
1998–2004, when the MOF issued treasury bonds and onlent the proceeds to SNGs for financing infrastructure, as part of the countercyclical
fiscal policy to mitigate the impact of the Asian financial crisis. In 2009,
provincial bonds would become a direct source of financing for SNGs.
The Chinese government began to prepare for expanding subnational market access well before the 2008–09 global financial crisis. As
noted, a series of reforms in budget management and in the state-owned
asset management system had laid the groundwork. The government
also conducted research on international experience in subnational debt
management32 and an audit of UDIC debt. However, allowing provinces
to directly issue bonds would require preconditions, many of which,
such as subnational credit ratings, were absent and would take time to
complete. Provinces had no experience with the securities markets, such
Transition from Direct Central Government Onlending to Subnational Market Access in China
as dealing with underwriters, dealers, and auctions. It also takes time to
build a unified management system of subnational bonds.
The issuance of provincial bonds thus took a gradual approach. Instead
of letting provinces directly issue bonds, the central government issued
RMB 200 billion (US$30 billion) in bonds in 2009 on behalf of the provincial governments. These bonds were booked directly as provincials’ own
debt. Essentially, subnational bonds in 2009 were issued by the MOF acting as the issuing agent through the existing channels of treasury bonds,
with the participation of SNGs in market activities such as auctions.
On the one hand, the reform could build a close link between debt
issuance and payment responsibility and allow SNGs to acquire experience in market access. On the other hand, the process of issuance made
full use of the rich experience of the MOF in the Treasury bond market,
including the ministry’s matured techniques and its established relationships with investors. The national Treasury market has developed
rapidly. As of 2011, the outstanding balance of T-bonds reached RMB
6.8 trillion, ranking second in Asia and sixth globally.33 The process also
lowered the financing cost for SNGs and improved issuance efficiency.
In addition, investors received a guarantee that they would receive principal and interest on time.
In general, the 2009 SNG bond is a tradable book-entry bond. The
provincial governments34 were the issuers and debtors. On behalf of
provinces, the MOF acted as the issuing agency and pays the debt s ervice
and the issuance fee from the respective provincial escrow account
within the MOF.35 The term of all bonds was three years, and the annual
payments and the interest rate were priced through public bidding. The
funds financed by subnational bonds cannot be used to finance current
expenditure,36 and should mainly invest in the areas that match central
government investment on public projects that had difficulty attracting
private investments.37 The bond proceeds cannot finance commercial
projects that can obtain private financing.38
Among the different models of subnational debt management
(Ter-Minassian and Craig 1997), China followed the model of Direct
Administrative Controls for the subnational bonds. The central government allocated the RMB 200 billion quota of bond issuance among
the SNGs. The allocation of the quota followed a designated formula,
391
392
Until Debt Do Us Part
i ncluding the required subnational matching funds for projects in which
the central government invested, funds required for subnational infrastructure projects, and the fiscal capacity of repayment. The required
subnationals’ matching funds represented the most important factor
in the formula. To calculate the fiscal capacity of repayment in the formula, considerations included the debt ratio, the estimated growth rate
of subnational revenue, and the fiscal capacity of each jurisdiction.
The middle and western regions had larger infrastructure demands
than those of the coastal regions. Consequently, the 2009 central government stimulus package gave more weight to projects in the middle and
western regions. The SNGs in these regions had fewer fiscal resources
to provide the matching funds. To achieve a better performance of the
stimulus package, it was necessary to provide more quotas to the middle
and western regions. The amount of issued bonds for the eastern, middle, and western regions was RMB 60.3 billion, RMB 64.7 billion, and
RMB 75 billion, respectively, accounting for 30, 32, and 38 percent of
the total bonds issued.
Within the approved quotas and the framework set by the central
government, SNGs are responsible for choosing and appraising projects,
formulating their budgets, and obtaining approval from their respective People’s Congress. The MOF then issues SNG bonds, following the
issuance plan agreed with the SNG, on the basis of “the ready one will
be issued” principle to raise funds for the SNG quickly, and promote the
effective implementation of the fiscal policy.
The MOF formulated a series of regulatory rules with respect to the
inclusion of debt service as part of subnational budget management.39
In particular, to strengthen budget administration, the MOF enacted
the “Regulation of the Budget Administration of the 2009 Subnational
Government Bond Funds,” which requires that the proceeds from the
issuance of the SNG bonds and the debt service be incorporated into
subnational budget management. The adjusted budget proposal should
be submitted for approval of the People’s Congress Standing Committee at the same subnational level. In addition, the departments or units
as the users of bond funds should incorporate the expenditure into the
department’s or unit’s budget.
The MOF also established an improved system for project application, approval, performance assessment, and supervision to enhance
Transition from Direct Central Government Onlending to Subnational Market Access in China
the efficiency of fund administration. Under the system, the SNGs must
accept inspection of how the supervision department and audit department use the funds. Those violating the relevant regulations, such as by
intercepting and misappropriating the bond funds from the designated
use, will be penalized in accordance with the Fiscal Offense Punishment
Act (State Council, No. 427 in 2004).
To ensure that the SNGs actually fulfill their repayment responsibility, the solvency of a city, county, department, and unit as the user of the
bonds was taken into account to assess the default risk. The repayment
schedules were projected to ensure the availability of sources for timely
payment of principal, interest, and issuance cost to the MOF. In addition, penalty rules were implemented. Each SNG, as a debtor, should
comprehensively arrange its integrated financial resources and carry the
debt responsibilities. In particular, it should pay the bond principal and
interest and issuance fees to the MOF on time. Those that failed to pay
on time must pay penalty interest to the MOF,40 and, if the principal,
interest, issuance cost, and penalty interest are overdue, the MOF could
withhold the amount during the annual settlement between the MOF
and the SNG.
Developments since 2010
The subnational bonds issued in 2009 greatly strengthened the fiscal
capacity of local governments to provide matching funds and expand
infrastructure investment. However, challenges emerged with respect to
subnational autonomy in spending the bond proceeds and the maturity
of the bonds. Although the regulations clearly required SNGs to allocate
bond funds to finance those projects also invested in by the central government, some SNGs allocated the bonds funds to local projects. Meanwhile, the bond maturity term of three years was less attractive to the
underwriters and put repayment pressures on some SNGs.
To address these issues, the MOF took the following additional steps
in 2010:
•
Bonds with a five-year maturity were issued to meet the market
demand and the fiscal capacity of SNGs in the medium term. The fiveyear maturity bonds totaled RMB 61.6 billion of RMB 200 billion
of bonds in 2010.
393
394
Until Debt Do Us Part
•
•
•
The MOF granted SNGs more fiscal autonomy in using the bond
funds, not only for the matching part of the projects the central government invested in, but also for the projects invested by the SNGs,
as long as these local projects followed the investment priorities specified by the central government.
The quota for allocating the bonds was adjusted, while keeping the
total issuance of subnational bonds at RMB 200 billion in 2010 (the
same as in 2009).41
Efforts were made to reduce the issuance cost by joint issuances of the subnational bonds for different jurisdictions and to
improve the liquidity by increasing the volume of each issuance.
The investor base of the subnational bonds has been expanded to
all book-entry underwriters of the treasury bonds, and the coordination between underwriters and SNG finance departments
has been strengthened. In 2011, the same amount of subnational
bonds was issued. Table 10.2 summarizes the subnational bonds
issued during 2009–11 (for 2011, table 10.2 includes the municipal
bonds directly issued by four cities as part of RMB 200 billion, as
explained next).
In 2011, the State Council authorized four cities (Beijing, Shanghai,
Shenzhen, and Zhejiang) to pilot municipal bond issuance.42 These
municipal bonds differ from UDIC bonds; the former are secured by
the full faith and credit of the municipal government issuer, and the latter are secured by the proceeds from the project and from specific land
assets.43 These municipal bonds also differ from the 2009–11 provincial bonds. The municipal bonds were issued in 2011 directly by four
cities without the central government acting as the issuing agency. It
is one step further toward SNGs being fully responsible for their debt
payments.
The municipal bonds directly issued by the four cities were included
in the quota set by the central government. In fact, the maturity, bidding
methods, and investor base of the municipal bonds were not different
from those of the provincial bonds. The yield of these municipal bonds
was 5–10 basis points lower than the Treasury bonds (see table 10.3);
this reflects inefficient price formation in debt markets, which are still at
the development stage.
Transition from Direct Central Government Onlending to Subnational Market Access in China
395
Table 10.2 Subnational Bonds Issued in China, 2009–11
RMB billions
2009
Region
Amount
2010
% of total
Amount
% of total
Amount
% of total
Sichuan
18.0
9.0
Guangdong
10.9
5.45
Henan
8.8
Jiangsu
8.4
Yunnan
8.4
4.2
7.5
3.75
7.9
3.95
Zhejiang
8.2
4.1
8.0
4.0
8.0
4.0
Hunan
8.2
4.1
8.9
4.45
8.9
4.45
Hubei
8.1
4.05
8.6
4.3
8.6
4.3
Anhui
7.7
3.85
8.9
4.45
9.0
4.5
Shanghai
7.6
3.8
7.1
3.55
7.1
3.55
Shandong
7.0
3.5
6.9
3.45
7.3
3.65
Liaoning
6.6
3.3
6.0
3.0
6.0
3.0
Ganshu
6.5
3.25
5.5
2.75
5.9
2.95
Guangxi
6.5
3.25
5.5
2.75
6.0
3.0
Guizhou
6.4
3.2
5.4
2.7
5.4
2.7
Shan’xi
6.3
3.15
6.3
3.15
6.8
3.4
Jiangxi
6.2
3.1
6.5
3.25
7.0
3.5
Hebei
6.0
3.0
6.9
3.45
7.3
3.65
Helongjiang
6.0
3.0
6.9
3.45
6.9
3.45
Chongqing
5.8
2.9
4.9
2.45
5
2.5
Inner Mongolia
5.7
2.85
5.9
2.95
5.9
2.95
Beijing
5.6
2.8
5.4
2.7
5.4
2.7
Jilin
5.5
2.75
6.3
3.15
6.3
3.15
Xingjiang
5.5
2.75
6.0
3.0
6.0
3.0
Shanxi
5.3
2.65
5.1
2.55
5.1
2.55
Fujian
3.4
1.7
6.1
3.05
5.5
2.75
Ningxia
3.0
1.5
1.8
0.9
2.6
1.3
Hainan
2.9
1.45
2.5
1.25
2.9
1.45
Qinghai
2.9
1.45
3.3
1.65
3.9
1.95
Tianjing
2.6
1.3
2.5
1.25
2.5
1.25
200
100
200
100
200
100
Total
18
2011
9.0
13.5
6.75
9.1
4.55
9.1
4.55
4.4
9.3
4.65
9.3
4.65
4.2
8.9
4.45
8.9
4.45
Source: MOF.
Note: Within the RMB 200 billion of bonds in 2011, Shanghai, Guangdong, Zhejiang, and Shenzhen directly issued
RMB 7.1 billion, RMB 6.9 billion, RMB 6.7 billion, and RMB 2.2 billion, respectively.
396
Until Debt Do Us Part
Table 10.3 Municipal Bonds Directly Issued by Four Cities in 2011
3-year maturity
City
5-year maturity
Date of issuance
Volume
Yield (%)
Volume
Yield (%)
Shanghai
November 15, 2011
3.6
3.10
3.5
3.30
Guangdong
November 18, 2011
3.45
3.08
3.45
3.29
Zhejiang
November 21, 2011
3.3
3.01
3.4
3.24
Shenzhen
November 28, 2011
1.1
3.03
1.1
3.25
Sources: Data compiled based on DF Daily 2011 and Shanghai Security 2011.
Institutional Reforms
Institutional reforms started even before the 2009 decision to issue provincial bonds, including fiscal budget management and transparency,
and separating management from ownership in government-owned
enterprises (section two). The new bond instrument has aimed at creating a framework for medium-term capital budgeting for infrastructure investments. Past budgeting practice did not separate financing of
capital budgeting from that of current expenditure. Long-term capital
investments will need to be financed through debt for intergenerational
equity. With the new instrument, SNGs can use the debt to finance
capital outlays under newly developed budgeting procedures. Other
institutional reforms since 2009 have covered, among other things,
UDICs and developing a monitoring system for fiscal risks stemming
from subnational debt. Provinces and cities have also launched reform
experiments.
A major reform was put forward by the State Council in 2010 on the
regulation of UDICs.44 The directive authorized the audit of UDIC debt
and classified the UDIC debt into three categories: (a) debt issued for
public purposes and implicitly securitized by budget revenues, (b) debt
issued for public purposes but securitized by the revenues generated by
the project being financed by debt, and (c) debt issued for commercial
purposes.
The classification of the debt is significant in laying the groundwork
for the future development of different debt instruments, such as general obligation bonds and revenue bonds, and would limit the debt
financing to public investments, which has become a general principle
Transition from Direct Central Government Onlending to Subnational Market Access in China
in the subnational debt regulatory framework in many countries (Liu
and Waibel 2008). The directive also specified the management of the
UDIC, differentiating UDICs that are financially self-sustainable from
those that rely on budget support. Furthermore, bank loans to UDICs
will need to follow credit risk analysis, and guarantees provided to
UDICs would be regulated.
The government is also studying international experience in developing indictors to monitor risks from subnational debt and establishing debt limits to guide subnational entities’ investment and borrowing
plans. International experience shows that there is a trade-off in establishing debt limits; setting the thresholds too tight can hamper growth
by severely restricting subnational infrastructure investment, while loosening thresholds can endanger macroeconomic and financial stability by
encouraging excessive subnational borrowing (Liu and Pradelli 2012).
Provinces and cities have also piloted reforms. Many city governments, including Guiyang, Kunming, Shanghai, and Xian, have audited
their own UDIC debt and are developing institutional frameworks to
better link debt financing with capital investment plans and the master
plan of the city and the medium-term fiscal framework.
Experience and Implications of Expanding
Market Access
The immediate objective of the transition to market access was to establish
a formal financing channel through which debt instruments could be used
quickly to finance subnational capital expenditure, particularly to finance
SNG matching funds for the large-scale projects identified by the central
government as key to domestic integration and for which the central government also invested its own funds. The long-term objective is to develop
transparent and sustainable channels of financing for SNG capital investments and to support economic growth.
Transitional Instrument of Subnational Bonds
The issuance of provincial bonds was successful in establishing a new
instrument to finance subnational capital expenditure. It helped smooth
the economic cyclicality through the central stimulus package by providing SNGs with matching funds for the projects in which the central
397
398
Until Debt Do Us Part
government co-invested. Meanwhile, it promoted regional d
evelopment,
especially in the less developed regions. For example, Xinjiang allocated
the bond proceeds mainly for building hospitals and schools in seismic
areas, and Sichuan ensured smooth reconstruction after the Wenchuan
earthquake in May 2008. More important, China took the first step in
establishing a budget framework to guide capital budgeting management, because the subnational bonds are incorporated into the budget
management, in particular the financing of the capital budget. The SNGs
for the first time have direct market interaction with creditors and are
responsible for the debt repayments.
Notwithstanding the achievements, the issuance of provincial bonds
in 2009–11 was transitional in nature. The subnational bonds took
advantage of the central government’s strong credit, since the MOF was
the issuer on behalf of SNGs. This essentially lowered the cost of borrowing, which was important when implementing a large-scale fiscal
stimulus package. The average interest of the three-year bonds issued
in 2009 was only 1.91 percent and that of five-year bonds was only
2.96 percent,45 which was significantly lower than that of commercial
banks. Meanwhile, there was no significant difference in the interest
rates across jurisdictions. Subnational bond interest rates did not reflect
the diverse levels of economic development of different jurisdictions
and the underlying credit conditions of subnational debtors.
The instrument was only the first step in clarifying the relationship
between debt issuance and debt service obligations, in that the debtors
of subnational bonds are the provincial governments, and the central
government remains the issuer. Thus, the central government has to
assume responsibility for the debt in case of default,46 although the SNG
escrow accounts with the MOF and the annual transaction settlement
would greatly reduce the risks to the central government. The pilot in
municipal bond issuance directly by four cities without sovereign guarantee represents a further step in linking debt issuer and debt service
obligation.
The 2009–11 reform could be regarded as the beginning steps toward
achieving the implicit objective of developing formal financing channels
of subnational capital investments. Fundamentally, the factors contributing to the fast growth of subnational debt in China can be classified
into the demand side and the supply side.
Transition from Direct Central Government Onlending to Subnational Market Access in China
399
Strong Demand for Subnational Debt in China
A number of factors have contributed to the strong demand for subnational debt instruments since the 1990s. These factors include rapid
urbanization, rapid growth in urban land value, rapid economic growth,
and the large financing gap of subnational government fiscal accounts.
Most of these factors are expected to continue, underlining the significance of reform undertaken by the Chinese government in reforming
the management of subnational debt financing.
Urbanization. China has been undergoing rapid urbanization and
industrialization since the 1990s. The percentage of urban residents in
the total population increased from 26 percent in 1990 to 48.3 p
ercent
in 2009; urban residents more than doubled, from 302 million to about
645 million during the same period (figure 10.3). This has significantly
challenged the ability of SNGs to provide public services, such as power,
water and sanitation facilities, parks, bridges and roads, subways, and
telecommunication networks. Fiscal transfers and taxation alone cannot adequately finance the scale of infrastructure investments required
to accommodate the pace of urbanization. In a supply-demand analytic framework, urbanization shifts the demand curve of subnational
700
600
500
400
300
200
100
Year
Urban residents (%)
Source: National Bureau of Statistics 2010.
Urban population (millions)
20
10
08
20
20
06
04
20
20
02
00
20
98
19
96
19
94
19
19
19
92
0
Urban population, millions
60
55
50
45
40
35
30
25
20
15
10
5
0
90
Urban residents, %
Figure 10.3 Rapid Urbanization in China, 1990–2010
Until Debt Do Us Part
borrowing to the right, resulting in the large-scale demand for subnational debt.
The land-asset effects. The use of land assets to secure financing is
prevalent in China. In fact, land-asset revenues have been one of the
most important extrabudgetary financing sources for SNGs in China.
Under Chinese law, governments hold exclusive ownership of land and
are able to exercise power over land asset-based financing. As shown in
figure 10.4, the annual amount of land-transfer fees collected in China
has grown rapidly,47 especially in 2007 and 2009. SNGs in China have
successfully financed infrastructure through land asset instruments. As
noted, land asset-based financing is an important ingredient of SNG
finance in many countries (Petersen and Kaganova 2010). In China,
investment financing from proceeds from land leasing and public bank
lending securitized by land and property valuation accounts for 80–
90 percent of SNG infrastructure financing (Liu 2008).
However, the land financing policy created a huge increase in the perceived wealth of SNGs. Since all lands are publicly owned, SNGs perceive
appreciation of the value of their assets due to the rising real estate prices
in recent years.48 The perception of the growth in wealth is further augmented by the one-time nature of land transactions in China that allows
SNGs to collect 50- or 70-year leasing fees up front.49 International
Figure 10.4 Annual Land Transfer Fee in China, 2004–09
1,800
1,600
1,400
RMB, billions
1,200
1,000
800
600
400
200
Year
Source: Ministry of Land and Resources, China.
09
20
08
20
07
20
06
20
05
20
04
0
20
400
Transition from Direct Central Government Onlending to Subnational Market Access in China
experience shows that during the high-growth period of urban development, publicly owned land often has been used as collateral for borrowing to finance subnational public investment. The expectation that
land values will increase with urban growth has made land an attractive
asset for loan collateral, for both public borrowers and private lenders.
However, this practice magnifies the risks, because land values decline
in periods of economic stress, when it is most difficult to finance loan
repayments. There is a potential for heightened systemic risk when the
entire subnational sector relies heavily on land and land values to provide security for subnational borrowing (Peterson and Kaganova 2010).
Growth expectations. Growth expectations come from the optimistic
view of local officials about the prospect of development of the local
economy and fiscal revenue. As shown in figure 10.5, China has experienced decades of rapid economic growth. Consequently, subnational
government revenues in China show a strong growth trend. Subnational
officials might perceive that this growth would continue for a relatively
long period and that more revenues will be generated in the future.
Thus, subnational officials tend to overborrow, because they believe that
the resulting deficit will be temporary and will be covered in the foreseeable future (see, for example, Wang, Xu, and Li 2009).
Figure 10.5 GDP Growth Rates and Subnational Revenue, 1991–2010
40
30
Percent
20
10
0
–10
–20
–30
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
–40
Year
Real GDP
Source: National Bureau of Statistics 2011.
Note: GDP = gross domestic product.
Nominal GDP
Subnational revenue
401
402
Until Debt Do Us Part
Subnational financing gap. One of the major fiscal issues in China is
the increasing disparity between subnational government expenditure
responsibilities and revenues50 (as manifested in figure 10.1), which has
contributed to the use of debt instruments to close the financing gap.
Several factors with respect to intergovernmental fiscal relations have
contributed to the imbalance of expenditure and revenue (MartinezVazquez and Qiao 2010).
First, SNGs bear a heavy expenditure burden (table 10.4), especially
county-level governments taking the main responsibility of basic service provision such as basic education, health care, and social welfare.
Notably, the assignment of social security services to subprovincial
governments helped generate operational deficits and implicit liabilities. In addition, with the extensive cosharing or concurrent responsibilities, the upper-level governments have assigned unfunded or
insufficiently funded mandates to the lower levels.
Table 10.4 Relative Shares of Expenditure at Different Government Levels in
China, 2003
percent
Central
Provincial
Prefecture
County
and under
Total
30
18
22
30
Capital investment
44
23
22
11
Agriculture expenditure
12
46
11
30
8
15
18
60
63
23
9
5
Expenditure
Education
Scientific research
Health care
3
22
32
43
Social security
11
39
32
18
Government administration
19
11
22
48
Public security and procuratorial
agencies, court of justice
National defense
Foreign affairs
Foreign aid
Other
Source: Martinez-Vazquez and Qiao 2010.
5
25
34
35
99
1
0
0
87
13
0
0
100
0
0
0
29
16
25
31
Transition from Direct Central Government Onlending to Subnational Market Access in China
Second, while decentralizing public service provision responsibilities, the tax-sharing reform in 1994 allowed SNGs to share major
taxes, such as the value-added tax and the personal and corporate
income tax.51 Although they have some taxation powers, such as a
business tax, in general, SNG taxation power is limited. Moreover,
the property tax system, which generally serves as an important revenue source for SNGs in international practice, is still in its embryonic form. Table 10.4 shows that SNGs bore the increasing burden of
decentralized expenditure responsibilities (over 90 percent in education and health), but the share of their revenues from various sources
has never exceeded 50 percent of total government revenue for the
past decade.
Third, although the intergovernmental transfer system mitigates the
fiscal imbalance to some extent, there are still significant fiscal gaps for
some SNGs, particularly those in less developed jurisdictions. As one of
the most important components in the transfer system, the tax rebate
depends heavily on historical record instead of on the true fiscal gap
between responsibilities and revenues.
In addition, the lack of accountability, imperfect budgetary management, and feeding finance52 in some jurisdictions also contributed
to the increasing fiscal gaps (Qiao and Shah 2006; Zhu 1999). The current criteria for performance evaluation put a relatively high weight on
GDP growth, so subnational officials are motivated to expand public
investment and attract private investment. Cai and Treisman (2004)
give specific examples of this competition for investments, which may
compromise subnational tax collection and lead to more subnational
debt financing. The central government has taken steps to improve the
effectiveness and transparency of the budget.
The Supply Side of Debt Instruments
The structure of the financial market has an important bearing on the
growth and risk control of subnational debt in China. With rapid economic growth, China has made significant strides in the development
of financial markets. There are remaining challenges, two of which—
the governance structure and the excessive liquidity in the financial
market—are particularly relevant for the further development and
reform of the subnational credit market.
403
404
Until Debt Do Us Part
On the issue of governance structure in the financial sector, market discipline will need to play an important role in regulating the debt financing
of SNGs. Market discipline will require fiscal transparency from debtors, a
risk control system for creditors, and the expectation of no bailout in the
event of default. The institutional capacity to manage fiscal transparency
is gradually developing in China, and the independent credit rating system
is in the early stages of development. The MOF is developing an in-house
credit-rating system to audit and monitor the size of the subnational debt
liabilities. The National Development and Reform Commission has worked
with Standard and Poor’s to undertake the credit ratings of select UDICs.
On the risk control of creditors, many loan decisions in the past have
been made without proper risk assessment, due to a lack of sufficient
risk regulations in the banking system. The lack of comprehensive data
for subnational fiscal accounts has contributed to weak risk control in
assessing the quality of lending to the subnational sector. Moreover, facing competition, banks may ignore signals of possible insolvency and
tolerate the financial risk of subnational insolvency since the soft budget
constraint problem strengthens their expectation of a bailout (Wu, Wu,
and Liu 2008). The State Council Directive of June 10, 2010, aims at
strengthening credit risk analysis. The Development Bank of China, a
major subnational lender, has increased emphasis on risk control.53
The interventions of some local officials in the loan decisions have
also encouraged the growth of subnational debt. Local officials in China
still have various channels through which to influence bank lending
decisions. Most financial institutions in China are government owned
and need the support of SNGs for their local operations (Qiao 2012). In
particular, subnational government officials may play a role in promoting executives of local branches of some state-owned financial institutions (Wang, Xu, and Li 2009).
China has made significant progress in reforming its financial markets, which has resulted in increased competition among governmentowned banks, development of a sovereign debt market, development
of an interbanking system, and development of nonbanking financial
institutions. Nonetheless, interest rates remain controlled, making it difficult to properly price risks and returns. Consequently, the true costs
of capital are not reflected and the supply curve of debt services has
become relatively flat.
Transition from Direct Central Government Onlending to Subnational Market Access in China
International experience shows that an efficient central government bond market is important for an SNG bond market to function
efficiently. China has the potential to improve the efficiency of the
government bond market, which underpins the growth of diversified
and sophisticated fixed-income markets. Future reform areas include
strengthening coordination among government entities in developing
and implementing regulations, improving the Inter-Bank Bond Market
and the repo market, and developing a diverse investor base.54
Demand for investment opportunities by individual and institutional
investors appears to be increasing due to high savings rates and excessive
market liquidity. It is well known that one of the driving forces of the
Chinese economy is its high household savings rate (Xie 2011). Increasing personal income and the high savings rate have led to an increasing
trend of household deposits (Yu 2006). Thus, strong demand for investment instruments exists among individual and institutional investors.
The increasing savings deposits accumulated in financial institutions
induces their strong investment demand.
The Budget Law restricts SNGs from running deficits without
approval of the State Council, so SNGs set up UDICs that borrow
mainly from state-owned banks. Since there were no clear regulatory
and operational rules concerning UDIC borrowing, both SNGs and
banks could be influenced by the perception of the potential for bailouts
(Jia 2012).
The effects of all these factors can be viewed more clearly in the
demand-supply graph of debt services, as illustrated in figure 10.6. In
the figure, lines D and S represent the initial debt demand and supply
curves, respectively. Line D shows the SNG demand for debt to finance
capital expenditure driven by the growing urbanization process. Efficient capital markets supply funds according to correctly priced risks
and returns, represented in the supply curve S. Lines S and D jointly
form the desirable equilibrium E0. Other demand-side factors distort E0
by shifting the demand curve to the right, indicated by the new demand
curve D’.
The supply-side factors have two impacts on the original supply
curve S. These factors first move the supply curve to the right, since they
increase the debt quantity for any given capital costs. Moreover, the factors have an extra impact on preventing the markets from pricing risks
405
Until Debt Do Us Part
Figure 10.6 Formation of Debt Equilibrium, China
D’
Price/interest rate
406
D
A
S
Demand
E3
Supply
E0
E1
S’
E2
O
Qa
Overborrowing
Debt-quantity
Source: Qiao 2012.
and returns correctly, consequently shifting the supply curve flatter.
The two effects form the new supply curve S’. The new demand curve
and supply curve determine the new equilibrium E1. The new equilibrium lies to the far right of the desirable equilibrium, indicating a large
increase in the debt service costs. Overborrowing by SNGs is the difference between quantities of the equilibrium E1 and E0.
The Budget Law establishes a binding ceiling of debt quantity that is 0
or close to 0 over the market, forcing the entire market to be hidden. In
figure 10.6, the ceiling policy is indicated by line A, which sets a binding
quantity of Qa, and all quantities above Qa are hidden.
Summary
Subnational debt in China is unlikely to pose systematic macroeconomic
risks. The subnational debt outstanding in China was estimated at RMB
10.7 trillion at end-2010, about 27 percent of GDP. China’s sovereign debt
accounted for 17 percent of GDP at end-2010.55 The combined public
(sovereign and subsovereign) debt was below 50 percent of GDP, lower
than in many other countries. China’s central government has been running low fiscal deficits in the past 10 years, and 2.5 and 2 percent of GDP
in 2010 and 2011, respectively.56 Furthermore, the institutional reforms
(see section three) aim at developing a comprehensive regulatory framework to strengthen the management of subnational debt.
Transition from Direct Central Government Onlending to Subnational Market Access in China
The rapid urbanization and industrialization and the large subnational fiscal gap demanded subnational borrowing, and the subnational
borrowing significantly improved infrastructure and accelerated economic growth. A more efficient market of subnational government debt
will facilitate the sustainable financing of infrastructure.
On the demand side, the SNGs have shown a relatively high degree
of administrative discretion in expenditure and are capable of influencing decisions of financial institutions. On the supply side, the banking
system operates in an environment of excessive liquidity and tends to
yield to pressures from the demand side. These have encouraged the
overspending and overborrowing behavior of SNGs in China, bringing unregulated risks. Consequently, it is necessary to improve both the
overall fiscal system and the local incentive system to effectively block
the informal borrowing.
The market of subnational government debt also needs more transparency. Since the direct borrowing by SNGs is restricted by the budget
law, borrowing by off-budget vehicles creates a hidden market that lacks
transparency, setting hurdles for the functioning of fiscal constraints
and market discipline. The situation deteriorates under imperfect budget management and financial governance, which induce the potential
expectation of a bailout. In this market, SNGs conduct debt financing
without an effective regulatory mechanism, and investors make supply
decisions without prudent considerations of risks. Thus, allowing direct
borrowing by SNGs and market access by UDICs, under transparent and
rule-based regulatory frameworks, would help make subnational debt
finance in China more sustainable in the long run.
Concluding Remarks
China successfully established a new instrument of subnational bonds
to finance subnational capital expenditure through the transition from
direct central government onlending to market access. The reform was
timely in meeting subnational needs to help finance the subnational
matching part of investment projects in which the central government
co-invested in response to the 2008–09 global financial crisis. In particular, recognizing that certain preconditions did not exist for direct market
access by SNGs, China adopted a pragmatic and innovative approach by
407
408
Until Debt Do Us Part
initiating the subnational bonds by the central issuance. The approach
significantly lowered the financing costs for SNGs, enabled them to start
acquiring market access skills, and linked the SNGs as debtors with their
debt service obligations. Piloting municipal bonds, without the central government as the issuer, is an additional step for SNGs to directly
assess the market, which would provide experience and lessons for further reforms.
Urbanization and industrialization will continue unabated, requiring continuing capital investments by SNGs in large-scale urban
infrastructure to support sustained economic growth. Subnational
debt instruments will continue to play a vital role in economic transformation. Although bank loans are currently the most important
debt instrument, China already has the largest subnational bond market among developing countries, and the market is likely to continue
to expand.
China has made important progress in institutional reforms and
moving toward a more transparent fiscal framework. The reforms
in fiscal management (including the single Treasury account and
expenditure reforms) and in separating management from ownership of public enterprises have laid the groundwork for the piloting
of provincial bonds. The new bond instrument to finance capital outlays under newly developed budgeting procedures will facilitate the
development of a framework for medium-term capital budgeting for
infrastructure investments. The audit of, and the ongoing efforts to
better classify, UDIC debt would facilitate the development of different bond instruments with different risks and securitization profiles.
Continuing reforms need to address two main challenges. The first
challenge is to transform direct lending from the central government
to SNGs into SNG market borrowing. Issuing provincial bonds is a
first step in that direction. Since the central government is the issuing
agency, one challenge is to mitigate the potential bailout expectations of
both the SNGs and investors in the long run. The second challenge is
to develop regulatory frameworks for UDIC direct borrowing from the
financial markets and managing the contingent fiscal risks from UDICs.
Subnational governments may circumvent the regulations by issuing
UDIC bonds. Reform options would include incorporating the liabilities
of UDICs into subnational budgets, providing debt service provisions in
Transition from Direct Central Government Onlending to Subnational Market Access in China
the budget for UDICs that are not financially self-reliant, and disclosing
contingent liabilities through financial statements as annexes to subnational government budgets.
In addition to developing regulatory frameworks for subnational
debt—such as ex-ante rules regulating types and purposes of debt and
procedures for issuing debt, and recourse for insolvent subnationals
and UDICs—reforms in broader areas will support sustainable market
access. These would include strengthening intergovernmental fiscal systems and public financial management, enhancing fiscal transparency,
deepening financial sector market reform, and improving the efficiency
of the government debt market (or, more broadly, fixed-income markets) and the capability of SNGs to manage their own issuance program.
The success of subnational debt financing requires a strong institutional
foundation in terms of subnational financial planning, budgeting, debt
management, and a credit rating system.
China has the potential to accelerate subnational debt market development, while managing risks to the macroeconomic frameworks and
financial system. China’s strengths include (a) a stable macroeconomic
framework, including an impressive growth record (economic growth
is a key determinant of debt sustainability); (b) large domestic savings,
which provide capital supply to the financial markets; (c) rapid urbanization, which can be facilitated by financial market long-term financing; (d) a decentralized fiscal structure; (e) infrastructure companies
organized along infrastructure networks and the adoption of cost recovery goals, both of which are critical for developing a deep revenue bond
market; and (f) a track record of reforms.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. The term subnational refers to all tiers of government below the central government. The category also includes special purpose vehicles or investment companies created by SNGs.
2. “Auditing Report 2010,” No. 35, National Audit Office of the People’s Republic
of China.
409
410
Until Debt Do Us Part
3. SNGs have better information regarding their debt and may not have incentives
to reveal that information.
4. The land-use-right transfer fee is a one-time payment made by land users to
obtain urban land-use rights for a given period of time, usually about 70 years
for residential use and 50 years for commercial use.
5. “Announcement on Issuance of 2009 Subnational Bonds, State Council” 2009,
No. 2. These bonds are equivalent to general obligation bonds; that is, debt is
secured by the general revenues of a province, including taxation and fiscal
transfers.
6. “Announcement on the Experiment of 2011 Municipal Bonds Issuance,” Ministry of Finance (Treasury) 2011, No. 141.
7. Loans from foreign governments or from international financial institutions.
In accordance with regulations, on behalf of the Chinese government, the Ministry of Finance could take out a loan and give the proceeds to central agencies
and SNGs for domestic expenditure.
8. Article 21 of the Road Law (1998) states that in raising funds for highway construction, the government at all levels may raise funds for highway construction, including collecting charges or seeking loans from domestic and foreign
financial organizations or foreign governments according to law and relevant
provisions of the State Council.
9. 1998 was the first time that China implemented the proactive fiscal policy.
The State Council decided that the MOF should onlend a certain amount of
external national debt to provincial jurisdictions to help finance construction
projects in order to expand domestic demand and promote stable economic
growth.
10.
The investments included urban infrastructure, environmental protection,
developing and upgrading urban and rural power grids, agriculture, forestry,
water conservancy, and transportation.
11. The UDICs started in 1992. That year, to promote the development of Pudong
New District in Shanghai, the Shanghai Chentou UDIC, with the permission of
the central government, issued a bond for Pudong New District Construction
in the amount of RMB 500 million.
12. “The Decision on the Reform of Investment System,” State Council, No. 20,
2004.
13. The United States continues to be the largest subnational bond market. From
2007 to 2011, annual average issuance of total subnational bonds was about
US$450 billion (Source: Thomson Reuters), and at the end of 2011, outstanding debt was US$3.743 trillion (“Flow of Funds Accounts, Flows and Outstandings,” Federal Reserve Board, Fourth Quarter 2011).
14. “Auditing Report 2010,” No. 35, National Audit Office of the People’s Republic
of China.
15. “Notice about Strengthening the Management of UDICs,” (Document No.19),
issued by the State Council on June 13, 2010.
Transition from Direct Central Government Onlending to Subnational Market Access in China
16. For more on land asset-based financing, see Peterson and Kaganova (2010).
17. According to the China Index Academy, land-leasing contract revenues set a
record in 2009, with Hangzhou (RMB 105.4 billion in contract revenues),
Shanghai (RMB 104.3 billion in contract revenues), and Beijing (RMB 92.8 billion in contract revenues) leading the way. In Beijing’s case, the land-leasing
contract value in 2009 was equal to 45 percent of total fiscal revenue. (Source:
Xinhua News Agency, February 5, 2010. See also Peterson and Kaganova 2010).
18. For example, for an entire new commercial and residential zone, or for a hightech zone (see Peterson and Kaganova 2010).
19. This is often used during the high-growth period of urban development (see,
for example, in France, in Peterson and Kaganova 2010). The expectation that
land values will increase with urban growth has made land an attractive asset
for loan collateral, both for public borrowers and private lenders.
20. The exchange rate at the time of writing of the RMB to the U.S. dollar was RMB
6.30 to US$1. The RMB has appreciated continuously since 2005, at about 3–5
percent per year.
21. See chapter 2 “Restructuring of Legacy Debt for Financing Rural Schools in
China” by Liu and Qiao in this volume.
22. UDICs construct and maintain most transport infrastructure, with the exception
of the rail network, which is the responsibility of the Ministry of Railways. (Data
from the World Bank and Development Research Center of the State Council,
the People’s Republic of China, 2012; and Amos, Bullock, and Sondhi 2010).
23. In accordance with the relevant provisions of the state, the treasuries at various
levels of government must manage promptly and accurately the collection, allocation, withholding, and turnover of budgetary revenues, and the appropriation of budgetary expenditures.
24. This facilitates the separation of the government function from the commercial
function, so the public administration risks can be eliminated and the management of the public enterprise reflects commercial risks.
25. “Interim Regulations on Supervision and Management of State-owned Assets
of Enterprises,” State Council, No. 378, 2003.
26. The land-use-right transfer fee is a one-time payment made by land users for
obtaining urban land-use rights for a period of time, usually about 70 years for
residential use and 50 years for commercial use.
27. See Peterson and Kaganova 2010.
28. In the United States, subnational bond financing is subsidized through a federal
income tax exemption.
29. For more, see Liu 2010.
30. See note 5.
31.
National Development and Reform Commission, http://www.sdpc.gov.cn
/xwzx/xwtt/t20090521_280383.htm (in Chinese).
32. The preparation included an international conference organized by the MOF
in 2008 in Hangzhou and a set of policy research reports by the MOF on
411
412
Until Debt Do Us Part
international experience in managing subnational debt (see, for example, Li,
Xu, and Li 2009; and Zhang et al. 2008).
33. “International Conference on the Development of Sound Secondary Market
for Government Securities in China,” February 2012, Beijing.
34. Including provinces, autonomous regions, four municipalities, and five specifically designated cities.
35. The bonds were named by the debtor governments specifically as “2009 government bond of XX province (xx batch).”
36. “Regulation of Budgetary Management of 2009 Subnational Bonds,” MOF,
2009.
37. Such as the housing project for low-income families; rural livelihood projects
and rural infrastructure; health care, education, and culture, and other social
welfare infrastructure; ecological construction; earthquake recovery and reconstruction; and other projects related to people’s livelihood.
38. “Regulation of Budgetary Management of 2009 Subnational Bonds,” MOF,
2009.
39. The rules include the “Regulation of Budgetary Management of 2009 Subnational Bonds,” “Regulations on the Issuance and Payment of 2009 Subnational
Bonds Issued by the MOF on Behalf of SNGs,” the “Financial Budget Accounting Regulations of the Subnational Bonds Issued by the MOF on Behalf of
Subnational Governments,” and “Management Regulations of Project Arrangements for 2009 Subnational Bond Funds.”
40. Penalty interest = overdue payment × (coupon rate × 2 ÷ days in a year) ×
overdue days. Liquidated damages = overdue payment × (coupon rate × 2 ÷
days in a year) × overdue days.
41. The amounts issued to the eastern and western regions were reduced by RMB
1.2 billion and RMB 4 billion, respectively, and the amount allocated to the
middle region was increased by RMB 5.2 billion, based on the adjusted capital
investment plan.
42. “The Announcement of Insurance of Municipal Bonds,” 2011, MOF.
43. Therefore, the municipal bonds issued by the four cities are similar to general
obligation bonds. The UDIC bonds are secured by the revenues generated by
the project or securitized by land assets; in reality, the lenders view the bonds as
implicitly guaranteed by the government that owns the UDIC.
44. “State Council Directive on Management of Urban Development and Investment Corporations,” No. 19, June 10, 2010.
45. Source: Ministry of Finance.
46. “Regulations on the Issuance and Payment of 2009 Subnational Bonds Issued
by the MOF on Behalf of SNGs.”
47. The leasing fees collected by SNGs for leasing land.
48. Real estate prices are rising for various reasons. On the one hand, the unprecedented scale of urbanization creates a rising demand for real estate, which
causes property values to appreciate, giving SNGs a strong incentive to
Transition from Direct Central Government Onlending to Subnational Market Access in China
participate in the land market. On the other hand, the fact that SNGs reap
the benefit of land asset-based financing also contributes to the rising cost
of housing, pushing the price higher and further strengthening the wealth
effects.
49. In China, SNGs typically charge land users a transfer fee for 50-year usage for
commercial development and 70-year usage for residential development.
50. For more on the issue of subnational fiscal autonomy and accountability, see
Martinez-Vazquez and Qiao 2010.
51. Provincial governments receive 25 percent of value-added tax revenue and
40 percent of personal and corporate income tax revenue. Provincial governments decide the tax-sharing formula with their own subprovincial government
units. The administration of shared taxes is centralized.
52. The fiscal expenditures of some less developed areas are concentrated on
administration costs, especially on the expenses of local employees.
53. Based on an interview with an official in a local branch of the China Development Bank in 2010. The local branch has a risk analysis department that is separate from its lending activities, and one risk factor has been factored in during
the lending decision—the total amount of UDIC debt in a local jurisdiction
against the total revenues of that local government.
54. See World Bank and International Monetary Fund 2011.
55. Data for subnational debt are from “Auditing Report 2010,” No. 35, National
Audit Office of the People’s Republic of China. Data for sovereign debt are from
the China Statistical Yearbook (National Bureau of Statistics 2011).
56. Data for 2010 are from the China Statistical Yearbook (National Bureau of Statistics 2011). Data for 2011 are calculated by authors based on MOF data.
Bibliography
Amos, Paul, Dick Bullock, and Jitendra Sondhi. 2010. High-Speed Rail: The Fast
Track to Economic Development? Washington, DC: World Bank.
Cai, Hongbin, and Daniel Treisman. 2004. “State Corroding Federalism.” Journal of
Public Economics 88 (March): 819–43.
Canuto, Otaviano, and Lili Liu. 2010. “Subnational Debt Finance: Make It Sustainable.” In The Day After Tomorrow: A Handbook on the Future of Economic Policy
in the Developing World, 219–38, ed. Otaviano Canuto and Marcelo Giugale.
Washington, DC: World Bank.
DF Daily. 2011. “New Lower Yield for the Directly Issued Municipal Bonds.” November 22.
http://finance.sina.com.cn/stock/t/20111122/011510856886.shtml.
Federal Reserve Board of the United States. 2011. “Flow of Funds Accounts, Flows
and Outstandings.” Fourth Quarter.
“International Conference on the Development of Sound Secondary Market for
Government Securities in China.” 2012. San Ya, February 23–24.
413
414
Until Debt Do Us Part
Jia, Kang. 2012. “Transparency, Hidden Rules, and Hidden Subnational Debts.”
DF Daily, March 4, http://finance.ifeng.com/money/roll/20120304/5698869
.shtml.
Li, Ping, Hongcai Xu, and Cheng Li. 2009. Subnational Debt Management: International Experience and Comparison. Beijing: China Finance and Economics Publisher, March.
Liu, Lili. 2008. “Creating a Regulatory Framework for Managing Subnational
Borrowing.” In Public Finance in China: Reform and Growth for a Harmonious Society, ed. Jiwei Lou and Shuilin Wang, 171–90. Washington, DC:
World Bank.
———. 2010. “Strengthening Subnational Debt Financing and Managing Risks.”
Review of Economic Research, Ministry of Finance, Beijing, August 16, 46 F-9.
Liu, Lili, and Baoyun Qiao. 2013. “Restructuring of Legacy Debt for Financing Rural
Schools in China.” In Until Debt Do Us Part: Subnational Debt, Insolvency, and
Markets, ed. Otaviano Canuto and Lili Liu, 81–108. Washington, DC: World Bank.
Liu, Lili, and Juan Pradelli. 2012. “Financing Infrastructure and Monitoring Fiscal Risks at the Subnational Level.” Policy Research Paper 6069, World Bank,
Washington, DC.
Liu, Lili, and Michael Waibel. 2008. “Subnational Borrowing, Insolvency and Regulations.” In Macro Federalism and Local Finance, ed. Anwar Shah, Washington,
DC: World Bank.
Martinez-Vazquez, Jorge, and Baoyun Qiao. 2010. “Assessing the Assignment of
Expenditure Responsibilities.” In China’s Local Public Finance in Transition, ed.
Joyce Yanyun Man and Yu-Hung Hong, 21–40. Cambridge, MA: Lincoln Institute of Land Policy.
National Bureau of Statistics. 1993. China Statistical Yearbook 1993. Beijing: China
Statistics Press.
———. 2010. China Statistical Yearbook 2010. Beijing: China Statistics Press.
———. 2011. China Statistical Yearbook 2011. Beijing: China Statistics Press.
Peterson, George E., and Olga Kaganova. 2010. “Integrating Land Financing into
Subnational Fiscal Management.” Policy Research Working Paper 5409, World
Bank, Washington, DC.
Qiao, Baoyun. 2012. “China’s Approach to Stimulating and Financing Rapid
Regional Infrastructure Development and Economic Growth: A Perspective of
Debt Financing for Capital Expenditure.” In Fiscal Decentralization in Indonesia a Decade after the Big Bang, 289–317, Ministry of Finance, Republic of
Indonesia, Directorate General of Fiscal Balance. Jakarta: University of Indonesia Press.
Qiao, Baoyun, and Anwar Shah. 2006. “Local Government Organization and
Finance in China.” In Local Governance in Developing Countries, ed. Anwar
Shah, 137–68. Washington, DC: World Bank.
Shanghai Security. 2011. “Shenzhen Municipal Bond Continued the Low Yield.”
November 29, http://finance.jrj.com.cn/2011/11/26010011653277.shtml.
Transition from Direct Central Government Onlending to Subnational Market Access in China
Ter-Minassian, Teresa, and Jon Craig. 1997. “Control of Subnational Government
Borrowing.” In Fiscal Federalism in Theory and Practice, ed. Teresa Ter-Minassian
156–74. Washington, DC: International Monetary Fund.
Wang, Xianbin, Xianxiang Xu, and Xun Li. 2009. “Provincial Governors’ Turnover
and Economic Growth: Evidence from China.” China Economic Quarterly 8
(4): 1301–28.
World Bank and Development Research Center of the State Council, the People’s
Republic of China. 2012. China 2030: Building a Modern, Harmonious, and
Creative High-Income Society. Conference Edition. Washington, DC: World
Bank.
World Bank and International Monetary Fund. 2011. “Financial Sector Assessment Program.” Washington, DC, November.
Wu, Wenfeng, Chongfeng Wu, and Xiaowei Liu. 2008. “Political Connection
and Market Valuation: Evidence from China’s Individual-controlled Listed
Firms.” Economic Research Journal (7): 130–41.
Xie, Yong. 2011. “Research on Factors Impacting Chinese Household Saving.”
Ph.D. thesis, Nanjing University.
Yu, Wentao. 2006. “The Analysis and Countermeasures in Rapid Growth of Chinese Household Savings.” Journal of System Reform (10): 14–19.
Zhang, Zhihua, Ya Zhou, Lifeng Yin, Yi Liu, Xiaoming Yan, and Zhihao Chen. 2008.
“International Experiences on Scale Control and Risk Warning of Subnational
Debt.” Review of Economic Research (22): 8–10.
Zhu, Boming. 1999. “The Research on ‘Feeding Finance.’” Public Finance Research (3):
43–45.
Government of China Documents
“Announcement on the Experiment of 2011 Municipal Bonds Issuance.” 2011. Ministry of Finance (Treasury), No. 141.
“Announcement of Insurance of Subnational Bonds.” 2009. Ministry of Finance.
http://www.gov.cn/gzdt/2009-03/17/content_1261497.htm.
“Announcement on Issuance of 2009 Subnational Bonds, State Council.” 2009. No. 2.
“Auditing Report 2010,” No. 35, National Audit Office of the People’s Republic of
China.
“Budget Management Regulations of 2009 Subnational Bonds.” 2009. Ministry of
Finance, No. 21.
“Financial Budget Accounting Regulations of the Subnational Bonds Issued by the
Ministry of Finance on Behalf of Subnational Governments.” 2009. Ministry of
Finance, No. 19.
“Interim Regulations on Supervision and Management of State-owned Assets of
Enterprises.” 2003. State Council, No. 378.
“Management Regulations of Project Arrangements for 2009 Subnational Bond
Funds.” 2009. Ministry of Finance, No.121.
415
416
Until Debt Do Us Part
“Regulations on the Issuance and Payment of 2009 Subnational Bonds Issued by the
Ministry of Finance on Behalf of Subnational Governments.” http://www.cnki
.net.cn/Article/CJFDTotal-YSKJ200905018.htm.
“State Council Directive on Strengthening the Management of Local Government
Financing Platform Companies.” 2010. No. 19, June 13.
“The Decision on the Reform of Investment System.” 2004. State Council, No. 20.
11
The Philippines: Recent
Developments in the Subnational
Government Debt Markets
Lili Liu, Gilberto Llanto, and
John Petersen
Introduction
This chapter reviews access of Philippine subnational governments
to the credit markets, impediments to such access, and recent developments in Philippine subnational finances. The Local Government
Code of 1991 commenced decentralization and defined the structure
of subnational government units in a unitary system. In this chapter, the term Local Government Unit (LGU) in the Philippines, used
interchangeably with subnational governments, includes all tiers of
the government under the central government.1 Our findings have
benefited from several recent reports that bear on the various issues
covered in this chapter.2
In the Philippines, LGU borrowing is low compared to borrowing by
subnational governments in many other countries. As will be discussed
in the chapter, LGUs are carefully monitored by the central government
both on their individual debt transactions, which are almost exclusively
done with four Government Financial Institutions (GFIs),3 and by regular reporting to the Philippine Department of Finance (DOF). There is
no evidence that LGU borrowing has been used to cover operating deficits or to finance unusually large, speculative projects.
417
418
Until Debt Do Us Part
The LGUs appear to live under a “hard” budget constraint, and the
sector as a whole typically runs a budget surplus. Part of the smaller
appetite of LGUs for taking on debt has to do with the assignment of
service responsibilities to the various levels of local governments. Most
major infrastructure projects are controlled and funded at the national
government level. Aside from a few major cities and the nation’s capital
region around Manila, there is a diffuse scattering of small and often
rural governments that are highly reliant on transfers from the central
government. In addition to the small scale and diffusion of local governments, there are also significant institutional and managerial barriers to their planning and managing major building projects.
Subnational governments in the Philippines were largely unaffected
by the 2008–09 global financial crisis. LGUs, heavily reliant on national
government transfers, felt little impact (they were more affected by two
large tropical storms). While future national government payments to
them will be slightly affected (the distribution formula has a three-year
lag), the LGUs have continued to operate with an annual surplus.
Overall, local governments in the Philippines are light borrowers and
appear to restrict lending to relatively small projects or to meeting occasional cash flow needs. By and large, the LGU sector has relatively smallscale and pedestrian (albeit individually important) capital financing
needs. The low level of indebtedness is attributed to the limited functions
assigned to LGUs that require infrastructure spending, the reluctance of
local governments to borrow, and the impact of various financial oversight mechanisms. Some areas are changing, however, such as the use of
project financing and restricting the security on loans to specific projects.
Notwithstanding a decade and a half of policy planning and innumerable reports, implementation has been slow in following through
on an initial planning framework designed to move LGUs into private financing markets. Earlier, the Local Government Unit Guarantee Corporation (LGUGC) spearheaded the development of the LGU
bond market. More recently, it has spurred Private Financial Institution
(PFI) direct lending to LGUs, water districts, and electric cooperatives
through the use of its guarantees. These efforts, while innovative, have
not been mainstreamed in recent years. A main challenge has been the
reluctance of GFIs and other government agencies to open the subnational credit market for LGU borrowing.
The Philippines: Recent Developments in the Subnational Government Debt Markets
The basic question that remains is how to construct a financing
framework to help LGUs attain “genuine and meaningful local autonomy to enable them to attain their fullest development as self-reliant
communities and make them more effective partners in the attainment
of national goals.”4 The DOF developed the LGU Financing Framework
in 1996, which was later confirmed by the government.5 The development of clear and consistent government policies to encourage competition in the subnational credit market will in the long term help the
implementation of the framework.
Another key aspect in the deferral of implementation of policies
lies in the very nature of the LGUs themselves. While there has been
innovation in devising programs to foster the use of credit to support
development, the local governments have been reluctant to borrow.
This appears to be due in large part to a natural conservatism. Local
governments appear to have little appetite for credit financing, with the
exception of a relatively small number, which have tapped both loans
and bonds to finance various local projects.6
On the one hand, this attitude has avoided fiscal difficulties that
might have occurred as a result of profligacy. On the other hand, the
cautiousness has stymied local development initiatives. According to
the discussions within the country, next steps include bolstering LGU
credit relationships with the private sector. However, the dominant role
of GFIs in financing LGUs may limit the extent of the financing opportunities by the private sector.
The remainder of this chapter is structured as follows. Section two
examines the structure and finances of subnational governments in the
country, including the growth and patterns of local government spending and revenues, and presents a recent history of LGU borrowing and
its levels of indebtedness and the regulatory framework for subnational
unit borrowing. Section three analyzes a range of factors, including the
legal borrowing limitations and the various financial oversight mechanisms that have led to low demand for debt instruments by subnational
governments in the country. Section four reviews the development of
subnational credit markets, the status of PFI lending, and the composition of debt instruments. Section five reviews recent innovation in the
subnational credit market and analyzes prospects. Section six provides
conclusions.
419
420
Until Debt Do Us Part
Subnational Government Finance and
Borrowing Framework
The Philippines is a unitary state with a hierarchical system of governance. Subnational governments are part of the state and are directly
under the control of the national government, though with certain constitutional protections. The subnational government sector consists of
three levels: the provinces and major cities, the municipalities, and the
barangays (neighborhood organizations). The country, with a population of approximately 90 million, has more than 1,700 local governments (not counting the 42,025 barangays), including 80 provinces, 138
cities, and 1,496 municipalities.7
The Local Government Code of 1991 was revolutionary in its impact
on decentralization. It assigned greater responsibilities for service provision to subnational governments and also entitled them, under the
Internal Revenue Allotment (IRA) scheme, to receive 40 percent of the
national government’s income and value added tax revenues, which are
distributed on a formula basis. The Code also gave local governments
expanded powers for setting local tax rates and collecting own-source
revenues. The mainstays of local revenues are the property tax, the business tax, and taxes on vehicles.
The implementation of decentralization and the realization of what
was envisioned in the Code have been slow. Two decades after enactment of the Code, the size of subnational governments measured by
spending remains small. Total LGU spending increased from an average of 1.6 percent of Gross National Product (GNP) during 1985–91 to
about 3 percent in the late 2000s.8
Local Government Spending and Revenues
Local governments allocate the biggest portion of their budgets to
general public services, which are basically the general administration
services needed for the daily routine of running a local government.
Expenditure for economic services is the second-biggest expense. A
review of local public expenditure management is needed to achieve
more efficient allocation of resources. There appears to be relatively
low investment in human capital (education, health, and nutrition)
and in infrastructure relative to other expenditure items.9 The central
The Philippines: Recent Developments in the Subnational Government Debt Markets
421
government and government-owned and government-controlled corporations continue to implement major infrastructure projects, but
with greater emphasis on public-private partnerships (PPP) in infrastructure. Overall, LGU spending has averaged only around 3 percent
of GNP. Table 11.1 presents a distribution of LGU expenditure.
On the revenue side, subnational governments’ own revenues
accounted for about 32 percent of total revenues in 2009, and subnational governments are highly dependent on fiscal transfers from
the central government. Figure 11.1 illustrates the percentage composition of sources of revenues of all local governments. There were
no significant changes in the composition of sources of revenues in
the 2000s. The fiscal transfers, principally the IRA, account for almost
two-thirds of LGU revenues.
Table 11.1 Distribution of Total Expenditure, All Local Government Units, 2001–09
2001
2002
2003
2004
2005
2006
2007
2008
2009
40.51
41.34
40.41
40.02
39.63
40.36
41.74
44.14
53.91
Education,
culture,
and sports/
manpower
development
7.09
6.53
6.85
6.61
6.95
6.86
6.53
5.94
6.11
Health, nutrition,
and population
control
11.50
11.72
10.85
10.97
10.18
9.80
9.78
9.76
11.35
Labor and
employment
0.16
0.15
0.12
0.06
0.07
0.07
0.07
0.06
0.07
Housing and
community
development
4.38
4.42
2.40
2.05
2.18
2.05
2.01
2.13
3.28
Social security/
social services
and welfare
3.02
2.83
2.57
2.39
2.39
2.35
2.45
2.41
5.16
18.55
16.74
15.76
15.76
15.75
15.04
15.22
15.09
18.55
General public
services
Economic
services
Debt service
2.41
2.39
2.87
2.73
3.27
3.21
3.23
3.29
1.59
Other purposes
12.37
13.88
18.18
19.42
19.59
20.26
18.97
17.17
0.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
Expenditures
Source: Bureau of Local Government Finance.
100.00
422
Until Debt Do Us Part
Figure 11.1 Distribution of Total Income, All Local Government Units, 2009
1.3%
1%
2%
22%
10%
64%
Own tax revenue
Own nontax revenue
Internal revenue allotment
Other shares
Extraordinary receipts/AIDS
Interlocal transfers
Source: Bureau of Local Government Finance.
Note: AIDS = national government assistance to local governments.
Figure 11.2 presents revenue sources across provinces, cities, and
municipalities for 2009. As can be seen, the aggregate revenue numbers
mask big differences among the categories of local jurisdictions. Cities
derived about 40 percent of their revenues from their own revenue
sources in 2009 compared with only about 8 percent for municipalities and 10 percent for provinces. The cities have larger tax bases and,
consequently, enjoy more buoyant own-source revenue opportunities.
However, most LGUs (that is, the provinces and municipalities) have
narrower tax bases and thus do not raise proportionately as much ownsource revenues. They have remained dependent on fiscal transfers, principally the IRA, for funding local development activities.10
The IRA program of the formula-based revenue sharing led to local
governments largely substituting the new revenues from the central
government for own-source revenues, especially the local property
tax. Between 1990 and 1996, local own-source revenues declined from
50 percent of total local revenue to 30 percent, which is about the same
today. The large vertical fiscal gap has been filled by IRA transfers,
which comprised around 65 percent of total LGU incomes in 2009. The
dependence on the IRA results in lesser local fiscal autonomy, which
The Philippines: Recent Developments in the Subnational Government Debt Markets
423
Figure 11.2 Composition of Revenues by Type of Local Government Unit, 2009
Sources of revenue
National transfers
Nontax revenue
Other taxes
Business tax
Real property tax
Tax revenue
0
10
20
30
40
Provinces
Cities
50
60
70
80
Municipalities
Percentage of revenues
Source: Bureau of Local Government Finance.
c reates opportunities for greater control by the central government,
contrary to the envisaged situation of local governments able to use
own resources to respond to local needs and to match local outputs
with local preferences.11 In other countries, greater reliance on ownrevenue generation has given subnational governments more fiscal
autonomy. Granting subnational governments more revenue-raising
power aims at creating a closer link between expenditure accountability and the use of revenues to finance such expenditure (Bird 2010).
Meanwhile, fueled by IRA payments, local governments’ share of
total government spending in the Philippines between 1990 and 1996
grew from 6 to 16 percent.12 Thus, although their position as the final
deliverer of public services grew, the local governments’ relative share
of direct spending remained small compared to that of the central
government.
One motivation for the decentralized intergovernmental structure as
reflected by the Code was to enable subnational governments to assume
a greater share of the burden of financing infrastructure. It was thought
that this might be accomplished by permitting subnational governments
broad powers to borrow without the approval of the national government. To that end, the Philippine DOF, with considerable support from
International Financial Institutions (IFIs), led the way on initiatives to
90
424
Until Debt Do Us Part
expand local governments’ access to credit, following a policy articulated in 1996.13
Although the Code grants subnational governments the general
power to borrow, there are regulatory restrictions on borrowing activities. These restrictions are specified in the Code itself and also arise
from regulations in the banking and financial sector. A brief synopsis of
the major components of the regulation of local government borrowing
and indebtedness, which is discussed in greater detail below, is provided
in table 11.2.
Governments borrow for a variety of reasons. Generally, long-term
borrowing for long-lived capital improvements is recognized as a legitimate use of debt as long as the indebtedness incurred aligns with a locality’s ability (and ongoing willingness) to repay during the economic
life of the capital investments. Borrowing to fund persistent shortfalls
in current revenues (aside from unforeseen emergencies) is frowned
upon.14 Consistent application of this behavioral norm is an important
element of a “hard” budget constraint.
The Local Government Code allows subnational governments to use
credit financing for two purposes: liquidity and capital projects. Meeting liquidity needs involves credit financing of a local government’s
current spending in advance of expected releases of intergovernmental
(primarily IRA) payments or the receipt of taxes. Borrowing by local
Table 11.2 Local Government Borrowing and Debt Limitations: The Philippines
Debt service ratio limit
Debt service not to exceed
20 percent of “regular
income,” which includes
intergovernmental
payments. Of those
payments, not more than
20 percent can be used
for debt service. All LGU
debt is effectively general
obligation. Some water
district borrowing (water)
is based only on revenues.
Source: Petersen and Soriano 2008.
Note: LGU = Local Government Unit.
Outstanding debt
amount limit
None
Other restrictions
• Bank loans for current and long-term
investment needs; use of intercept of transfer
payments as loan security (i.e., the central
government fiscal transfers to LGUs can be used
or intercepted for loan payments)
• Bonds restricted to “self-supporting” (revenueproducing) investments
• Bond issues subject to central government
review for meeting debt guidelines
• Localities must budget for committed debt
service payments for their budgets to be valid.
The Philippines: Recent Developments in the Subnational Government Debt Markets
governments for liquidity has been modest, accounting for only 2 or
3 percent of their total annual receipts. However, subnational governments are not allowed by law to incur operating budget deficits and
must appropriate in their annual budgets amounts sufficient to pay debt
service for indebtedness incurred.15
The Local Government Code (Section 324) imposes a limit on subnational governments’ borrowing capacity, stipulating that their appropriations for debt service (payments of interest and principal) should not
exceed 20 percent of their “regular income” in any given year. Regular
income is defined by the Bureau of Local Government Finance (BLGF)
under the DOF, which certifies the debt service and debt capacity calculations, as the combined total of the three-year average of Locally Own
Sources Income, the IRA payments estimated by the Department of
Budget and Management, and the three-year average of national wealth
payments. The total gives the “ARI” or “average regular income.”16 The
(ARI) × (.20) equals the maximum allowable debt service ceiling.17
Given the high dependency of subnational governments on central
transfers, what portion of revenues that can be pledged as security for
debt service becomes important? A tighter definition is used with regard
to the amount of IRA payments that can be pledged to debt service
because of the widespread use of the IRA “intercept” (or a deposit offset) as a security on subnational loans. Only in the case of large cities
does the distinction between “the regular income” and the IRA payment
make a difference, since for municipalities and provinces, the IRA payments dominate the revenue stream.
As noted, regulation by the Code requires that a subnational government must budget for all its contractually due debt services; otherwise, its budget is considered void, and it cannot lawfully spend funds.
Furthermore, the BLGF provides oversight of subnational government
lending by calculating the required debt service.
Regulation of subnational bond issuance was indirectly implied
by Section 296 of the Local Government Code, which subjected such
debt to regulation by the Securities and Exchange Commission and the
Bangko Sentral ng Pilipinas (central bank). However, at the outset of
devolution after 1991, these regulatory provisions were not energetically exercised. Early on, the Securities and Exchange Commission held
that local government bond issues were exempt from its registration
425
426
Until Debt Do Us Part
procedures, but in late 2000, the DOF successfully requested the Commission to “delegate” its approval powers to it for purposes of developing a registration procedure.18
In addition, the New Central Bank Act (Republic Act 7653) requires
that, as a condition of borrowing, the monetary board render an opinion on the impact of the borrowing on monetary aggregates, the price
level, and the balance of payments. For a sovereign guarantee, there is
a more rigorous test, and approval is required from the secretary of
finance. No subnational government has borrowed with such a guarantee, nor has any borrowed in foreign currency.19
There is no formal subnational government insolvency system in
the Philippines. The prevention of any potential defaults is through
the screening (and de facto approval) of borrowings by the BLGF,
and the GFI deposit offsets and intercepts fiscal flows to subnational
governments. The intercept can be used directly only by the Municipal
Development Fund Office (MDFO), which has not invoked it because
subnational governments have proven to be good borrowers. The MDFO
has worked closely with BLGF in tracking subnational borrowing capacity and debt service capacity, and has made good use of the information
in screening subnational loan applicants. Government policy does not
allow the direct use of the IRA intercept by government or private banks.
However, all central government payments to subnational units are made
via the Land Bank of the Philippines, which does employ assignments of
IRA and deposit offset agreements to secure loans. Furthermore, subnational governments are required to keep deposits in the Land Bank and
the Development Bank of the Philippines—both GFIs—except under
special circumstances that require specific exceptions.
The global financial crisis of 2008–09 did not adversely affect the
financial sector in the Philippines, much less the subnational governments. In general, domestic banks, investors, and subnational governments had no exposure to sophisticated financial instruments such as
derivatives, although a few (domestic) commercial banks had negligible
amounts in their investment portfolios. Instead, the Philippine subnational governments have maintained budget surpluses in general, with
their revenues unaffected by international conditions (see table 11.3).
All three levels of subnational governments have maintained budgetary surpluses. Overall, the Philippine economy runs a trade surplus,
The Philippines: Recent Developments in the Subnational Government Debt Markets
427
Table 11.3 Local Government Finances: Key Ratios by Type of Unit, 2010
Type of Local
Government Unit
Surplus
as % revenue
Debt service
as % revenue
Transfers
as % revenue
Borrowings
as % revenue
Total revenue
(pesos millions)
Cities
13.4
1.3
41.3
5.2
126,763
Municipalities
11.7
0.8
78.4
2.1
99,270
Provinces
16.0
1.6
76.2
3.7
71,596
Overall
13.5
1.2
62.0
3.8
297,629
Source: Bureau of Local Government Finance.
Note: Borrowings = total receipts from loans and borrowings, debt service = debt service (interest expense and other
charges), surplus = total revenue – (total current operating expenditure + total nonoperating expenditures), transfers =
IRA + other shares from national tax collections, and total revenue = total current operating income + total nonincome
receipts.
which is helped greatly by the growing level of international remittances
from an estimated 8 million Filipino overseas workers. Philippine gross
domestic product (GDP) continued to grow during the 2008–09 global
recession. Philippine nominal GDP grew annually by 7.1 percent in
2007, 4.1 percent in 2008, and 1.1 percent in 2009. Subsequent growth
in GDP was 7.5 percent in 2010 and 4 percent in 2011.20
Table 11.3, which is based on 2008 data, indicates that, with the
exception of the cities, the smaller municipalities and the provinces are
highly reliant on intergovernmental transfers (most of which consist of
the IRA transfers). The LGU share of national internal revenue taxes was
fixed by law at 40 percent of national internal revenues in the third year
prior to the allocation year. However, the amount of IRA transfer varies
over time, depending on the revenue effort of the central government.
In 2009, a year after the financial crisis, the country’s national revenue
effort (measured as national revenues as a percentage of GDP) declined
to 14.6 percent of GDP from 16.2 percent in 2008. In 2010, revenue
effort further deteriorated to 14.1 percent. This means that the current IRA allocation was computed on the basis of central government
collection of national internal revenues during the financial crisis year,
which was a relatively bad year for revenue effort, with GDP growing at
1.1 percent in 2009. Thus, IRA payments are expected to decline somewhat—but, there has been a considerable lag.
Meanwhile, as discussed next, LGUs have surpluses to buffer the
fluctuations in IRA payments. In addition, as seen in table 11.3, the
level of LGU borrowing is negligible in proportion to receipts, and
428
Until Debt Do Us Part
annual debt service requirements are just as low, equaling less than
3 percent of revenues in the aggregate. It needs to be noted that the
actual surpluses are smaller than the reported “surpluses” in table 11.3
because, until 2011 when reforms were implemented, the government
reporting systems did not capture expenditures from “continuing
appropriations” that LGUs carry over from capital outlays that span
more than one fiscal year.21
Table 11.4 provides a four-year display of the annual budget surpluses of subnational governments by type of unit. These have been
steady among the various types of units.22 As noted, the IRA payments
to subnational governments are based on a three-year lag of national
government revenues, which delays the impact of any national receipts
on the IRA payments.
Central oversight of local unit debt is generally predicated on the
nature of the market for such debt. Because there is a limited private
market for Philippine LGU debt, there effectively has been de facto
supervision of borrowing by the GFIs, which hold the vast majority of
LGU debt. More formally, oversight is exercised by the BLGF through its
review of the debt service ceiling and represents a de facto approval of
borrowing.
LGU debt is monitored by the BLGF under the DOF. An agreement
signed in 2002 requires that the GFIs, the central bank, and the LGUGC
submit data on LGU debt to the BLGF. LGUs also provide data on debt
and debt service through their Statement of Income and Expenditure,
which they submit quarterly to the BLGF. The central bank also monitors GFI loans to LGUs and the purchases of LGU bonds.23
Table 11.4 Budget Surpluses of Local Government Units, 2005–08
billion pesos
Type of unit
2005
2006
2007
2008
Provinces
6.2
5.5
4.7
8.8
Cities
13.5
19.2
14.9
21.2
Municipalities
Total
7.3
8.2
7.2
10.7
27.0
32.9
26.8
40.7
Source: BLGF Statement of Income and Expenditure.
Note: After adjusting for the expenditures from “continuing appropriations,” Local Government
Unit annual surpluses ranged from half to one-third of the reported surpluses (see note 23).
The Philippines: Recent Developments in the Subnational Government Debt Markets
Subnational Demand for Debt Instruments
The Philippine subnational debt is low compared to many other countries. The best way to make comparisons is to examine the level of
debt as a percentage of the nation’s GDP, since this takes into account
the varying sizes of the underlying economy. The reported Philippine
subnational debt is less than 1 percent of the national GDP. That ratio
changes little by adding the indirect debt of subnational water utilities.24 By comparison, the average ratio of subnational debt to GDP
for developing and transitioning countries was 5 percent in the mid2000s.25 The low percentage suggests that Philippine subnational governments neither do much capital spending nor use much borrowing
to finance those needs. Most capital spending appears to be for “development” purposes and is relatively small scale.26 The contrast is even
greater when compared to developed countries (the Organisation for
Economic Co-operation and Development [OECD] nations). For the
developed countries, the average subnational debt to GDP in 2006 was
6.7 percent.27
The disincentives for LGUs to contract debt for public infrastructure
investments contribute to what seems to be a low demand by subnational governments for debt instruments in the Philippines.28 The disincentives are mainly explained by the binding constraints that weaken
the capacity of local governments to provide basic devolved services and
drive economic growth.29 The high level of fragmentation in the subnational government structure contributes to the lack of economies of
scale in infrastructure provisions. The local government system comprises a large number of small jurisdictions at each level of subnational
government. The fragmentation can be overcome by forming interjurisdictional cooperation in infrastructure provision.30 However, there is a
lack of noticeable pressure from local citizens for local governments to
invest in better infrastructure and service delivery. The service delivery
system is multitracked in almost all sectors. National government agencies continue to play major roles in the delivery and finance of local services, including using discretionary funds.
The current IRA formula does not compensate for the varying degrees of fiscal capacities of LGUs. In fact, the large IRA transfers from the central government have a disincentive effect on local
429
430
Until Debt Do Us Part
tax efforts; LGUs that receive a larger IRA tend to be lax in their tax
efforts.31 Local officials have strong incentives to lobby for resources
directly from the national government. A recent study shows that a
model-based Good Governance Index is not strongly associated with
election results.32
The poorer LGUs have a low fiscal capacity to leverage borrowing.
Real property taxation potentially offers a revenue-rich tax base, but
there are challenges to greater use of that source: land titling issues, lack
of cadastral surveys, unwillingness of local assessors to update assessment levels, and resistance of the local propertied class to an increase
in property taxes. Weak local economies give rise to low local business
tax collections. As a result, the weaker, rural LGUs heavily depend on
the IRA, the national government’s fiscal transfer, to finance local development activities, and on the “pork barrel” of legislators for livelihood
projects and the usual infrastructure projects such as farm-to-market
roads, barangay halls, and others.
An ongoing national government project on improving land and
property valuation seeks to provide LGUs with the tools for doing
proper land valuation and assessment.33 According to BLGF, many LGUs
are starting to realize the great revenue potential arising from updated
land values and better assessment practices. The project has led to the
establishment of Philippine Valuation Standards, patterned after the
international best practices on valuation, the development of IT (information technology) systems, and measures to support the formulation
of a market-based schedule of property values.
The higher-income LGUs are experiencing pressure for better infrastructure and services, stemming from their transformation into growing urban centers. The growth of global business processing outsourced
to Philippine-based companies and call centers, among others, is transforming the bigger LGUs, which are seeing the need for better local
infrastructure, including more reliable and competitively priced electricity supply, and public service delivery. These higher-income LGUs
borrow mostly from the GFIs, and receive a larger share of the IRA due
to their larger population and land areas.34
The national government has adopted PPP as its main strategy
for infrastructure provision at the national and local levels. The PPP
approach is a potential source of demand for more debt financing at the
The Philippines: Recent Developments in the Subnational Government Debt Markets
local level if LGUs are to manage those PPP projects assigned to their
mandate and the above-mentioned binding constraints to local service
delivery relaxed.
Moreover, on the demand side, using PPP for infrastructure development at both the national and local government levels, encouraged by
the government policy, will require the building of local capacities to
deal with private sector investors and lenders under this mode of procurement.35 LGU demand for borrowing, especially for infrastructure,
is expected to increase. PPP for local infrastructure and an increase in
LGU demand for borrowed funds for this purpose will require effective coordination with the national government, especially its oversight
agencies and infrastructure agencies.36 There is also the issue of how
effective the PPP approach will be as an alternative for small-scale and
non-revenue-producing projects.
Subnational Credit Market and Composition
of Debt Instruments
The composition of Philippine subnational debt consists mainly of debt
owed either to higher levels of government or debt in the form of loans
from government-owned banks. Although development of the subnational bond market gained momentum in the late 1990s, subnational
bonds have not become a main part of the debt portfolio.
Public Financial Institution Lending to LGUs
The Local Government Code of 1991 has the potential to open several
avenues for local governments to access credit finance from bank credits
and “other similar forms of credits,” and also from bonds and “other
securities.” Notwithstanding the potential, lending by the GFIs continues to be almost exclusively the source of loan funds for subnational
governments. Subnational governments naturally made initial credit
requests to the GFIs, since the GFIs hold the cash accounts of LGUs.
The main sources of domestic credit financing are two GFIs, the Land
Bank and the Development Bank of the Philippines, and two specialized onlending institutions, the Local Water Utilities Administration
(LWUA)37 and the Municipal Development Fund (MDF), which is run
by MDFO.38 The LWUA channels development assistance to local water
431
432
Until Debt Do Us Part
supply projects and has offered long-term loans that match those of the
underlying development assistance loans.
In the early years after the Local Government Code was implemented, the Philippine National Bank, which was later privatized,
and the Land Bank were the largest providers of credit to local
governments. In 1995, the Philippine National Bank held about
6 billion pesos in loans to local governments and the Land Bank held
about 5 billion pesos. The Development Bank of the Philippines was
just starting to lend to local units. The MDF had about 2 billion pesos
in loans and the LWUA had about 8 billion pesos in loans to the water
sector.39
The GFIs, reopening their lending windows to local governments
after the widespread defaults of the 1980s, focused on those with higher
incomes, as evidenced by the large average loan size in their local government loan portfolios. Interest rates on these loans were about the
same as those on their prime commercial loans, suggesting that they
assigned a low-risk premium to local governments. The average tenures
were longer than those for commercial loans, at about two to four years.
After 1995, the growth of lending rapidly accelerated for both the
Land Bank and the Development Bank of the Philippines, in part
because of the rapid withdrawal of the Philippine National Bank from
the local government credit market following the bank’s privatization.40
Also, both the Land Bank and the Development Bank of the Philippines
got access to foreign loans for relending to LGUs and enhanced their
depository relationship with LGUs.
The GFIs have actively used the depository relationship and government reporting to create credit and investment instruments. They base
lending decisions for capital projects on the IRA and revenue flows of
the LGUs rather than on the revenue flows of the project. They make
available short-term credit facilities tied to future budget releases that
allow LGUs to draw funds in advance of revenues.41 They also enable
LGUs to arbitrage on interest rates and on financial reporting by, for
example, granting loans secured on their deposits and allowing the local
governments to earn spreads on their investments, while still reporting
high deposit balances. These practices help the GFIs manage the risk of
lending to LGUs, while enabling the LGUs, to venture into commercial
borrowing and financing of capital projects.
The Philippines: Recent Developments in the Subnational Government Debt Markets
433
The LWUA, which lends to the local water districts, suffered ongoing structural problems that prevented it from expanding its participation in financing local water supply projects. Lending by the MDFO also
grew slowly, reaching 2.7 billion pesos in 1999 and 3.7 billion pesos in
outstanding loans by the third quarter of 2010. Among other possible
sources, the government pension funds, which had shown early interest,
were content to invest in high-yield government obligations and made
heavy commitments to the commercial property sector and equity
investments. These factors impeded their participation in the local government credit market.
Table 11.5 summarizes the present structure of the Philippine local
government debt market. Overall, with a relatively small domestic bond
market, the provision of credit to LGUs in the Philippines is overwhelmingly done by the GFIs. The Philippines has a largely b
ank-dominated
credit system. Generally, as financial systems mature, they tend to
develop alternatives to the reliance on banks for credit. This usually
entails the growth of savings-based institutions and various forms of
insurance that have longer-term investment horizons. As will be discussed, the Philippines has a formal policy of promoting private sector
financing of those LGUs that are higher income and that have self-
supporting projects. However, notwithstanding earlier efforts to achieve
that goal, the current state of affairs is a continuing dominance by the
GFIs over LGU credit access, as shown in table 11.6.
Out of a total outstanding LGU debt of 68.02 billion pesos, as of
September 10, 2010, about 86 percent was owed to GFIs, and 9 billion
pesos, or 13 percent, was owed to two private banks, which by law
Table 11.5 Structure of Philippine Local Government Debt Markets
Institutions
Predominantly loans by
government-owned banks.
Also, some bond issues in
capital market using bond
insurance that also relies
on transfer aid intercept
provisions. Some special
onlending funds.
Source: Petersen and Soriano 2008.
Cost of borrowing
Most interest rates at market
levels. Government banks
can use intercept/offset on
intergovernmental transfers
to secure loan repayments.
Local governments usually
must use government banks
for deposits.
Debt instruments
Bank loans and bonds of 7-to-10years’ maturity. Some subsidized
onlending using donor funds for
special purposes up to 15 years.
434
Until Debt Do Us Part
Table 11.6 Outstanding Loans and Bonds of LGUs (as of September 10, 2010)
Amount (billions pesos)
Loans from GFIs
Land Bank of the Philippines
% of total
58.29
85.69
43.25
63.59
Development Bank of the Philippines
3.26
4.79
MDFO
11.77
17.31
8.94
13.15
Philippine National Banka
3.67
5.39
Philippine Veterans Bankb
5.28
7.76
Loans from PFIs
Bonds outstanding
LGU Guarantee Corporation
Philippine Veterans Bankc
Total
0.79
1.16
0.25
0.37
0.54
0.79
68.02
100
Source: Bureau of Local Government Finance.
Note: GFIs = Government Financial Institutions, LGU = Local Government Unit, MDFO = Municipal Development Fund
Office, PFIs = Private Financial Institutions.
a. Formerly a GFI that, after its privatization, has retained the ability to hold LGU deposits.
b. A privately owned bank whose board of directors is appointed by the president, with ability to hold LGU deposits. Not
shown is an estimated 20 billion pesos in LWUA loans outstanding to water districts.
c. This amount may include bank loans that are guaranteed by the Philippine Veterans Bank, as well as bonds.
(Philippine Veterans Bank) or by special authority from the Monetary
Board (Philippine National Bank) are authorized to accept deposits
from LGUs. It appears that just over 1 percent of LGU debt is owed
to other PFIs, which are primarily banks that have bought the bonds
issued by LGUs and guaranteed by the LGUGC or the Philippine
Veterans Bank.
The Philippine debt service limit is similar to that found in many
countries and is, perhaps, a little less generous than that found in
others. Many countries place a maximum limit on debt service as a
ratio to annual revenues or expenditures.42 A few countries limit the
amount of annual borrowings to a fraction of a government’s total
revenues (which may make sense in terms of short-term debt but is
not rational when it comes to long-term debt). Overall, the limitation on general obligation borrowing (borrowing is secured by full
faith and general revenues such as sales and property taxes of the local
government) makes sense. But it does not address the case where an
LGU might not pledge its “general revenues” to the repayment of the
The Philippines: Recent Developments in the Subnational Government Debt Markets
debt but relies on enterprise earnings or other “non-general” revenues,
where borrowing is secured by specific revenues or self-sustaining revenue generated by the project that is being financed by the debt (this is
addressed further, below).43
The statutory debt limit is not, however, the deterrence to LGU borrowing in the Philippines. Given the modest demands for loan funds
by LGU borrowers, the debt limit is seldom reached. Generally speaking, for the LGU sector as a whole, it appears that total debt service
payments, in the aggregate, are equal to only 2–3 percent of “regular”
income (depending on how tightly that concept is defined). Furthermore, annual LGU borrowing equals only 2–3 percent of total LGU
receipts (recently, 4–6 billion pesos in aggregate borrowing, compared
to around 200 billion to 230 billion pesos in total revenues and receipts).
According to the BLGF, there have been few cases of LGUs being near
(or closely approaching) the debt service limit. Overall, the excellent
repayment record of LGUs does not evidence any systematic fiscal strain
or imprudence.
The debt capacity limitation, however, leaves one potential LGU borrowing opportunity unchartered. That is, while the debt limit makes
perfect sense for those borrowings that are secured on general revenues
(including the IRA), it does not contemplate the situation where an LGU
might not want to borrow against its general revenues but rather against
some specifically pledged revenues or assets. That would be a “revenue
bond” or “limited obligation,” in the parlance of the bond markets,
where there is not a pledge of general revenues. Currently, there seems to
be no explicit regulatory provision for that type of borrowing by LGUs.
However, that approach is what the water districts in the Philippines are
now using with their “water revenue loans” that are backed by pledged
water revenues and “step-in-provisions” that allow the LWUA to take
over operation in case of a default.44
BLGF has been recognizing “trust funds” in its tabulation of LGU
accounts. The “trust fund” to conduct commercial operations within
a government is a positive development. The other is the creation of
“special districts” or “authorities” that have their own governing body,
operate under a trust agreement (contract) with the lenders, and are
insulated from the day-to-day budget and political issues of the LGUs. In
these cases, either the record of existing operations or strong feasibility
435
436
Until Debt Do Us Part
studies are needed. The role of the rating agencies is also often critical,
since the ratings provide general benchmarks on the credit quality and
debt marketability. Sometimes, there is general obligation support for
debt service until the project becomes self-sustaining, at which time its
obligations no longer count against the debt limitation.45 However, as
noted by Canuto and Liu (2010), without proper governance structure
and financial transparency, the special district types of arrangement can
carry fiscal risks and contingent liabilities to the government owners of
such special districts.
The adoption of the special fund doctrine as a means of financing
LGU capital needs will require legally enforceable contracts where the
LGU (or a special fund or district created by it) will be required to run
the project as a commercial enterprise and agree to do so with the creditors. Provisions will be needed to protect the investors in case the issuer
or borrower defaults on its obligations. The current revolving lending
program in the water district areas (which involves PFIs and does not
rely on the IRA guarantee) needs to be carefully evaluated since such
an evolving financing technique might have broader applications in the
Philippines.
Private Credit and Capital Markets
LGUs have been able to access private credit markets but, until recently,
only indirectly through bond issues, which have been purchased mostly
by private banks and guaranteed by the LGUGC (or the Philippine
Veterans Bank). Aside from isolated cases, there have been no direct PFI
loans to LGUs. Most of the LGU bonds to date have had seven years’
maturity, consistent with the short-to-medium-term nature of the
banks’ funds. However, LGU bond flotation has been infrequent since
mid-2006. Direct lending by PFIs to LGUs has commenced, using loan
insurance and liquidity facilities.46 These deals are being done in conjunction with a new IFI-sponsored water revolving fund project. These
new programs are aiming at getting PFIs to lend directly to LGUs. The
LGUGC is currently packaging three such loans.
The PFIs are well aware that LGU lending and depository activities represent a profitable area for the GFIs; they would be anxious to
compete if they could overcome the impediments. It is realized that
the intercept on IRA payments (or some adaptation of it) is likely
The Philippines: Recent Developments in the Subnational Government Debt Markets
ecessary to sustain the high repayment rates of the LGUs.47 As noted,
n
it has been difficult for PFIs to get an assignment of the IRA payments
(which requires GFI involvement since they act as the LGUs’ depository).48 There are also continuing concerns about the ability and willingness of LGUs to pay on their debt, especially given the three-year
election cycle. LGUs under the existing “IRA intercept/offset” enforced
credit regime, however, have a nearly perfect record of paying their
obligations to GFIs.
The Philippines received much international attention for its early
efforts to build a municipal bond market, and its innovative use of
bond insurance is accomplishing that deed. There was a flurry of
activity in the issuance of municipal bonds in the late 1990s and early
2000s. This was made possible by the creation in 1997 of the LGUGC
as a “private” insurer49 of LGU loans and bonds. Between May 1999
and December 2010, the LGUGC guaranteed 19 bonds amounting
to 3.25 billion pesos issued by 16 LGUs.50 Projects financed include
tourism-related infrastructure such as the Tagaytay City Convention
Center and the C
aticlan-Boracay Jetty Port. Public markets, commercial
centers, public terminals, slaughterhouses, housing projects, a hospital,
an academic center, a gymnasium, and an integrated solid waste management system have also been financed. Over 2.7 billion pesos of the
LGUGC-guaranteed bonds had already been redeemed by December
2010, and it is estimated that only 590 million pesos of bond principal
remained outstanding.51
However, insuring bonds was the first step taken by LGUGC and has
not been the limit of its sphere of activity. By 2006, it began to insure
bank loans, which did not entail the issuance of marketable bonds, but
was geared to individual bank loans or syndicates of bank loans.52 Over
the last five years, this activity has increased, usually involving water districts and other public corporate borrowers. Table 11.7 presents LGUGC’s
overall activity through the end of 2010. As indicated, by the end of 2010,
LGUGC’s annual new insurance deals were running about 900 million
pesos a year, and overall outstanding guarantees amounted to 2 billion
pesos, of which an estimated 589 million pesos represented outstanding
LGU bond issues.
The Philippine Veterans Bank also has guaranteed LGU bonds since
2003, amounting to 505 million pesos.53 The bonds had a seven-year
437
438
Until Debt Do Us Part
Table 11.7 LGU and Other Entity Outstanding Debt (with LGUGC
Guarantee), 2010, by Type of Unit
Type of LGU debt
Million pesos
LGU bonds
589
Water district loans
748
Municipal enterprise loans
151
LGU bank loans
518
Total
2,006
Source: LGUGC, December 31, 2010.
Note: LGU = Local Government Unit, LGUGC = Local Government Unit Guarantee
Corporation.
maturity and a two-year grace period. However, most bonds had been
fully redeemed, and by the end of 2009, the Philippine Veterans Bank had
only approximately 100 million pesos in guaranteed bonds outstanding.
On a combined basis, the LGUGC and the Philippine Veterans Bank
have guaranteed the issuances of LGU bonds amounting to 3.5 billion
pesos, of which only about 500 million pesos (about 15 percent) remain
outstanding. There have been no defaults, and early issues have been
paid and retired. The rest were redeemed early, long before the end of
the maturity period. Most of the redemptions were due to refinancing
of the bonds via loans made by GFIs.54
When the LGUGC was started in the late 1990s, there was an initial
flurry of bond issues and the hope that the bond market would “take
off.”55 However, by the mid-2000s, the pace of bond sales slowed greatly.
Private creditors were concerned that the GFIs were protective of their
LGU business and sought to exclude others from lending to LGUs. The
GFIs have a great informational advantage. Because of their holding of
the LGU deposits, they are able to track the LGUs’ financial activities
and identify potential bond deals.56
In addition, the GFIs were reluctant to enter into assignments of
IRAs and other LGU revenues to the PFIs engaged in lending to LGUs,
and, in the view of LGUs, bond issues involve additional costs and
administrative approval delays.57 Such costs and delays can be avoided
by seeking GFI lending that is secured by offsets to IRA transfers
(which the LGUs were required to deposit in GFIs).58 According to the
PFIs, this requirement has created an uncompetitive situation in which
The Philippines: Recent Developments in the Subnational Government Debt Markets
the GFIs have strong advantages due to government regulation.59 It is
important to raise an even broader policy constraint for PFIs, which is
the Monetary Board policy restricting PFIs from serving as depository
banks for LGUs, except for special cases in which the Monetary Board
issues a waiver allowing a specific LGU and PFI branch to engage in
such a banking relationship. This is a fundamental barrier to direct PFI
lending to LGUs.
Credit Ratings, Project Finance, and Debt Instruments
There continues to be a lack of timely and generally accessible financial information on LGUs and “independent” published ratings. The
LGUGC publishes underlying ratings on the bonds it insures or is
intending to insure. These information deficiencies have not been a
major problem for the current GFI lenders, which took IRA payments
as the primary security.60 However, if the subnational capital market is
to expand, the availability of operational data, well-prepared financial
feasibility reports, and credit ratings would become important.
LGU loans receive the same treatment as commercial loans under the
bank capital adequacy rules. However, municipal bonds sold with the
LGUGC guarantee receive more favorable treatment.61 The Agri-Agra
capital requirements for bank investments continue to be a source of
demand for holding LGU loans as assets, since they qualify in meeting
the requirements.62 It is reported that there is sizable demand by longterm investors for LGU bonds and loans.63 In the past, some of the private bank trusts did come up against diversification limitations in the
case of the LGU bonds.
There is also potential in long-term fixed-income demand from
institutional investors such as life insurance, pension systems (public
and private), and individual trusts. These investors have tended to invest
in national government securities. Overall, however, the level of institutional financial investment in the country is not high. The national
government has dominated the bond markets (over 95 percent of all
bonds), and the growth has occurred most in the commercial paper
area. However, the development of the fixed-income exchange by the
Philippine Dealing & Exchange Corporation is changing that situation.
This should increase liquidity and improve the demand for bonds in the
Philippine market. The new exchange has developed a “yield curve” for
439
440
Until Debt Do Us Part
government securities that can be used as a benchmark for other fixedincome obligations. However, LGU bonds can be listed on the exchange
only if they are of sufficient volume.
There appears to be some headway on attracting PFI investment
to the LGU sector. The Philippine Water Revolving Fund (PWRF)
was established to mobilize private funds to the water sector and has
qualified several PFIs to participate.64 The Development Bank of the
Philippines and PFIs can cofinance the loan with the LGUGC providing credit guarantee. The MDFO (in the case of LGUs) and the
Development Bank of the Philippines (in the case of water districts)
can make available a stand-by line of credit to refinance the PFI loan
if the private lender decides not to extend the tenure beyond the current 10-year tenure. As of February 2011, five private commercial
banks had loaned 1.069 billion pesos to nine water districts. Cofinanced loans amounting to 747 million pesos have been given to two
water districts.
PFIs seem more comfortable with utility-type investments that are
revenue producing.65 For example, water district lending would appear
to be favorable if the credit security issues can be resolved. Many water
districts seem to be more professionally managed.66 The election cycles
may impact the behavior of local government officials, who may favor
high visibility and fast-payoff projects.67
Overall, the PFIs are interested in authorities or districts that are insulated from day-to-day local administration and are professionally managed as enterprises. In such cases, debt issues are controlled by provisions
in the underlying project and its loan contracts (“revenue bonds”) and
do not depend on general revenues (and the IRA payments).68
The development of new legal structures to support project financing
deserves greater study, since there appears to be some headway already
made in the formulation of LGU accounts. There is now recognition of
LGUs having trust accounts that are not part of the regular annual budgetary appropriation, and which function as restricted accounts during the
term of the trust. LGU economic operations that can be “ring-fenced”—
in the sense that enterprise revenues must be dedicated to —payment for
the enterprise’s operations and debt repayment—are potentially feasible
if there is a “special fund” doctrine.69
The Philippines: Recent Developments in the Subnational Government Debt Markets
Extending loan maturity is important to finance long-term infrastructure assets. The Philippine bond and bank loan market, aside from the
IFI funds that are usually provided for long-term maturity, is relatively
short to medium term and, in the case of the PFIs, typically carry variable
interest rates. Being restricted to the short maturities is a classic difficulty
that subnational governments face in many emerging economies (and
also in developed economies, especially when there is a substantial threat
of inflation). Short maturities and variable rate structure prevent an erosion of asset value to lenders if interest rates rise rapidly.
The classic way to modify the terms of debt in the market under such
conditions is to provide options that can protect both the lender and the
borrower. In the case of lenders, they will want the ability to protect the
value of their investment if interest rates increase. Borrowers, however,
will want protection if interest rates decrease and, with a call option, they
are permitted to refinance debt at lower interest rates.
These options were not used in the Philippine LGU financial markets
until recently.70
Recent Innovations and Prospects in Credit Market Access
In the Philippines, fostering a competitive and diversified subnational
credit market has been a long-standing policy goal.71 It has also been
in a protracted phase of experimentation. The private sector financial
markets are playing a minor role in LGU financing, except in selective
cases. Nonetheless, there are emerging opportunities for that activity to
occur.
In the Philippines, as in many other countries, subnational government credit needs have been met for many years by bank loans. Various countries have moved toward a more competitive market structure
with the participation of private creditors (Canuto and Liu 2010). Some
countries, including the Philippines, have continued to rely mainly on
GFI for subnational credit financing. In many cases, this approach was
initially necessitated by weak domestic credit and capital markets. Also,
local governments were subordinate in the political and economic
structure and depended on the central government for guidance and
support.
441
442
Until Debt Do Us Part
The Local Government Code of 1991 sought to significantly change
the relationship of local governments to the central government. It
sought to empower local governments to act with greater responsibility
and autonomy and provided them with substantial transfers (the IRA
payments) that now represent around 65 percent of all LGU revenues.
The Code also provided LGUs with substantial powers to borrow. Meanwhile, the LGU financing framework was developed in 1996 to implement the credit-related dimensions of the LGU Code. It envisioned a
combined effort by public and private financial entities to provide capital to the LGU sector. The GFIs were to be the lead entities in the case of
the higher-income localities and self-supporting projects, and to gradually bring in private sector financial institutions.72
The overarching objective was to move local governments in the
direction of sustainability and reliance on private market sources to the
greatest degree possible.73 A key role was assigned to the GFIs to act as a
conveyor belt for those LGUs that are becoming financially stronger (or
have stronger self-supporting projects), and helping them graduate into
the private markets. This largely has not happened. The GFIs—because
of their superior credit position, their ability to intercept LGU deposits,
and their capacity in dealing with LGUs—have incentives to make loans
to LGUs. The transition to PFI financing has been slow to materialize.
The development of competitive subnational credit markets will
need to address both demand- and supply-side constraints. On the
demand side, it is critical to strengthen the local finance and accountability systems for citizens to demand better services. As noted by section three, several factors contribute to the LGUs’ low demand for debt
instruments. These factors include a lack of pressure from citizens for
better service delivery, difficulty in developing interjurisdictional cooperation in infrastructure provision to take advantage of economies of
scale, inadequate debt and fiscal capacity, weak technical capacity to
develop projects, and the available alternatives of accessing “pork barrel” projects from the national legislators. These factors collectively
make the risks of borrowing by LGUs outweigh the potential rewards
for LGU infrastructure investment and financing.
The higher-income LGUs are experiencing pressure for b
etter
infrastructure and services, stemming from their transformation
into growing urban centers. They are likely to demand more debt
The Philippines: Recent Developments in the Subnational Government Debt Markets
instruments; thus, relaxing the binding constraints can help the
higher-income LGUs become more viable borrowers. The national
government has adopted PPPs as its main strategy for infrastructure provision at the national and local levels, which creates potential demand for more debt financing at the local level, particularly in
major urban centers where the stronger fiscal capacity can make the
PPP model more attractive to private players, including private financers. Developing strict sectoral financing policies will help reduce adhoc allocation of central government funds and reduce the incentives
for LGU lobbying for the central discretionary resources.
In the Philippines, municipalities and lenders rely heavily on the
IRA, in effect making all the municipal debt homogenized as a national
credit. In the early state of developing municipal debt markets, debt
issued based on fiscal transfers and central government guarantees may
help start the market. But the continuing development of the market
requires the use of financing instruments that rely more on own-source
revenues and credit differentiations among LGUs. There are financing
instruments that can forge closer links between own revenue of local
governments and their capacity to access the market, which in turn help
strengthen local accountability.
Revenue bonds have been used extensively in various countries,
including the United States, as a powerful instrument to finance subnational infrastructure by linking project finance with benefit taxation.74
The debt service is secured by revenue streams produced by the project
financed by the bond instrument. The recent innovation in the water
districts in the Philippines has indicated a similar direction—the water
districts in the Philippines are securing their “water revenue loans” by
pledged water revenues and “step-in-provisions.” The water districts
legally exist outside the LGU accounts and act as separate governmentowned-and-controlled corporations. The water district design might be
adapted into a broader notion of economic development and/or infrastructure districts that would have a degree of insulation from the dayto-day administration of the LGUs.
Another often used instrument—tax increment financing—also
helps link LGUs’ own revenue with infrastructure financing. The
instrument is used for financing infrastructure and other communityimprovement projects in many countries, including the United States.
443
444
Until Debt Do Us Part
Tax increment financing uses future gains in taxes to finance current
improvements, which are projected to create the conditions for future
gains. The completion of an infrastructure project such as power and
water often results in an increase in the value of surrounding real estate,
which generates additional tax revenue.75 Tax increment financing dedicates tax increments within a certain defined district to finance the debt
that is issued to pay for the project. It creates funding for public or private projects by borrowing against the future increase in these propertytax revenues.
Opening up the subnational credit market to private competition
helps lower the financing cost. Attracting PFI investment to the LGU
sector has recently gained headway in the Philippines. The PWRD was
established with the participation of both public and PFIs to cofinance
loans to water districts and LGUs. An important aspect of the subnational credit market is the competition among debt instruments. The
private (corporate) bond market is small in many developing countries,
and it remains small in the Philippines. Yet, properly secured, infrastructure financing would appear to be an ideal use of long-term bond
issues. There is also potential in long-term fixed-income demand from
institutional investors such as life insurance, pension systems (public
and private), and individual trusts.
Conclusion: Constraints and Opportunities
in the Philippines
The Philippines represents an emerging economy that continues to
chart its own unique course when it comes to developing its subnational
debt markets. The Philippines has been innovative in its efforts to extend
the legal possibilities for local governments to take initiative in the use of
credit and in the design of credit market techniques to make this possible. The Local Government Code, with its broad array of borrowing
powers granted to LGUs, and the creation of the LGUGC insurance
company to bolster local credits, are pioneering efforts.
Nonetheless, while a few urban and sophisticated Philippine local
governments have been able to take advantage of these initiatives,
many of the country’s LGUs remain passive in their efforts to move
ahead and proactively use the new powers to improve their condition.
The Philippines: Recent Developments in the Subnational Government Debt Markets
There are binding constraints that weaken the capacity of LGUs for
better infrastructure service delivery. The constraints relate to the local
finance and accountability systems. As has been noted, this low level of
accessing the debt markets has averted serious fiscal problems for local
governments, but it has also resulted in limited local initiatives to promote economic growth.
Fostering competitive subnational credit markets has been a longstanding policy goal in the Philippines. Notwithstanding the Philippines’s
efforts to develop more diversified subnational credit markets, the transition to PFI financing has been slow to materialize, and the GFIs continue
to be the main lenders to LGUs. The development of competitive subnational credit markets will need to address both demand- and supplyside constraints. On the demand side, it is critical to strengthen the local
finance and accountability systems for citizens to demand better services.
On the supply side, removing constraints to private participation in the
market will increase competition and help lower the cost of financing.
It is also helpful to experiment with those financing instruments that
can forge closer links between own revenue of local governments and
their capacity to access the market, which in turn help strengthen local
accountability. The recent experiments of encouraging greater partnerships between the local governments and the private sector credit markets
could pave the way for a more competitive and diversified subnational
credit market.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. The terms subnational government and local government are used interchangeably in this chapter. While the country’s legal framework for decentralization is reflected in the Local Government Code of 1991, the term subnational
government is used to maintain consistency with usage in other chapters in
this book volume.
2. Studies consulted include the following: ARD Inc. (2000), ARD Government
Finance Group (2001), Janet Tay Consultants (2007), Llanto (2007), Pellegrini
and Soriano (2002, 2006), and World Bank and Asian Development Bank
(2005).
445
446
Until Debt Do Us Part
3.
The government financial institutions mainly consist of the Development
Bank of the Philippines and the Land Bank of the Philippines and two specialized onlending institutions: the Local Water Utilities Administration and the
Municipal Development Fund. Section four provides details on government
financial institutions.
4. In “Declaration of Policy,” Section 2, Chapter 1, Title One, Book I of Republic
Act No. 7160, the Local Government Code of 1991.
5. World Bank and Asian Development Bank 2005.
6. Llanto 2011.
7. As of March 31, 2012. Data from the National Statistical Coordination Board,
National Economic and Development Authority.
8. Manasan (2005) and authors’ calculations.
9. Llanto 2011.
10. Llanto 2011.
11. Llanto 2011.
12. Petersen 2004, 463.
13. Llanto et al. 1998.
14. The borrowing for long-term capital investment is called the “Golden Rule.”
Liu and Waibel (2008) review how various developing countries have adopted
the Golden Rule in regulating their subnational debt financing, as part of their
regulatory reform in managing subnational finance.
15. Section 303 of the Local Government Code of 1991.
16. Very little appears to be left out of revenues, except extraordinary items and,
perhaps, some national grants. The IRA and national wealth payments make up
98 percent of all national payments. No local own-sources seem to be included
(except for extraordinary items). For practical purposes, it appears that about
99 percent of all own-source and national-source revenues are included.
17. To calculate the net remaining debt ceiling, from that maximum amount is subtracted the annual debt service (which is called “amortization”) on existing debt.
To calculate “borrowing capacity,” the net remaining debt ceiling is multiplied
by an annuity factor that corresponds with the maturity terms of the principal
repayments and the interest rate on the proposed debt. In other words, the debt
capacity certification looks at the particulars of the proposed loan or bond issuance to see whether, after the borrowing is consummated, there is remaining
debt capacity.
18. Petersen 2004, 464.
19. As a policy, DOF will not issue a sovereign guarantee for an LGU loan directly
sourced from a foreign lender, including multilateral institutions.
20. See National Economic and Development Authority (Philippines), 2011 http://
www.neda.gov.ph/econreports_dbs/NIA/GNP_GDP.
21. For example, an LGU that is unable to fully complete a capital investment project in a given year can carry over the remaining appropriation to the following
year and complete the expenditures then. However, the official reporting systems
The Philippines: Recent Developments in the Subnational Government Debt Markets
prior to 2011 did not capture these expenditures from “continuing appropriations,” hence, resulting in underreported expenditures and overreported surpluses. A recent LGU data analysis done by the World Bank estimated that the
aggregated annual LGU surpluses after adjusting for the expenditures from “continuing appropriations” range from half to one-third of what had been reported.
22. Unfortunately, BLGF does not provide balance sheet information. Also note
that borrowing receipts are reported as “income,” which is technically incorrect.
However, adjustments to the local government income numbers in the aggregate
have little impact on the revenue figures because of the low level of borrowing.
23. In that regard, the Bangko Sentral ng Pilipinas has asked the BLGF to review
projects to be financed by LGU bonds. However, the BLGF, aside from doing the
basic debt burden and capacity measures, does not appear to employ such technical capability at present.
24. The direct debt figure is a bit low because it does not include outstanding debt
of water districts or local water utilities, which is considered as indirect debt of
subnational governments. Water utility debt amounted to about 17.7 billion
pesos as of end-December 2007. However, even if the number were twice as
large, the percentage of GDP would still be only slightly over 1 percent. As of
end-December 2009, LGU debt as reported by BLGF was 70 billion pesos. There
was another 21 billion pesos in loans made to the water districts by Local Water
Utilities Administration.
25. Petersen and Soriano 2008. The sample comprised 20 emerging and transitioning countries and their estimated subnational debt as of 2006.
26. An interesting contrast is provided by China, where urban development and
finance corporations owned by subnational governments borrow from financial
markets to finance mainly infrastructure investments. There are varying estimates of the size of subnational debt in China, averaging around 16 percent of
GDP (Liu 2010).
27. The United States, with its federal system, provides a contrast to most other
countries. Combined debt of the subnational governments equaled about 20
percent of GDP (U.S. Federal Reserve System, Flow of Funds). In the United
States, the vast majority of public services are assigned to the subnational
(state and local) governments, and most basic infrastructure facilities (airports, education facilities, highways, ports, sanitation, sewerage, and water)
are owned and operated by subnational governments. Private sector provision
of infrastructure is dominant in the electricity sector and telecommunications.
Overall, state and locally owned facilities represent about 60 percent of ownership, and the private sector represents about 40 percent (see Petersen and Vu
2011, 4).
28. Unless otherwise noted, this section draws mainly from a World Bank mission
conducted during January–February 2011.
29. The analysis of the binding constraints draws from the World Bank (2010),
which provides a more detailed analysis of these constraints.
447
448
Until Debt Do Us Part
30. The examples can be found in the special vehicle arraignments in the United
States (see for example chapter 14 by Liu, Tian, and Wallis in this volume),
and intermunicipal cooperation arrangements in France covering a range
of services such as water supply, household waste collection, and sewerage.
France has a large number of small municipalities (see chapter 6 by Liu, Gaillard, and Waibel in this volume).
31. See Manasan 2007.
32. See Virola et al. (2007), whose study used 2004 and 2007 gubernatorial elections data.
33. Under the Land Administration and Management Project Phase II of the
World Bank-AusAID. DOF has a property valuation office, which is spearheading efforts to assist interested LGUs with proper land valuation and
assessment policies and techniques. The DOF Property Valuation Office will
be superseded by a National Valuation Authority under a bill sponsored by
the government creating the Authority, if it is voted into law by the Congress.
34. These higher-income LGUs receive a larger share of the IRA, but the share of
IRA in their total revenues is smaller than the lower-income LGUs, as noted in
section two, mainly due to the greater capacity of higher-income LGUs to raise
own revenues.
35. LGUs are reportedly reluctant to undertake large infrastructure projects with
the private sector, partly due to their lack of experience in working with the
private sector on such projects.
36. The central government departments and agencies include the Department
of Public Works and Highways, the Department of Transportation and Communications, and regulatory bodies such as the Toll Regulatory Board, and the
Energy Regulatory Commission.
37. LGUs would first have to organize a water district. LWUA was established to
lend to water districts.
38. The Municipal Development Fund is not a GFI per se but is lumped with the
Land Bank and the Development Bank of the Philippines to form the three
biggest government lending institutions providing loans to LGUs. Technically,
MDFO is a bureau of the Department of Finance and is not a “free-standing”
GFI institution. All GFIs and MDFO are subject to DOF oversight.
39. Llanto et al. 1998, 4. At an exchange rate of 50 pesos per US$1 (in 1995).
40. To this day, the exact status of the Philippine National Bank seems uncertain.
While the bank has private ownership, it is still listed as a GFI by the Department
of Finance for purposes of dealing with the LGUs (see “Department of Finance
Annual Report” 2010).
41. It is estimated that about 80 percent of lending to LGUs is for capital projects
and the other 20 percent is for cash flow purposes (borrowing in anticipation
of collections of taxes or aid payments).
42. For a review of fiscal rules with respect to debt service limits, see Liu and
Waibel (2008).
The Philippines: Recent Developments in the Subnational Government Debt Markets
43. International statistics vary on the definitions. Generally, the U.S. numbers for
subnational units include both limited and unlimited obligations as local debt.
The other OECD countries usually do not count locally owned enterprises that
are self-sustaining as local government debt, and the limitation caps do not
cover that debt. Practices vary and this can lead to some of the differences in
local debt numbers.
44. Discussions with Philippine bankers over the years indicate cautious interest
in lending to LGUs without resorting to the IRA pledge. For example, a selffinancing project could “pay for itself ” without using up the borrowing capacity of the LGU. This can be done through the use of a “trust fund” that would
have assigned revenues, such as from an enterprise’s operations. (This is called
the “special fund doctrine” in the United States and is the basis of the “revenue
bond” structure.) Typically, debt limitations do not apply to this form of debt,
which is viewed as “commercial,” although carried out by a government unit.
That is true in the United States and in the European Union. The test for issuance is purely a market test: Will investors feel secure in holding the debt, given
the security that is pledged? They have no recourse to general revenues. Not
surprisingly, there are many bonds that are “mixtures.” In the case of a loan
default, we understand that the LWUA has a “step-in provision” that it will
operate the utility instead of the defaulting water district.
45. These obligations are called “double-barreled bonds” since they are revenue
bonds when they are self-supporting but can become general obligations when
they are not self-sustaining. The general practice is to not count them against
debt limitations when they are self-sustaining.
46.
A large private commercial bank has total LGU loan approvals of around
322 million pesos, up from 61 million pesos in 2007, without the benefit of a
depository relationship with LGUs. It lends only to 1st and 2nd class LGUs under a
loan guarantee provided by the LGUGC, and further secured by a deed of assignment of the IRA that is signed by the LGU borrower. At present, the LGUGC has
given a loan guarantee of 3 billion pesos to LGU loans of this commercial bank.
Only the MDFO is allowed to use the IRA intercept before IRAs (or other national
government payments) are deposited. When it comes to GFIs and the private
banks, the loan security is the deposit accounts of the LGUs. Those deposits consist
of the IRA, own-source revenues, and other LGU receipts. More accurately, banks
can seize those deposits, but not the IRA funds, per se. This is formally known as
a right of offset, where the banks can make up any shortfall in the loan payments.
47. The repayment rate on LGU loans is high—about 98 percent. This level of
repayment is also enjoyed by the MDFO, which has a full-fledged intercept (as
opposed to the offset privilege used by the GFIs).
48. GFI loan officers are reportedly under pressure to keep their LGU clients.
49. At its inception, 51 percent of the LGUGC was owned by the Bankers Association of the Philippines and 49 percent by the Development Bank of the Philippines. In 2002, the Asian Development Bank bought a 25 percent ownership.
449
450
Until Debt Do Us Part
50. Caloocan City issued three bonds amounting to 620 million pesos on December 5, 2000. The 185 million pesos in bonds sold for a public market was fully
redeemed by May 14, 2003.
51. LCUCG records as of December 2010.
52. As a matter of policy, the LGUGC will insure 100 percent of the debt service
payments on bonds. It will insure only up to 85 percent of the debt service on
insured bank loans.
53. From 2003 to 2006, several cities issued bonds backed by the Philippine Veterans Bank. These cities included Batangas, Masbate, and Tacloban. The bonds
financed, among other things, a fishing port, a cold storage facility, a public
market, and a transport terminal.
54. There were some timing factors. Interest rates in the Philippines fell rapidly
in the mid-2000s, which allowed the GFIs to refinance the earlier municipal
bond deals. However, it was believed by PFI market participants that the refinancing done by the GFIs on attractive terms was intended to deter market
entry.
55. The 200-million-peso bond flotation of Cagayan Province for the construction
of a commercial center in Tuguegarao City was guaranteed by a private insurance company, with a sister company acting as trustee bank.
56. The GFIs have extensive branches (particularly the Land Bank). About 65 percent of all Land Bank deposits are those of governmental units, including the
LGUs. The Land Bank has a central role in the payments mechanism: it acts as
the paying agent for all national government transfers to the LGUs, so the vast
bulk of payments (including IRAs) are made through its accounts.
57. Bonds and loans require approval by the local sanggunian (municipal councils).
However, bond issues must also be reviewed by the Philippine central bank.
58. Other program lenders have noted that the use of the IRA offset against debt
service (popularly known as the intercept) discourages LGU borrowing from
other programs that required meeting various program requirements. Historically, the costs of bond issues have typically ranged from 3 to 5 percent of
the bond size, including the LGUGC insurance premium of 1–2 percent. See
Exhibit C of Appendix E.1 of ARD (2001).
59. In the case of IFI loans, the national government imposes a surcharge of
4 percent on top of the IFI Libor-based loan rate. When the Libor is 4 percent, then the all-in IFI loan rate is 8 percent for the GFI. To this rate, the
GFI may add a markup of 3 or 4 percent, which means a loan rate of 11–12
percent to the LGU. That rate might be compared to a conventional market
rate (say, on long-term mortgages). But, as is recounted in this report, the
GFIs are the main lender to LGUs and have LGU deposits and can use the
“intercept” to back up their loans. Thus, from a credit perspective, the intercepted-reinforced LGU loan is virtually riskless and close to the credit quality
of the national government’s own obligations, although they are clearly not
as liquid.
The Philippines: Recent Developments in the Subnational Government Debt Markets
60. The Land Bank has its own internal rating system for LGU borrowers, which
is proprietary. The LGUGC screen ratings (which are an overall guide and not
attached to a particular loan) are public information.
61. The LGUGC is attempting to get its insured loans made to LGUs and water districts (as opposed to its insured bonds) also eligible for the favorable Agri-Agra
treatment.
62. The Agri-Agra requirements are much like community investment requirements in the United States that are aimed at encouraging bank investment in
local businesses.
63. Based on January 2011 field interviews with a number of private financial
institutions.
64. The Japan Bank for International Cooperation, the United States Agency for
International Development, the Development Bank of the Philippines, MDFO,
and LGUGC worked together and established the PWRF.
65. This recognizes that some LGU-sponsored projects may not involve the pledge
of IRAs (or other general tax receipts) but would be secured solely on project earnings. Recent loans involving water districts evidently carry this “limited
recourse” provision, where the reserves and income accounts are pledged by
assignment to the repayment of debt.
66. The management of water districts is overseen by a board of directors, who serve
overlapping six-year terms. Directors can only be removed for “cause” and do
not serve “at the pleasure” of the mayor or governor. As a practical matter, the
water districts are generally insulated from the day-to-day politics of the LGUs.
67. This lack of repute among investors (no matter what the credit record) could be
an argument for intermediation such as a bond bank, where individual “names”
are submerged into the portfolios.
68. The security can be provided through the loan contract by empowering LWUA
to take over the operation if there is a default (the step-in provision). This is
already the practice—the step-in provision is in the agreements with water
districts.
69. The special fund doctrine means that the revenues in the funds are dedicated to
a particular purpose and cannot be used for other purposes. In revenue-bond
transactions, the operation of the fund is restricted and carried out in accordance with the loan contract (an indenture). Such funds can be subsidized by
general funds but can only transfer revenues to the general fund as is consistent
with the contract. Often, the indenture is closed, meaning that all revenues and any
surpluses must go to retiring debt. An alternative is to create the fund as a separate
legal entity (a special district). The water districts in the Philippines are separate
corporations and provide a model for that approach.
70. Water district financing under the revolving fund will have a put option for PFI
banks. Evidently, at five to seven years, private banks can exercise a put option,
which obliges the Development Bank of the Philippines or MDFO to absorb or
“buy back” the loan. LGUGC is providing the insurance “wrapper.”
451
452
Until Debt Do Us Part
71. See, for example, ARD Government Finance Group (2001), Chapter 6.
72. This included the use of BOT (build-operate-transfer)-type financing and the
use of municipal bonds. The lower-income LGUs and more social or environment-oriented projects would be supported by the MDFO, as it was then
called, with an emphasis on long-term loans, matching grants, and technical
assistance.
73. See Appendix 1-A in Pellegrini and Soriano (2002).
74. For the development of revenue bonds in the United States, see chapter 14 by
Liu, Tian, and Wallis in this volume. The adoption of the special fund doctrine
as a means of financing LGU capital needs will require a regulatory framework
that establishes financial rules, enforces contracts, and ensures transparency
and disclosure.
75. Sales-tax revenue may also increase, and jobs may be added, although these factors and their multipliers usually do not influence the structure of the tax incremental financing.
Bibliography
ARD Government Finance Group. 2001. Capacity Building in Local Government Unit
Finance. TA 3349-PHI, Final Report to the Asian Development Bank, Manila,
September.
ARD Inc. 2000. Strengthening Local Government Finance. ADB TA 3145–PHI, Asian
Development Bank, Manila, April.
Bird, Richard. 2010. “Subnational Taxation in Developing Countries: A Review of the
Literature.” Policy Research Working Paper 5450, World Bank, Washington, DC.
Bureau of Local Government Finance of the Philippines. (various years). Annual
Report on Local Government Revenues and Expenses. Manila.
Canuto, Otaviano, and Lili Liu. 2010. “Subnational Debt Finance: Make It Sustainable.” In The Day After Tomorrow: A Handbook on the Future of Economic Policy
in the Developing World, 219–38, ed. Otaviano Canuto and Marcelo Giugale.
Washington, DC: World Bank.
Department of Finance. 1996. LGU Financing of Basic Services and Infrastructure
Projects: A New Vision and Policy Framework. Manila.
———. 2000. “Joint Memorandum Circular: Establishing a Monitoring Framework
and Disclosure System for Local Government Bond Flotations.” Manila, Third
Draft, November.
———. 2009. Annual Report DOF Bulwark of Strength and Stability. Manila.
———. 2010. “Department of Finance Annual Report.” Manila.
Financial Executives Institute of the Philippines. 2000. Municipal Bonds: A Manual.
Makati: Financial Executives Institute of the Philippines.
Gavino, Carlos. 1998. “LGU Financing: Present Sources, Availability, and Terms.”
Coordinating Council of the Philippines Assistance Program and U.S. Agency
for International Development, Manila.
The Philippines: Recent Developments in the Subnational Government Debt Markets
Janet Tay Consultants. 2007. “Local Government Finance and Budget Reform.” ADB
TA 4556-PHI, Asian Development Bank, Manila, November.
Liu, Lili. 2010. “Strengthening Subnational Debt Financing and Managing Risks.”
Review of Economic Research, Ministry of Finance, Beijing, August 16, 46
F-9.
Liu, Lili, and Michael Waibel. 2008. “Subnational Borrowing, Insolvency and Regulations.” In Macro Federalism and Local Finance, ed. Anwar Shah, 215–41.
Washington, DC: World Bank.
Liu, Lili, Norbert Gaillard, and Michael Waibel. 2013. “France’s Subnational Insolvency System.” In Until Debt Do Us Part: Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and Lili Liu, 221–60. Washington, DC: World Bank.
Liu, Lili, Xiaowei Tian, and John Joseph Wallis. 2013. “Caveat Creditor: State Systems of Local Government Borrowing in the United States.” In Until Debt Do
Us Part: Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and
Lili Liu, 539–90. Washington, DC: World Bank.
Llanto, Gilberto M. 2007. “On the Rationalizing of Credit Programs for the Water
Sector.” DAI—Philippine Water Revolving Fund (PWRF) Support Program,
Manila, April.
———. 2011. “Fiscal Decentralization in the Philippines: Status and Emerging
Policy Issues.” UN-HABITAT, Nairobi, Kenya, October.
Llanto, Gilberto M., Rosario Manasan, Mario Lamberte, and Jaime Laya. 1998.
Local Government Units’ Access to the Private Capital Markets. Manila: Philippine Institute for Development Studies.
Manasan, Rosario G. 2005. “Local Public Finance in the Philippines: Lessons in
Autonomy and Accountability.” Philippine Journal of Development, Second
Semester, XXXII (2): 31– 102.
———. 2007. “Decentralization and the Financing of Regional Development.”
In The Dynamics of Regional Development: The Philippines in East Asia, ed.
Arsenio M. Balisacan and Hal Hill, 275–315. Manila: Edward Elgar.
National Economic and Development Authority (Philippines). 2011. http://www
.neda.gov.ph/econreports_dbs/NIA/GNP_GDP/nia2010fy.pdf.
Pellegrini, Anthony, and Cecilia Soriano. 2002. A Study to Revisit the LGU Financing Framework and Its Implementation. Final Report, Department of Finance,
Philippines, August.
———. 2006. Status Report/Issue Paper on LGU Financing in the Philippines.
Asian Development Bank, Manila, Draft, October.
———. 2004. “The Philippines.” In Subnational Capital Markets in Developing
Countries, ed. Mila Freire and John Petersen, 461–85. Washington, DC: World
Bank; New York: Oxford University Press.
Petersen, John E., and Cecilia Soriano. 2008. “Sources of Local Government Unit
Financing.” Asian Development Bank, Manila, September.
Petersen, John E., and Ha T. T. Vu. 2011. “Infrastructure Investment in the United
States: Grading the Decision-Making Process.” Paper presented at the World
453
454
Until Debt Do Us Part
Bank—KDI School Conference on Fiscal Policy and Management, Seoul,
Korea, November 14–16.
Tan, Roberto B. 2011. “Developing Subnational Debt Market: The Philippine Case.”
Presentation at the World Bank Seminar on Debt, Insolvency, and Markets:
Lessons Learned and Emerging Issues, June 20.
U.S. Federal Reserve System, Board of Governors. (various years). Flow of Funds for
the United States. http://www.federalreserve.gov/releases/z1/.
Virola, Romulo A., Severa B. de Costo, Noel S. Nepomuceno, Kristin Agtarap, Ma.
lvy T. Querubin, and Mai Lin C. Villaruel. 2007. “Governance Statistics: Did
Performance Matter in the 2007 Elections?” National Statistical Coordination
Board, Manila.
World Bank. 2010. “Approach Paper on a World Bank Strategy for Supporting
Decentralization Reforms.” Other Public Sector Study 68623, Washington, DC,
August.
World Bank and Asian Development Bank. 2005. “Decentralization in the Philippines.” Washington, DC, November 5.
12
Russian Federation: Development
of Public Finances and
Subnational Debt Markets
Galina Kurlyandskaya
Introduction
Since the formation of the modern Russian state in 1991, the development of the subnational debt market in the Russian Federation has gone
through three distinct phases: the 1990s, 2000–08, and 2008 onward.
How and why subnational governments (SNGs) accessed the financial
markets throughout the three phases has been substantially shaped by
the evolving macroeconomic conditions in Russia and the development
of an intergovernmental fiscal system since 1991.
During the first phase, the 1990s, the system of intergovernmental
relations was highly centralized. The central government controlled
subnational spending standards and norms, set prices for housing
and utilities services, regulated wages of government employees, and
arbitrarily determined the shared taxes assigned to each region. Intergovernmental fiscal transfers were negotiated between the federal Ministry of Finance and regional governments. During this period, Russia
went through an unstable and uncertain macroeconomic situation in
the early 1990s, followed by stabilization from 1992 to 1997. But the
high fiscal deficit of the federal government and extensive use of internal and external borrowing to cover budget deficits contributed to the
455
456
Until Debt Do Us Part
accumulation of federal government debt. This was followed by growing federal debt, financial crisis, and federal default on debt obligations
in the late 1990s.
It is within this context in the 1990s that the subnational debt
market began its early development and reached its peak in capital
market development in 1997, and was then followed by defaults by
most regional governments during 1998–2000. The growing demand
on SNGs, unfunded federal mandates, and political decentralization
contributed to the growing demand for debt instruments by SNGs. At
the same time, there was a complete lack of debt regulation, and SNGs
contracted debt through informal negotiations with the federal government. SNGs also lacked experience and capacity in managing debt
risks. Debt was issued to finance recurrent expenditures, mostly with
short-term maturities. With a rapidly deteriorating macroeconomic
environment in Russia in the late 1990s, refinancing risks facing SNGs
rapidly rose, and the federal government, with its own macroeconomic
woes, was not in a position to provide support to SNGs. Fifty-seven of
89 regions defaulted on their debt from 1998 to 2000.
The subnational debt market entered its second period of development in 2000, and its development up to 2008 was helped by macroeconomic stabilization and success in Russia. Improved m
acroeconomic
fundamentals contributed to positive changes in intergovernmental
relations and incentives for new principles of financial management
for the regions and municipalities. Gradually, financial capabilities
of the Russian regions began to improve. Substantial legislative
reforms—significant amendments to the Tax Code and the adoption
of the Budget Code—were undertaken, aimed at the creation of a new
formula-based system of relations among the tiers of government to
replace the outdated, nontransparent, and informal arrangements.
The 2006 legislation on local self-government established a uniform
two-tier system of local self-government. Since the second half of the
2000s, the federal government has been paying more attention to the
quality of public finance management in the regions and municipalities, which, in the long term, underpins the access by SNGs to marketbased financing.
The Budget Code regulates subnational debt. Specifically, it specifies
(a) sources for budget deficit financing; (b) limits on the size of fiscal
Russian Federation: Development of Public Finances and Subnational Debt Markets
deficit, debt, and debt service; (c) regulations on external borrowing
and on guarantees; and (d) sources of deficit financing and structure
and types of debt instruments. In addition, the Budget Code, and especially the 2007 amendments, regulates the system of granting federal,
regional, and municipal guarantees. The Budget Code also establishes
the structure of regional and municipal debt, its types, and maturity.
Favorable terms of international trade and successful domestic economic development in Russia, especially between 2004 and 2008, have
largely neutralized some negative aspects of the reform of intergovernmental fiscal relations for some of the Russian regions and municipalities. Their financial positions have strengthened considerably, together
with revenue growth.
The debt load of the Russian regions remained low at the end of
2007. The regional debt was unevenly distributed across regions; five
large regions accounted for about half of regional debt. Another feature
of the debt of Russian regions was its short-term character. Short-term
bank loans were a major debt instrument for most Russian regions. The
majority of SNGs had little experience in debt management and had
only short credit histories. During 2000–08, regions and municipalities
showed increasing interest in obtaining credit ratings from rating agencies. Access to capital markets, however, is limited to the most creditworthy SNGs.
The global financial crisis of 2008–09 severely struck Russian public
finances in 2009, though the impact varied across regions and municipalities. The strong regions that relied on their own tax capacity or
exports were among the most severely affected, while the regions with
a greater dependence on federal transfers appeared to be less affected.
Bank loans and federal government loans became the main debt instruments of SNGs. The major problem for the Russian regions was not the
absolute size of a debt load, but its payment structure. Reduced debt
maturity terms created a substantial risk for refinancing and high debt
service costs—factors that put pressure on their budgets.
A key difference between the 1998 and 2008–09 crises was the lack of
defaults of regions on their debt obligations during the latter, owing to
additional support from the federal government and the liquidity accumulated during previous years by the regions. Nonmarket instruments
for financing the deficit of subnational budgets predominated in 2010.
457
458
Until Debt Do Us Part
Since 2011, subnational fiscal positions have improved, and the Russian
economy has gradually recovered. The debt markets have recovered, and
borrowing costs have been reduced. But activity in the domestic bond
market remained moderate until 2011, when the market expanded. The
debt repayment profile continued to be short due to the large share of
short-term bank loans in the debt structure.
This chapter analyzes the development of the subnational debt market in Russia over the last 20 years, and shows how subnational debt
market development is influenced and shaped by key macroeconomic
developments and the evolving structure of the intergovernmental fi
scal
system. The chapter is organized as follows: section two describes the
development of the subnational bond market from birth to expansion
and from crisis to recovery over the 10-year period, 1991–2000. Section
three summarizes key reforms in the intergovernmental fiscal system
from 2000 to 2008 and how they have shaped the regulatory framework
for subnational debt borrowing. Section four discusses the impact of the
2008–09 global financial crisis on subnational fiscal performance and
subnational access to borrowing. Section five concludes with remarks
on the challenges to the continuing development of the subnational
debt market in Russia.
The Subnational Bond Market in Russia: Birth, Expansion,
Crisis, and Recovery, 1991–2000
The formation of the modern Russian state began in 1991. The new
state inherited its federal structure from the Soviet Union. The current structure of the government includes (a) the federal government,
(b) 83 regions, (c) self-governments of the first-tier municipalities
(520 larger cities and 1,793 rural districts), and (d) self-governments
of the second-tier municipalities (1,732 townships and 19,919 rural
communities).1 In the 1990s, the system of intergovernmental relations was highly centralized. The central government controlled
subnational spending standards and norms, set prices for housing and
utilities services, and regulated wages of government employees. The
shared taxes assigned to each region were arbitrarily determined by the
central government. Intergovernmental fiscal transfers were negotiated
between the federal Ministry of Finance and regional governments.
Russian Federation: Development of Public Finances and Subnational Debt Markets
Russia faced an unstable and uncertain macroeconomic situation in
the early 1990s, and the Russian government undertook a series of steps
to stabilize the economy. The steps included (a) the creation of the federal treasury payment system, (b) phasing out of monetary financing to
cover federal spending, (c) reducing wage and other spending arrears
in the government sector, (d) liberalization of exchange rate regulation,
(e) setting substantial foreign currency reserves, and (f) curbing inflation. However, the annual deficits of the federal government in the second half of the 1990s reached 7–10 percent of gross domestic product
(GDP). Consequently, the government made extensive use of internal
and external borrowing to cover budget deficits, which contributed
to the accumulation of debt. In addition, nonfinancial instruments—
barters and offsets—were heavily used.
Russian regions and municipalities also faced difficulties. The budgeted deficit was rather high in each region, and in many regions it
reached 50 percent of revenues, including transfers, while revenue
collection was low and unstable. Regions and municipalities actively
resorted to short-term borrowing to finance their expenditures. In the
mid-1990s, bank loans and bond issues became popular. At the same
time, regulation of the subnational debt markets was erratic and under
political influence. There were no restrictions on the amount of borrowing or the purpose of borrowed funds.
In practice, regions and municipalities borrowed mainly for current
expenditures, on disadvantageous terms. They did not develop payment
or debt refinancing plans. As a result, subnational entities had accumulated a significant amount of short-term debt between 1993 and 1996.
In addition, in the 1990s, the Russian regions relied heavily on such
unconventional short-term debt instruments as accumulation of budgetary overdue payables. In 1998, budget payables of SNGs amounted
to 15.3 percent of subnational spending, or 4.7 percent of GDP. The
budgetary payables, however, were nearly fully offset by overdue taxes to
SNGs, thus creating the mechanism of noncash spending.2
Development of Regional and Municipal Bond
Markets in the 1990s
During 1992–93, the bond issuance process was experimental in nature.
The first bond was issued by the regional government of Khabarovsk
459
460
Until Debt Do Us Part
Krai in March 1992. The Ministry of Finance registered only five bond
issues by regions and municipalities in 1992 and eight in 1993. Bonds
were placed and circulated on isolated regional markets with inadequate
bond market infrastructure. A clear legal framework did not exist, investors demonstrated weak interest, and market infrastructure was lacking.
From 1994 until the macroeconomic crisis in the late 1990s, regions
and municipalities became more active in using bond instruments as
an alternative to bank financing. The Ministry of Finance registered
28 issues of regional and municipal bonds in 1994 and 73 in 1995. In
1995 and 1996, some regions used bonds to finance 50 percent of their
budget deficit. Regional and municipal bills became more popular (as
a convenient instrument for which registration was not required); they
were used by the authorities to untie the knot of nonpayments and
finance budget expenditures. However, after enactment of the law “On
the Promissory Note and the Bill of Exchange” in 1996, the government
banned issuance of guarantee bills. As a result, they were replaced by
regional and municipal bonds, whose issues had increased dramatically.
The peak was registered in 1997, with 313 issues of regional and municipal bonds worth 29.5 billion rubles (almost US$5 billion).3 Hyperinflation and the lack of a regulatory framework were the main challenges
to the development of the subnational bond market at that time.4
In 1996, Russian SNGs for the first time entered the external market
for borrowing. The regional governments of Moscow and St. Petersburg
issued Eurobonds.5 Borrowing in a foreign currency was motivated by
significantly more attractive borrowing terms (lower rates, longer terms
of loans, large amounts of borrowing); SNGs assumed that these terms
would continue given the then fixed exchange rate system (the fixed
exchange rate was abandoned in the late 1990s macroeconomic crisis).
During 1992–96, the Russian subnational bond market demonstrated the following features6:
• Issuers (Russian regions and municipalities) did not provide pro-
spectuses, including the purpose of borrowing, of acceptable quality.
Therefore, market participants had no reliable information on the
financial and economic status of a jurisdiction. The bulk of the funds
received from the bond issue was used to cover temporary cash gaps
or simply to increase revenues.
Russian Federation: Development of Public Finances and Subnational Debt Markets
• When issuers selected agents for placing and maintaining bond pro-
ceeds, most of them gave strong preference to local financial institutions, which were small and unreliable. Thus, strong investors, facing
high risks dealing with regions and municipalities, refused to enter
the regional market.
• No generally accepted and guaranteed payment mechanisms existed
for issued bonds. Often, debt service costs were not shown as a separate line item in regional and local budgets.
• A significant portion of bonds was used as a barter instrument for tax
payments to the regional budget, which only increased the flow of real
financial resources out of a region.
By 1997, the macroeconomic situation in Russia had improved: GDP
began to show positive growth, and inflation and interest rates declined.
Foreign capital flows increased into the sovereign debt market in
response to the reduction of political risk after the 1996 presidential
elections and the progressive external financial liberalization. In general,
the attitude of domestic and foreign investors to Russian debt securities
improved—for the first time since credit ratings were assigned to them.7
The activity of regional and municipal authorities in the debt securities market reached its peak in 1997, with the city of Moscow the largest
and most active borrower. The total number of registered issuances of
regional and municipal bonds increased more than eightfold from 1996
to 1997 (from 39 to 313). On January 1, 1998, the Ministry of Finance
of the Russian Federation registered 446 issues, totaling an equivalent
of US$8.3 million. About 70 percent of the total number of bond issuances during 1992–97 was registered in 1997.8 One of the features of the
regional and municipal bond market in 1997 was again that a significant portion of issuances had no particular purpose. Most of the bonds
(60 percent) were short or medium term. Conditions in the securities market worsened by the end of 1997; interest rates began to rise
and became unstable, while the maturity of bonds slightly increased.
As a result, the pressure on regional and municipal budgets increased
(figure 12.1), making debt policy planning more difficult.
Despite this, regions and municipalities continued to issue securities until May 1998. From January to May 1998, the Ministry of Finance
registered 59 bond issues. However, the average monthly value of bond
461
Until Debt Do Us Part
Figure 12.1 Composition of Consolidated Subnational Borrowing in the Russian
Federation, 1995–2000
8
Deficit/surplus as % of revenues
6
4
2
0
–2
–4
–6
–8
–10
00
20
99
19
98
19
97
19
96
19
95
–12
19
462
Year
Bank loans
Federal loans
Regional bonds
Cash residuals
Deficit (–)/surplus (+)
Source: Author’s estimates based on Russian Federation Treasury data.
issues fell (in U.S. dollar equivalent) from US$350,000 in 1997 to
US$260,000 in 1998.9
The macroeconomic crisis in Russia in 1998 was triggered by both
external and internal factors. The global economic slowdown that
began in 1997 was responsible for the decline in world demand for oil,
which had a negative impact on the Russian federal budget. Internal
factors—weak monetary and fiscal policy, a fixed exchange rate, excessive government borrowing—also contributed to the crisis in August
1998. The federal government defaulted on virtually all domestic obligations and was forced to cancel the fixed exchange rate. The crisis led
to a sharp devaluation of the ruble, a fall in real GDP by 5.3 percent, a
jump in annual inflation to 84 percent, and the collapse of the banking
system.10
From June to August 1998, 28 regional and municipal governments
continued issuing their bonds, though the capacity of the Russian financial market was falling dramatically as a result of an abrupt outflow
of foreign capital from Russia and a withdrawal of resources from the
Russian Federation: Development of Public Finances and Subnational Debt Markets
securities market by nonresident investors, followed by conversion of
these resources into hard currency. During June–August 1998, the first
defaults of regions and municipalities on their bonds were reported.
However, the cause of the defaults was not only the worsened financial
situation but also the reluctance of issuers to pay their obligations.11
The majority of borrowing in the precrisis era was short term and
was used to finance the current budget deficit; only a small part of
borrowing was to finance capital investments. Regional governments
were more active in the bond market than municipalities. Before
1998, only a few subnational entities had obtained credit ratings from
international rating agencies.12 The regional governments of Moscow
and St. Petersburg managed to get ratings from several international
agencies.13 In that period, issuers operating in the domestic debt market
got ratings to improve their image and demonstrate their openness to
the investor community. However, credit ratings had almost no effect
on the amount and cost of borrowing. At the same time, it was essential
for the regions wishing to access foreign debt markets to obtain credit
ratings from international agencies. Prior to 1998, there were only three
regional governments that borrowed in the foreign debt market—the
cities of Moscow and St. Petersburg and Nizhny Novgorod Oblast.
Despite the rapid development of the regional and municipal bond
market in the 1990s, its share was only a small part in the overall Russian
securities market. In 1997, bonds issued by regions and municipalities
accounted for only 6.6 percent of the total bond market in Russia. The
growth of Russian Government Treasury Bills and Federal Loan Bonds
was rapid, and new types of government securities were emerging, targeting small and individual investors. But the institution of underwriting had yet to be formed for both the federal government and SNGs.
Lessons of the Crisis and Beginning of the Recovery, 1998–2000
The events of 1998 demonstrated the risks of an unregulated debt market. External borrowing was attractive at a fixed exchange rate, but the
devaluation significantly increased the debt load of SNGs that borrowed
in foreign currency. Even those regions that relied solely on domestic
borrowing could not avoid default, since they had not received expected
transfers and shared taxes from the federal government, whose situation was also grave. Between 1998 and 2001, 57 of 89 Russian regions
463
464
Until Debt Do Us Part
declared default on their debts. In fact, only the federal cities—Moscow
and St. Petersburg—continued to make payments on their debt obligations.14 Figure 12.2 summarizes defaults by type of debt instrument.
A considerable share of these defaults was related to nonpayment on
so-called agrobonds (as part of subnational bonds in figure 12.2). In
the mid-1990s, many Russian regions received financial assistance from
the central government for their agricultural sector. However, later the
central government decided to convert this assistance into bonds. Many
Russian regions considered this to be unfair and refused to repay after
the 1998 crisis. At that time, the credit culture in Russia was relatively
weak. For example, a newly elected governor or mayor could question
debts raised by previous administrations.
The subnational defaults offer the following lessons: (a) borrowing in
a foreign currency in the absence of reliable hedging instruments in an
unstable macroeconomic environment is extremely risky, (b) unfettered
market access by subnational borrowers can outpace the development
of sound revenue systems and adequate security, and (c) it is essential
to create a regulatory system for regional and municipal borrowing and
debt market development (Alam, Titov, and Peterson 2004).
Figure 12.2 Subnational Defaults by Type of Debt Operation
Bank loans in
national
currency, 2%
Federal loans,
17%
Bank loans in
foreign currency,
51%
Subnational
bonds, 17%
Eurobonds,
13%
Source: Standard & Poor’s Global Credit Portal 2004.
Russian Federation: Development of Public Finances and Subnational Debt Markets
The economy began to recover in 1999. Real GDP grew by 6.4 percent
in 1999 after falling by more than 50 percent during 1991–98. While
inflation peaked at 84 percent in the postcrisis year of 1999, it fell to below
20 percent in 2001. Foreign currency reserves grew from US$12 billion in
1998 to US$37 billion by the end of 2001.15 Several factors contributed to
the strengthening of the Russian economy and the competitive position of
the country in the world market, including devaluation of the ruble, subsidized domestic energy prices, and rising world prices of oil and gas. These
strengthened the export sector, which remained stable.
Fiscal performance also improved with increased tax revenues and
restricted expenditure growth. In 2000, the first federal budget surplus of 1.2 percent of GDP was recorded. In 2001, the surplus rose to
2.4 percent of GDP. The ratio of public debt to GDP decreased dramatically—to 43 percent in 2001 from 145 percent in 1998 due to
the strengthening of the ruble, inflation effects, partial debt payment,
and a write-off of US$10.6 billion in debt under a London Club debt
restructuring agreement.16
Eventually, starting in 2000, Russian regions and municipalities began
to restore their solvency. Assets of regional and local governments grew
rapidly. According to the Central Bank of Russia, on November 1, 2001,
bank deposits of regions and municipalities amounted in US$1.8 billion equivalent, while their liabilities to the banking system were US$730
million equivalent. Arrears of the regions also decreased significantly.17
During 1998–2000, increased financial resources were mainly used to
refinance existing debt rather than to accumulate new debt. At the same
time, by 2001 the surplus of regional budgets had reduced the subnational
securities market by more than US$170 million equivalent. Changes in
debt structure in favor of bank loans had also contributed to this reduction (figure 12.1). Nevertheless, despite the decline in the bond market
as a whole, new subnational borrowers entered the market, while interest
rates gradually returned to precrisis levels.
Public Finance Reform, Debt Regulations, and
Subnational Bond Market Development, 2000–08
Improved macroeconomic fundamentals in the postcrisis period contributed to positive changes in intergovernmental relations and incentives for
465
466
Until Debt Do Us Part
new principles of financial management for the regions and m
unicipalities.
New legislative and administrative initiatives aimed at the creation of a
new formula-based system of relations among the tiers of g overnment
to replace the outdated, nontransparent, and informal arrangements.
During the early 2000s, the legislative framework was reformed, with
significant amendments to the Tax Code and the adoption of the Budget Code. Uniform standards for the allocation of tax sources were
established; expenditure responsibilities were clearly assigned among all
levels of government, accompanied by the reduction of unfunded mandates; and fiscal discipline has gradually been strengthened. The federal
government has established a system of incentives for the development
of financial management in the regions and municipalities.
On the whole, the macroeconomic situation between the early 2000s
and the 2008 crisis can be characterized as successful. A favorable external economic environment contributed to the growth of the Russian
economy by an average of 7 percent per year. Inflation and unemployment gradually declined, while gold and foreign currency reserves
demonstrated growth. Under these conditions, the federal budget had
significantly improved and showed a surplus instead of a deficit. Thus, it
was possible to dramatically decrease borrowing at the federal level and
reduce the amount of debt to 8 percent of GDP.
Gradually, the financial capabilities of the Russian regions began
to improve. In the majority of them, revenues grew faster than expenditures, which enabled the governments to improve current budget performance and, ultimately, to direct more resources to capital
projects to maintain and restore worn-out infrastructure. During the
2000s, the credit culture of Russian SNGs strengthened significantly.
Debt books were cleaned, and all questionable debts inherited from
the 1990s were eliminated. Many regions continued to be dependent
on federal government transfers, but the level and quality of support
from the federal budget had changed significantly due to reforms in
fiscal relations.
Public Finance Reform in the 2000s
For the federal government, the improved economic situation had
provided an incentive to implement harmonized reforms in pub
lic finances aimed at a common goal: improving the quality of public
Russian Federation: Development of Public Finances and Subnational Debt Markets
finance management at all levels of government—federal, regional, and
municipal. Several important laws had been enacted, the most important of which was the Budget Code of the Russian Federation (January
1, 2000), which established the basic principles for the system of federal
and subnational finances. The main provisions of the Budget Code in
the area of debt relations will be discussed later.
The reform of intergovernmental fiscal relations was one of the most
important for the Russian regions and municipalities. It addressed the
following targets: (a) clear assignment of expenditure responsibilities
across the tiers of government, (b) elimination of unfunded mandates
to a lower level of government without provision of necessary financial resources, (c) allocation of taxes and tax shares among the levels of
government on a long-term basis, and (d) formula-based equalization
transfers to regional governments that took into account per capita fiscal capacity of a region and differences in costs of public services across
regions.
Municipal governments were recognized as independent participants
in fiscal relations, which became one of the fundamental features of
the local self-government reform, as reflected in Federal Law #131 of
October 6, 2003, “On General Principles Underlying the Organization
of Local Self-Government in the Russian Federation” (“Law on Local
Self-Government”).
The law established a uniform two-tier system of local self-government. The first tier included townships and rural villages, and the second tier included municipal districts consisting of several townships or
villages and cities. The latter referred to major cities, such as capitals
of regions and several large, developed cities with dense populations
within a region.
The provision of the law stating that henceforth all levels of local
government were full-fledged participants in economic and financial
relations was revolutionary. It meant that all municipalities (including
even rural settlements with small populations) were to set up municipal governments, employ municipal office staff, formulate and execute
budgets, and conduct an independent debt policy. The law assigned a
set of expenditure responsibilities to each level of local self-government
(local government issues) and the Budget Code specified their revenue
sources.18
467
468
Until Debt Do Us Part
Before the reform, even large municipalities, including the capitals
of regions, depended financially on transfers from regional governments. Municipal financial authorities were territorial subdivisions of
the regional finance department. Most small municipalities carried out
mandated financial activities given to them by higher levels of government. The 2003 “Law On Local Self-Government” actually prohibited
such practice, and higher levels of government can no longer mandate
activities to small municipalities.
The new system was to fully come into force on January 1, 2006.
However, given the complexity of changes and the need for training of
municipal officials, municipalities were given a three-year transitional
period. Though some “brave” municipalities moved to a new system in
2006, most of them took advantage of the transitional period and completely switched to the new principles in January 2009, when the reform
was finally implemented.
Since the second half of the 2000s, the federal government has been
paying more attention to the quality of public finance management
in the regions and municipalities. A Fund for Regional and Municipal Finance Reform (the Fund), to be administered by the Ministry of
Finance, was established as part of the federal budget. The Fund was
allocated among regions and municipalities on a competitive basis. Two
separate competitions were held annually—one among the Russian
regions, the other among the municipalities. Regions and municipalities
participating in the contest had to prepare and submit to the Ministry
of Finance a plan of the public finance reform in their jurisdictions. The
areas to be reformed were delineated by the Ministry of Finance and
included (a) improvement of the quality of public services, (b) introduction of a program-oriented budget, (c) raising revenue sources,
(d) reforming the extended state (municipal) sector, and (d) debt management and estimation of debt capacity.
Regions (municipalities) were to identify the specific mechanisms of
reform options and assess the implementation risks of planned activities. Annually, no more than eight winners were determined among both
regions and municipalities whose programs were the most convincing.
Also, they received funds only as a result of successful completion of
the first and second stages of their plan. The received resources could
be spent on funding the plan implementation, which usually covered a
Russian Federation: Development of Public Finances and Subnational Debt Markets
wide spectrum of activities, including the purchase of computer hardware and software, organization of refresher courses for public (municipal) employees, fulfillment of social obligations, and debt repayment.
Assessing Reform Results
The reforms were certainly an important step toward a better quality
of public finance management in the regions and municipalities. The
reform proved to be efficient in the following areas19:
• The revenue sources and expenditure responsibilities have been clearly
assigned across levels of government and fixed in the legislation. A unified two-tier system of local self-government has been introduced across
all Russia and secured by law.
Unfunded mandates have been mostly eliminated. When federal legislation imposes additional expenditure responsibilities on lower-tier
governments, the federal government allocates adequate funds to fulfill
those responsibilities. This allocation takes place in a timely manner.
A transfer formula has been introduced to equalize the fiscal capac-
•
•
ity of regions and municipalities. This was especially important in
view of the huge economic disparity among regions and the resulting uneven revenue capacity. The equalization formula, which has
not changed much since then, is based on a proportional increase
of per capita revenue capacity of regions below the national average.
The formula also takes into account the differences in costs of public
goods delivery.
Russian regions and municipalities have been encouraged to improve
the quality of public finance management. The federal government has
allocated grants to regions and municipalities introducing best principles in public finance management, including debt management
procedures. These grants have been allocated on a competitive basis.
However, the reform was followed by a number of backward steps in the
following areas:
• The allocation of responsibilities among the levels of government
has been modified every other year. These modifications prevent
the regions and municipalities from pursuing a predictable and
long-term fiscal policy. Moreover, these modifications are being
469
470
Until Debt Do Us Part
introduced by the federal government at the end of the fiscal year,
when regional and municipal budgets for the next period have
already been formulated. Regional and municipal governments have
to amend their budgets in early January.
• The revenue sources of SNGs depend on transfers. Eighty-five
percent of the regions receive transfers from the federal government
in the form of equalization grants, and every region receives earmarked grants. The federal government has at its disposal 200 types
of earmarked grants that are provided to regional governments.
Each region might receive all 200 types or fewer. Since grants are
a more or less permanent source of income, the regions have less
incentive to develop their own revenue base. At the same time, the
value of each grant is unpredictable before the fiscal year begins.
However, the federal government has tended to reduce financial
support to regions. Municipalities can obtain transfers only from
regional governments. Regional transfer mechanisms usually replicate the federal transfer policy regarding regions.
As for municipalities, their revenue sources are indeed very limited. There are only two local (municipal) taxes: the personal property tax and the land tax. Their tax compliance is very poor because
of problems with taxpayer registration. In addition, the administration of the land tax requires a cadastral valuation of land, which is
within the power of the federal government, and the establishment
of a fair market land price in the relevant documents. The portion
of the federal shared taxes assigned to municipal governments by
federal law (the personal income tax being the largest) improves the
situation in regional economic centers. About 20 percent of regions
grant additional shares of the regional taxes in favor of municipalities to improve municipal tax capacity.
• Implementation of the regional and municipal finance reform
programs is often a mere formality and does not contribute to the
actual quality of public finance management. For some regions and
municipalities, the sole purpose of writing and implementing programs has become a way to obtain additional funds from the federal government. They hired consulting firms in order to develop
and implement a detailed program. If some regional administrations were really motivated to improve the system of public finances,
Russian Federation: Development of Public Finances and Subnational Debt Markets
others demonstrated a formal approach to the reform and did not
bother with the details of the reform plan proposed by the consultants. Some regions abolished regulations whose adoption was part
of the implementation of program activities after the money from
the federal government had been obtained.
Despite these shortcomings, the reforms meant the emergence in
Russia of an institutional environment to be further developed. With
the adoption of the Budget Code, the sphere of public debt management regulations has also undergone significant changes.
Major Provisions on Debt Management in the Budget Code
The Budget Code contains provisions for regulating the subnational
debt by specifying (a) the limits on the size of fiscal deficit, debt, and
debt service; (b) regulations on external borrowing and guarantees;
and (c) sources of deficit financing and structure and types of debt
instruments.
As prescribed in the Budget Code, the budget deficit of a region
should not exceed 15 percent of its annual revenues, excluding all
intergovernmental transfers. Local budget deficits should not exceed
10 percent of annual revenues, excluding all intergovernmental transfers and (or) revenues from shared federal and regional taxes. The 2007
amendment to the Budget Code tightened those limits for highly subsidized regions and municipalities. The highly subsidized regions are
those in whose budgets the share of intergovernmental transfers (except
particular earmarked transfers) exceeded 60 percent of revenues (except
particular earmarked transfers) of the regional consolidated budget. For
such regions, the deficit should not exceed 10 percent of their revenues,
excluding all intergovernmental transfers. For highly subsidized municipalities (where intergovernmental transfers exceed 70 percent of their
revenues, except particular earmarked transfers), the deficit limit was
established at 5 percent of annual revenues, excluding all intergovernmental transfers.
The Budget Code stipulates that the outstanding debt of a region or
municipality should not exceed its annual revenues, excluding intergovernmental transfers. The 2007 amendments provide that the outstanding debt of the highly subsidized regions and municipalities should not
exceed 50 percent of their annual revenues, excluding intergovernmental
471
472
Until Debt Do Us Part
transfers. The Budget Code also stipulates that the debt service of a region
or municipality must not exceed 15 percent of expenditures of the relevant year. Table 12.1 summarizes fiscal and debt rules for SNGs as stipulated in the Budget Code and its amendments.
Before 2000, foreign borrowing by a region or a municipality was not
expressly prohibited by law, but in practice, it could only be undertaken
on authority of a special presidential decree. The right to place loans
abroad before 2000 was granted by the Russian president to the cities
of Moscow and St. Petersburg, Nizhny Novgorod, Moscow, Sverdlovsk,
Leningrad, Orel and Samara Oblasts, Krasnoyarsk Krai, and the Republic of Komi. These regions had issued external debt in various currencies
to finance their budget deficits.
The Budget Code enacted in 2000 set severe restrictions on external
borrowing. Since 2000, the regions could only borrow in foreign currency to refinance existing foreign debt. Currently, only Moscow has
maintained its presence in the Eurobond market.20 All other regional
and municipal governments issue obligations only in rubles.21 The
2007 amendments to the Budget Code granted the regions an opportunity to borrow in foreign currency starting January 1, 2011. The
prerequisite for this is financial independence of a region from federal
support; that is, the share of federal transfers (excluding funding federal mandates) should not exceed 5 percent of own revenues. Also, the
Table 12.1 Fiscal and Debt Rules for Subnational Governments
Regions
Highly subsidized
regions
Municipalities
Highly subsidized
municipalities
Deficit/revenue (excluding
transfers) ceiling
10%
10%
10%
5%
Debt service/expenditure
ceiling
15%
15%
15%
15%
Rule
Total debt/revenue
(excluding transfers)
ceiling
Term of borrowing ceiling
A ban on external (foreign
currency) borrowing
100%
50%
100%
50%
30 years
30 years
10 years
10 years
Allowed to refinance
foreign debta
Allowed to refinance
foreign debt
Source: The Budget Code of the Russian Federation, Articles 92.1, 99, 100, 104, 107, and 111.
a. Only those Russian Federation subjects in whose budgets the share of intergovernmental transfers was less than
5 percent during the past three years are allowed to take new loans from January 1, 2011.
Russian Federation: Development of Public Finances and Subnational Debt Markets
Russian government would develop a foreign borrowing procedure for
the regions. However, at the time of writing, the procedure had not yet
been developed.
The Budget Code, and especially the 2007 amendments, regulate the
system of granting federal, regional, and municipal guarantees. Two
types of guarantees have been established: recourse and nonrecourse
guarantees. The recourse guarantee is a guarantee issued by a region or
municipality to take on an obligation to pay the debt of a third party,
with the right to claim the debt from the debtor. The called guarantee,
according to the Budget Code, is recorded as an account receivable and
not as an expenditure item. The nonrecourse guarantee is a guarantee issued by a region or municipality to take on an obligation to pay
the debt of a third party, without the right to claim the debt from the
debtor. The called nonrecourse guarantee is subject to recording as an
expenditure item. In practice, both types of guarantees are widely used
by regions and municipalities.
The Budget Code also established the sources for budget deficit
financing and limited the size of the budget deficit, debt, and its service.
The sources of financing the budget deficit of a region or municipality
include the difference between (a) the funds received from the placement of securities and funds allocated for repayment, (b) received and
repaid loans of credit institutions, and (c) received and repaid loans
from government and from international financial institutions. It also
includes changes in the balance of funds on accounts and other sources
of financing the budget deficit. More specifically, other sources include
proceeds from the sale of shares and other equity, foreign exchange rate
differences, and funds allocated for execution of government guarantees.
The Budget Code also established the structure of regional and
municipal debt and its types and maturity. The debt structure of regions
and municipalities is limited to government securities, intergovernmental loans, loans from financial institutions, and guarantees. The maximum maturity of debt obligations is set at 30 years for the regions and
10 years for municipalities. At the same time, the federal government is
not liable for debt of regions and municipalities that are not guaranteed
by the federal government.
The 2007 amendments to the Budget Code included a requirement for regions and municipalities to record and register their debt
473
474
Until Debt Do Us Part
bligations in their respective financial statements. The recorded inforo
mation includes the types of debt, the date of contracting the debt, and
the date of repayment in full or in part, forms of security, and arrears.
The information from regional and municipal debt books is subject
to mandatory transfer to the Ministry of Finance, which monitors the
regional and municipal observance of the constraints imposed by the
Budget Code. If a region or municipality violates such constraints, it
will be unable to incur new debt until the situation once again meets the
requirements of the Budget Code.
The limit on the size of debt raises a question about the evaluation of
a subnational’s creditworthiness. The size of debt itself is not the main
factor affecting the creditworthiness of the government. Much more
important are the structure of debt and the cost of debt service, namely,
the period of repayment, the currency of borrowing, the interest rate,
and the amortization of debt payment. The limit on budget deficit also
raises an issue for financing large-scale infrastructure projects with
major capital requirements. A severe limitation on the total amount of
borrowing during one financial year may have a negative impact on the
effectiveness of the investment policies pursued by regional and municipal governments, since regional and municipal governments may have
to scale down the size of investment and lose the economies of scale in
infrastructure networks.
Since guarantees are subject only to debt limits but not to current
deficit limits, guarantees (nonrecourse) are being used by regional and
municipal governments to finance their own expenditures through issuing guarantees in excess of deficit limits set by the Budget Code to the
companies controlled by the relevant regional or municipal government
that borrow on behalf of the government to finance government infrastructure projects. The funds to repay the company’s debt come from
grants provided to these companies from regional or municipal budgets. This is a way to avoid restrictions on current deficits as long as the
total debt stays within the legal limitations.
The Budget Code does not specify restrictions on the use of borrowed funds. Even though best practices recommend using borrowed
funds for funding only capital investments, Russian SNGs still often
borrow to finance current expenses such as payroll and maintenance
costs.22 Unfortunately, the Budget Code does not prevent borrowing
Russian Federation: Development of Public Finances and Subnational Debt Markets
to finance current deficits. Therefore, the most frequent explanation
in financial statements under the “purpose of borrowing” entry is “to
finance budget deficit,” without further specification of costs and projects to be covered by the identified resources.23
The Subnational Debt Market during 2000–08: Growing Interest
in Credit Ratings
Favorable terms of international trade and successful domestic economic development in Russia, especially between 2004 and 2008, have
largely neutralized negative aspects of the reform of intergovernmental fiscal relations for some of the Russian regions and municipalities.
Their financial positions have strengthened considerably together with
revenue growth.
Between 2003 and 2008, the aggregate revenues of SNGs more than
tripled in absolute terms (while consumer prices nearly doubled). At
the same time, there was a tendency in regional budgets to increase
the share of current and social expenditures at the expense of capital
spending.24 This was largely due to the social security reform started
in 2005, during which the federal government encouraged regions and
municipalities to increase public sector wages by 50 percent in real
terms during 2006–08. However, the two-year period (2005–06) of
nearly 15 percent real growth per year of budget revenues was over in
2007. To balance their budgets, SNGs had to curtail the growth of current spending. This was a difficult task, given the inertia of the budgetary process and decisions already taken to increase the wages of public
sector employees.
The growing burden of current expenditures on regional budgets
has been exacerbated by the reduction of intergovernmental transfers to
regions (table 12.2). There is a large disparity in economic development
among Russian regions; in fact, a major feature of Russian fiscal federalism is the large disparity in the revenue capacity of the regions. In 2007,
for example, the wealthiest region was 38 times richer than the poorest
region. Federal equalization transfers reduced the gap to eight times the
fiscal capacity.25
From 2003 to 2005, the share of total federal grants to the regional
governments declined from 19 percent of the aggregate income of
regions of Russia to 15 percent and stayed stable until 2008. In 2008,
475
476
Until Debt Do Us Part
Table 12.2 Annual Growth of Subnational Revenue, Including Fiscal Transfers, 2004–08
percent
2004
2005
2006
2007
2008
Real growth, subnational (regional +
municipal) revenues, including transfers
11.9
14.0
16.2
-3.9
33.9
Real growth GDP
7.2
6.4
8.2
8.5
5.2
Real growth, aggregated federal equalization
transfers to regional governments
-8.6
4.5
9.8
2.0
11.0
Real growth, aggregated total federal
transfers to regional governments
-4.2
2.4
27.5
-8.5
58.9
Source: Author’s estimations based on data provided by the Russian Statistical Agency, http://www.gks.ru, and the Russian
Federation Treasury, http://www.roskazna.ru.
Note: GDP = gross domestic product.
total grants grew to 18 percent of consolidated regional revenues and
in 2009, reached 25 percent. As summarized by table 12.3, the federal
government gradually shifted the focus in the area of intergovernmental
fiscal relations from the equalization schemes to the stimulation of economic development of regions through special purpose transfers. The
share of earmarked transfers for execution of federal mandates also rose.
However, the debt load of the Russian regions remained low at the
end of 2007 (figure 12.3). According to the Ministry of Finance, total
debt of the Russian regions (excluding municipal debt) at the end of
2007 was US$18 billion equivalent, or about 11 percent of aggregate
regional revenues, including fiscal transfers to regional governments
in 2007.26 The regional debt was unevenly distributed across regions;
five regions—the city of Moscow, Moscow Oblast, Samara Oblast, the
Republic of Tatarstan, and the Republic of Sakha (Yakutia)—accounted
for about half of the regional debt.
Another feature of the Russian regions’ debt was its short-term character. Short-term bank loans were a major debt instrument for most
Russian regions (excluding the city of Moscow and Moscow Oblast,
which were the two largest borrowers on the securities market at the
time). What is more, most regions (excluding the city of Moscow) placed
bonds for a term not exceeding three years. The majority of SNGs had
little experience in debt management and had short credit histories.
During 2000–08, regions and municipalities showed increasing interest in obtaining credit ratings from the rating agencies. The number of
Russian Federation: Development of Public Finances and Subnational Debt Markets
477
Table 12.3 Intergovernmental Fiscal Transfers from the Russian Federation to Regions as a
Percentage of Each Type of Transfer in Total Transfers, 2003–10
percent
Transfers
2003
2004
General grants
58
54
Of which equalization grants
45
43
Earmarked transfers for cofinancing
regional programs
15
26
Earmarked transfers for execution of
federal mandates
27
19
Other transfers
Total
2005
2006
2007
2008
2009
61
61
39
47
27
10
2010
40
35
39
37
31
29
25
28
17
38
38
36
30
20
22
16
19
27
0
0
2
1
0
11
6
6
100
100
100
100
100
100
100
100
Source: Author’s estimations based on regional fiscal reports on the Russian Federation Treasury website, http://www
.roskazna.ru.
Figure 12.3 Federal and Subnational Debt as a Share of GDP
160
140
120
Percent
100
80
60
40
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
00
99
20
97
98
19
19
19
19
96
0
Year
Federal debt
Subnational debt
Source: Author’s estimates based on Russian Federation Treasury data.
Note: GDP = gross domestic product.
ratings assigned to Russian regions by Fitch Ratings increased from only
4 in 2003 to 25 in 2008. However, the majority of ratings of Russian
regions and municipalities belong to the so-called “speculative-grade”
(from “BB” to “D” categories, according to the international ratings scale
by Fitch); that is, they indicated a fairly high level of credit risk. Only a
478
Until Debt Do Us Part
few regions, including the federal cities of Moscow and St. Petersburg,
the Republic of Tatarstan, and the Khanty-Mansi and Yamal-Nenets
Autonomous Okrugs (according to Fitch Ratings and Standard &
Poor’s), had ratings of “investment grade” (“BBB-” and above).27
Access to capital markets is limited to the most creditworthy SNGs.
As shown in figure 12.4, Moscow Oblast accounted for over one-third of
bond issuance, followed by the city of Moscow at 30 percent.
Lack of sustainable legal provisions and poor quality of debt management are the major constraints on increasing credit ratings of Russian
regions and municipalities.28 As a result, the average long-term ratings
of Russian regions in foreign currency are significantly lower than similar ratings in European countries. The difference between Russia and
Poland, the nearest country in rank, is four steps on the international
rating scale. Moreover, in Russia, there is the greatest gap between the
country’s sovereign rating (long-term rating “BBB” in foreign currency)29 and the median rating of Russian regions (long-term rating
“BB-” in foreign currency). In this case, the gap is four steps, whereas in
other European countries it does not exceed one or two steps.30
Figure 12.4 Regions’ Share of Total Regional Bond Debt Outstanding at the
End of 2008
Samara oblast,
7%
Republic of Sakha
(Yakutia),
3%
Yaroslavskaya
oblast, 2%
Nizhny Novgorod
oblast, 2%
Moscow oblast,
35%
Other regions,
21%
City of Moscow,
30%
Source: Author’s estimations based on regional fiscal reports on the RF Ministry of Finance website, http://www1
.minfin.ru/ru/public_debt.
Russian Federation: Development of Public Finances and Subnational Debt Markets
Despite the fact that the crisis in Russia began to develop in the
fourth quarter of 2008, when industrial production in the regions
began to decline, most of the regions demonstrated quite satisfactory fiscal indicators on a yearly basis. Due to a good economic situation in the first six months in 2008, the economic growth in most
regions showed a positive trend on a yearly basis. Budget indicators
also remained satisfactory. Only in some regions with large industries,
which were more vulnerable to the crisis, did economic decline become
apparent in 2008.31 This happened in Nizhny Novgorod Oblast, which
was significantly affected by the financial woes facing the automobile
industry (the industry has since recovered).32
The Global Economic Crisis and Its Effects:
Trends and Prospects
Impact in 2009
The global financial crisis that began to unfold in the fourth quarter of
2008 severely struck Russian public finances in 2009. The budget deficit
in most regions had increased, owing to a sharp decline in revenues and
an inability to cut expenditures. Federal support had helped mitigate
the financing gap in many regions but was not enough to fully compensate for falling revenues. As a result, the regions faced significantly larger
budget deficits. The fiscal deficit grew from US$1.8 billion equivalent
(or 0.8 percent of total aggregated revenue and 1.1 percent of revenue
net of transfers) in 2008 to US$8.7 billion equivalent (or 5.7 percent
of total aggregated revenue and 8.2 percent of revenue net of transfers) in 2009, of which about US$4.6 billion equivalent was accounted
by the city of Moscow.33 Part of the overall deficit was financed from
fiscal reserves accumulated during the previous years. However, the
larger part of the cumulative deficit in 2009 was covered by a more than
50 percent increase in borrowing by subnational entities. The new debt
had shorter maturities that increased refinancing risks. Soaring interest
rates also contributed to increasing debt service costs. Interest rates on
bank loans reached 20 to 22 percent in some regions.34
The impact of the crisis varied among regions and municipalities.
The strong regions, which relied on their own tax capacity (mainly, the
personal income tax and the corporate income tax), were among the
479
480
Until Debt Do Us Part
most severely affected by the crisis. These were the regions with welldeveloped industries. The regions that depended on exports (metallurgical and oil-producing regions) were also vulnerable. However, the
regions with a greater dependence on federal transfers appeared to be
less affected. On the whole, federal support to the regions during the
crisis, mainly in the form of additional general purpose transfers, was
the most important factor mitigating the consequences of the crisis for
SNGs. In 2009, transfers increased by US$13.2 billion equivalent, or by
35 percent compared with 2008.35 This was a crucial difference from the
1998 crisis, when the Federation provided no support to the regions and
left them on their own.
In 2009, bank loans and federal government loans became the main
debt instruments of SNGs. Issuance of domestic bonds almost stopped
in the first half of the year. The only exception was the city of Moscow,
which issued bonds as early as January 14, 2009, to borrow US$460 million equivalent at a 10 percent interest rate, a lower rate compared with
costs available for other regions in 2009. Throughout 2009, Moscow
was an active player in the regional bond market, which allowed the city
to maintain a long-term debt profile and a smooth structure of debt
repayment.
In the second half of 2009, those regions that were traditionally
financially stronger began to return to the bond market. But despite
this, the number of issuers decreased from 23 regions in 2008 to
10 regions in 2009. The number of securities issues dropped even
more substantially—from 35 in 2008 to 16 in 2009.36 Most other
regions attracted short-term loans with a one-year maturity and
interest rates that were as high as 20–22 percent per year in the first
quarter of 2009. Subsequently, when the financial markets began
to stabilize, a number of regions negotiated the reduction of interest rates for the remaining period to mitigate the significant growth
in debt service costs. In 2009, the aggregate amount of bank loans
attracted by the regions increased by US$1.7 billion equivalent compared to the previous year.37
In 2009, federal loans to regions became very important. Together
with additional subsidies, federal loans were another form of indirect
federal support to the regions. The advantages of federal loans compared with bank loans were a lower interest rate and a longer repayment
Russian Federation: Development of Public Finances and Subnational Debt Markets
horizon. In 2009, the federal government issued loans amounting to
US$4 billion equivalent to regional governments, an eightfold increase
over 2008.38
Starting in 2009, the federal loans were granted for three years. Previously, as mandated by the Budget Code, they were to be repaid within
one year. The interest rate on federal loans was one-quarter of the refinancing rate of the Central Bank of Russia39—a subsidized rate for the
regions to finance budget deficits when access to loans was limited and
interest rates in capital markets were extremely high. As mentioned, the
market rates peaked at 20–22 percent in the first quarter of 2009, but
they declined to 11–13 percent by the end of that year.40
Federal loans can be divided into two groups according to their
purpose. The first group of loans is the general purpose loans that
are provided to cover current budget deficits. They, as a rule, have an
amortized repayment structure; 60 percent of the principal is paid off
in the second year and the remaining 40 percent in the third year. Such
a structure smoothes out the peaks of debt repayment and makes the
debt profile more favorable.
The second group is special purpose loans for construction, reconstruction, and repair of public roads. In this case, the principal is
repayable at the end of the term. In 2009, the aggregate direct debt
of the Russian regions increased to a US23.8 billion equivalent, that
is, by almost 60 percent over the previous year. However, compared
to the total revenues of Russian regions in 2009, it constituted a mere
15 percent, which, by international standards, is quite manageable.
The major problem for the Russian regions was not the absolute size
of a debt load but its payment structure. Reduced debt maturity terms
created a substantial risk for refinancing and high debt service costs—
factors that put pressure on their budgets.
Also worth mentioning are the contingent liabilities of subnational
entities that include guarantees and financial debts of affiliated regional
and municipal enterprises in times of crisis. Despite the fact that on the
eve of the crisis the popularity of guarantees had somewhat declined,
SNGs still use them widely to support the local economy. During the
crisis, regions used different tactics regarding guarantees. Some opted
not to incur additional risk and in difficult financial conditions refused
to issue guarantees to enterprises so as not to impose additional risks
481
482
Until Debt Do Us Part
on the government. At the same time, many regions supported major
local taxpayers that faced great difficulties obtaining loans to develop
their businesses. In that case, it was important to carefully select companies according to the most stringent criteria to mitigate possible negative consequences for the budget. The tightened rules envisioned by
the Budget Code had also positively contributed to mitigating potential
serious fiscal risks from guarantees.
Owing to the financial crisis of 2008 and the sharp deterioration
in the construction sector, government mortgage agencies in several
regions (Republic of Khakassia, Republic of Tatarstan, and Tomsk
Oblast) were not able to pay their liabilities, which rquired provision of
additional resources of the respective regional governments.
Novosibirsk Oblast provides an example of guarantee use in the difficult year of 2009. It granted about US$80 million equivalent of nonrecourse guarantees to enterprises of the construction sector, a most
vulnerable industry during the crisis. As a result, the construction sector, a major component of the service economy of Novosibirsk Oblast,
showed an increase in housing construction in 2009 compared to
2008, which was extraordinary for the Russian regions in times of crisis. Moreover, despite the nonrecourse nature of guarantees, construction companies managed to pay their debt and not add pressure on the
regional budget. Thus, in this case, the guarantees helped to substantially support an important sector of the economy and prevent a significant deterioration of the oblast budget execution process, which had
been predicted by many.41
A distinctive feature of the crisis in 2008 that differed from the 1998
crisis was the lack of defaults of regions on their debt obligations. This
was due to the additional support from the federal government and
the liquidity accumulated in previous years by those regions that were
able to foresee a possible deterioration in the economic and financial situation and prepare for it. Accordingly, credit ratings of most
regions remained at the precrisis level. Those regions that were not able
to restrain the growth of current spending in 2009 (and significantly
increased their debts) were negatively evaluated by the rating agencies.
Their ratings were downgraded or, in many cases, the forecast on ratings
was changed to “negative.”
Russian Federation: Development of Public Finances and Subnational Debt Markets
2010: Public Finance Improved, Debt Structure
Remained Vulnerable
In 2010, after the 18-month crisis, a gradual recovery of the Russian
economy took place within the context of the global economic recovery.
In addition to a return to favorable international economic conditions
for Russia’s major exports, there was a restoration of economic activity in general. Economic growth recovered from negative 7.8 percent
in 2009 to positive 4.3 percent in 2010, and export growth increased
from negative 35 percent in 2009 to positive 32 percent in 2010. This led
to a significant increase in tax revenues, especially from the corporate
income tax, an important revenue source for most Russian regions. In
2009, its reduction for regional governments amounted to an average of
40 percent compared to 2008. The level of industrial production in 2010
exceeded the precrisis level in half the Russian regions.42
The economic recovery contributed to public finance improvement at
the subnational level. In 2010, the aggregate revenues of SNGs (excluding Moscow) increased by approximately 10 percent compared to 2009
(while the 2010 inflation rate was about 9 percent). At the same time, the
revenue structure of SNGs changed in 2010; the regional governments
became less dependent on federal transfers. Thus, the share of various
transfers accounted for 32.5 percent of total SNGs revenues in 2010 compared with 37 percent in 2009.43 Despite the gradual recovery of tax revenue, especially from the corporate income tax, financial independence
of the majority of Russian regional and local governments remains low.
In 2010, the increase in aggregate spending of the regions reached
a moderate 9 percent, which is close to the inflation rate. However, the
composition of expenditure changed: social spending increased, while
capital expenditures decreased. The share of capital expenditures in the
total regional budget expenditures decreased from 18 percent in 2008 to
13 percent during 2009–10. However, given the high demand for infrastructure development, lower capital spending in most regions should
be considered a temporary phenomenon. A moderate increase in expenditure containment and an increased revenue base resulted in a lower
consolidated deficit of regional budgets in 2010—1.6 percent of total
aggregated revenue (compared to 5.7 percent for 2009) and 2.2 percent
of revenue net of transfers (compared to 8.2 percent for 2009).44
483
484
Until Debt Do Us Part
In 2010, the debt markets recovered, and borrowing costs decreased.
Though the debt load of the regions continued to increase, its growth
rate declined. The aggregate regional debt remained moderate, at
20 percent of total revenue in 2010. At the same time, regional distribution of debt remained uneven. Ten regions accounted for about
57 percent of the total debt in nominal terms. The relative debt burden was also unevenly distributed; the median region’s debt constituted 16.5 percent of total income, while the maximum level of debt
amounted to 60 percent of revenues (Astrakhan Oblast). Due to federal support, Russian regions compared favorably with SNGs in other
European countries (figure 12.5).
Notwithstanding the low debt as share of regional revenue, the structure of debt in most regions has considerable room for improvement.
Federal loans became the main source of deficit financing of regional
budgets in 2010, accounting for 45 percent of the total direct debt
(excluding Moscow). However, since 2010, the terms for repayment of
federal loans was extended up to five years, which extends the maturity
profile of debt for the regions. The share of bank loans as direct debt
amounted to 31 percent of all SNGs debt (excluding Moscow). Interest
rates on bank loans declined significantly in 2010 to 7–8 percent per
Figure 12.5 Outlook on Credit Ratings: Russian Federation Regions and
Subnationals in European Countries, End-December 2010
Spain
Russian
Federation
Poland
Italy
Germany
France
Europe
0
20
40
60
80
Percent
Negative
Source: Fitch Rating 2011.
Stable
Positive
100
Russian Federation: Development of Public Finances and Subnational Debt Markets
year and were close to the refinancing rate of the Central Bank of Russia.
This has significantly reduced debt servicing costs. In addition, banks
started to offer loans for longer terms, that is, for two to three years.
However, not all regions are ready for longer-term loans because the
interest rate on them is usually slightly higher than on one-year loans.
Activity in the domestic bond market remained moderate. Only 13
regions issued bonds in 2010 compared to 10 in 2009. New bond issues
have also had a longer period—from three to five years—compared
with one to three years in 2009.
These were nonmarket instruments for financing the deficit of subnational budgets, which predominated in 2010. Also, the debt repayment
profile continued to be short, owing to the large share of short-term
bank loans in the debt structure. The overall structure of regional debt
differs fundamentally from the debt structure of the city of Moscow,
where bonds with the maturity stretched to 2022 constitute up to
90 percent of the debt portfolio.
Continuing Improvement in Subnational Fiscal Positions and More
Active Subnational Debt Activities in 2011
The fiscal outcomes for the regional governments improved in 2011; the
deficit of the consolidated budget of the Russian regions was 0.45 percent
of revenues (excluding transfers) in 2011, down from 1.53 percent in 2010.45
The deficit shrinking would have been stronger if not for the growth of
spending commitments of all levels of government before the presidential
election in 2012. On the spending composition, social outlays of the regions
have outstripped their capital spending; in 2011, capital expenditures
were reduced by 20 percent compared to 2008, while spending on social
programs demonstrated a 71 percent increase.46 Though in 2011, regional
incomes practically recovered due to the growth of tax revenues (in the
first place, the corporate income tax), their expenditures also increased and
almost reached their revenue level.47
In 2011, the share of federal loans in the debt structure of the R
ussian
SNGs continued to grow and reached more than 40 percent; these loans
were provided at 4 percent per year, or at about half the minimum
market rate.48
During 2011, the debt of the Russian regions grew 13.8 percent in
real terms, but the total debt stock remained manageable at the end of
485
486
Until Debt Do Us Part
2011.49 However, the structure of the debt profile remains risky. The
short-term structure of the debt implies refinancing risks and the potentially high cost of repayment and debt service in the coming years.50
While the average debt level for the regions was 18.3 percent of revenues (including intergovernmental transfers), in six regions, the debt
level is 50–70 percent. Thirteen Russian regions had to spend over
15 percent of their income on debt service. The city of Moscow
accounted for 20 percent of all regional debts. In 2011, the city executed
its budget with a surplus, and borrowing, therefore, was not needed.51
A similar situation was typical for other “wealthy” Russian SNGs (for
example, oil-producing Tyumen Oblast and Yamal-Nenets Autonomous
Okrug), many of which were able to generate large financial reserves.
In 2011, the total volume of domestic borrowing of the regions and
municipalities was reduced by 15 percent.52 The trend of relative growth
of borrowing from commercial banks continued, with a reduction by
one-third of the share of issues of securities. Securities accounted for
9.1 percent of the total amount of internal borrowing of consolidated
regional budgets compared to 21.3 percent of federal loans and 70 percent
of borrowing from commercial banks and international credit institutions, which reflected a shrinking bond market.53
At the same time, the SNGs became more active in bond markets.
The number of issuers increased; 21 regions and 5 municipalities registered bond prospectuses (compared to 17 regions and 6 municipalities
that issued bonds in 2010). The three-largest issuers were responsible
for 53.1 percent of total outstanding regional and municipal bonds in
2011. On the whole, the city of Moscow, which accounts for nearly onehalf of outstanding bonds, continues to dominate the bond market.54
According to the “Guidelines for the State Debt Policy in 2012–14,”
approved by the Government of the Russian Federation,55 there will be
fundamental changes in intergovernmental fiscal relations. The volume
of federal loans to regional governments will be significantly reduced,
and loans will be granted only “in case of emergency.” As a result,
the document states, “there will be a growing need on the part of the
Russian regions in market borrowing.” Thus, only two debt instruments
would be available to the regions—bank loans and publicly sold bonds.
In addition, the regions would have to return budgetary loans taken in
the crisis, the peak of payments being 2012.
Russian Federation: Development of Public Finances and Subnational Debt Markets
Challenges for Deepening Subnational Debt Markets
Key challenges facing the continuing development of subnational debt
market include the following.
There is a need to develop sustainable legal provisions. Despite the
fact that the system of intergovernmental fiscal relations began to take
shape in the early 2000s, it is still under development. This entails an
unpredictable change in intergovernmental fiscal design, which is a risk
for each region and municipality in Russia. In recent years, the federal
government has often decided to change the allocation of tax revenues,
expenditure responsibilities, and transfer rules. At the same time, the
share of flexible tax revenues in Russia—that is, revenues the regions
and municipalities might influence (such as establishing a tax base or tax
rate and administering the collection)—in most cases, does not exceed
10 percent of total revenues of the regional and local budgets.
The budget classification of Russia is subject to annual changes.
In addition, due to the format of the budget classification (which is
obligatory for every level of government), the grouping of revenues
and expenditures into recurrent and capital ones can be achieved only
through sophisticated and time-consuming analytical research. It is also
not possible to obtain clear information on a number of crucial indicators of government performance in the area of public finance management and to do estimations of the true level of the debt burden.
Investment and debt policies are rarely coordinated by SNGs.
According to best practices, all debt resources should be used only for
investment purposes. However, in Russia, only a few regions operate
under this scheme (the cities of Moscow and St. Petersburg, mainly).
Most SNGs do not even try to compare their investment needs and volume of borrowing. This means that debt can be used both for investment and for current spending. In many respects, the underdeveloped
legislation makes the situation even more serious and does not allow for
realistic medium- and long-term budgeting and a long-term investment
strategy.
The subnational debt portfolio remains short-term in nature. In the
debt portfolio of most SNGs, one-year bank loans dominate, which
leads to higher refinancing risk. In 2011, however, this situation did not
cause particular concern because of the availability of sufficient liquidity in the Russian banking system. However, when liquidity tightens, the
487
488
Until Debt Do Us Part
creditworthiness of SNGs with a high proportion of short-term debt
may face refinancing risks. Debt maturity and interest rate structure are
among the most important parameters determining the quality of the
debt portfolio of a region or municipality.
Bank financing of regions is dominated by a few state-controlled
banks. Due to the underdevelopment of the banking system in Russia,
bank financing of SNGs is dominated by a few state-controlled banks
(primarily Sberbank and VTB). The dimensions of these banks, which
are based on the past branch networks of the then Soviet-state banks,
allow them to compete successfully with smaller-transaction banks, in
terms of pricing. However, both Sberbank and VTB comply with strict
market rules vis-à-vis their customers regarding issuance and repayment of loans.
There is a lack of comprehensive accounting for liabilities of government enterprises when assessing the total debt of a SNG. Financial
debts of government enterprises (their bank loans and bond issues) are
contingent liabilities of a subnational entity. The Budget Code does not
include such contingent liabilities in the debt definition for regions and
municipalities. This means that, formally, SNGs are not responsible for
the debts of such enterprises, unless the debt is explicitly guaranteed by
a government. For this reason, the vast majority of SNGs do not record
the debts of their subsidiaries (partially or fully owned by a government) as a component imposing risks. Government enterprises submit
their often incomplete and inconsistent financial statements with significant delay. At the same time, as recent events have shown, the regions
must be involved in solving problems caused by the inability of affiliated companies to repay their debts.
Conclusions
The subnational debt market in Russia began to develop in the early
1990s. Unfunded federal mandates and political decentralization
contributed to the growing demand for debt instruments, including
foreign currency debt. At the same time, there was a complete lack
of debt regulations and SNG lacked experience in managing debt
risks. Debt was issued to finance recurrent expenditures, mostly with
short-term maturities. With a rapidly deteriorating macroeconomic
Russian Federation: Development of Public Finances and Subnational Debt Markets
environment in Russia in the late 1990s, refinancing risks facing
SNGs rapidly rose. From 1998 to 2000, 57 of 89 regions defaulted on
their debt.
Improved macroeconomic fundamentals during 2000–08 and substantial legislative reforms—significant amendments to the Tax Code,
the adoption of the Budget Code, and the 2006 legislation on local
self-government—contributed to positive changes in intergovernmental relations and to incentives to formulate new principles of financial
management for the regions and municipalities. The current system of
intergovernmental fiscal transfers allocation is formula based, extends
to all levels of SNG, and takes into account key socioeconomic parameters. However, the formula changes often. Thus, the horizon for prediction of the amount of transfers is one or two years.
Russian legislation puts strict constraints on total debt amount,
annual budget deficits, and debt service. Moreover, with revenue
growth, the financial positions of the regions and municipalities
have strengthened considerably. The debt load of the Russian regions
remains low. Most Russian SNGs have no currency or derivatives risks.
Debt management and budgeting processes have become more transparent. The Treasury and the Ministry of Finance of the Russian Federation
publish updated information and data on budget execution and debt
obligations on a regular and timely basis.
The 2008–09 global financial crisis struck Russian public finances in
2009, though the impact varied across SNGs. But there were no regional
defaults, owing to support from the federal government and the liquidity accumulated in prior years by the regions. Since 2011, subnational
fiscal positions have improved along with a gradual recovery of the
Russian economy. The debt markets have recovered, and borrowing
costs have decreased. Activity in the domestic bond market remained
moderate until 2011, when the market expanded.
There are continuing challenges in subnational debt market development. Most SNGs have a short-term debt profile dominated by one-year
bank loans, implying higher refinancing risk. Bank financing of regions
is dominated by a few state-controlled banks. There is a lack of comprehensive accounting for the contingent liabilities of government enterprises. Debt management will need to be an integral part of fiscal policy
and will need to be integrated into the budget planning and execution
489
490
Until Debt Do Us Part
process. It is necessary to coordinate debt and investment policies and
to extend the horizons of budget planning. Subnational entities should
develop a detailed medium-term fiscal framework that forecasts financing gaps. The borrowing plan should include information on attracting
the new loans, their rates, and the repayment schedule of new and existing debt obligations, including contingent liabilities.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Subnational governments in Russia include (a) governments of the subjects
of federation (the official term, but informally called “regions”), and (b) local
self-government or municipal governments (the official terms, but informally
called “local governments”).
2. Institute for Economies in Transition 2003; Pinto et al. 2000; the RF Treasury website, http://www.roskazna.ru; and the Russian Statistical Agency website, http://
www.gks.ru.
3. Author’s estimation based on RF Ministry of Finance data at http://www1
.minfin.ru/ru/public_debt/capital_issue.
4. Pakhomov 2009.
5. Presidential Decree #304 of April 8, 1997 (as amended on March 20, 1998), “On
the issue of external bonded loans by the executive authorities of the cities of
Moscow and St. Petersburg and Nizhny Novgorod Oblast.”
6. Arakelyan, Kashcheev, and Shchyurikov 2001.
7. Pakhomov 2009.
8. Author’s estimation based on RF Ministry of Finance data at http://www1
.minfin.ru/ru/public_debt/capital_issue.
9. Author’s estimation based on RF Ministry of Finance data at http://www1
.minfin.ru/ru/public_debt/capital_issue.
10. The website of the Russian Statistical agency, http://www.gks.ru.
11. Arakelyan, Kashcheev, and Shchyurikov 2001.
12. Among them the Republics of Sakha (Yakutia), Komi, and Tatarstan, as well as
Novgorod, Samara, and Irkutsk Oblasts and several other regions.
13. Moscow and St. Petersburg are also called Federal Cities, which have the status
of subjects of the RF (in other words, regional governments). All other cities
have the status of municipalities.
14. Fitch Ratings, various issues.
15. The Central Bank of Russia, http://www.cbr.ru, and the Russian Statistical
Agency, http://www.gks.ru.
Russian Federation: Development of Public Finances and Subnational Debt Markets
16. Author’s estimation based on RF Ministry of Finance data, www1.minfin.ru/
ru/public_debt and Central Bank of Russia http://www.cbr.ru.
17. Central Bank of Russia website, http://www.cbr.ru.
18. The Law on Local Self-Government refers only to municipalities. Hence, in this
chapter, “local government” is consistent with “municipalities.”
19. De Silva et al. 2009.
20. State Debt Committee of the City of Moscow, http://www.moscowdebt.ru.
21. C-Bonds Financial Information, http://www.cbonds.ru.
22. Fitch Ratings 2008a.
23. Based on author’s technical assistance to regional governments.
24. Fitch Ratings 2008a.
25. By comparison, disparities in subnational revenues (richest to poorest per capita) in the Philippines are equal to 35.4 before grants, 28.1 after grants; and in
Brazil, 9.3 before grants, 4.9 after grants (Braga et al. 2002).
26. Author’s estimations based on regional fiscal reports on the RF Ministry of
Finance website, http://www1.minfin.ru/ru/public_debt and RF Treasury site
http://www.roskazna.ru.
27. Fitch Ratings, various issues, http://www.fitchratings.com. Standard & Poor’s,
various issues, http://www.standardandpoors.com.
28. Fitch Ratings 2008b. According to Fitch’s methodology, the regions are assigned
both a local currency long-term rating and a foreign currency long-term rating,
even if the regions do not borrow in foreign currency (the level of these ratings is
the same for Russian regions). This is done to enable a comparison of the level of
ratings of the regions from different countries using Fitch’s unified international
scale.
29. As of June 1, 2012, the long-term rating of the RF on the Fitch Ratings international scale is “BBB” with a “Stable” outlook.
30. Fitch Ratings 2012.
31. http://www.roskazna.ru.
32. http://autopeople.ru/gaz.
33. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru
34. Fitch Ratings 2011a.
35. Author’s calculations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
36. C-Bonds Financial Information, http://www.cbonds.ru.
37. Fitch Ratings 2011b.
38. Author’s estimates based on RF Treasury data on the RF Treasury website,
http://www.roskazna.ru.
39. As of June 1, 2011, the refinancing rate of the Central Bank of Russia was 8.25
percent.
40. Fitch Ratings 2011b.
41. Fitch Ratings 2010.
491
492
Until Debt Do Us Part
42. Author’s estimation based on data on the websites of the RF Treasury, http://
www.roskazna.ru; and the Russian Statistical Agency, www.gks.ru.
43. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru. Thirty-seven percent is the average. In 15 regions,
the share of transfers was between 50 and 90 percent.
44. Golovanova 2010.
45. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
46. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
47. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
48. See point 2 of Article 13 of the Law on Federal Budget in 2011–12, December
13, 2010.
49. Author’s estimations based on data on the RF Ministry of Finance website,
http://www1.minfin.ru/ru/public_debt/subdbt/.
50. Gaidar Institute for Economic Policy 2012.
51. Author’s estimations based on data on the RF Ministry of Finance website,
http://www1.minfin.ru/ru/public_debt/subdbt/, and on the RF Treasury website, http://www.roskazna.ru.
52. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
53. Author’s estimations based on regional fiscal reports on the RF Treasury website, http://www.roskazna.ru.
54. Author’s estimations based on data on the RF Ministry of Finance website,
http://www1.minfin.ru/ru/public_debt/capital_issue/state_securities/.
55. The RF Ministry of Finance website, http://www1.minfin.ru/common/img/
uploaded/library/2011/08/Dolgovaya_politika_na_sayt.pdf.
Bibliography
Alam, Asad, Stepan Titov, and John Peterson. 2004. “Russian Federation.” In Subnational Capital Market, ed. Mila Ferrie and Johan Petersen, 571–92. Washington,
DC: World Bank.
Arakelyan, A., N. Kashcheev, and A. Shchyurikov. 2001. “Regional and Municipal
Borrowing.” Rynok Tsennykh Bumag (5): 145–147. http://www.old.rcb.ru/
archive/articles.asp?id=1706.
Braga, Tania, Natalia Golovanova, Stephen Laurent, and Francisco Tortolero. 2002.
“Problems of Fiscal Equalisation in Federal Systems: Future Challenges.” In
Future Challenges for Federalism in a Changing World—Learning from Each
Other, The Youth Perspective, ed. U. Abderhalden and R. Blindenbacher,
233–63. St. Gallen: International Conference on Federalism.
Buklemishev, O. V. 1999. The Eurobond Market. Delo: Moscow.
Russian Federation: Development of Public Finances and Subnational Debt Markets
De Silva, Migara O., Galina Kurlyandskaya, Elena Andreeva, and Natalia Golovanova. 2009. Intergovernmental Reforms in the Russian Federation: One Step Forward, Two Steps Back? Washington, DC: World Bank.
Fitch Ratings. 2008a. “Capital Expenditure of Russian Regions and Outlook
for Infrastructure Development.” Special Report, October 13. http://www
.fitchratings.com.
———. 2008b. “European Local and Regional Government Outlook 2008.” Special
Report. http://www.fitchratings.com.
———. 2010. Full Rating Report of Fitch Ratings on Novosibirsk Region. April 29.
http://www.fitchratings.com.
———. 2011a. “Outlook: Fiscal and Operational Challenges Remain for European
Subnationals.” Outlook Report. http://www.fitchratings.com.
———. 2011b. “The Impact of the Economic Crisis on Russian Regions: Responses
and Planning Key to Resilience.” Special Report, September 8. http://www
.fitchratings.com.
———. 2012. “Institutional Framework for Russian Subnationals.” Special Report,
April 11. http://www.fitchratings.com.
Gaidar Institute for Economic Policy. 2012. The Russian Economy in 2011. Trends
and Prospects. April 20. http://www.finansy.ru/t/page_obz_0.html.
Golovachev, D. L. 1998. Government Debt. Theory, Russian and International Practice. Moscow: CheRo.
Golovanova, N. V., ed. 2010. Fiscal Policy of the Subjects of the Russian Federation.
Moscow: Delo.
Hofman, Bert, and Susana Cordeiro Guerra. 2005. “Fiscal Disparities in East Asia:
How Large and Do They Matter?” In East Asia Decentralizes: Making Local Government Work, 67–83. Washington, DC: World Bank.
Institute for Economies in Transition. 2003. Analysis of Budgetary Arrears in the
Russian Federation; Ways of Payment and Methods to Prevent their Generation.
Moscow. http://www.iep.ru/files/text/usaid/zadol.pdf.
Pakhomov, Serguei B. 2009. Debt Management of Russian Regions and Municipalities.
Institutional Forms, Mechanisms and Technologies. Moscow: YuNP.
Pinto, Brian, Vladimir Drebentsov, and Alexander Morozov (with the input from
Galina Kurlyandskaya). 2000. “Dismantling Russia’s Nonpayment System.”
Technical Paper 471, World Bank, Washington, DC.
Standard & Poor’s Global Credit Portal. 2004. “Russian Experience: Defaults of
Regional Governments and Restructuring of Overdue Obligations, 1990–2003.”
February 27.
Yu, A. Danilov. 2002. Government Debt Markets: World Trends and Russian Practice.
Moscow: Higher School of Economics.
493
13
South Africa: Leveraging Private
Financing for Infrastructure
Kenneth Brown, Tebogo Motsoane,
and Lili Liu
Introduction
With the end of the apartheid era in 1994, the Republic of South Africa
entered a new stage of development with far-reaching institutional
reform. After the first democratic elections in 1994, a new constitution
was adopted that fundamentally changed the way the government was
structured and operated. The 1996 South African Constitution created three independent and interrelated spheres of government at the
national, provincial, and local levels. The national government was primarily tasked with formulating policy and delivering critical national
services such as police and defense services. Provincial governments
were made responsible for the delivery of health, education, and social
services, while local government, as the sphere closest to citizens, was
mandated with the delivery of basic services and amenities. Local government was established as an autonomous sphere of government with
executive and legislative powers vested in its Municipal Council.1
In the post-apartheid era, South African municipalities faced a dual
challenge of extending the delivery of basic services to all citizens, while
simultaneously improving the quality and efficiency of existing services offered to residents. The need for infrastructure investment was
495
496
Until Debt Do Us Part
immense, driven by huge backlogs of inadequate investment during
the apartheid regime, reflected in aging electricity networks and water
and sanitation systems. Rapid urbanization and the need to accelerate economic development also required the development of new
infrastructure.
From 1994 to 2000, the municipal sector was restructured and consolidated into 283 newly formed municipalities. The amalgamation
process integrated poor and wealthy urban communities, and created
cities that brought together business hubs, wealthy suburbs, and townships under one administration.2 Since the adoption of the Constitution
in 1996, a series of important legislative and institutional reforms have
been carried out to develop a framework for strengthening local government capacity in providing critical infrastructure and services.
The government’s 1998 “White Paper on Local Government” stressed
the importance of leveraging private sector finance to meet the infrastructure requirements of municipalities over the long term.3 The White
Paper proposed a three-pronged approach to deepen municipal credit
markets. First, it proposed national legislation to better define the borrowing powers of municipalities and the rules governing interventions.
A comprehensive framework for monitoring the financial position
of municipalities was also suggested as a way of promoting financial discipline. Second, the White Paper encouraged the use of credit
enhancement measures to improve the credit quality of municipalities
and accelerate lending to local government. Third, concessional lending through state-sponsored entities was seen as a viable alternative to
market-based lending in those cases where the quality of municipal
credit prevented municipalities from accessing the market.
The 1998 White Paper was followed by extensive stakeholder consultation between 1998 and 2003, leading to the enactment of the landmark
Municipal Finance Management Act (MFMA). The act sought to “secure
sound and sustainable management of the financial affairs of municipalities and other institutions in the local sphere of government and to
establish treasury norms and standards for the local sphere of government,”4 with the aim of improving the delivery of services by municipalities. As part of the financial management, the MFMA provides the
overarching regulatory framework for borrowing by local authorities.
The act provides a comprehensive set of ex-ante rules regulating the
South Africa: Leveraging Private Financing for Infrastructure
types of borrowing and the conditions under which such borrowings
can take place. Equally important, Chapter 13 of the act stipulates a procedural approach for dealing with municipalities in financial distress.
Since 2005, activity in municipal credit markets has risen rapidly.
All metropolitan municipalities have in the last decade borrowed funds
from the banking sector, capital markets, or both, to finance infrastructure development. Long-term borrowing increased rapidly in the runup to the 2010 FIFA (Fédération Internationale de Football Association)
World Cup, changing the landscape of municipal finance from a high
level of dependency on fiscal transfers to one where borrowing plays an
increasingly important role in financing capital expenditure. However,
there are continuing challenges, including the lack of a fully developed
secondary market, and incompatibility of short-to-medium-term debt
maturities with long-term assets of infrastructure,5 and the need to
crowd-in more private financing in the market.6
The infrastructure financing needs of South African municipalities
will remain substantial over the next 10 years, estimated at approximately R 500 billion (approximately US$59.3 billion).7 According to the
national government, existing sources of capital finance, namely, municipalities’ internally generated funds and intergovernmental grants, are
insufficient to meet the estimated demand. Expanding and deepening
the subnational credit market is viewed by the government as critical
to providing a long-term financing source. In addition, the government
has broadened the financing strategy to include other sources of capital finance, such as development charges, land leases, and public private
partnerships (PPPs).8 The national government also views sound financial management practices as essential to the long-term sustainability of
municipalities.9
This chapter reviews the South African strategy of leveraging private
financing for infrastructure and the accompanying legislative and institutional reforms. The rest of the chapter is organized as follows. Section
two examines institutional reforms since the 1996 Constitution, particularly the enactment of the landmark MFMA, which defines a framework for municipal finance and access to the financial market. Section
three presents the borrowing framework for municipalities—ex-ante
rules for municipal borrowing and an ex-post system for addressing
municipal financial distress. Section four discusses the development
497
498
Until Debt Do Us Part
of the municipal credit market since the enactment of the MFMA, its
progress, and challenges. Section five presents the government’s strategy
for leveraging private finance by linking four complementary elements:
debt financing, land asset-based financing, and PPPs from the financing
side, and enhancing borrowers creditworthiness from the demand side.
Section six provides concluding remarks.
Historical Context and Institutional Reforms
The New Constitution
The 1996 Constitution of the Republic of South Africa created a broad
legislative framework for a general system of governance and provided
the core institutional framework for the legislative, executive, and judicial branches of government. The Constitution elevated provincial governments and local municipalities from being merely creatures of statute
to constitutional authorities.11 Local government was established as an
independent sphere of government with executive and legislative powers
vested in its Municipal Council.12 Moreover, the Constitution entrenched
the autonomy of local government by prohibiting any actions by national
and provincial government that might compromise or impede the ability of a municipality to discharge its constitutional obligations.
Section 139 of the Constitution opted for an administrative solution
to dealing with municipalities in financial distress by allowing provincial
government to intervene in the affairs of local government when a local
government fails to fulfill its obligations. How these provincial interventions would be carried out, and their implications for the rights and obligations of borrowers toward their creditors, were clarified in subsequent
national legislation.13 In addition, the Constitution limited the power of
the national government to guarantee subnational debt by requiring that
any such guarantees be done in accordance with national legislation.
Such legislation could only be enacted after consideration of the recommendations of the Financial and Fiscal Commission, a body established
to safeguard the probity of public finance policies and legislation.14
The government’s 1998 White Paper on Local Government concluded
that there were too many municipalities in South Africa, and that many
were not financially viable. The 1998 Municipal Structures Act provided
a legislative framework for the consolidation and rationalization of
South Africa: Leveraging Private Financing for Infrastructure
municipalities in accordance with the new constitution. The act established three types of municipalities and the criteria for each type.15
Category A municipalities comprise the six largest municipalities
with exclusive municipal executive and legislative authority in its areas.
Category B municipalities comprise 231 local municipalities that share
municipal executive and legislative authority in its area with a category
C municipality within whose area it falls. Category C municipalities
comprise local municipalities that fell under a district municipality.16
The number of municipalities was reduced from 843 to 284. During this
process, a number of urban municipalities were transformed into metropolitan municipalities, and their fiscal accounts were consolidated,
enabling cross subsidization between richer and poorer areas. Following the 2011 local government elections, the number of municipalities
was further reduced to 278, comprising 8 metropolitan municipalities,
44 districts, and 226 local municipalities.17
The financial crisis of the then Greater Johannesburg metropolitan
municipality in 1997 became the first to test the provisions of Section
139 in the Constitution (box 13.1). Lessons from the crisis subsequently
Box 13.1 Section 139 Intervention in the Greater Johannesburg
Metropolitan Municipality
The Greater Johannesburg Metropolitan Municipality was created in 1995 with four
independent local councils under the overarching Greater Johannesburg Metropolitan
Council (GJMC). Each local council could approve its own budget, and the balanced
budget applied only to the aggregate budget of all councils.
Councils rolled out ambitious spending plans without adequate finance, assuming
that shortfalls would be offset by surpluses of other councils. The crisis hit the GJMC
in July 1997 with unpaid bills of R 300 million to Eskom, the national electricity supplier.
All local councils faced severe cash flow pressure due to low revenue collection and
overambitious capital budgets, and the GJMC itself had underfunded reserves of R 1.8
billion.
The Minister for Development Planning and Local Government made a legislative
intervention in late 1997 (the first time a provincial government used Section 139 of the
Constitution), supported by the National Treasury. An emergency loan was arranged
with the Development Bank of South Africa, and a Committee of experts was instrumental in bringing expenditure in line with revenues. The crisis led to broader reform of
the municipal governance structure in the country.
Sources: City of Johannesburg 2002; The Water Dialogue South Africa 2009; World Bank 2003.
499
500
Until Debt Do Us Part
influenced the drafting of the Municipal Financial Management Act and
its emphasis on ensuring that the deleterious effects of municipal financial crises on service delivery are contained.
White Paper on Local Government
The ending of national government guarantees on municipal borrowing placed the obligation for debt service with the subnational
governments themselves. The capital market would then need clarity on a framework for borrowing rules, including remedies in the
event of municipal financial distress and emergency. Since such a
framework was yet to be developed, municipal credit markets (and,
in particular, the bond market) started to collapse after 1996. No new
bonds were issued by any municipality until 2004 (after the enactment of the MFMA in 2003).18 Naturally, this limited the ability of
municipalities to finance infrastructure development through debt
financing.
The national government’s 1998 White Paper on Local Government
aimed to address these concerns. The White Paper and the 2000 Policy
Framework for Municipal Borrowing and Financial Emergencies make
it clear that government policy regarding municipal borrowing must be
based on a market system and on lenders pricing credit to reflect the
risks they perceive.19
The government’s 1998 White Paper on Local Government stressed
the importance of using capital markets to leverage private investment.
It notes that “Ultimately, a vibrant and innovative primary and secondary market for short and long term municipal debt should emerge. To
achieve this, national government must clearly define the basic ‘rules of
the game.’ Local government will need to establish its creditworthiness
through proper budgeting and sound financial management, including
establishing firm credit control measures and affordable infrastructure
investment programmes. Finally, a growth in the quantum, scope and
activities of underwriters and market facilitators (such as credit-rating
agencies and bond insurers) will be required. … The rules governing
intervention in the event that municipalities experience financial difficulties need to be clearly defined and transparently and consistently
applied. It is critical that municipalities, investors, as well as national and
provincial government, have a clear understanding of the character of
South Africa: Leveraging Private Financing for Infrastructure
their respective risks. Risks should not be unduly transferred to national
or provincial government.”
As reviewed by the South African National Treasury (2001), the
White Paper stresses the importance of both private sector investors and
capital markets. Private sector lenders and investors are important not
only because they bring additional funding to the national table but also
because they tend to have better expertise for evaluating projects and
credit risk and for managing outstanding loans than do public sector
lenders. Active capital markets, with a variety of buyers and sellers and a
variety of financial products, can offer more efficiency than direct lending for two reasons: (a) competition for municipal debt instruments
tends to keep borrowing costs down and creates structural options for
every need; and (b) an active market implies liquidity for an investor
who may wish to sell, and liquidity reduces risk, increases the pool of
potential investors, and, thus, improves efficiency.20
The White Paper provided the basic foundation for the formulation of more detailed policies and laws governing local government.
It included proposals on how local government would relate to the
national fiscus21 and general guidelines on financial structures for local
government. More important, the White Paper acknowledged the need
to leverage private sector finance to meet the infrastructure requirements of municipalities.22
The White Paper proposed a three-pronged approach to deepen
municipal credit markets. First, it proposed national legislation to better
define the borrowing powers of municipalities and the rules governing
interventions. A comprehensive framework for monitoring the financial position of municipalities was also suggested as a way of promoting
financial discipline. Second, the White Paper encouraged the use of credit
enhancement measures that could be used to improve the credit quality
of municipalities and accelerate lending to local government. Third, concessional lending through state-sponsored entities was seen as a viable
alternative to market-based lending in those cases where the quality of
municipal credit prevented municipalities from accessing the market.
Municipal Finance Management Act
From 1998, when the White Paper was issued, to 2003, a series of legislative reforms was carried out to pave the way for the development of
501
502
Until Debt Do Us Part
a unifying framework for the management of municipal finance. This
included introduction of the Public Finance Management Act of 1999
to regulate financial management within the public sector, in order
to ensure that the revenue, expenditure, and assets and liabilities of
national and provincial government would be managed effectively. The
act made the newly established National Treasury responsible for the
establishment of uniform treasury norms and standards, and required
that every government department or constitutional institution should
have an accounting officer. The accounting officer would be the chief
executive, and this individual would ultimately be responsible for the
institution’s finances. The act thus introduced greater accountability for
public finances.
The enactment of the MFMA 2003 marked the culmination of an
extensive consultation process among stakeholders. It necessitated two
constitutional amendments23 before the bill could be enacted. Since its
first tabling in Parliament in 2000, 41 committee hearings were held to
discuss and deliberate on the bill, which reflected the challenges associated with safeguarding the independence of local government while
allowing national and provincial governments to fulfill their policy
making and oversight functions.24 Three consecutive versions of the
Municipal Finance Management Bill were ultimately tabled before the
enactment of the final act in 2003.
The extensive consultation process was needed in order to synthesize the interests of the various parties—the Treasury, lenders, and local
government. A case in point is the challenge of addressing financial
distress in municipalities when the interests of borrowers and lenders
diverge, and the national government has multiple objectives: the fiscal sustainability of municipal government, delivery of essential public
services, and development of municipal capital markets. At the heart
of the procedures for dealing with municipal financial distress are debt
and fiscal adjustment.
However, under the original Section 139 of the Constitution (1995–
2002), few remedies existed to effect debt and fiscal adjustments for a
financially troubled local government. Budgets, spending, and taxes
were under the purview of the Municipal Council. Intervention into
local government affairs by provincial government was limited to cases
where an “executive obligation” was not fulfilled. The province could
South Africa: Leveraging Private Financing for Infrastructure
only issue a directive to the council or assume responsibility for the
obligation.
Various proposals were put forward to effect debt and fiscal adjustments for a financially troubled municipality. In July 2000, the Department of Finance (now the National Treasury) put forward the Policy
Framework for Municipal Borrowing and Financial Emergencies. It
clarified the powers and procedures of municipalities to raise debt. It
acknowledged that, with the ending of national government guarantees,
the system of municipal borrowing with national guarantees would need
to be replaced by local responsibility for raising market-based financing.
To assure that municipal borrowing from capital markets is effective
and efficient, the legal and regulatory environment must be clear and
predictable. Both borrowers and lenders must have good information,
and the risks from poor decisions must be appropriately assigned. It
also noted the need for a systematic approach to dealing with financial
emergencies of local government.25 It proposed the establishment of an
administrative agency overseen by the judiciary to manage the financial
recovery of local authorities.26
The first version of the Municipal Finance Management Bill was
tabled in Parliament in July 2000. This was followed by a revised bill published in August 2001 and reintroduced in Parliament in 2002. The basic
framework defining the municipal borrowing power and p
rocedures
was already articulated in the original bill. For example, Chapter 6 of the
original version regulated municipal borrowing and contained a number of important changes. The bill described the s pecific procedures for
securing short-term debt. A municipality was permitted to incur shortterm debt only if a resolution of the municipal council had approved
the debt agreement and the accounting officer has signed an agreement
that created or acknowledged the debt. Clause 45 of the bill therefore
put in place a system of checks and balances to ensure that short-term
financing is not abused by either the political or administrative arms of
the municipality.27
The debates and amendments focused on several issues, including
two main issues of particular concern to municipal borrowing.
The first issue concerns the borrowing power of municipalities. Specifically, it concerns the balance between the intervention power of other
spheres of government (national and provincial governments) and the
503
504
Until Debt Do Us Part
autonomy of local governments, as empowered by the South African
constitution, when a municipal government faces financial stress or
insolvency. The original bill envisaged the Municipal Financial Emergency Authority as an independent financial recovery service outside
the influence of the executive and legislative branches. The final version
of the bill reduced the powers of the Municipal Financial Emergency
Authority and shifted the responsibility for overseeing an intervention
to the Member of the Executive Council (MEC) responsible for local
government within the province. A national entity, in the form of the
Municipal Financial Recovery Service, would assist in implementing the
financial recovery plan, while the MEC for local government leads the
intervention. The revision tried to strike a balance between local autonomy and intervention in the South African system of decentralization.
The second issue concerns the protection of private creditors in the
event of municipal fiscal stress. Despite the need for capital markets to
finance infrastructure, long-term private lending to municipalities was
essentially flat from 1997 to 2001. Municipal debt owed to the private
sector did not change greatly during the period, generally remaining
between R 11 and R 12 billion. At the same time, debt owed to public
sector institutions, including the Development Bank of Southern Africa
(DBSA), grew significantly—from R 5.6 to R 8.1 billion. This increasing reliance on public sector debt was viewed as inconsistent with the
government’s policy goal of increasing private sector investment. While
new policies and legislation will not, by themselves, guarantee that private sector lending increases, there would be no chance of an increase
without clear policies and legislation, according to the government.28
The revised bill afforded additional protection to creditors. Credit
agreements for the refinancing of short-term debt could be upheld if
the creditor had acted in good faith when entering the agreement with
the municipality. Refinancing of long-term debt was permitted by the
bill under certain conditions (Section 3). The bill sought to promote an
open and transparent municipal credit market by providing within the
legislation assurances to lenders that they could rely on the written representation of the municipality signed by the accounting officer.29
Two constitutional amendments (South Africa Act No. 34 of 2001 and
South Africa Act No. 3 2003) paved the way for dealing with financial
distress within municipalities. The amendments make the debt issued
South Africa: Leveraging Private Financing for Infrastructure
by the current local council valid beyond the term of the council and
expand the power of other spheres of government to intervene in legislative aspects, such as the budget or the imposition of taxes. The MFMA,
enacted in 2003, contains a new framework for municipal finance and
borrowing. Chapter 13 of the act spells out detailed criteria for interventions and financial recovery plans, specifies the role of higher-level
governments and courts in the insolvency mechanism, and outlines the
fiscal and debt adjustment process. Only courts can stay debt payments
and discharge debt obligations.30
Intervention is potentially strong and can involve substantial loss of
local political autonomy. Types of interventions include the issuance of
directives, full loss of municipal autonomy in financial matters under
mandatory interventions, and dissolution of the Municipal Council in
extreme circumstances. Primary responsibility lies with the provincial
government, but the central government may intervene when the province is unable or unwilling to act.31
The South African experience demonstrates the complexity of subnational borrowing and insolvency legislation and the importance of
building political consensus among various stakeholders. Broad support may require concerted effort over a number of years. It took South
Africa two years to develop the basic policy framework (1998–2000),
another year for cabinet approval (2001), and an additional two years
of parliamentary debate on the constitutional amendments and on the
MFMA (2001–03).32
Regulatory Framework for Municipal Borrowing
The MFMA was enacted in 2003 to ensure the sound and sustainable
management of the financial affairs of local governments and their
institutions. The act is a comprehensive piece of legislation that regulates the preparation of municipal operational and capital budgets and
the management of revenue, expenditure, and debt. In addition, the act
enhances political and managerial accountability by clearly specifying
the roles and responsibilities of the mayors and accounting officers.
An essential part of the act was to provide a framework for municipal
borrowing, averting financial crises, addressing financial distress, and
ensuring the sustainable financial management of municipalities. The
505
506
Until Debt Do Us Part
act regulates municipal borrowing by providing a comprehensive set of
ex-ante rules and creating a sound framework for dealing with financial
distress.
Legal Provisions Governing Borrowing
The MFMA seeks to ensure the long-term fiscal sustainability and sound
governance of local government. Reforms around capital budgeting are
designed to bring greater certainty and transparency to municipal budgets by ensuring that the costs and benefits of a project over its lifetime
are fully disclosed. Specifically, Section 19 of the MFMA enforces prudent financial management by requiring that the total cost of the capital
project be disclosed, along with the implications of such capital expenditure on future operational costs and on municipal tariffs and taxes.
The act also places the onus on a municipality to ensure that the various possible types of funding available are considered and analyzed in
choosing the appropriate mix of financial sources.
Chapter 6 of the MFMA sets out the procedures for securing shortand long-term debt. Municipalities can incur debt, following the
approval of the municipal council and a signed debenture agreement
by the accounting officer.33 Long-term borrowing is restricted to financing capital expenditure to ensure that future generations are not held
accountable for operational expenditure incurred by the current generation. (From a public policy perspective, long-term borrowing relieves
current generations from bearing excessive costs by paying cash for
infrastructure that will serve many generations ahead.) The act adopts
a broad definition of debt, which it defines as “a monetary liability or
obligation created by a financing agreement, note, debenture, bond,
overdraft or by the issuance of municipal debt instruments; or a contingent liability such as that created by guaranteeing a monetary liability or
obligation of another.”34 By including contingent liabilities in the definition, the act promotes a comprehensive approach to managing and
monitoring both short- and long-term debt.
Refinancing of debt is strictly controlled, as follows: (a) long-term debt
is refinanced only if the existing long-term debt was lawfully incurred,
(b) refinancing does not extend the term of the debt beyond the useful
life of the assets for which the original debt was incurred, (c) the net present value of projected future payments (including principal and interest
South Africa: Leveraging Private Financing for Infrastructure
payments) after refinancing is less than the net present value of projected
future payments before refinancing, and (d) the discount rate used in projecting net present value must be in accordance with prescribed criteria.35
Chapter 6 makes allowances for the provision of security as collateral, but places strict conditions. A municipality may, by resolution of
the council, pledge security for any debt obligations of its own or its
municipal entity, but the act restricts the municipality’s ability to pledge
any infrastructure involved in delivering minimum levels of basic services. Such infrastructure can only be pledged subject to the constraint
that in the event of default, the creditor may not sell or change the asset
in any way that will affect the delivery of basic services.36
The act permits municipalities to issue guarantees, provided they
receive the approval of the National Treasury, and only if such a guarantee is backed by cash reserves for the duration of the guarantee, or
if the municipality’s exposure to risk in the event of a default by the
guaranteed entity is insured by a comprehensive policy. Checks and balances introduced include the provisions in Section 51 of the act, which
explicitly prohibit national or provincial governments from guaranteeing municipal debt, except to the extent granted by Chapter 8 of the
Public Finance Management Act of 1999.37
Legal Provisions Governing Resolution of Financial Distress
Chapter 13 of the MFMA governs the resolution of financial distress
and emergencies of municipalities. It provides a framework for debt
relief and restructuring and the types of, and criteria for, provincial
and national interventions. More important, the MFMA recognizes
the rights of municipal creditors and the role of the courts in enforcing credit agreements. Thus, the act aims to foster greater confidence
in the regulatory framework of local government, which over time will
improve the ability of local government to access capital markets or
commercial loans at lower rates.
Triggers for financial distress and emergencies. Section 135 of the
MFMA places the primary responsibility for avoiding, identifying, and
resolving financial problems in a municipality with the municipality itself. To facilitate the timely identification of any such problems,
Section 71 of the MFMA makes it mandatory for the municipality’s
507
508
Until Debt Do Us Part
accounting officer to produce monthly budget statements no later than
10 days after the end of every month, and requires the accounting officer to report to the Municipal Council on any anticipated or actual
shortfalls, overspending, and overdrafts. The act provides for a supervisory role for the National Treasury.
Although the MFMA does not provide an explicit legal definition of
financial insolvency, it does make reference to instances when it is either
mandatory or discretionary for the provincial government to intervene
in the event of a financial crisis. Intervention in the financial affairs of
a municipality becomes mandatory “as a result of a crisis in its financial affairs” or when a municipality “is in serious or persistent material
breach of its obligations to provide basic services or to meet its financial commitments, or admits that it is unable to meet its obligations or
financial commitments” (Section 139). That is, an intervention by the
provincial government must occur not only if the municipality fails to
pay its creditors, but also if it fails to supply basic services.
There are, however, other instances when a municipality must notify
the provincial government and the relevant minister. The act categorizes
such intervention as discretionary and, in such cases, it would then be
up to provincial government and officials to decide whether or not to
intervene. These instances are outlined in Sections 135, 136, 137, and
138. They require that the municipality notify the provincial government if any of the following occur38: (a) the municipality fails to make
payments when they are due, (b) the municipality defaults on its financial obligations due to financial difficulties, (c) current expenditure
exceeds current revenue for two consecutive financial years, or the deficit exceeds 5 percent in a particular year, and (d) the municipality does
not produce its financial statements on time, or its accounts are not
signed off by the Auditor General.
Early warning system. The MFMA outlines a comprehensive system of
monitoring and reporting, serving as an early warning system to identify financial problems in municipalities. Each layer of reporting allows
financial problems to be identified, analyzed, and addressed. Periodic
reporting is prescribed by Section 71, which requires the accounting officer of a municipality to report the differences between any budgeted and
actual expenditure, revenue, and borrowings. All material differences
South Africa: Leveraging Private Financing for Infrastructure
must be accompanied by an explanation, and the report must be submitted to the mayor and the relevant provincial treasury by no later than
10 days after the month ends. Provincial treasuries are required to consolidate reports and submit a statement on the state of municipalities.39
Hence, both the provincial and national treasury are able to identify current or potential financial problems and take remedial action to assist
the municipality through less intrusive means.
Notwithstanding the reporting provisions in the MFMA, Section
135(5) places the responsibility on the municipality to report serious
financial problems to the MEC for local government and finance in
the province. Similarly, should the MEC for local government become
aware of any serious financial problems, the MEC must assess the situation and determine whether an intervention in terms of Section 139 of
the Constitution is warranted.40
Fiscal adjustment. The Municipal Financial Recovery Service is a legal
mechanism created to administer the financial recovery of municipalities. Established through Section 157 of the MFMA, the Municipal
Financial Recovery Service is responsible for preparing a financial recovery plan and monitoring its implementation at the request of the MEC
of finance in the province concerned.41 The Municipal Financial Recovery Service may also assist in identifying the causes of financial problems and potential solutions. Prior to its implementation, the recovery
plan must be submitted to the municipality, MECs for local government
and finance, organized government in the provinces, organized labor,
and suppliers or creditors of the municipality. Comments received
from these stakeholders must be taken into account when finalizing the
financial recovery plan.42 Under the current legislative framework, the
Municipal Financial Recovery Service falls within the National Treasury,
and its staff are employed within the public service.43
To secure the municipality’s ability to deliver basic services and
meet its financial commitments, the financial recovery plan contains
a minimum set of activities that the municipality must perform to
restore its financial health and service delivery obligations.44 In the
case of mandatory intervention, the financial recovery plan’s interventions must set out spending limits and revenue targets, outline budget
parameters, and identify specific revenue-raising measures. The plan
509
510
Until Debt Do Us Part
creates a binding legislative and executive obligation on the municipal
council. This provision was included to counter the potential risk of a
newly elected municipal council implementing its own spending priorities and creating further financial strain on the municipality.
Debt relief and restructuring. Chapter 13 of the MFMA provides for
the resolutions of financial problems in municipalities, and Part 3, in
particular, provides for debt relief and restructuring. “If a municipality is unable to meet its financial commitments, it may apply to the
High Court for an order to stay, for a period not exceeding 90 days, all
legal proceedings, including the execution of legal process, by persons
claiming money from the municipality or a municipal entity under the
sole control of the municipality” (Section 152(1) of the MFMA). The
act provides for a voluntary form of liquidation while protecting the
municipality and preventing creditors from incurring further losses.
Similarly, under the provisions of Section 153 of the MFMA, the
Court may suspend or terminate the municipality’s financial obligations and settle claims (in accordance with Section 155), under certain
conditions, including that the provincial executive has intervened in
terms of Section 139, a financial recovery plan to restore the municipality to financial health has been approved for the municipality, and
that the financial recovery plan is likely to fail without the protection of
such an order. More important, in an attempt to protect the delivery of
basic services, the court must ensure that all assets not necessary to the
delivery of basic services have been liquidated in accordance with the
financial recovery plan.
The court must be satisfied that (a) the municipality cannot currently meet its financial obligations to creditors, and (b) all assets not
reasonably necessary to sustain effective administration or to provide
the minimum level of basic municipal services have been or are to be
liquidated in accordance with the approved financial recovery plan for
the benefit of meeting creditors’ claims (Section 154).
Section 151 of the MFMA guarantees the legal rights of a municipality’s creditors and their recourse to the courts. When the court issues an
order to settle claims against the municipality, the MEC for local government must appoint a trustee to prepare a distribution plan. Preference in a distribution plan is given to secured creditors, and, thereafter,
South Africa: Leveraging Private Financing for Infrastructure
the preferences as outlined in the Insolvency Act (1936) are applied. Any
distribution plan must be approved by the court prior to settlement.
Fiscal monitoring. The National Treasury started systematic monitor-
ing of local government fiscal positions in 2009. The 2011 report, “State
of Local Government Finances and Financial Management,” shows
improvements in local government fiscal management, as demonstrated by an increase in unqualified audit reports as a share of total
audit reports during that year. The report also evaluates seven areas of
fiscal management, from cash management to debt growth, and identifies 66 of 283 municipalities under financial stress. Not all of the stress
was related to debt problems. Some problems, as identified in the Auditor General’s reports, emanated from weak financial management, poor
governance, and low levels of capacity within municipalities. The 2011
report also noted that 19 municipalities and 3 district municipalities
(about 6 percent of the country’s population) were under constitutionally mandated Section 139 interventions. As analyzed in the next section, the MFMA has revitalized municipal credit markets. The findings
from the implementation experience of Section 139 of the MFMA will
help strengthen the regulatory framework.
Development of Municipal Credit Market
Changes in the legislative and regulatory framework will invariably
impact the working of municipal credit markets. The MFMA regulates
both short- and long-term borrowing by municipalities and determines
the permissible uses of borrowing, and places certain obligations on the
municipality in raising long-term debt. These factors have influenced
the demand side of municipal credit markets and the landscape of local
government borrowing. Regulatory reforms also improve the credibility
of financial information, giving potential lenders more accurate information on the financial position of municipalities. This allows them
to assess the credit quality of local governments and price their risks
accurately. According to lenders, the promulgation of the MFMA and
the concomitant reforms in financial management and reporting have
enhanced the credibility of information produced by municipalities,
enabling commercial lenders to profile municipal risks more accurately.
511
Until Debt Do Us Part
Having a legal framework that dealt with financial emergencies was also
an important consideration by lenders in the extension of credit toward
local government.45
Municipal Borrowing
Total municipal borrowing (total closing balances in outstanding municipal borrowings) grew from R 18.7 billion in 2005 to R 38.1 billion in
2010, representing an average annual growth of 15 percent (figure 13.1).46
Private sector lending to municipalities outpaced public sector lending
except in 2009, when the global financial crisis impacted the domestic
lending market.
Figure 13.1 Trends in the Municipal Borrowing Market, South Africa, 2005–10
30
25
R, billions
20
15
10
10
09
20
20
20
08
07
20
06
20
05
5
20
512
Year
Public sector
Private sector
Source: South African National Treasury 2011c, with data from the National Treasury local government
database.
Metropolitan Borrowing
Capital expenditure in South Africa’s six metropolitan municipalities,
which cover 35 percent of the population, tripled during 2004/05 to
2009/10.47 World-Cup-related expenditure accounted for a significant
South Africa: Leveraging Private Financing for Infrastructure
portion of this increase, particularly for the cities of Cape Town,
eThekwini, Johannesburg, and Nelson Mandela Bay.
The six original metropolitan municipalities used external borrowing to finance a large portion of this increase in capital expenditure.
External borrowing was the highest source of funding of capital expenditure from 2005/06 to 2007/08, with government transfers becoming
the most significant source starting in 2008/09 (figure 13.2).
Given the increased capital expenditure and borrowing activity, the
cumulative amount of long-term debt has increased markedly since 2005.
However, the increase relative to revenue is less dramatic ( figure 13.3);
metropolitan revenues increased strongly from 2005/06 to 2008/09, and,
thus, borrowing relative to revenue remained at around the same level
of 35 percent of total revenues. Long-term debt as a share of revenues
30
50
45
40
35
30
25
20
15
10
5
0
R, billions
25
20
15
10
5
0
20
09
/1
20
08
/0
9
20
07
/0
8
/0
7
06
20
20
05
/0
6
20
04
/0
5
0
Percent
Figure 13.2 Metropolitan Municipality Capital Expenditure, South Africa,
2004/05–2009/10
Year
Capex
7,567,704
9,188,683
11,268,969 17,018,685 25,490,729 22,721,404
External loans
2,668,137
4,014,656
4,440,292
6,273,537 8,868,887
8,961,320
Grants and
subsidies
3,150,497
3,569,452
3,737,871
6,195,073
11,647,871
8,756,004
External
Loans/Capex
35.3 %
43.7 %
39.4 %
36.9 %
34.8 %
39.4 %
Grants/Capex
41.6 %
38.8 %
33.2 %
36.4 %
45.7 %
38.5 %
Source: http://www.mfma.treasury.gov.za.
Note: The data cover six metropolitan municipalities: Cape Town, Ekurhuleni (East Rand), ethekwini (Durban),
Johannesburg, Nelson Mandela (Port Elizabeth), and Tshwane (Pretoria).
513
Until Debt Do Us Part
Figure 13.3 Metropolitan Municipality Borrowing, South Africa, 2004/05–2009/10
50
100
45
40
80
35
30
60
25
20
40
Percent
R, billions
15
10
20
5
0
0
09
/1
20
/0
9
08
20
07
/0
8
20
/0
7
06
20
05
/0
6
20
04
/0
5
0
20
514
Year
Revenue
51,925,949 54,960,591 61,077,481 72,059,720 83,355,499 82,804,490
Borrowings
15,178,497 16,703,274 17,657,582 21,639,150 27,870,734 34,120,510
Borrowi ngs /
revenue
29.2 %
30.4 %
28.9 %
30.0 %
33.4 %
41.2 %
Source: Annual financial statements of six metropolitan municipalities: Cape Town, Ekurhuleni (East Rand),
ethekwini (Durban), Johannesburg, Nelson Mandela (Port Elizabeth), and Tshwane (Pretoria).
rose above 40 percent in 2009/10, owing to weaker revenue and increased
borrowing.
The city of Johannesburg was the most active metropolitan borrower
(figure 13.4); at the end of 2009/10, its cumulative long-term debt
was R 10.6 billion, accounting for over a third of overall metropolitan
municipality outstanding borrowing of R 34.1 billion for that year.
This was followed by eThekwini and Tshwane, whose long-term debt
was R 8.7 and R 5.6 billion, respectively, at the end of 2009/10. Nelson
Mandela Bay’s share of total debt by metropolitan m
unicipality increased
from 2 percent in 2008/09 to 4 percent in 2009/10 as its borrowing
increased from R 442.4 to R 1.46 billion, largely due to the raising of
new loans for 2010 World-Cup-related infrastructure. Previous research
reveals that this dramatic increase in debt contributed partly to the city’s
subsequent financial woes.48
South Africa: Leveraging Private Financing for Infrastructure
Figure 13.4 Outstanding Debt of Metropolitan Municipalities, South Africa,
2004/05–2009/10
40
50
35
45
40
35
25
30
20
25
15
20
Percent
R, billions
30
15
10
10
0
20
09
/1
/0
9
08
498,834
442,395
1,461,015
Ekurhul eni
1,533,666
1,348,348
1,182,431
1,127,824
2,076,914
2,881,085
Ts hwa ne
2,256,577
2,733,854 3,356,646
3,401,190
4,701,642
4,927,395
Ca pe Town
2,180,030
2,164,352 2,068,949 3,290,175
4,133,283
5,566,231
eThekwi ni
3,866,672 4,284,596
4,656,173 5,412,084
6,161,492
8,674,686
20
222,597
20
342,383
20
374,518
20
20
07
/0
8
06
04
/0
7
0
05
/0
6
5
0
/0
5
5
Year
Nel s on
Ma ndel a Ba y
(Port Elizabeth)
Joha nnes burg 4,967,034
5,829,741
6,170,786 7,909,043 10,355,008 10,610,098
Source: Annual financial statements of six metropolitan municipalities: Cape Town, Ekurhuleni (East Rand),
ethekwini (Durban), Johannesburg, Nelson Mandela (Port Elizabeth), and Tshwane (Pretoria).
Municipal credit markets in South Africa remain relatively undeveloped, with a limited amount of borrowing instruments available
to municipalities through which to raise financing. Amortizing loans
from domestic commercial banks are the principal borrowing instrument used by the metropolitan municipalities. Bonds are becoming an
increasingly important source of borrowing, with bonds (amounting
to R 15.2 billion) accounting for 55.5 percent of outstanding debt in
2009/10 (figure 13.5). Johannesburg, in 2004, was the first South African
metropolitan municipality to enter the bond market, followed by Cape
Town in 2008, and, most recently, by Ekurhuleni in 2010.
Debt service costs as a share of revenue is a critical measure of the
debt sustainability of a government. Based on international experience,
515
Until Debt Do Us Part
Figure 13.5 Debt Composition of Metropolitan Municipalities, South Africa,
2004/05–2009/10
35
30
25
R, billions
20
15
10
5
0
09
/1
20
/0
9
08
20
07
/0
8
20
/0
7
06
20
05
/0
6
20
04
/0
5
0
20
516
Year
Bank loans
12,668,497 12,973,273 13,927,583 14,782,285 17,969,917 18,950,891
Bonds
2,510,000 3,730,000 3,730,000 6,856,865 9,900,817 15,169,619
Source: Annual financial statements of six metropolitan municipalities: Cape Town, Ekurhuleni (East Rand),
ethekwini (Durban), Johannesburg, Nelson Mandela (Port Elizabeth), and Tshwane (Pretoria).
rudential guidelines suggest that debt service costs are often capped at
p
no more than 15 percent of municipal total revenues.49 For the six original
metropolitan municipalities in South Africa, aggregate annual debt service costs (including interest and principal repayments) increased from
R 2.8 billion in 2004/05 to R 6.1 billion in 2009/10, and the ratio of debt
service to revenue increased from 4.5 percent in 2007/08 to 7.4 percent in
2009/10 (still well below the prudential limit of 15 percent) (figure 13.6).
Borrowing by Secondary Cities
The growth of secondary cities reflects the rapid urbanization in South
Africa. The 19 secondary cities comprise 1.8 million households and a
population of 6.25 million, or 13 percent of the country’s population.50
Many of these secondary cities are likely to become the next generation of metropolitan municipalities. Secondary cities are critical urban
nodes, and the demand for public services infrastructure within these
cities has increased significantly. While, traditionally, secondary cities
have largely relied on fiscal transfers to finance capital expenditure,
South Africa: Leveraging Private Financing for Infrastructure
0
9/
1
20
0
8/
09
20
0
20
07
/0
8
6/
07
20
0
5/
06
20
0
20
0
Percent
8
7
6
5
4
3
2
1
0
7
6
5
4
3
2
1
0
4/
05
R, billions
Figure 13.6 Debt Service Costs, South Africa, 2004/05–2009/10
755,230
1,227,975
Year
Loan principal
1,122,775
repaid
Total finance
1,943,147
costs
Debt service
5.9 %
cost/revenue
1,992,604
1,372,170
2,101,242
2,262,561 2,452,026 3,514,435 3,681,506
7.4 %
6.0 %
4.5 %
5.7 %
2,420,297
7.4 %
Source: Annual financial statements of six metropolitan municipalities: Cape Town, Ekurhuleni (East Rand),
ethekwini (Durban), Johannesburg, Nelson Mandela (Port Elizabeth), and Tshwane (Pretoria).
borrowing from municipal credit markets has become an important
source of finance to augment capital budgets.
Aggregate borrowings by the secondary cities increased over the last
five years, from R 2.4 billion in 2004/05 to R 4.2 billion in 2009/10. Most
secondary cities have been conservative borrowers relying largely on fiscal transfers. However, a small number of secondary cities, particularly
uMhlathuze, George, Rustenburg, and Msunduzi, borrowed aggressively during this time to augment capital budgets, with an increase in
borrowing from 2004/05 to 2009/10 of 1,749, 523, 395, and 70 percent,
respectively, though from a low base (table 13.1).
Two of these 19 secondary cities, Msunduzi and uMhlathuze, ran into
financial trouble as a result of this rapid increase in long-term debt and
debt service costs relative to their revenue increase. For Msunduzi, the
provincial government staged a constitutionally mandated Section 139
intervention, and uMhlathuze municipality adopted a voluntary recovery plan. Financial recovery plans were implemented in both cases.51
To summarize, the MFMA is viewed by lenders as the most critical
factor in revitalizing the municipal credit markets.52 Borrowing by metropolitan municipalities tripled between 2004/05 and 2008/09, which
suggests a willingness by market participants to lend to metropolitan
517
518
Until Debt Do Us Part
Table 13.1 Secondary City Long-Term Borrowing, South Africa, 2004/05–2009/10
R, millions
City
uMhlathuze
2004/05
2005/06
2006/07
2007/08
2008/09
2009/10
51,097
134,954
429,379
411,670
767,236
893,888
Msunduzi
356,834
336,123
315,412
421,126
463,577
607,435
Madibeng
316,610
347,094
373,393
394,221
445,137
486,051
George
61,626
141,142
227,313
310,108
403,515
384,016
Rustenburg
70,112
89,473
88,330
156,649
359,459
346,941
Emalahleni
258,895
242,690
226,485
210,280
300,339
272,243
Drakenstein
106,305
92,491
56,799
142,312
186,167
250,987
Steve Tshwete
101,930
124,809
113,443
134,424
152,393
167,503
Matlosana
190,097
180,377
170,657
160,937
151,590
141,105
Mogale City
327,035
205,125
185,800
155,299
153,134
119,931
Govan Mbeki
114,310
111,423
108,536
105,649
102,762
99,875
Emfuleni
125,167
120,811
116,455
112,099
105,254
99,492
Newcastle
12,740
33,437
66,565
78,037
78,045
84,877
Sol Plaatje
59,806
56,635
53,464
49,950
64,964
66,435
Polokwane
92,492
92,492
92,492
92,492
92,492
50,000
Mbombela
100,706
93,604
85,260
77,653
65,758
58,151
8,356
33,580
33,597
38,204
29,768
38,183
Matjhabeng
54,140
48,987
43,834
38,681
39,095
29,591
Tlokwe
32,808
25,013
17,218
32,498
22,483
22,686
2,441,895
2,511,089
2,805,261
3,073,118
3,983,998
4,220,219
Stellenbosch
Total
Source: Secondary city annual financial statements.
municipalities and secondary cities. Commercial loans have been the
mainstay of municipal lending, but bond markets are an increasing
source of funding for metropolitan municipalities, which reflects an
increasing confidence from debt capital markets in local government
and its regulatory framework. From the perspective of municipalities,
the MFMA has also brought regulatory certainty by specifying the borrowing power of municipalities and the procedural rules for incurring
debt. More important, the act regulates capital budgeting, thus ensuring that borrowed funds are used for the development of infrastructure.
The act also addresses the concern of investors by developing remedies
in the event of municipal financial distress and emergency.
Analysis of South African municipal borrowing and debt cannot be
separated from the consolidated public debt of South Africa. Debt limits
South Africa: Leveraging Private Financing for Infrastructure
for subnational governments must take into account the fiscal space
available for the total public sector, that is, national and subnational. For
any given resources available to repay the total public debt, the borrowing
space is ultimately split between national and subnational entities (Liu
and Pradelli 2012). National government debt has been managed countercyclically and is mostly denominated in domestic currency.53 Total
debt outstanding has declined from about 58 percent of gross domestic
product (GDP) in 1998/99 to about 36 percent in 2010/11, albeit with
an increase from 2008/09 to 2010/11, due to countercyclical fiscal policies. Yields on foreign currency debt are lower than those on domestic
currency debt and are mostly long term.54 The market estimates a low
default risk, which, like those of its peers, varies with global risk aversion. South African government debt has attracted nonresident interest
despite the exchange rate risk.
Broadening the Strategy for Leveraging Private Financing55
Over the next 10 years, the municipal infrastructure financing needs
of South Africa will remain substantial—an estimated R 500 billion
(US$59.3 billion) (figure 13.7), of which R 421 billion (US$49.9 billion)
is required to finance new infrastructure and rehabilitation, and R 79
billion (US$9.4 billion) is required for the eradication of backlogs.56
According to the national government, revenues to municipalities from
own revenues and national fiscal transfers are insufficient to meet the
scale of municipal infrastructure investments. Thus, the government
has laid out a strategy of leveraging private finance through multiple
sources—borrowing, development charges, land leases, and PPPs—to
mobilize additional resources to fund infrastructure investments. At the
same time, sound financial management practices are essential to the
long-term sustainability of municipalities.
While municipalities need to explore ways of leveraging primary
sources of finance to mobilize additional resources for funding infrastructure investments, the capacity of municipalities to leverage private finance differs significantly. The investment needs of the 140
municipalities that are anchored by smaller cities and large towns
(so-called B2 and B3 municipalities) amount to about R 98 billion
(US$11.6 billion).57 These municipalities often find it difficult to access
capital markets, either because the scale at which they wish to borrow
519
Until Debt Do Us Part
Figure 13.7 South African Municipal Infrastructure Investment Requirements,
2010–19
300
250
200
R, billions
150
100
50
Growth
Backlogs
M
o
un stly
ici ru
pa ra
lit l
ies
m
T
m ow
un nici ba
pa se
lit d
ies
0
se Me
co tr
nd os
ar an
yc d
iti
es
520
Rehabilitation
Source: World Bank 2009.
makes lending expensive, or because weaknesses in their financial management make them a poor credit risk for lending institutions.
The investment requirement of the 70 mostly rural municipalities (so-called B4 municipalities) is estimated to be R 131 billion
(US$15.5 billion)58 over the next 10 years; however, the borrowing
capacity of these municipalities is very limited. Since average household
incomes in these municipalities are very low, their ability to collect revenues from property rates and service charges is limited. Consequently,
these municipalities will continue to rely mainly on government transfers to fund their capital budgets. Generally, borrowing to finance their
infrastructure needs is not an option, unless provided on special terms
by development finance institutions.
Deepening Municipal Credit Markets
As noted, private sector lending to municipalities outpaced public sector
lending from 2005 to 2009. During the recession of 2009–10, total public
South Africa: Leveraging Private Financing for Infrastructure
sector lending exceeded private sector lending for the first time since 2005.
Private lenders became more risk averse, with total debt from late 2008
to the end of the third quarter of 2010 remaining flat. In addition, the
Infrastructure Finance Corporation Limited, a major lender to municipalities, withdrew from the market in 2009, citing declining margins
due to c ompetition from public sector lenders. In contrast, public sector
lending—almost entirely from the DBSA accelerated during this period,
resulting in total public sector lending exceeding private sector lending.
The municipal bond market remains small and underdeveloped,
accounting for only 2 percent of total government bonds listed on
the Johannesburg Stock Exchange. Bonds have been issued by three
metropolitan municipalities (Cape Town, Ekurhuleni, and Johannesburg). Municipal bond repayments are typically structured with a large,
lump-sum (or “bullet”) payment at the end of the repayment period.
This creates a spike in municipal debt repayment profiles that requires
careful management to minimize the risk of default. Ideally, the debt
service profiles of municipalities should be growing broadly consistent
with revenue growth. Deferring higher levels of debt service to later
years can indicate current fiscal pressure. If adequate reserves (a s inking
fund) are not set aside over the period of the bond, the municipality
could be forced to refinance the final bullet payments with additional
debt. International experience shows that the development of serial
maturities is crucial for market development and for managing refinancing risks and maturity profiles.
Although there has been a recent recovery in private lending to
municipalities, there is a concern that both the historical and current
level of private lending to municipalities is still limited, notwithstanding
the legislative and policy reforms that have been introduced to stimulate
private sector participation (see section three). Recent research indicates
that the development of the municipal credit market is being limited by
the following five factors:
• Lack of a developed secondary bond market. A secondary market would
enhance the liquidity of bond instruments because it enables municipal
bondholders to trade the instrument. The limited size of the m
unicipal
bond issuances to date is itself an obstacle to the development of a
secondary market. The South African bond market is dominated by
521
522
Until Debt Do Us Part
ension funds and insurers that invest funds with the intention of holdp
ing until maturity. The lack of a developed secondary municipal bond
market means investors with shorter time horizons are reluctant to buy
long-term instruments whose term matches the economic life of infrastructure investments.
• Short maturities on loans. The short maturities offered by banks
means that municipalities cannot obtain loan tenures that are in line
with the life span of assets. Municipalities are compelled to finance
long-life assets with medium-term funds. This means that rates and
tariffs have to be higher in the medium term, and funds have to be
used to fund higher debt service costs rather than services over the
period of the loans.
• Creditworthiness. Borrowing should be used to finance infrastructure
that will generate income for the municipality, either directly through
tariff income or indirectly through higher property rates income. Currently, many municipalities are using borrowing to fund social infrastructure, which costs money to operate but does not expand their
revenue base. This negatively impacts the creditworthiness of municipalities and, together with many municipalities’ overall poor financial
performance, has reduced their capacity to incur further debt.
• Lack of treasury management capacity. Treasury management skills
and capacity vary significantly across municipalities. Most municipalities do not have clear borrowing strategies that support their
infrastructure investment programs. Improving treasury management capacity within municipalities will help optimize their borrowing activities, including their debt profile.
• The role of the DBSA. While the increased lending by the DBSA to
municipalities is a welcome development, going forward it needs to
explore strategies for partnering with the private sector to crowd-in
lending to local government in line with its mandate. Also, the DBSA’s
loan book should reflect an appetite for risk that is somewhat different
from that of private sector institutions and more c ommensurate with
lending to municipalities at the lower end of the market.
Through the Regulatory Framework for Municipal Borrowing (1999)
and the MFMA (2003), the government has already implemented a
range of measures to facilitate municipal borrowing, as presented in
South Africa: Leveraging Private Financing for Infrastructure
sections two and three of this chapter. With the ending of the sovereign
guarantee for municipal debt, except those approved following Chapter 8
of the Public Financing Management Act (1999), the MFMA provides
legal recourse to investors through Chapter 13 of the MFMA.
Section 48 of the MFMA states that a municipality may provide any
appropriate security for its debt obligations, and presents a range of
options in this regard, including pledging specific revenue streams,
ceding rights to future revenues, and so on. These provisions are supported by a provision in the annual Division of Revenue Act that
allows municipalities to pledge future conditional grants as reflected
in the medium-term expenditure framework. It is important that
these credit enhancements are carefully designed and implemented to
reduce moral hazard, and that they do not impede the delivery of basic
services.
There is no legal provision that allows the national government or
provincial governments to lend funds directly to municipalities. The
national development finance institutions (such as the DBSA) are
responsible for lending to municipalities, in accordance with their mandates, and may provide interest rate subsidies in accordance with their
developmental role. The government is committed to facilitating the
development of secondary markets for municipal debt to enhance the
liquidity of the municipal credit market, lower the risk of lenders, and,
thus, lower the cost of borrowing for municipalities.
Facilitating Municipality Access to Private Finance
The government is also exploring ways of enabling municipalities with
no, or only limited, access to financial markets to access private finance.
Pool finance for secondary cities. The basic idea of pool finance is to
create an instrument for secondary cities with similar credit qualities
that will allow them to pool their financing needs and approach the
financial markets collectively.
Secondary cities have large funding requirements (borrowing was
R 4.1 billion [US$500 million]59 at the end of 2010), adequate own revenues, and good institutional capacity. However, they lack the finance
expertise to issue bonds independently, and the scale of their financing
needs makes it uneconomical to approach the bond market separately.
523
524
Until Debt Do Us Part
It is envisaged that this bond pooling instrument would reduce transaction costs of the underwriting process due to increased economies of
scale.
Such bond pooling would be cost-effective for secondary cities since
they would benefit from the longer maturities and lower debt costs generally associated with bonds. In addition, bond pools can be structured
to achieve higher credit ratings in the primary market, which would further reduce the cost of the debt.
DBSA fulfilling its developmental role. Development finance institu-
tions in some developing countries have been instrumental in lending to municipalities with good potential but whose balance sheets are
comparatively weak, thus developing the lower end of the capital market. The government and the DBSA have agreed that the DBSA should
increase its support for municipalities in line with its developmental
mandate. This will entail increasing lending to those municipalities
that currently do not have access to credit markets. It is also envisaged
that the DBSA will increasingly play the role of market facilitator and,
thereby, crowd-in private finance, instead of acting as a primary lender
and effectively crowding out private finance. Steps that the bank is being
encouraged to take in this regard include:
• Championing a model that involves private sector cofinancing of the
projects it invests in
• Providing technical support to municipalities to build their capacity
to participate in credit markets generally, and not simply to facilitate
the DBSA’s own lending activities
• Facilitating municipalities’ entry and participation into private capital markets by underwriting municipal borrowing or offering limited
guarantees to municipalities
• Managing the development of a bond pooling instrument for secondary cities (using the DBSA’s extensive treasury expertise)
• Encouraging the development of the secondary market in municipal
bonds by selling its current holdings of metropolitan municipality
bonds to secondary investors that are more likely to trade them.
To support these initiatives, the government has raised the DBSA’s
callable capital by R 15.2 billion to R 20 billion, thereby increasing its
South Africa: Leveraging Private Financing for Infrastructure
lending capacity to R 140 billion. The government is also exploring ways
to reduce the DBSA’s exposure when lending to municipalities that are
a credit risk.
Developing the treasury function capacity in municipalities. Generally,
the treasury function capacity of municipalities is weak, even among
some metropolitan municipalities. The result is that municipalities are
not managing their borrowing optimally. This leads to municipalities
either underutilizing their borrowing capacity or borrowing excessively
and getting into financial difficulties. It is also reflected in the unevenness of many municipalities’ debt profiles. The National Treasury will be
exploring ways to strengthen municipalities’ treasury functions, which
may include providing specific training, developing appropriate guidelines, and providing technical advice to municipalities on how to optimize their borrowing strategies.
Development Charges, Land Leasing, and PPPs
Development charges. A development charge is designed to pass on the
up-front costs to the responsible developers, who will then pass it on to
their customers. The municipal infrastructure required to support new
property developments is typically very costly. There are essentially two
approaches to financing it.
In the first approach, the municipality borrows the required funds on
the strength of its balance sheet and then repays the debt with income
derived from all ratepayers and customers of the municipality, including
those that benefit from the new development. In the second approach,
the property developer is required to pay a development charge equivalent to the up-front cost of the new municipal infrastructure (and the
cost of using the capacity of existing infrastructure) and passes these
costs on to whomever buys into the development. Essentially, the new
landowners finance the cost of the infrastructure, which may be through
commercial debt, such as home loans in the case of residential property
developments.
One instrument that brings together the debt instrument and benefits taxation is the use of tax incremental financing,60 which helps link
local governments’ own revenue with infrastructure financing. Applying the “benefit” principle of public finance means that those who
525
526
Until Debt Do Us Part
enefit more from a product or service should pay for it in proporb
tion to the value they derive from it. Tax incremental financing is used
for financing infrastructure and other community improvement projects in many countries, including the United States. Tax incremental
financing uses future gains in taxes to finance current improvements,
which are projected to create the conditions for future gains. The completion of an infrastructure project, such as power and water, often
results in an increase in the value of the surrounding real estate, which
generates additional tax revenue. Tax incremental financing dedicates
tax increments within a certain defined district to finance the debt that
is issued to pay for the project. It creates funding for public or private
projects by borrowing against the future increase in these property tax
revenues.
A development charge is designed to pass on the up-front costs of
the new municipal infrastructure associated with specific developments to the responsible developers, who, in turn, will pass it on to
their c ustomers—the users of the new infrastructure. These users derive
a direct benefit from the provision of infrastructure, since its value is
reflected in their property valuations.
Development charges are, thus, an important component of a
sustainable system of municipal infrastructure finance and, if used
judiciously, can play an important role in accelerating the overall development of municipal infrastructure. This is because, without these
charges, the infrastructure required for new developments would have
to be financed within the confines of the municipality’s capital budget.
This means that the new infrastructure would need to be prioritized
relative to other municipal projects, which may result in it being delayed
for many years, particularly where municipalities’ scope to borrow is
limited due to weak balance sheets and poor credit ratings.
When the municipality decides to invest in the new infrastructure, it
would mean delaying other capital projects. It would also mean that the
costs related to specific developments are unfairly borne by all residents
in general, since the municipality would raise the required funds from
its entire rates and tariffs base.
It is generally accepted that using development charges is economically efficient in that the user pays. Their absence creates d
istortions in
the economy, particularly through underpricing the cost of d
evelopment
South Africa: Leveraging Private Financing for Infrastructure
in some municipalities and contributing to the underprovision of
municipal infrastructure more generally. This, in turn, acts as a significant constraint to growth and job creation.
Development charges are not a general revenue source for municipalities. Rather, they are a one-off fee that must be used to cover the
cost of municipal infrastructure associated with a new development.
They do not cover the ongoing operating costs of the services that the
infrastructure is used to provide or the future cost of the rehabilitation
or replacement of the infrastructure. These costs ought to be funded
through property taxes and user fees. Development charges are also not
intended to cover the cost of infrastructure that is internal to a development, such as sewerage or water connections to private stands or infrastructure within the boundaries of a new development. These costs are
always borne fully by the landowner.
Development charges are imposed to meet the costs of bulk and connector infrastructure, such as water mains that bring services to the boundary
of the development, and infrastructure costs associated with the utilization
of existing capacity or the need to expand the capacity of water storage and
treatment facilities, substations, and sewerage treatment works.
The use of development charges has declined in recent years. Among
the metropolitan municipalities, development charges were 2 percent of
the value of buildings completed in 2004/05. This declined to 1.7 percent
in 2009/10. Implementation is also uneven across municipalities. Both
the decline and uneven implementation can be ascribed to weaknesses
in the regulatory framework that make them administratively complex.
The National Treasury has done extensive work in relation to
municipal development charges and is in the process of developing
a framework that will set norms and standards to ensure that these
charges facilitate (and do not stifle) new property developments. Certain municipalities have already begun revising their policies related to
development charges, in line with National Treasury’s research findings.
All municipalities are encouraged to do the same.
Land-based financing strategies. Land assets are an important ingredient
of subnational government finance in most developing countries. Land
frequently is the most valuable asset on the asset side of subnational
balance sheets. Direct sales of land by subnational governments are the
527
528
Until Debt Do Us Part
clearest example of “capital” land financing. In addition, there are other
instruments for converting public land rights to cash or infrastructure.
Land may be used as collateral for borrowing, a practice that has a long
history of financing urban investment. Today, land often is the most
important public contribution to PPPs that build metro (subway) lines,
airports, or other large infrastructure projects. Beyond physical land,
rights to more intensive land development—a higher Floor Space Index
or higher Floor Area Ratio—may also be sold by public development
agencies. These “excess density rights” in effect represent the publicly
controlled share of privately owned land. The development rights have
economic value that can be sold by public authorities, as has happened
in Mumbai, São Paulo, and the United States.61
Due to the recent rapid growth in land prices, municipal land
sales have become an attractive way to mobilize finance for municipal
infrastructure (and sometimes also to finance operating deficits). However, this use of municipal-owned land undermines the
long-term financial health and wealth of the municipality. Even
when a municipality invests the funds in municipal infrastructure,
it is exchanging an appreciating asset (land) for a depreciating asset
(infrastructure). As a principle of good stewardship, municipalities
should always use the proceeds of municipal land sales to purchase
other land for the municipality in order to maintain and grow the value
of the municipality’s land portfolio and to facilitate the realization of
its spatial development strategy.
Apart from selling land, there are a range of other land-based strategies to raise finance for infrastructure investments that municipalities
can explore. First, municipalities can use municipal land as security
for raising loans to fund infrastructure related to the development of
that land or other infrastructure. This is fairly common practice among
municipalities.
Second, municipalities can use leaseholds on municipal land. The
experience of other developing countries is that this strategy has the
greatest potential where there is rapid urban growth, such as in the metro
politan municipalities and cities. The municipality will sell the development rights to the municipal land to a developer subject to the proposed
development being in line with the municipality’s spatial development
framework. The parties may agree that part of the proceeds of the sale
South Africa: Leveraging Private Financing for Infrastructure
should be used to provide infrastructure to the approved development.
The developer’s rights to the property are spelled out in a leasehold
agreement. Typically, this agreement should require the lessor to pay a
rental at least commensurate with the rates that would be raised on the
developed property. The leasehold agreement will have a specific term
(20, 40, or 99 years), depending on the type of development. Usually, the
developer is allowed to sell the leasehold to a third party under certain
circumstances. Once the term expires, all rights in the property revert to
the municipality. The leasehold system enables a municipality to partner with private developers to accelerate the development of inner-city
land while retaining ownership of the land.
Third, municipalities can use land-use exchanges. The basic idea is
that certain municipal offices or functions (such as stores, workshops,
or vehicle depots) are located on land that can and should be used for
alternative, higher-value purposes. Where this is the case, the municipality should explore relocating these offices or functions to suitable
alternative locations (often on the city outskirts), and so release the
high-value land for development.
In many instances, inner-city land is owned by either other spheres of
government or state-owned enterprises. Municipalities need to engage
with these property owners to explore ways in which they, too, can facilitate development through similar land-use exchanges.
Land-use exchanges may involve land swaps, lease swaps, or simply buying land with the funds generated from either selling or leasing
the vacated land. The net result should be a more appropriate use of
land that fosters development. The best known example of this kind of
development is the Victoria and Alfred Waterfront in Cape Town, where
a harbor was turned into a shopping mall and tourist destination.
International experience shows that the fiscal risks from land-based
financing will need to be managed prudently.62 Land sales often involve
less transparency than borrowing. Many sales are conducted off-budget, which makes it easier to divert proceeds into operating budgets.
Capital revenues from sales of land assets exert a much more volatile
trend and could create an incentive to appropriate auction proceeds
to finance the operating budget, particularly in times of budget shortfalls during economic downturns. Furthermore, land collateral and
expected future land-value appreciation for bank loans can be linked
529
530
Until Debt Do Us Part
with macroeconomic risks. It is critical to develop ex-ante prudential
rules, comparable to those governing borrowing, to reduce fiscal risks
and the contingent liabilities associated with the land-based revenues
for financing infrastructure.
Public-private partnerships. PPPs are important service delivery
mechanisms that facilitate rapid infrastructure development. They
allow municipalities to take advantage of private sector expertise and
experience. There are different types of PPPs that involve models for
risk sharing between the municipality and its partners. In many cases,
the private party is in a better position to raise debt and equity to
finance the project. Municipalities can take advantage of private sector expertise and experience in the construction of the infrastructure.
Furthermore, the development of PPPs for economically justifiable
projects eases the pressure on the municipality’s budget and allows
better allocation of funds toward addressing social needs of the
community.
There are fiscal risks associated with PPPs.63 Often, subnational
governments provide explicit or implicit guarantees for market borrowings of public enterprises that form partnerships with private
investors. Challenges arise from implicit guarantees, which influence
creditors’ risk assessment. Moreover, there is a lack of standardized
accounting, recording, collecting, and disclosing of such debt incurred by
off-budget financing vehicles in many developing countries. These tasks
are challenging because of an array of complex arrangements of PPPs.
Subnational-owned enterprises may have different quasi-fiscal relations
with the budgets of their owners—subnational governments. Adding
to the complexity is the wide variety of legal contractual relationships
in PPPs. There is no standard uniformity in these contractual relationships; they vary across and within sectors.
Enhancing Creditworthiness of Municipalities64
Sound financial management practices are essential to the long-term
sustainability of municipalities. Generally, municipalities are encouraged to access private finance on the strength of their balance sheets
and their credit ratings.65 Municipal financial management involves
managing a range of interrelated components: planning and budgeting,
South Africa: Leveraging Private Financing for Infrastructure
revenue, cash and expenditure management, procurement, asset management, reporting, and oversight. Each component contributes to
ensuring that expenditure is developmental, effective, and efficient and
that municipalities can be held accountable.
The reforms introduced by the MFMA are the cornerstone of the
broader reform package for local government outlined in the 1998
White Paper on Local Government. The MFMA, together with the
Municipal Structures Act (1998), the Municipal Systems Act (2000), the
Municipal Property Rates Act (2004), and the Municipal Fiscal Powers and Functions Act (2007), sets out frameworks and key requirements for municipal operations, planning, budgeting, governance, and
accountability.
Since 2008, the National Treasury has paid attention to strengthening municipal budgeting and reporting practices. Key initiatives have
been the introduction of the Municipal Budget and Reporting Regulations in 2009, the enforcement of in-year financial reporting processes,
and firmer management of conditional grants in accordance with the
annual Division of Revenue Act. These reforms have been supported by
strengthening the National Treasury’s local government database and by
publishing an increasing range of local government financial information on the National Treasury’s website. The National Treasury is currently working on a number of reform initiatives, including a standard
chart of accounts for municipalities, strengthening revenue and cash
management policies, and finalizing the regulations for financial misconduct to facilitate the enforcement of the provisions dealing with
financial conduct in Chapter 15 of the MFMA.
Conclusions
South Africa developed a comprehensive regulatory framework for the
financial management practices of local government within the fundamental changes in the country’s political structure and municipal system. Designing the regulatory framework in a federal system, where the
subnational spheres of government are autonomous, was a consultative
process in South Africa. The legislative process coordinated institutional
and policy reforms and synthesized different interests of the national
government, provincial government, local government, and creditors.
531
532
Until Debt Do Us Part
A key challenge was to reach a proper balance between the autonomy
of local government, as granted by the 1996 Constitution, and the
national government’s obligation to ensure fiscal sustainability of local
governments.
The South African experience shows that the benefits of having a
strong regulatory framework are numerous; in particular, the regulatory
framework has provided certainty and clarity on rules and procedures,
giving confidence to the capital markets to finance much-needed infrastructure to its citizens, thereby improving the quality of their lives.
The enactment of the law has helped revitalize municipal credit markets. Borrowing by metropolitan municipalities tripled and borrowing
by secondary cities doubled between 2004/05 and 2008/09, suggesting a
willingness by market participants to lend to metropolitan municipalities. Historically, commercial loans have been the mainstay of municipal
lending, but bond markets are now an increasingly popular alternative
source of funding for metropolitan municipalities. This reflects the
increasing confidence that the capital markets have in the regulatory
framework and local government finance.
Notwithstanding the expanded activities of municipal credit markets, the markets would need continuing expansion and deepening to
support substantial infrastructure investment demands. South Africa
faces infrastructure financing requirements over the next decade, estimated at approximately R 500 billion. The demand for municipal
infrastructure is spread across all municipalities but is greatest in the
metropolitan municipalities and secondary cities. The municipal credit
markets face challenges: the secondary market for municipal securities
is almost nonexistent; there is a mismatch between the long-term asset
life of infrastructure and the relative short maturities; and the capacity of many municipalities to manage a debt portfolio and access markets is weak. The DBSA also faces the challenges of crowding-in private
creditors.
The government envisions multiple strategies for leveraging private
financing for infrastructure investments. The government is exploring ways of deepening and broadening the municipal capital markets
through developing a bond pooling instrument for secondary cities and
building municipal capacity in managing a debt portfolio and accessing markets. It is encouraging the DBSA to fulfill its developmental role
South Africa: Leveraging Private Financing for Infrastructure
and become a market facilitator and, thereby, crowd-in private finance,
instead of acting as a primary lender and effectively crowding out private finance.
Going beyond the development of competitive municipal credit
markets, the government is also exploring ways of mobilizing private
financing for infrastructure through development charges, land-based
financing, and PPPs. International experience has demonstrated the
enormous potential of these instruments in leveraging private financing, provided that the fiscal risks from land-based financing and PPPs
are prudently managed.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. Section 151 of the 1996 Constitution of the Republic of South Africa, http://www
.justice.gov.za/legislation/constitution/constitution.htm.
2. Nyalunga 2006.
3. Section 3 of the 1998 White Paper on Local Government, http://www.info.gov
.za/view/DownloadFileAction?id=108131.
4. Preamble to the MFMA, No. 56, of 2003, http://mfma.treasury.gov.za/Legisla
tion/lgmfma/Pages/default.aspx.
5. From a public policy perspective, matching the debt maturity term with the
asset life of infrastructure is consistent with intergenerational equity (Liu 2008).
6. South African National Treasury 2011c.
7. World Bank 2009. R 8.43 = US$1, September 3, 2012.
8. South African National Treasury 2011c.
9. South African National Treasury 2011b.
10. Sections two and three draw from Liu and Waibel (2008, 2009) and background
materials and work by DNA Economics, Pretoria, South Africa, commissioned
by the World Bank.
11. South African Local Government Association, http://www.salga.org.za/pages/
Municipalities/About-Municipalities.
12. Section 151 of the 1996 Constitution of the Republic of South Africa.
13. Municipal Finance Management Act, 2003.
14. Section 218 of the 1996 Constitution of the Republic of South Africa.
15. As of December 31, 2000.
16. Before the 2011 local government election, there were originally six Category
A metropolitan municipalities based in the six largest cities in South Africa,
533
534
Until Debt Do Us Part
namely, Cape Town, Ekurhuleni (the East Rand), ethekwini (Durban), Johannesburg, Nelson Mandela Bay (Port Elizabeth), and Tshwane (Pretoria). In
2011, Mangaung (Bloemfontein) and Buffalo City (East London) were also
declared as Category A municipalities. Category B, or local municipalities, cover
the areas that fall outside the Category A municipalities. In 2010, there were 231
local municipalities and 44 district municipalities.
17. South African National Treasury 2011a.
18. In 2004, the City of Johannesburg went to market following its recovery from a
financial crisis.
19. South African National Treasury 2011c.
20. South African National Treasury 2001c, 192–93.
21. The national fiscus refers to the government’s fiscal activity and includes revenues, expenditures, and debts.
22. Section three of the 1998 White Paper on Local Government.
23. Second Amendment Act 2001, which allowed municipalities to borrow; Second Amendment Act of 2003, which legalized provincial intervention in local
government.
24. Wandrag 2009.
25. Department of Finance, South Africa 2000, 2.
26. This model was informed by insolvency practices in the private sector, where
the Master of the High Court appoints an insolvency practitioner to sequester
an estate.
27. Republic of South Africa 2002.
28. South African National Treasury 2001, 189–90.
29. Clause 49(2) of the Municipal Finance Management Bill (B1D-2002).
30. Liu and Waibel 2009.
31. Liu and Waibel 2009.
32. Liu and Waibel 2009.
33. Section 45 of the MFMA (2003).
34. Definitions in Chapter 1 of the MFMA (2003).
35. Section 46(5) of the MFMA (2003).
36. Section 48 of the MFMA (2003).
37. Sections 70(b) and 66(3)(c) of the Public Finance Management Act allow for
guarantees to be granted in special cases by the national Finance Minister in
consultation with the national minister responsible for a specific portfolio,
that is, the Minister of Cooperative Governance and Traditional Affairs, who
is responsible for local government. Any guarantee issued by the Minister of
Finance binds the effectively national revenue fund.
38. A discretionary intervention may be initiated if any of the above-mentioned
conditions are met in a municipally owned entity.
39. In the case of certain larger metropolitan municipalities and secondary cities,
such reports must be submitted to the National Treasury.
40. Section 136(1) of the MFMA.
South Africa: Leveraging Private Financing for Infrastructure
41. In case of discretionary intervention, the recovery plan may be prepared by
the Municipal Financial Recovery Service or by a suitably qualified person
appointed by the provincial executive.
42. Section 141(3)(c) of the MFMA (2003).
43. As noted, the Municipal Financial Recovery Service as established in the act can
be thought of as an administrative support structure, in contrast to the quasijudicial structure proposed in the previous versions of the Municipal Finance
Management Act.
44. Section 142 of the MFMA.
45. Jitsing, Chisadza, and Condon 2012.
46.
Data from the 2004/05 to 2009/10 audited annual financial statements
were been collected and analyzed and cover the original six metropolitan
municipalities.
47. Data on metropolitan municipalities up to 2010 are for six metropolitan c ities:
Cape Town, Ekurhuleni (East Rand), ethekwini (Durban), Johannesburg, Nelson
Mandela (Port Elizabeth), and Tshwane (Pretoria). In May 2011, the number of
metropolitan municipalities was increased from six to eight (see note 16).
48. Jitsing, Chisadza, and Condon 2012.
49. See Liu and Waibel 2008; Liu and Webb 2010.
50. Source: STATSSA Community Survey 2007.
51. As mentioned, 22 municipalities (6 percent of the country’s population) are
under section 139 intervention. The lessons to be learned from these cases, and
from the Msunduzi and uMhlathuze financial recoveries, will help strengthen
implementation of the MFMA.
52. Interviews by DNA Economics of South Africa during 2010–11 with commercial banks for the National Treasury demonstrated that lenders view the MFMA
as the most important factor in revitalizing the municipal credit markets.
53. Source on the assessment of national government debt: International Monetary
Fund (2011).
54. Inflation-linked long-term debt and fixed-income long-term debt accounted
for 81 percent of national government debt in 2010 (International Monetary
Fund 2011).
55. Unless otherwise indicated, this section draws mainly from reports by the South
African National Treasury (2011b, 2011c).
56. World Bank 2009. R 8.43 = US$1, September 3, 2012.
57. R 8.43 = US$1, September 3, 2012.
58. R 8.43 = US$1, September 3, 2012.
59. R 8.43 = US$1, September 3, 2012.
60. Tax incremental financing is based on the authors’ own research.
61. Peterson and Kaganova 2010.
62. This draws from Peterson and Koganova (2010).
63. The discussion of fiscal risks from PPPs draws from Canuto and Liu (2010) and
Irwin (2007).
535
536
Until Debt Do Us Part
64. This section draws from South African National Treasury 2011b.
65. South African National Treasury 2011c.
Bibliography
Canuto, Otaviano, and Lili Liu. 2010. “Subnational Debt Finance: Make It Sustainable.” In The Day After Tomorrow: A Handbook on the Future of Economic Policy
in the Developing World, 219–39, ed. Otaviano Canuto and Marcelo Giugale.
Washington, DC: World Bank.
City of Johannesburg. 2002. Addendum 2 of the Annual Report. http://www.joburgarchive.co.za/city_vision/AnnualReport02Ch8.pdf.
Department of Finance, South Africa. 2000. Policy Framework for Municipal
Borrowing and Financial Emergencies. http://www.info.gov.za/view/Download
FileAction?id=70340.
International Monetary Fund. 2011. South Africa 2011 Article Consultation. Country
Report 11/258, International Monetary Fund, Washington, DC.
Irwin, Timothy. 2007. Government Guarantees. Washington, DC: World Bank.
Jitsing, A., S. Chisadza, and N. Condon. 2012. Municipal Credit Markets in South
Africa. Report prepared for the South African National Treasury by DNA Economics, Pretoria.
Liu, Lili. 2008. “Creating a Regulatory Framework for Managing Subnational Borrowing.” In Public Finance in China: Reform and Growth for a Harmonious Society, ed. Jiwei Lou and Shuilin Wang, 171–90. Washington, DC: World Bank.
Liu, Lili, and Juan Pradelli. 2012. “Financing Infrastructure and Monitoring Fiscal Risks at the Subnational Level.” Policy Research Paper 6069, World Bank,
Washington, DC, May.
Liu, Lili, and Michael Waibel. 2008. “Subnational Borrowing, Insolvency and Regulations.” In Macro Federalism and Local Finance, ed. A. Shah, 215–41. Washington, DC: World Bank.
———. 2009. “Subnational Insolvency and Governance: Cross-Country Experiences and Lessons.” In Does Decentralization Enhance Service Delivery and
Poverty Reduction?, ed. Ehtisham Ahmad and Giorgio Brosio, 333–75. Cheltenham, U.K.: Edward Elgar.
Liu, Lili, and Steven Webb. 2010. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper 5409,
World Bank, Washington, DC.
Nyalunga, Dumisani. 2006. “The Revitalization of Local Government in South
Africa.” International NGO Journal 1 (2): 15–20.
Peterson, George E., and Olga Kaganova. 2010. “Integrating Land Financing into
Subnational Fiscal Management.” Policy Research Working Paper 5409, World
Bank, Washington, DC.
Republic of South Africa. 2002. The Municipal Finance Management Bill. B1D-2002.
http://www.info.gov.za/view/DownloadFileAction?id=66873.
South Africa: Leveraging Private Financing for Infrastructure
———. 2009. “Companies Act No. 71 of 2008.” Government Gazette No. 32121.
http://www.info.gov.za/view/DownloadFileAction?id=98894.
South African National Treasury. 2001. Intergovernmental Fiscal Review. Pretoria.
———. 2011a. Annexure to the 2011 Budget Review. Explanatory Memorandum to
the Division of Revenue. Pretoria: South African National Treasury.
———. 2011b. “Financial Management and MFMA Implementation.” In Local
Government Budget and Expenditure Review. Pretoria: South African National
Treasury.
———. 2011c. “Leveraging Private Financing.” In Local Government Budget and
Expenditure Review. Pretoria: South African National Treasury.
South African Local Government Association. About Local Government.” http://
www.salga.org.za/pages/Municipalities/About-Municipalities.
The Water Dialogue South Africa. 2009. “Johannesburg Case Study.” In The Water
Dialogues. http://www.waterdialogues.org/south-africa/documents/Johannesburg
CaseStudy-FullReport.pdf.
Wandrag, R. 2009. “The Quest for Financial Discipline at Local Government Level:
The Regulation of Municipal Borrowing and Financial Emergencies.” Law,
Democracy & Development 7 (2): 243–68.
World Bank. 2003. WDR 2004: Making Services Work for Poor People. Washington,
DC: World Bank.
———. 2009. “Municipal Infrastructure Finance Synthesis Report.” World Bank,
Washington, DC.
537
14
Caveat Creditor: State Systems
of Local Government Borrowing
in the United States
Lili Liu, Xiaowei Tian, and
John Joseph Wallis
Introduction
Economists, political scientists, and development specialists have long
been interested in what happens when sovereign governments fail to
repay loans, since the only laws the creditors can access to force repayment are the laws the government itself promulgates and enforces. The
problem is no less challenging when nonsovereign governments fail
to repay their debts. Presumably, higher-level governments can create
and enforce rules for debt issue, debt repayment, and debt adjustment
for their political subdivisions. However, the common pool problems
and associated moral hazards pose a special challenge to fiscal adjustment and debt restructuring.1 The ongoing fiscal challenges in numerous developed countries, exacerbated by the global financial crisis
of 2008–09, have brought problems of insolvency and sovereign and
nonsovereign debt to the forefront of policy debates. Many developing
countries face similar challenges with their subnational governments.
Our focus is the historical development and current structure of
insolvency rules for United States local governments. In many countries, all subnational governments are nonsovereign governments. But
in the United States, state governments also possess sovereignty.2 State
539
540
Until Debt Do Us Part
sovereignty with respect to state debts is explicitly recognized in the
Eleventh Amendment to the national constitution.3 All of the governments below the state level, what Americans call “local government,” are
not sovereign, but rather are created by and subject to the laws of each
respective state. In 2007, there were 89,476 local governments comprising
3,033 counties, 19,492 municipalities (cities), 16,519 towns and townships, 14,561 school districts, and 37,381 special purpose districts.4 While
these governments are widely divergent in structure and purpose, for this
chapter, they are all included under the category of “local government.”
The United States has by far the largest subnational government capital market in the world. In 2007, local governments issued
US$225 billion in bonds, and total local government debt outstanding was US$1.5 trillion, while state governments issued US$161 billion
in bonds and had US$936 billion in bonds outstanding.5 In c ontrast,
in 2007, subnational bonds issued by all countries outside the United
States totaled roughly US$130 billion. The amount of subnational
bond issuance outside of the United States expanded rapidly in the
late 2000s, particularly in Canada, China, the Federal Republic of
Germany, and Japan, but the subnational bond market in the United
States remains larger than the rest of the world combined.6
If the policy goal of developing countries is to promote credible and
responsible subnational government borrowing to finance infrastructure, then the experience of the United States offers instructive lessons.
Rather than one unitary government, the United States encompasses
50 different regimes of local governance structure. Not only are state
governments sovereign, they each design and constitute a fiscal system
for their own local governments. A few states regulate local governments closely and have well-established institutions for monitoring and
regulating local government fiscal performance. Other states do relatively little in the way of active monitoring. States also vary in the ways
that they allow local governments to make decisions about borrowing,
taxing, and spending. Many states have ex-ante limits on the amount of
local government borrowing and the level of taxation and expenditures.
States also place limits on the kind of functions that local governments
can perform and the purposes for which bonds may be issued. Understanding what happens when local governments become insolvent, or
Caveat Creditor: State Systems of Local Government Borrowing in the United States
face the possibility of a fiscal crisis, is impossible without reference to
the larger set of state-level fiscal and constitutional institutions.
The intent of this chapter is to explain how the systems evolved
and how they work. These issues matter enormously for developing
countries. Two aspects are particularly important. The first is that the
sequence of historical development matters for understanding how
the system works. Local insolvency problems reached a crisis point in
the late 19th century when a significant number of local governments
defaulted on bonded debts. It was not clear how the liabilities of insolvent, democratically elected governments could be enforced, since
enforcement almost inevitably involved imposing burdens on taxpayers
and voters that they themselves might not consent to. Rather than unilaterally forcing local governments to repay debts, a set of institutions
developed that clearly outlined the powers and responsibilities of local
governments with regard to issuing debt, and then left it up to private
capital markets to assess the risk of lending to local governments. This
led to the second key aspect of the American systems: active monitoring and disciplining of most local borrowing is accomplished through
private markets. The markets do not, however, operate in isolation.
A series of institutions—some public, some private, and others mixed—
have evolved to make local borrowing sustainable and credible. The first
half of the chapter traces the historical development of the American
systems, and the second half more closely analyzes the systems that are
currently in place in the American states. Three dimensions of insolvency systems shape our approach. First is the distinction between exante and ex-post policies. Ex-ante elements of an insolvency system
come into play even before a local government decides to borrow, such
as limits on the amount a local government can borrow and procedural
rules on how they borrow. E
x-post elements come into play after a fiscal
crisis begins and/or a default on debt has occurred.
Second is the distinction between passive and active policies. More
than half of the American states have insolvency systems that rely
largely on ex-ante constraints on decisions that local governments
make about borrowing. These systems are passive, in the sense that the
state does not take direct action or intervene in the operation of local
governments in normal circumstances. Although there are some active
541
542
Until Debt Do Us Part
monitoring systems in place, only a few states actually have systems that
actively interfere with local government fiscal decisions.
Third, the systems work because the passive constraints provide
guidelines, which private citizens, bondholders, bond underwriters,
capital markets, and the courts can use to discipline and shape the interests and behavior of local governments. The chapter will describe systems in which institutions leverage up the ability of public governments
and private markets to clearly identify the risks of borrowing and lending, and the revenue sources available to repay debts.7
More than half of the American states have completely passive exante insolvency systems. Those systems are the norm. Slightly less than
half of American states have insolvency systems that involve any ex-post
elements that engage after a fiscal crisis is identified or begins. Part of the
ex-post systems have access to Chapter 9 proceedings under the federal
bankruptcy code for municipal governments that was created in 1937.8
Twenty-three states do not allow local governments to avail themselves
of federal bankruptcy procedure and nine states only do so under limited conditions.9 Slightly less than a third of the states have more active
systems for monitoring, regulating, and, ultimately, intervening in local
government insolvency crises.
Despite its visibility, the Chapter 9 procedures are rarely used. From
1980 through 2009, an average of fewer than 8 cases were filed annually nationwide. Given that there are approximately 89,000 local governments, this is a take-up rate of less than 1 in 10,000 per year. From 1937
to 2011, there were roughly 600 Chapter 9 cases.10 This chapter considers why it is that states, rather than the national government, regulate
local governments. This is even true in areas where the national government has explicit constitutional permission to regulate bankruptcy.
Many local governments in the United States have been undergoing fiscal strain or crises during the recession that began in 2008. Two
local governments have defaulted on their bonds (at the end of 2011).
What changes the current crisis will bring to local insolvency systems is
hard to predict. All the institutions that govern local governments are
endogenous at the state level. Even when rules regarding local government independence are enshrined in a state constitution, that constitution is subject to change. As shown throughout the chapter, insolvency
systems in the United States continuously adapt. Constitutions and laws
Caveat Creditor: State Systems of Local Government Borrowing in the United States
do not change every year, but they do change over time. Both state and
local governments regularly reconsider how they should interact, and it
is expected that some changes will occur in the coming years.
Section two of the chapter presents several conceptual issues with
regard to sovereignty and self-government in the American political system. To appreciate the dynamic nature of insolvency systems, section
three begins with the early history of state constitutional provisions
regarding local government borrowing, largely the passive ex-ante elements implemented during the 19th century, and then follows with the
history of changes in the 20th century that leveraged the ability of private markets to coordinate with public borrowers. Section four examines state insolvency systems currently in force, including passive and
active systems. Section five looks closely at three states, North Carolina,
Pennsylvania, and Ohio, which have active intervention systems. S ection
six provides empirical results on the difference in revenues, expenditures, and debt associated with different types of insolvency systems.
Section seven concludes with a review of lessons learned about the role
of market discipline in the American states.
Conceptual and Constitutional Issues
The constitutional structure of American government is complicated. States are governed by the national constitution but are explicitly sovereign governments that enjoy all powers not explicitly granted
to the national government in the national constitution (the Tenth
Amendment). Particularly important for government debt, the Eleventh
Amendment explicitly makes states immune from legal cases brought
by citizens of other states or nations that they do not consent to. The
relationship between states and local governments is even more complicated in ways that bear directly on local government borrowing and
insolvency.
The national constitution does not affect the relationship between
state and local governments.11 In legal and constitutional terms, states
play the dominant role in structuring local governments and managing municipal insolvency. But the source and scope of local government
powers have long been subject to controversy and change.12 Actual practices fall between two conceptual extremes.13 At one extreme is Dillon’s
543
544
Until Debt Do Us Part
Rule, formulated by John Dillon of the Iowa Supreme Court,14 which
holds that municipalities are simply the administrative instrumentalities created by the states to implement state policies that only enjoy
powers the state has granted to them expressly and incidental thereto.
States enjoy broad powers to create, alter, or abolish their local governments, change their boundaries, and modify or eliminate their powers.15
The other extreme is the Cooley Doctrine, formulated by Thomas
Cooley of the Michigan Supreme Court, which holds that municipalities are the creations of their constituents. Consequently, municipalities enjoy some degrees of sovereignty. Municipalities have a right of
self-government, and local constituents fundamentally determine the
local government activities. In an important conceptual sense, all states
are Dillon’s Rule states. States always retain the possibility of exerting complete power over local governments within the constitutional
framework of the nation. State constitutions can always be amended,
so D
illon’s Rule is always a possibility.16 On the other hand, almost all
states choose to allow local governments some freedom along the lines
of the Cooley Doctrine, although the extent of local autonomy and
choice varies widely.
The autonomy of the Cooley Doctrine is also related to concepts of
sovereignty—in this case, the sovereignty of voters. To appreciate the
implications of voter sovereignty for government borrowing, we need
to understand that Americans draw a distinction between their governments and their citizens. Legitimate actions taken by governments are
obviously binding on citizens, but actions taken by governments that
go beyond the authority granted to governments by their citizens and
embodied in constitutions are problematic. A government that takes an
action beyond its allotted powers cannot legally bind its citizens to support the action. If the action is borrowing money that needs to be repaid
from tax revenues in the future, then voter-citizen-taxpayers may have a
claim that the government was acting “beyond its powers.”
Exactly what powers local governments possess is complicated by
Dillon’s Rule, since the source of legitimate local government power is
the state government and the sovereign power of the citizens. The continuing evolution of the state’s relationships with local governments is
far too complicated to go into in detail here, but three aspects need our
attention.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
The first is how state governments deal with local governments. Initially, all relationships between state and local governments were, in the
legal terminology, “special,” in the sense that states could deal with individual counties, municipalities, school districts, or other local governments on a case-by-case basis. Because of the political problems that
special treatment involved, many states began prohibiting the state legislature from dealing with local governments on an individual basis. In
many states, state constitutions began requiring that all local governments be subject to the same “general” laws that affect all municipalities in the same way, or what are called “general incorporation laws” for
local governments.17 These constitutional provisions and laws began
to appear in the 1850s. States retain sovereignty over the structure and
actions of local governments but can only exercise that sovereignty in
a way that applies equally to all local governments. The relationship
between state and local governments varies widely across states.
This can have important implications in a fiscal crisis. For example,
the Ohio Constitution of 1851 prohibits special legislation for all corporate bodies.18 As a result, when Cleveland’s fiscal crisis reached a peak
in 1979, the state was required to respond with legislation that governs
a wide range of municipalities: cities, villages, counties, and school districts. The Ohio legislature could not pass a law that applied only to
Cleveland. In contrast, New York State is not constrained by a special
legislation ban. New York was able to adopt special legislation that only
applied to New York City to remedy the city’s fiscal crisis in 1975.
The second aspect of sovereignty is how much latitude states allow
their local governments in structuring their internal governance and
deciding which functions to perform. The first general incorporation
laws in the 1850s tended to be quite narrow, specifying exactly (or within
a narrow range) what local governments could do. This one-size-fits-all
rule had obvious costs, and beginning in the 1880s, many states allowed
local governments some measure of “home rule.” Home rule grants local
governments limited rights to self-government and may limit how states
can intervene in local affairs. Home rule laws and provisions may place
certain aspects of local government structure and behavior beyond the
reach of state governments. Although home rule is sometimes equated
with the Cooley Doctrine, the actual structure of home rule in a state
is usually somewhere between the Dillon and Cooley extremes. States
545
546
Until Debt Do Us Part
regulate some aspects of local governments directly, while allowing local
governments sovereignty in other dimensions.
A third aspect of sovereignty is that some states have constitutional
prohibitions on special state commissions that can take over municipal
functions.19 The special commission bans were intended to protect local
autonomy by curbing the ability of the states to take over important
municipal functions, which are vested in democratically elected local
officials.
Table 14.1 lists the dates when states adopted mandatory general incorporation acts for local governments and the dates when states adopted
some form of home rule for local governments. Because of this evolution,
Table 14.1 Year of First General Law for Municipalities and First Home Rule Law,
United States
State
Statehood
General Law
Home Rule
Alabama
1819
—
—
Alaska
1959
—
1959
Arizona
1912
1912
1912
Arkansas
1836
1868
—
California
1850
1879
1879
Colorado
1876
1876
1912
Connecticut
1788
1965
—
Delaware
1787
—
—
Florida
1845
1861
—
Georgia
1788
—
—
Hawaii
1959
1959
1959
Idaho
1890
1889
—
Illinois
1818
—
1970
Indiana
1816
—
—
Iowa
1846
—
1968
Kansas
1861
1859
1960
Kentucky
1792
1891
—
Louisiana
1812
1974
1974
Maryland
1788
1864
1954
Massachusetts
1788
—
—
(continued next page)
Caveat Creditor: State Systems of Local Government Borrowing in the United States
Table 14.1 (continued)
State
Statehood
General Law
Home Rule
Michigan
1837
1909
1909
Minnesota
1858
1896
1896
Mississippi
1817
1890
—
Missouri
1821
—
1875
Montana
1889
1922
1973
Nebraska
1867
1866
1912
Nevada
1864
1864
1924
New Hampshire
1788
1966
—
New Jersey
1787
—
—
New Mexico
1912
—
1970
New York
1788
1894
—
North Carolina
1789
1916
—
North Dakota
1889
1889
1966
Ohio
1803
1851
1912
Oklahoma
1907
1907
1907
Oregon
1859
—
1906
Pennsylvania
1787
1874
1922
Rhode Island
1790
1951
1951
South Carolina
1788
1896
1973
South Dakota
1889
1889
1963
Tennessee
1796
—
1953
Texas
1845
1876
1912
Utah
1896
1896
—
Washington
1889
1889
1889
West Virginia
1863
1936
1936
Wisconsin
1848
1848
1924
Wyoming
1890
1889
—
Source: Hennessey 2009.
Note: — = No law passed as of 2009.
which is described in more detail in the next section, by the late 19th century, almost every state had in place constitutional provisions that governed local government borrowing. Those powers varied from state to
state and, in some states, from local government to local government.
A critical implication of these structures for local government borrowing
547
548
Until Debt Do Us Part
and insolvency was that voters and taxpayers could be held liable only
for commitments that their local government made that fell within the
local government’s lawful authority and functions. In some cases, specific
voter approval of a bond issue and related project is required.
The legal concept that governs is quo warranto (by what authority).
Local governments could borrow only for purposes for and by methods which they were authorized to borrow; otherwise, the taxpayers
were not under an obligation to repay the money. This would have
enormous implications for the dynamic development of local government finances, to which we now turn.
Historical Context
The origins of state system intervention in local government finances lie
deep in American history. Understanding changes in the institutional
framework of local government finance in 20th-century America is not
possible without understanding the 19th-century framework. Therefore, this section covers both the colonial period to the late 19th century
and the late 19th century to the present.20
From Colonies to the Late 19th Century
During the colonial period, cities enjoyed substantial autonomy from
colonial governments. Fourteen colonial cities were given royal charters. The charters granted extensive economic powers to cities, as well as
modes of governance that were not transparently democratic. Albany, for
example, possessed a monopoly on the fur trade of western New York.
Teaford (1975) argues that, following contemporary British practice in
the colonial period, local charters were regarded as sacrosanct.21
The independence of local governments from state government
intervention did not last after the American Revolution. States asserted
their rights to regulate local governments. State governments rescinded
or replaced the charters of all 14 colonial cities. In the cases of New
York; Philadelphia; Norfolk, Virginia; and Newport, Rhode Island,
states replaced city charters over the objections of the existing city governments. In the landmark 1819 Supreme Court decision about the
nature of corporate charters, Dartmouth v. Woodward, Justice Story
distinguished public corporations from private corporations. The case
Caveat Creditor: State Systems of Local Government Borrowing in the United States
is famous for articulating the principle that corporate charters are
contracts and that states are bound to honor their contracts. But the
decision explicitly recognized that states could change the charters of
public corporations, including municipal charters, at will.
There was never a period after winning its independence from
Britain when local governments in the United States were presumed
to be independent from state governments. Dillon’s “rule” that all local
governments are creatures of the state was not a ruling handed down by
Dillon, but a simple recognition of the facts on the ground. While states
varied widely in how they structured and regulated local governments,
certain patterns can be observed over time in state-local government
relationships with regard to finance and administration. These changes
began in the 1840s and continue to the present day.
In the early 1840s, eight states and the Territory of Florida defaulted
on their sovereign debts. Five of the states eventually repudiated all or
part of their bonds, and several other states renegotiated with bondholders.22 In the aftermath of the default crisis, almost half of the existing states wrote new constitutions. Eleven of the 12 new constitutions
contained “procedural debt restrictions.” These procedures allowed state
governments to borrow money, but legislatures were required to calculate the amount of new taxes necessary to finance bond repayment and
to submit a referendum to the voters, in which a majority must approve
the higher taxes before bonds could be issued. These “bond referendums” are common in American elections today. Although some states
capped the total amount of debt that could be outstanding, most states
only altered the procedure for issuing debt.
Imposing procedural restrictions on state bond issues raised the
political cost of borrowing at the state level. State legislatures were now
required to raise taxes before they borrowed and to obtain voter approval
of the increase. In response, borrowing shifted to the local level.23 Local
governments began borrowing large amounts of money and, in the
1870s, local governments began defaulting on their debts, particularly
on debts incurred to build or support railroads. States responded to the
wave of local defaults by extending procedural bond restrictions to local
governments. Table 14.2 lists the dates when states first extended fiscal
restrictions to local governments up through the 1890s, and table 14.3
lists the dates to the present.
549
550
Until Debt Do Us Part
Table 14.2 State Constitutional Provisions Governing Local
Debt and Borrowing Provisions, United States, 1841–90
State
Provision
Yeara
Alabama
1
1875
Arkansas
1
1874
California
1
1879
Colorado
1
1876
Connecticut
1
1877
Delaware
0
n.a.
Florida
1
1868, 1875
Georgia
1
1877
Idaho
1
1889
Illinois
1
1870
Indiana
1
1851, 1881
Iowa
0
n.a.
Kansas
0
n.a.
Kentucky
0
n.a.
Louisianab
1
1879
Maine
1
1868, 1878
Maryland
1
1867
Massachusetts
0
n.a.
Michigan
1
1850
Minnesota
1
1879
Mississippi
1
1875
Missouri
1
1875
Montana
1
1889
Nebraska
1
1875
Nevada
1
1864
New Hampshire
1
1877
New Jersey
0
n.a.
New York
1
1846, 1874, 1884
North Carolina
1
1876
North Dakota
1
1889
Ohio
1
1851
Oregon
1
1857
(continued next page)
Caveat Creditor: State Systems of Local Government Borrowing in the United States
551
Table 14.2 (continued)
State
Provision
Yeara
Pennsylvania
1
1873
Rhode Island
0
n.a.
South Carolina
1
1868, 1884
South Dakota
1
1889
Tennessee
1
1870
Texas
1
1876
Utah
0
n.a.
Vermont
0
n.a.
Virginia
0
n.a.
Washington
1
1889
West Virginia
1
1872
Wisconsin
1
1848, 1874
Wyoming
1
1889
Source: The provisions in the table are taken from the 1880 and 1890 Census
reports, supplemented by the constitutional texts on the National Bureau
of Economic Research (NBER)/Maryland Constitution project, http://www
.stateconstitutions.umd.edu.
Note: Local provisions include some type of restriction or regulation on the
issue of debt by local governments. These include procedural restrictions,
such as referendums, absolute dollar limits, and percentage valuation limits.
n.a. = not applicable.
a. The table list a “1” if the state had any provisions, a “0” if it did not. The
dates refer to the first year a state adopted a debt restriction or limitation,
and subsequent years where significant changes occurred. The dates are not
absolutely accurate, in the sense that they do not consider the confederate
or reconstruction constitutions in southern states. Several reconstruction
constitutions had debt limits, which were ignored, and interpreting those
limits is problematic.
b. Louisiana wrote constitutions in 1845, 1852, 1861, 1864, 1868, and 1879, 1898,
and 1913. The table refers only to the original 1845 provisions and the modifications made in 1879.
Table 14.3 State Constitutional Provisions on Local Government Debt Issue, United States
First
yeara
GO
limitsb
Other
debt
limitsc
Procedure
restrictionsd
Alabama
1901
1
1
1
Alaska
1959
1
1
1
1
1
1972, 1974, 1980
1
1
1924, 1926, 1984, 1986, 1988, 1992
State
Arizona
1912
Arkansas
1874
1
Other yearse
1927, 1960, 1965, 1967, 1972, 1977
(continued next page)
552
Until Debt Do Us Part
Table 14.3 (continued)
State
First
yeara
GO
limitsb
Other
debt
limitsc
Procedure
restrictionsd
California
1879
1
1
1
1892, 1900, 1906, 1914, 1918, 1922, 1940, 1949,
1950, 1972
Colorado
1876
1
1
1888, 1972
Connecticut
1877
1
Delaware
1897
Other yearse
1955, 1965
Florida
1912
1
1
1
1924, 1930, 1952, 1963, 1968
Georgiaf
1877
1
1
1
1945, 1976, 1983
Hawaii
1959
1
1
1
1968, 1978
Idaho
1890
1
1
1
1950, 1964, 1966, 1968, 1974, 1976, 1978, 1996
Illinois
1870
1
1
1
1904, 1970
Indiana
1881
1
Iowa
1857
1
Kansas
—
Kentucky
1891
1
1
1909, 1994
Louisiana
1898
1
1
1904, 1906, 1908, 1913, 1914, 1916, 1920–27,
1974, 1989, 1994
Maine
1878
1
1
1913, 1951, 1962
Maryland
1867
1
1
1
1933, 1934
1
1
1893, 1899, 1905, 1909, 1910, 1917, 1928, 1964
Massachusetts
—
Michigan
1850
1
Minnesota
1872
1
Mississippi
—
Missourif
1875
1
Montana
1889
1
Nebraska
1875
1
Nevada
—
New Hampshire
—
New Jersey
—
1879, 1924
1
1
1
1905, 1945, 1960, 1974, 1988, 1990, 2002
1
1950, 1973
1
1920, 1972, 1978
New Mexico
1911
1
1
1964, 1982, 1988, 1996
New York
1846
1
1
1
1894, 1905, 1907, 1909, 1917, 1938, 1945,
1951, 1953, 1963, 1973
North Carolina
1868
1
1
1
1936, 1946, 1962, 1971, 1973, 1976, 1977, 1986
North Dakota
1889
1
1
1
1920, 1981
(continued next page)
Caveat Creditor: State Systems of Local Government Borrowing in the United States
553
Table 14.3 (continued)
State
First
yeara
GO
limitsb
Other
debt
limitsc
Ohio
1912
1
1
Oklahoma
1907
1
1
Oregonf
1857
Pennsylvania
1874
Rhode Island
South Carolina
South Dakota
1889
Tennesseef
Procedure
restrictionsd
Other yearse
1974
1
1960, 1962, 1963, 1957,1958, 1963, 1963, 1966,
1976, 1986, 1998
1
1
1910, 1912, 1916, 1920
1
1
1911, 1915, 1918, 1951, 1961, 1966, 1969
1951
1
1
1986
1896
1
1
1977, plus
1
1
1954
1
1904, 1909, 1933, 1947, 1956, 1958, 1960, 1962,
1965, 1966,
1967, 1969, 1970, 1978, 1981, 1982, 1987, 1989,
1997, 1999
—
Texas
1876
1
Utah
1895
1
1
1911, 1975, 1991
Virginia
1902
1
1
1
1928, 1971, 1981
Washington
1889
1
1
1
1952, 1972, 1981
West Virginia
1872
1
1
1950
Wisconsin
1872
1
1
1909, 1929
Wyoming
1889
1
1
1919, 1953, 1961
f
1
Source: NBER/Maryland Constitution project, http://www.stateconstitutions.umd.edu.
Note: — = State has no constitutional provision regarding local borrowing.
a. The “First year” is the first year that state provision with respect to local governments appear.
b. “GO limits” are 1 when the state has some limit on the amount of general obligation debt that local governments can
issue. These limits can be absolute dollar amounts or relative limits (percentage of assessed value, percentage of tax
revenue, and so forth).
c. “Other limits” are 1 when the state government limits the amount of other types of debt that local governments can
issue, largely revenue bonds and forms of nonguaranteed debt.
d. “Procedural restrictions” are 1 when local governments are required to go through a specific procedure to approve a
debt issue, like a bond referendum or a super majority.
e. “Other years” are either new constitutions or amendments to the constitution that change the nature of the constitutional provisions.
f. Information for these states may be incomplete because of problems with the constitutional texts.
By the end of the 19th century, most state constitutions had
some provisions that regulated the fiscal behavior of local governments. Prominent among the provisions were requirements that local
governments hold referendums to approve tax increases before they
borrowed, and limits on the total amount of taxation, spending, or
borrowing local governments could engage in.
554
Until Debt Do Us Part
As described in the previous section, some states also implemented a
major institutional change in the 1850s, when state constitutions began
to mandate that state legislatures pass “general incorporation acts” for
creating municipal (and other local) governments. A general incorporation act provided a standardized form of corporate charter, which local
governments could implement through simple administrative procedures without the explicit approval of the state legislature.
There are two important implications of the general incorporation
acts. First, the active involvement of state legislatures was taken out of
the process of creating and structuring local governments. Second, in
principle, every municipal charter would be exactly the same within a
state—the charter is included in the general incorporation act. The general incorporation acts represented a major step in the creation of stable
political institutions governing local governments, clarifying rules, and
minimizing the extent of state-level political discretion over local institutions. Again, some states did not adopt general acts.
The problem with the general acts was the one-size-fits-all nature
of the chartering procedure. Local governments varied considerably in
size and circumstances. Beginning in the 1870s, states began inserting
“home rule” provisions in their constitutions, allowing some or all local
governments to structure their own charters within the limits laid out
by the state government. These home rule acts are better thought of as
“liberal general incorporation acts,” similar to the new incorporation
acts that states began creating for business corporations in the 1880s.
The home rule charters were more liberal in allowing local governments
to choose between a wider set of options for structuring their charters
and governments. The home rule reforms were consistent with the
desire to limit state political interference with local institutions. Within
the broader limits established under home rule, local governments were
essentially independent.
It must be emphasized, however, that home rule provisions in constitutions and home rule legislation passed by state legislatures typically
include restrictions on local governments as well: limits on local government borrowing, spending, and taxation; restrictions on the kinds
of activities that local governments could engage in; and restrictions
on the form of administration a local government could adopt. These
constitutional and legislative provisions are a central part of the passive rules governing local government borrowing. Table 14.1 provides
Caveat Creditor: State Systems of Local Government Borrowing in the United States
information on the dates states banned special incorporation for local
governments, mandated general incorporation for local governments,
and allowed home rule.
The systems in place by the late 19th century in all states were passive.
There was no active monitoring of local government fiscal conditions
by state governments, nor were there any mandated actions that state
governments took when a local government got into fiscal straights. The
systems worked fairly well because they were embedded in two larger
sets of social institutions and processes that actively monitored and disciplined local governments: citizens and bond markets. State constitutions and laws set the parameters within which local governments could
operate. Both citizens and bond markets could use those parameters as
a way to discipline local governments through the voting booth, courts,
and markets. Individual citizens could, and did, bring cases against local
governments in the courts when they felt that their local government
had overstepped its authority. Bond markets could discipline local borrowing through interest rates and bond ratings. So even though the
state insolvency system was passive, in the sense that the state did not
actively monitor or regulate local borrowing, the system enabled active
monitoring of local governments by citizens and markets. The system in
place in the late 19th century was far from perfect, however, and subsequent institutional changes would sharpen the ability of both voters and
markets to discipline local governments.
The various systems seem to have worked well, at least in a comparative context, as measured by the size of local government borrowing relative to state and national borrowing. Table 14.4 lists government debt
by level of government—national, state, and local—for selected dates
from 1838 to 2002. At the turn of the 20th century, local government
debt was larger than national and state debt combined. American local
governments were leaders in infrastructure and education investments.
They played a key role in financing the emergence of a modern industrial economy in the United States.24
Private and Public Institutions from the Late 19th Century
to the Present
The wave of local government defaults in the 1870s led many states to
require local governments to hold bond referendums to authorize local
government borrowing that obligated the general funds of the local
555
556
Until Debt Do Us Part
Table 14.4 Government Debt by Level of Government, Nominal Amount, and Shares,
United States, 1838–2002
Year
State debt
(US$ million)
1838
172
Local debt
(US$ million)
25
National debt
(US$ million)
State
share (%)
Local
share (%)
3
86.0
12.5
National
share (%)
1.5
1841
190
25
5
86.4
11.4
2.3
1870
352
516
2,436
10.7
15.6
73.7
1880
297
826
2,090
9.2
25.7
65.0
1890
228
905
1,122
10.1
40.1
49.8
1902
230
1,877
1,178
7.0
57.1
35.9
1913
379
4,035
1,193
6.8
72.0
21.3
1922
1,131
8,978
22,963
3.4
27.1
69.4
1932
2,832
16,373
19,487
7.3
42.3
50.4
1942
3,257
16,080
67,753
3.7
18.5
77.8
1952
6,874
23,226
214,758
2.8
9.5
87.7
1962
22,023
58,779
248,010
6.7
17.9
75.4
1972
59,375
129,110
322,377
11.6
25.3
63.1
1982
147,470
257,109
924,600
11.1
19.3
69.6
1992
369,370
584,774
2,999,700
9.3
14.8
75.9
1997
456,657
764,844
3,772,300
9.1
15.3
75.5
2002
642,202
1,042,904
3,540,400
12.3
20.0
67.8
Source: Wallis 2000.
government (table 14.2). Two institutional responses in the late 19th
century, one private and one public, came to play a much larger role in
the market for local debt in the late 19th and early 20th centuries and
continue to be important institutions in modern-day American local
government finance.
The private institution was the bond counsel.25 The prevalence of quo
warranto defenses by taxpayers and local governments in default on their
bonds led the intermediaries in the bond market to require assurance
as to the valid, binding, and enforceable nature of the bonds. Financial
houses that marketed local government bonds began to require legal
assurance that the bonds were authorized in accordance with the law and
that they were valid, binding, and enforceable agreements of the local
governments. This function was performed by a bond counsel that was
Caveat Creditor: State Systems of Local Government Borrowing in the United States
retained by the issuer to render a legal opinion to such effect upon which
bondholders could rely.
Almost all local government bonds today come with extensive disclosure documents26 about the nature of the bond issue, the revenues
available for its payment, other information that an investor would find
important to making an investment decision, and a bond counsel opinion. The bond counsel does not provide private insurance for public
debt, since a bond opinion does not address whether the local government borrower would be unable to or might refuse to honor its debts.
The bond counsel opinion typically addresses the lawful issuance of the
bonds, the inclusion in the official statement of an accurate description
of the bonds, the nature of the local government’s payment obligation,
and whether interest paid on the bonds is exempt from federal income
taxation.
The public institutions that developed were the special district (also
known as special governments, special funds, and special purpose
vehicles) and revenue bonds. As local governments were increasingly
required to hold bond referendums to authorize bond issues, more
special purpose local governments, which provided a function such as
water, sewerage, irrigation, and transportation, began to develop. Geographically, these special districts often spanned several existing local
governments and were sometimes gerrymandered, so that a majority of
the voters in the district benefited from the function that the special district provided and, therefore, would support a bond issue (and higher
taxes or user fees) at a bond referendum, if one was required by the laws
of the particular state.
Revenue bonds were similar in effect to the special district. Revenue
bonds did not obligate the general funds of a local government. General
obligation bonds, or “GO” debt, are typically subject to the referendum
procedure. In contrast, revenue bonds were to be paid from specific revenue sources. Sometimes these revenues were connected to a specific
function of the government, such as user fees for water service being
used to pay bonds that financed the water system, but sometimes the
revenues were simply a distinct revenue source dedicated to bond service.27 Courts in many, but not all, states held that revenue bonds were
not subject to the debt procedures that required bond referendums,
since the general taxpayer was not obligated to service the debt.
557
558
Until Debt Do Us Part
The development of revenue bonds was less a way to circumvent
state debt limits and procedures than a method of linking the beneficiaries of the project that the bond is to finance with the cost of project finance and the sources of payment.28 By linking user-paid fees and
taxes with financing cost, the revenue bonds were able to address the
challenges facing the states in the early 19th century—statewide voters in a democratic system were unlikely to vote for broad taxation to
finance a project that benefits only a location-specific population. The
off-budget financing of railroads and canals in the early 1800s was a way
of addressing this issue.29 But those arrangements encumbered taxpayers with contingent liabilities that ultimately became due, and the states
eventually defaulted.30 The state constitutional amendments aimed at
resolving contingent risks did not completely address the disconnect
between benefits and costs. The revenue bond instruments in the late
1800s solved the problem of linking infrastructure benefits to willingness to pay. Revenue bonds may be outside the state debt limitations but
are subject to their own sustainability criteria.
As the number and types of local governments and the types of local
government bonds proliferated and became more complex, the role
of bond counsel grew in importance in order to determine for investors whether a local government was legally authorized to issue a particular bond offering and had complied with state law authorization
requirements.31
States continued to adjust the constitutional and legal institutions
governing local governments in the early 20th century (table 14.1).
The next round of institutional changes, which began in the 1930s in
response to the stock market crash in 1929 and the depression that followed, involved two new national laws. The federal Securities Act of
1933 and the Securities and Exchange Act of 1934 introduced broad
regulations of securities markets in such areas as disclosure, fraud, and
dealer-broker registration.32 Congress perceived that full and fair disclosure to investors, fair and efficient markets, instilling investor confidence in those markets, and assisting the process of capital formation
were fundamental reasons to regulate securities and securities markets.33
Unlike private stocks and bonds, the securities issued by state and
local government are generally excluded from the regulations of the
1933 and 1934 acts. The exception comes in the case of fraudulent
Caveat Creditor: State Systems of Local Government Borrowing in the United States
activities. Direct federal regulation of the process by which state and
local governments raise funds to finance their government activities
would have placed the federal government in the position of gatekeeper
to the financial markets for state and local governments. This would
have undermined long-standing concepts of state sovereignty and local
voter sovereignty. Disclosure requirements for both municipal and corporate securities are essential to the appraisal of risks and returns by
investors. However, federal law does not dictate the types of disclosures
required by state or local governments.
The second new development in the Great Depression was the enactment of Chapter 9 of the Bankruptcy Code in 1937. The motivation for
the code was to resolve the holdout problem in negotiations between local
governments and their creditors.34 During the wave of defaults in the
Great Depression, protracted negotiations between millions of investors
and municipal debt issuers had proven to be costly and inefficient. In the
absence of well-understood and enforceable procedures, a single investor or a group of individual investors can prevent the debt restructuring
agreement reached between the debtor and majority of investors. The
Chapter 9 procedures enabled a debt restructuring agreement between the
majority of bondholders and the debtor that overcome the objections of
individual minority investors through the power of the court.35 Chapter 9
not only provides ex-post legal procedures for resolutions, but also frames
expectations of investors and debtor on potential risks of default.
What Chapter 9 does not do is violate the sovereignty of local voters.
Chapter 9 filings have strict conditions, framed by the U.S. Constitution,
that grant states the power to manage their political subdivisions. States
cannot be forced to allow their local governments access to Chapter 9
procedures, as noted earlier. Federal bankruptcy courts cannot force
local governments to raise taxes, cut expenditures, or sell assets, because
those are actions of local governments that can only be imposed by voters. Chapter 9 exists to coordinate the negotiation process between local
governments and their creditors. Chapter 9 is not a legal instrument
that creditors can use to force local governments to repay their debts.
As shown in chapter 8 by De Angelis and Tian (2013) in this volume,
Chapter 9 is rarely used. From 1937 to the present, there were roughly
600 Chapter 9 cases. From 1980 through 2009, an average of fewer than
8 cases were filed annually nationwide.36
559
560
Until Debt Do Us Part
Both institutional innovations in the 1930s were implemented by the
national government, but it would be a mistake to think of these changes
as national regulation of local government debt issuances. Securities
market regulation directly affected private actors, such as brokers and
dealers, in the market for local government debt, not the public actors
(although antifraud provisions do affect public debtors). However, both
state and local governments are indirectly affected by regulations that
impact the municipal bond market.
The market for local government debt continued to evolve during
the 20th century. The last major institutional change was the establishment of the Municipal Securities Rulemaking Board (MSRB).37 The
MSRB was created by Congress in 1975 to provide oversight and regulation of broker-dealer firms engaged in the municipal securities business. The MSRB does not regulate state or local governments.38 The
Dodd-Frank Act recently expanded its regulatory authority to cover
municipal advisors. Some MSRB rules have resulted in reduced transaction costs and increased information flows.39
A number of industry groups have also contributed to improvements in disclosure by state and local governments. For example, the
Government Finance Officers Association and the National Federation
of Municipal Analysts have developed many municipal bond disclosure recommendations that have become accepted industry practice.
Most states and large local governments follow the Generally Accepted
Accounting Principles for state and local governments established by
the Governmental Accounting Standards Board.40
Most of the institutional changes in the 20th century that influence local government creation and repayment of debt are directed
at the market for local government bonds, not the governments that
issue the bonds. The insolvency systems that developed in the United
States are primarily passive systems. Most states do not actively regulate or m
onitor local governments. Each state has different rules for
what local governments can do, how much they can tax, how they
borrow, and, in general, how much independence from state control
they have. As a result, private investors in local government bonds
need to be aware of the legal powers that a local government has
before they invest. In the simplest terms, if a local government does
not lawfully authorize a bond, the bond is void ab initio (from the
Caveat Creditor: State Systems of Local Government Borrowing in the United States
beginning), and the local government does not have a legal obligation
to repay the bond. Voters and taxpayers are liable only for the lawfully
issued obligations of the local government. This does not preclude
bondholders from pursuing legal actions against fraudulent activities, but owners of a bond cannot enforce the debt obligations if the
bond is void ab initio.
Shifting the liability for ensuring that a local government has the
authority to issue debt and the types and amount of revenues available
to service debts from the public to the private sector occurred gradually during the late 19th century. The shift was the result of the unique
development of American federalism and is both historically rooted
and path dependent. The outcome, however, was to create incentives
for private markets to actively monitor and discipline local government
borrowers. The institutional changes that followed in the 20th century,
several of them originating in the national government, are not directed
at the behavior of local governments but, instead, are intended to make
information flows in the private capital markets for local government
operate more effectively.
This does not mean that all states have passive insolvency systems,
however. In the next sections, we consider the development of active
systems in a small number of states in the late 20th century.
State Intervention and Monitoring
of Local Governments
In legal and constitutional terms, states play the dominant role in structuring local governments and managing municipal insolvency. But
the source and scope of local government powers have long been subject to controversy and change.41 As seen, actual practices fall between
two extremes of Dillon’s Rule and the Cooley Doctrine.42 A few states
actively monitor local governments on an ongoing basis and have institutions in place to deal with local government financial crises and insolvency. This section begins our examination of those states. We want to
emphasize, however, that these states represent a minority of the states;
they are not “typical.”
The emergence of active fiscal monitoring began with North C
arolina
in 1931, but it appears that North Carolina was ahead of other states.43
561
562
Until Debt Do Us Part
Adoption of systems in other states was motivated by local fiscal crises
in the 1970s. The well-publicized fiscal crises of New York City and
Cleveland had national influence beyond New York and Ohio, serving as
a “wake-up call” for other states.44 Florida enacted its Local G
overnment
Financial Emergencies and Accountability Act in 1979.45 Ohio enacted a
comprehensive municipal fiscal emergency statute in 1979, as well.46 In
the 1980s, after years of decline of western Pennsylvania communities,
Pennsylvania enacted a Municipalities Financial Recovery Act to assist
distressed local governments.
The 1970s was a period of slow economic growth and recurrent
crises, like the first OPEC (Organization of the Petroleum Exporting
Countries) oil embargo of 1973. The 1970s and 1980s were a period of
substantial reform in state finances, as well, including the introduction
of rainy day funds and tax and expenditure limitations and modifications of state constitutions to increase the flexibility of debt restrictions
to explicitly exclude revenue bonds from the limit.47
Active state regulation of local government fiscal activity involves
three parts: monitoring, crisis definition, and intervention. Most states
require local governments to adopt standard accounting standards
(which are not necessarily those promulgated by the Governmental
Accounting Standards Board) and to regularly file reports on financial activity with a state agency.48 Few states actually do much with the
information, however. Research by Mackey (1993) and Coe (2007) suggest that as of 2003 only 17 states actively monitored local finances on a
regular basis.
Of the 17 states that actively monitor local finances, 9 actually have
a system in place to predict whether local governments are headed
for a fiscal crisis. Table 14.5 lists the states and their intervention systems. States in the table are ranked by the level of activity they regularly
exhibit. The first column, “Coe Predict,” has a 1 if states attempt to predict local fiscal conditions. The second column, “Coe Intervene,” has a
1 if the state has in place policies for intervening in local government
affairs. These nine states have the most active regulation of local fiscal
activity. We examine three of these states, North Carolina, Ohio, and
Pennsylvania, in the next section of the chapter.
The third column, “Kloha Monitor,” has a 1 if the state monitors local
activity in any way at all. Kloha, Weissert, and Kleine (KWK) (2005)
Caveat Creditor: State Systems of Local Government Borrowing in the United States
563
Table 14.5 State Monitoring of Local Fiscal Conditions, Home Rule, and Local Debt
Restrictions, United States
State
Coe predict
Coe
intervene
Kloha
monitor
Kloha early
warning
Honadle
formal
Honadle
none
States that monitor local fiscal conditions and attempt to predict fiscal crisis
Florida
1
1
1
1
1
0
Kentucky
1
1
1
0
0
0
New Jersey
1
1
1
1
1
0
New Mexico
1
1
1
0
1
0
North Carolina
1
1
1
1
1
0
Ohio
1
1
1
1
1
0
Pennsylvania
1
1
1
1
1
0
Maryland
1
0
1
1
0
0
New Hampshire
1
0
1
1
0
0
States that monitor local fiscal conditions but do not predict fiscal crisis
Alaska
0
0
1
0
1
0
Connecticut
0
0
1
0
0
0
Illinois
0
0
1
0
0
1
Massachusetts
0
0
1
0
0
0
Michigan
0
0
1
0
1
0
Nevada
0
0
1
0
0
1
New York
0
0
1
0
0
0
West Virginia
0
0
1
0
1
0
States that neither monitor local fiscal conditions, predict fiscal crisis, nor intervene in
a fiscal crisis
Alabama
0
0
0
0
0
1
Arizona
0
0
0
0
0
0
Arkansas
0
0
0
0
0
0
California
0
0
0
0
0
0
Colorado
0
0
0
0
0
0
Delaware
0
0
0
0
0
0
Georgia
0
0
0
0
0
1
Hawaii
0
0
0
0
0
0
Idaho
0
0
0
0
0
0
Indiana
0
0
0
0
0
1
Iowa
0
0
0
0
0
0
Kansas
0
0
0
0
0
1
(continued next page)
564
Until Debt Do Us Part
Table 14.5 (continued)
Coe predict
Coe
intervene
Kloha
monitor
Kloha early
warning
Honadle
formal
Honadle
none
Louisiana
0
0
0
0
0
1
Maine
0
0
0
0
0
1
Minnesota
0
0
0
0
0
0
Mississippi
0
0
0
0
0
0
Missouri
0
0
0
0
0
1
Montana
0
0
0
0
0
1
Nebraska
0
0
0
0
0
0
North Dakota
0
0
0
0
0
1
Oklahoma
0
0
0
0
0
1
Oregon
0
0
0
0
0
1
Rhode Island
0
0
0
0
1
0
South Carolina
0
0
0
0
0
1
South Dakota
0
0
0
0
0
1
Tennessee
0
0
0
0
1
0
Texas
0
0
0
0
0
0
Utah
0
0
0
0
0
1
Vermont
0
0
0
0
0
1
Virginia
0
0
0
0
0
1
Washington
0
0
0
0
0
0
Wisconsin
0
0
0
0
0
1
Wyoming
0
0
0
0
0
1
State
Sources: Coe 2007; Honadle 2003; Kloha, Weissert, and Kleine 2005.
conducted a phone survey of all 50 states to determine whether the
states had any monitoring system in place.49 KWK also asked whether
the states had an “early warning” system in place. If KWK report that the
state did, column 4 reports a 1 for “Kloha Early Warning.” Some confidence in the survey methodology can be found in the fact that KWK
find nine states with early warning systems, and they are all states that
Coe finds to have prediction systems.
The next element is the process by which a “fiscal crisis” is determined.
Honadle (2003) surveyed all 50 states to determine how and whether
states had procedures for determining when a crisis was occurring. She
Caveat Creditor: State Systems of Local Government Borrowing in the United States
found that 11 states had formal definitions of a fiscal crisis and 20 states
had no definition. In between the extremes, 14 states had working definitions of a crisis that had not been formalized, and 8 states relied on
local governments to define a local fiscal crisis.50 The table reports a 1 if
the state had a formal definition, “Honadle formal.” Likewise, the table
reports a 1 if the state had no definition under “Honadle none.” The
states with informal definitions are not indicated in the table.
The final step is intervening or assisting local governments that are in
crisis. Coe (2007) found that seven states had active intervention systems
in place. These systems worked pre- and postcrisis. States can also offer
assistance to local governments. Mackey found that 13 states had statutory provisions for providing state assistance, and an additional 6 states
had provided assistance on an ad-hoc basis. Mackey’s sample includes
41 states, 9 of which did not respond. Mackey’s results have not been
included in the table because of the problem with missing observations.
The arrangement of the table reflects the intensity with which states
monitor and intervene in local fiscal affairs but not the extent to which
they assist local governments that are in trouble. The first panel of the
table includes all of the nine states that monitor local fiscal conditions
and attempt to predict fiscal crisis. Seven of these nine states also have
systems in place to intervene in local finance. These states have the most
active monitoring and intervention systems. The second panel includes
the states that monitor local fiscal conditions but do not predict fiscal
crisis. The final panel includes those states that do not monitor local fiscal conditions, predict fiscal crisis, or intervene in a fiscal crisis.51 These
states have completely passive systems, at least ex ante, and most states
fall into the lower panel. They do not monitor local fiscal conditions nor
do they have any systematic plans in place for intervening in or assisting
local governments. This does not mean that states in the lower panel
are unwilling to assist local governments. They may well be, but they
do not have institutional arrangements in place to do so. To the extent
that these states are active, it is only in an ex-post sense, after a crisis has
already developed.
Section six of the chapter does some simple empirical tests to determine whether fiscal behavior in these three types of states differed significantly over the late 20th century. We find that they did. But first, we
want to look more carefully at three of the most active states, North
565
566
Until Debt Do Us Part
Carolina, Ohio, and Pennsylvania, to get a better idea of what “active”
monitoring and intervention in local government actually means.
State Intervention Cases
This section presents a more in-depth review, up to 2010, of the state
intervention system in North Carolina, Ohio, and Pennsylvania—
chosen from the active end of the state intervention spectrum. The
intervention systems in the three states were put in place to respond to
the debt stress of their respective local governments. However, the crisis
response differs across the three states. Fiscal monitoring and insolvency
resolution concern the source and scope of local government powers, so
home rule provisions are critical because they grant local governments
the levels of self-government rights, limit state intervention power, and,
consequently, determine how states can intervene in local affairs.
North Carolina is the most active state manager of local finance in
the United States.52 Ohio, a home-rule state, has a fiscal watch program
to provide early warning and a fiscal emergency program to deal with
entities mired in fiscal crises, but the state lacks broad intervention
power. With strong home rule limiting the power of the state’s intervention, Pennsylvania has used a state-appointed coordinator to provide a
fiscally distressed local government with a financial recovery plan; however, the local government can reject the adoption of the plan.53
During the Great Depression, North Carolina suffered from the second-largest number of municipal bond defaults in the United States.54
The Local Government Commission (LGC) was created by the state
legislature in 1931 to control local debt and assist fiscally distressed
municipalities.55 Ohio’s original municipal fiscal emergency law was
enacted in 1979 in response to the financial crisis in Cleveland. After
years of struggle in a recession and a massive auto industry layoff, in
1978, Cleveland became the first major American city to default on
its debts since the Great Depression.56 In 1987, when communities in
western Pennsylvania were hit by job losses due to the decline of the
steel industry, the state legislature enacted the Municipalities Financial
Recovery Act, known as Act 47, to assist fiscally distressed municipalities.57 Designed for small municipalities, Act 47 could not address the
financial difficulties experienced by Philadelphia when it reached the
Caveat Creditor: State Systems of Local Government Borrowing in the United States
brink of bankruptcy in 1991.58 Additional state assistance was thus provided through new legislation that created the Pennsylvania Intergovernmental Cooperation Authority (PICA).
The North Carolina constitution enables the state legislature to
define the powers and duties of municipalities. According to the state
constitution, the state can destroy, restructure, and create any of its
subdivisions.59 The state constitution, therefore, placed almost entire
control over municipalities in the hands of the state. Intergovernmental relationships in North Carolina embody a strong version of Dillon’s
Rule. In this constitutional context, North Carolina’s LGC possesses
strong intervention power in local financial management when the local
fiscal situation starts to deteriorate.
The LGC can issue orders to raise taxes or other revenues to meet
debt payment.60 These orders are as enforceable as those issued by
local officials. The LGC, under the State Treasurer, is actively involved
in almost all phases of local financial management. The LGC approves
almost all local debt issues and monitors local fiscal conditions. The
state has the authority to take over the financial management of local
governments that do not comply with the LGC directives, experience
fiscal stress, or fail to report financial operations on a timely basis.61
The LGC approves almost all traditional types of municipal debt,
including general obligation bonds, revenue bonds, installment and
lease purchase obligations, and the use of swap agreements. The LGC
uses indicators to measure debt affordability such as per capita debt,
debt as a percentage of assessed valuation, annual debt service payments as a percentage of the general fund budget, and the fiscal health
of enterprise operations and comparison of user charges levied.62 After
approving debt issues, the State Treasurer’s office handles the sale and
transactions of all local debt and monitors debt servicing.63
The North Carolina approach to financial oversight and monitoring
is proactive. The LGC monitors the finance of about 1,230 local governments and public authorities through annual reviews.64 The LGC requires
an independent auditor, mandates standard audit contracts, approves the
selection of local government auditors and audit contracts, and permits
final payment to an auditor only after approving the financial report.65
If financial problems are discovered, the LGC sends a letter to the local
entity, expressing concerns and offering suggestions for improvement.
567
568
Until Debt Do Us Part
Local units that receive a letter are required to report a detailed plan of
corrective actions to resolve the problems.66
In contrast to North Carolina, Pennsylvania’s constitution prohibits the state legislature from enacting laws that regulate the affairs
of a specific municipality. It also prohibits the state from establishing
any commission with the power to intervene in municipal functions or
delegating such powers to any commission.67 In order to comply with
this constitutional provision, the state cannot unilaterally intervene in
a municipal insolvency. State participation must be conditional on local
agreement or assent.68
The implementation of Act 47 in 1987 to assist fiscally distressed
municipalities is administered by the Department of Community and
Economic Development (DCED). Act 47 enables the DCED to compile
fiscal data of municipalities to monitor their fiscal condition and determine their distress status, using 11 indicators.69 Any single indicator can
trigger a declaration of distress status by the DCED.
Once a municipality is declared financially distressed, the DCED
appoints a coordinator responsible for developing and implementing a financial recovery plan.70 Act 47 grants municipalities the right to
reject such a plan. When labor unions challenged the constitutionality
of Act 47,71 the court held that the state-appointed coordinator did not
constitute a special commission and that municipalities retained their
decision-making authority through their right to reject the plan.72 The
ability of municipalities to reject the state plan is a critical result of the
general law provision: the state cannot treat local governments differently without their permission.
The state created a special commission, the PICA, in 1991, to address
the financial problems of near-default in Philadelphia. The PICA was
created under the intergovernmental corporation clause in the state
constitution. The state uses financial incentives to encourage local cooperation and to punish those who refuse to cooperate with the state by
withholding state funds.73 The PICA weathered legal challenges from
unions.74 The court held that the PICA did not interfere with municipal powers because the city voluntarily entered into a cooperative
agreement with the PICA.75 In a manner similar to Pennsylvania, the
Ohio Constitution sets forth that “Municipalities shall have authority to exercise all powers of local self-government.”76 Furthermore, the
Caveat Creditor: State Systems of Local Government Borrowing in the United States
c onstitution prohibits the state from adopting special legislation for a
specific municipality.77 As a result, in response to Cleveland’s fiscal crisis
in 1979, Ohio adopted a general statute that governs a wide range of
municipalities: cities, villages, counties, and school districts.
The 1979 local emergency law created the fiscal emergency program
and institutionalized the Financial Planning and Supervision Commission to assist fiscally distressed local governments. The State Auditor can
place a local government in the emergency program if the latter crosses
the threshold of any of six financial conditions relating to arrears, fund
deficits, liquidity, default on debt, payroll arrears, and tax transfers.78
The law was amended in 1996 to add a fiscal watch program providing early warning to faltering entities whose fiscal conditions approach
emergency status.79
After the State Auditor declares a local government in fiscal emergency, a financial planning and supervision commission is formed. The
municipality must submit a financial recovery plan for commission
approval. The commission then ensures the timely implementation of
the plan. The financial recovery plan lays out substantive fiscal adjustments to restore financial stability, eliminate fiscal emergency conditions, and avoid future reoccurrence. Five out of seven voting members
of the commission are locally based, including two local officials and
three locally nominated members. The state cannot mandate changes;
instead, it recommends strategies and oversees the implementation.80
The Ohio state intervention system is characterized by weak state intervention power.81
The commission has broad authority to make recommendations
regarding all financial matters, including cost reductions or revenue
increases, making and entering into all contracts and agreements necessary to the performance of its duties, and ensuring a balanced budget and its implementation. The commission is empowered to review
all revenue and expenditure estimates, and to approve and monitor the
monthly levels of expenditures and encumbrances consistent with the
financial plan. The commission also reviews the amount and purpose
of any debt issues, and provides technical support on the structure and
terms of debt obligations.82 The termination of an entity’s fiscal watch
status occurs when either the warning conditions no longer exist, as
determined by the State Auditor, or fiscal conditions continue to
569
570
Until Debt Do Us Part
deteriorate. In the latter case, the State Auditor may declare a fiscal
emergency.83 Local governments’ participation in the fiscal emergency
program averaged 5.1 years.84
Ohio does not actively formally monitor local governments, unless
a government enters the fiscal watch or emergency program.85 The
absence of monitoring is well illustrated by the fact that most fiscally
distressed municipalities bypassed fiscal watch and directly entered into
the fiscal emergency program. Early detection of fiscal problems is on a
voluntary basis. Municipalities, however, may discover they are in trouble later than might be optimal. The state intervention system in Ohio
is reactive rather than proactive. In practice, most cases are initiated by
local entities through request, although the fiscal emergency or watch
program could also be triggered by the Governor or the State Auditor.86
North Carolina, Pennsylvania, and Ohio offer revealing insights into
the nature of active state intervention into local fiscal affairs. If a state
constitution recognizes the independent fiscal power of local governments, the ability of the state in the fiscal adjustment process of the
local government is limited. Special legislative bans prohibit legislation
targeted at a specific municipality. Creation of financial control boards
may not be feasible if special commissions are prohibited. Local consent
may be required for state intervention, and states need to provide incentives for local cooperation.
North Carolina is an outlier at the very active end of the spectrum
of state and local relationships, even in this small sample of three states.
In most states, North Carolina’s aggressive intervention in local government affairs would be unconstitutional, given their state constitutional
frameworks. Since local governments, in general, have been successful
in borrowing to finance infrastructure during the 20th century, it does
not appear that North Carolina’s active state intervention has broad relevance to other states in efforts to achieve local fiscal responsibility.
Empirical Results
In order to determine whether the insolvency systems shown in
table 14.5 had a significant impact on local government finance, we
performed a simple set of difference-in-differences estimates on state
and local fiscal activity in 1972, 1992, and 2007. Data for those years
Caveat Creditor: State Systems of Local Government Borrowing in the United States
(and others) are readily available from the U.S. Census of Governments.
Data for 1992 and 2007 come after states began active monitoring and
intervention programs.
We took three basic measures of state and local fiscal behavior. “Total
revenue” is total revenue from “own” sources for state governments, all
local governments, and state and local governments combined. Own
revenue is revenue collected directly by the state or local governments
and does not include revenue from intergovernmental grants. “Total
expenditure” is total “direct” expenditure. Again, direct expenditure
excludes expenditures for intergovernmental grants and counts only
expenditures that are made by state or local governments directly. “Total
debt” outstanding combines both long- and short-term debt, but for
most state and local governments, the preponderant share of total debt
is long-term debt.87
Each of these variables is measured in two ways. The first takes the
local share of the combined state and local total. The local share of
revenue is local government own revenues divided by combined state
and local own revenues (excluding any grant revenue from the federal
government). Local shares of expenditure and debt were calculated in a
similar manner. The other measure is revenues, expenditures, and debt
per capita. These are measured in nominal terms for each year (the differencing takes care of the changes in price levels).
Table 14.6 reports the difference-in-differences results for the relative size of local fiscal activity in the state and local fisc.88 The first
panel of the table gives the local share of combined state and local revenues, expenditures, and debt for each of the three years: 1972, 1992,
and 2007. The first column gives the average share for the “Predict
States,” the states that Coe (2008) indicates are predicting whether a
fiscal crisis will occur. The second column gives the average share for
the “Monitor states,” which are the states that Kloha, Weissert, and Kleine (2005) find actively monitor local finances but do not attempt to
predict or intervene in local affairs. The third column gives the average
local share for the “Control states” (the untreated states), which do not
monitor or predict. The fourth column gives the average local share
for “All states.”
When interpreting the numbers, it is worth noting that the raw data
are on own revenues and expenditures for each level of government.
571
572
Until Debt Do Us Part
Table 14.6 Difference-in-Differences Estimates, Local Share of State and Local Totals,
United States
Local share
Predict states
Monitor states
Control states
All states
1972
Revenue share
0.411
0.433
0.426
0.428
Expend share
0.560
0.545
0.541
0.541
Debt share
0.641
0.616
0.712
0.693
Revenue share
0.387
0.379
0.398
0.393
Expend share
0.513
0.501
0.519
0.515
Debt share
0.583
0.488
0.571
0.560
Revenue share
0.411
0.385
0.397
0.397
Expend share
0.499
0.501
0.501
0.501
Debt share
0.537
0.469
0.557
0.539
1992
2007
First difference 1972–92
Revenue share
-0.024
-0.054
-0.028
Expend share
-0.047
-0.044
-0.021
Debt share
-0.058
-0.129
-0.141
First difference 1972 to 2007
Revenue share
Expend share
Debt share
0.000
-0.048
-0.030
-0.061
-0.044
-0.039
-0.104
-0.148
-0.155
Difference-in-differences
1972–92
Revenue share
0.004
-0.025
-0.025
-0.023
0.083
0.012
Revenue share
0.030
-0.018
Expend share
-0.022
-0.004
0.051
0.008
Expend share
Debt share
1972 to 2007
Debt share
Source: Census of Governments.
Since local governments are, in every case, net recipients of grants from
the states, and grants are excluded from the “own” and “direct” categories of revenues and expenditures, local governments have a smaller
share of state and local revenue than of state and local expenditures.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
The second panel of the table gives the first difference of the various
measures between 1972 and 1992, and between 1972 and 2007. Most of
the active monitoring came online in the 1970s and 1980s, and should
be reflected in both differences.
The third panel of the table gives the difference-in-differences estimates, comparing the difference in the “Predict states” and “Monitor
states” to the “Control states.”
Table 14.6 illuminates two striking results.
The first is the relative decline in the local share of the state and local
fisc on almost all measures over both time periods. Local revenues as a
share of state and local revenues rose slightly after 1992, but on every
other measure, the local fisc got relatively smaller.
The second is that the decline in the local share of debt issued
occurred much more slowly in the “predict” states, although it still
occurred. The change in local share of state and local debt was 8 percent
higher in “predict” states than in the “control” states during 1972–92,
and 5 percent higher over the longer period 1972 to 2007. This is a significant impact.
The table does not give measures of statistical significance, because
the 50 states are the relevant universe. This is not a sample of states.
Tables 14.7, 14.8, and 14.9 perform similar difference-in-differences
estimates for per capita revenues, expenditures, and debt for local, state,
and combined state and local totals. The tables have the same format as
table 14.6.
All three tables share the same striking, and unexpected, result. The
“monitor” states, that is, the states that monitor but do not actively
predict or intervene in local affairs, exhibit much faster growth in
every measure of state and local fiscal activity.
Unlike the results in table 14.6, which show that the share of state
and local borrowing is devolving to local governments in the “predict”
states relative to other states, tables 14.7, 14.8, and 14.9 show that per
capita revenues, expenditures, and debt are all growing faster in the
“monitor” states than in the “control” or “predict” states, and this goes
for both state and local governments. The effect of being a “monitor”
state is about twice as large for state-level measures as it is for local-level
measures. And unlike the table 14.6 results, the effect is stronger rather
than weaker over the longer period, 1972 to 2007, than over the shorter
period.
573
574
Until Debt Do Us Part
Table 14.7 Local Total Revenue, Expenditure, and Debt per Capita, United States,
1972, 1992, 2007
Predict
states (US$)
Monitor
states (US$)
Control
states (US$)
Revenues
280
358
298
312
Expenditures
446
636
453
489
Debt
444
690
469
512
Local total
All
states (US$)
1972
1992
Revenues
1,435
1,845
1,454
1,513
Expenditures
2,188
2,794
2,195
2,290
Debt
2,107
2,846
2,039
2,180
2,449
2,829
2,300
2,412
2007
Revenues
Expenditures
4,156
4,928
4,195
4,305
Debt
3,947
4,588
3,670
3,867
Revenues
1,154
1,487
Expenditures
1,742
2,158
1,741
Debt
1,663
2,156
1,570
Revenues
2,168
Expenditures
3,709
4,292
3,741
Debt
3,503
3,898
3,201
First difference 1972–92
1,156
First difference 1972 to 2007
2,471
Difference-in-differences
1972–92
Revenues
-2
Expenditures
331
0
416
93
587
Revenues
166
468
Expenditures
-32
551
Debt
302
697
Debt
1972 to 2007
Source: Census of Governments.
2,002
Caveat Creditor: State Systems of Local Government Borrowing in the United States
Table 14.8 State Total Revenue, Expenditure, and Debt, United States, 1972,
1992, 2007
Predict
states (US$)
Monitor
states (US$)
Control
states (US$)
All
states (US$)
Revenues
391
444
397
409
Expenditures
346
559
391
424
Debt
268
462
222
269
Revenues
2,302
3,365
2,233
2,427
Expenditures
2,075
2,935
2,054
2,199
Debt
1,659
3,093
1,601
1,850
Revenues
3,453
4,888
3,543
3,742
Expenditures
4,202
5,152
4,183
4,341
Debt
3,420
5,562
2,943
3,448
State total
1972
1992
2007
First difference 1972–92
Revenues
1,911
2,922
1,836
Expenditures
1,729
2,376
1,663
Debt
1,391
2,631
1,379
First difference 1972 to 2007
Revenues
3,062
4,445
3,146
Expenditures
3,856
4,592
3,792
5,100
2,721
Debt
3,151
Difference-in-differences
1972–92
Revenues
75
Expenditures
66
712
11
1,252
-84
1,299
Debt
1,085
1972 to 2007
Revenues
Expenditures
Debt
64
801
430
2,379
Source: Census of Governments.
575
576
Until Debt Do Us Part
Table 14.9 Combined State and Local Total Revenue, Expenditure, and Debt,
United States, 1972, 1992, 2007
State and local
total
Predict
states (US$)
Monitor
states (US$)
Control states
(US$)
All
states (US$)
1972
Revenues
671
802
695
721
Expenditures
793
1,195
844
913
Debt
712
1,152
691
781
3,721
5,183
3,672
3,923
Expenditures
4,264
5,729
4,249
4,488
Debt
3,766
5,940
3,640
4,031
Revenues
5,901
7,717
5,843
6,154
Expenditures
8,358
10,080
8,377
8,646
Debt
7,367
10,150
6,613
7,314
1992
Revenues
2007
First difference 1972–92
Revenues
3,050
4,382
2,977
Expenditures
3,471
4,533
3,405
Debt
3,054
4,788
2,949
First difference 1972 to 2007
Revenues
5,230
6,915
5,148
Expenditures
7,565
8,884
7,533
Debt
6,654
8,998
5,922
Difference-in-differences
1972–92
Revenues
73
1,404
Expenditures
66
1,129
Debt
105
1,838
82
1,767
1972 to 2007
Revenues
Expenditures
Debt
32
1,351
733
3,076
Source: Census of Governments.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
The impact of being a “monitor” state is economically significant.
State and local expenditures per capita, for example, rose by US$8,884
between 1972 and 2007. Being a “monitor” state increases state and local
expenditures by US$1,351. We cannot tell whether this is a causal effect
or not. States with rapidly growing expenditures may have implemented
monitoring systems, or local governments in states with monitoring
systems may face lower costs of raising revenue. It is also interesting that
in “monitor” states, own revenues rise faster than direct expenditures
(although interpreting this result for the overall fisc depends on changes
in grants from the national government, which are excluded from these
numbers).
The results in these tables are only suggestive. Clearly, many things
were happening in the states over this 35-year period, and no attempt
was made to control for selection. Some states became “predict” or
“monitor” states because they wanted to shift their fiscal structure, or
perhaps because their fiscal structures were shifting for completely
different reasons, so no causal interpretation should be placed on the
estimates.
Nonetheless, they do show that states systematically differ along the
dimension of insolvency systems and that the presence of the systems is
correlated with fiscal outcomes. Of particular interest is that the local
share of state and local activity increased in “predict” states between
1972 and 2002. In other words, states most actively monitor and intervene in local government fiscal matters in those states in which local
governments are growing in relative importance (or local governments
are shrinking at a slower rate). The association of active monitoring and
a growing local share of revenues and expenditures may suggest that
these states are consciously adapting institutions in a way that makes
local governments more important. The interaction among institutions
that govern how states regulate local taxing, spending, and borrowing
with levels of state and local taxing, spending, and borrowing is a fertile
area for future research.
Lessons: Market and Voter Discipline
Earlier sections of this chapter looked closely at the structure and development of the American local government insolvency system, from the
577
578
Until Debt Do Us Part
early 19th century to the present, in order to draw conclusions relevant
to the contemporary developing world. Because of the wide variety
of American experience across time and states, several lessons can be
drawn, some in conflict with each other, rather than a single set of institutional recommendations. Nonetheless, there are strong commonalities across the states.
The United States has by far the largest local government capital market in the world. In 2007, local governments issued US$225 billion in
bonds, and total local government debt outstanding was US$1.5 trillion.
The market works well, particularly in the context of a global comparison. Local governments borrow significant amounts of money to finance
infrastructure investments and have very low rates of default.
Local governments in most states face restrictions on how they borrow and what they can borrow for and, in some states, how much they
can borrow. For the most part, these restrictions are on the procedures
that local governments must follow to approve borrowing and how debt
service obligations are related to specific revenue sources (particularly
in the case of revenue bonds).
The central feature of the American experience is the importance
of ex-ante and passive insolvency systems. Only one-third of the states
have a system in place for monitoring local governments, and less than
20 percent have institutions and policies that enable or require state
action in the face of a local government fiscal crisis. The lack of active
state programs does not mean that local government borrowing and
debt servicing are not actively monitored by the larger society. Instead,
it highlights how the interaction of ex-ante institutional rules, voters,
capital markets, and courts play the key role in monitoring and limiting
local government borrowing.
The way the systems actually work can be difficult to grasp, since it
appears that local governments are offered a loophole to issue debt they
do not have to repay. State laws and constitutions authorize local governments to issue debt following certain procedures. In order to protect
citizen sovereignty, if a local government does not authorize its bond
issues in a lawful way, the bonds are not legally binding obligations, and
the local government is not obligated to pay them. Very few states, however, actually monitor local governments to prevent local governments
from issuing debt in unauthorized ways. Instead, the bondholders find
Caveat Creditor: State Systems of Local Government Borrowing in the United States
themselves in a position where the courts will not enforce their claims as
creditors against the local government if the local government has overstepped its bounds or violated procedures for authorizing and issuing
debt. As a result, the role of the bond counsel is important in assuring
the lawful authorization of the bonds that is the requirement for their
validity and enforceability.
The result of the framework for debt issuance has not been local governments that borrow wildly in unauthorized ways and then default,
but rather a steady increase in the capacity of private capital markets to
assess the creditworthiness of local governments and inform potential
borrowers of the actual conditions under which local debt is issued and
will be repaid. The institutional developments such as the bond counsel
and MSRB all make the provision of information to private market participants more credible and transparent. The national government has
not violated the sovereign powers of states to tax, spend, and borrow as
they wish, nor have they impaired the ability of states to establish systems for their local governments.
Whether developing countries have strong enough public and private
institutions to take advantage of passive systems is a question that cannot
be easily answered. The United States developed its framework for subnational debt through a series of incremental changes in institutions over a
long period of time. Establishing the legal precedent that local taxpayers
were not responsible for servicing debts that were incurred in an unauthorized manner or through defective procedures was a long, drawn-out
process undertaken at the end of the 19th century. Passive insolvency
systems also clearly require the existence of strong and credible rule of
law in order to work. Institutions in many developing countries are yet
to be developed to the extent necessary to discipline and regulate both
public debtors and private creditors in such a subtle way. Moreover, the
system operating in one country cannot be applied without care to other
country contexts. Nonetheless, it is a worthy goal to work toward creating clear interests among creditors to support strengthening both the
rule of law and incentives for private market development. One part of
the market process in the United States—transparency and disclosure of
the credit risks of all issuers— generally would be relevant.
The passive systems establish a close relationship between borrowing
and taxation.89 As noted, most of the constitutional reforms that followed
579
580
Until Debt Do Us Part
the 1840s required states, and local governments after the 1870s, to raise
current taxes when they issued debt. Forcing voters and taxpayers to
simultaneously raise taxes when they borrowed money led voters to pay
closer attention to the benefits of the expenditures that local governments
proposed. Again, this was a change at the margin that affected incentives
and likely had a positive effect on the dynamics of the political economy
process. A similar set of incentives was set in motion when special districts
and revenue bonds became widespread at the end of the 19th century.
By formally recognizing that local governments were creatures of the
states, Dillon’s Rule supported the premise that local governments could
borrow only when they were explicitly authorized to do so, for specific
functions, and often in limited amounts. In principle, states possess the
authority to unilaterally change the structure of any local government.
In practice, however, Americans learned that allowing state politics to
manipulate local politics was bad for democratic outcomes, so they
moved toward “general” laws governing local governments. This was
an institutional change that had to arise endogenously in the American setting, but it offers illustrative lessons on how an institution can be
adopted organically in a developing country.
Changing the way central governments interact with subordinate
governments changes the political economy dynamic among the levels
of government. Allowing subordinate governments to determine their
own debt and tax policies within defined limits must be paired with
the obligation to raise local taxes to service those debts. Such a policy
will only be self-sustaining if it applies equally (generally) to all local
governments. If individual local governments can approach the central
government for special treatment, or if the central government can single out individual local governments for special treatment (either positive or negative), then the incentives to create and enforce credible rules
are eroded. If all cities know that the same rules equally apply to all of
them, then all the cities collectively have a strong incentive to make sure
that a state enforces the rules equally across all cities and that the rules
work for most of the cities. The incentive to press for general and efficient rules works via the political economy incentives facing local governments. Those incentives disappear if the central government treats
subordinate governments differently. General rules, and the incentives
they create, cannot credibly develop if the central governments treat
Caveat Creditor: State Systems of Local Government Borrowing in the United States
states and cities differently. This is one of the larger lessons developing
countries can draw from the American experience.
The importance of general laws for local governments also helps
explain why so few states have active insolvency systems that require
states to intervene in local government finances. Ex-post intervention
creates the ex-ante expectation of action. Unless the state government is
willing to intensively regulate many dimensions of local political decisions, it may be better not to get involved at all on a systematic basis. In
most states, local crises are dealt with on an ad-hoc basis or not at all
by state governments. In contrast, as the case of North Carolina shows,
strict state monitoring comes with the loss of some local autonomy.
The economic recession that began in 2008 has caused a fiscal crisis
for many state and local governments in the United States. The crisis
has not yet passed, and it remains to be seen whether it will engender institutional changes along the lines of the 1930s or the 1970s and
1980s. The movement toward more active state monitoring systems in
the 1970s and 1980s led to more, not less, local government borrowing. Local governments with more debt are more susceptible to economic downturns, regardless of the state monitoring systems in place.
What has worked well in the United States is not fixed fiscal rules, but
the adoption of institutions that enable political and economic markets,
voters, and capital markets, to more clearly assess the cost and benefits
of government borrowing. It is hoped that institutional change in the
future will continue to enhance that ability.
Notes
The findings, interpretations, and conclusions expressed in this work are those
of the authors and do not necessarily reflect the views of The World Bank, its
Board of Executive Directors, the governments they represent, or any other
institutions with which the external authors may be affiliated.
1. See Canuto and Liu 2010a; Liu and Waibel 2009; and Liu and Webb 2011. Many
developing countries still rely on bank loans as the main source of finance. The
subnational capital market remains small (see Canuto and Liu 2010a).
2. In several other federal countries, such as Brazil, Canada, and India, states and
provinces have considerable political and fiscal power granted by their respective national constitutions.
3. The Eleventh Amendment reads: “The Judicial power of the United States
shall not be constructed to extend to any suit in law or equity, commenced or
581
582
Until Debt Do Us Part
prosecuted against one of the United States by Citizens of another State, or by
Citizens or Subjects of any Foreign State.”
4. United States Census of Governments 2007.
5. United States Census of Governments 2007. A complete census of U.S. governments is taken in years ending in 2 and 7, so the 2007 Census is the most recent
complete census. Bonds issued by state and local governments are sometimes
lumped together and called municipal bonds (or bonds issued by subnational
governments). From 2007 to 2011, annual average issuances of total subnational bonds were about US$450 billion (data source: Thomson Reuters), and
at the end of 2011, outstanding debt was US$3 trillion (Federal Reserve Board,
Flow of Funds Accounts of the United States, June 2012, Table L.104, p. 63).
6. See Canuto and Liu (2010b) for bond issuance by subnational governments
outside the United States during 2000–09. In 2009, subnational governments
outside of the United States issued about US$309 billion in new bonds, while
state and local governments issued US$368 billion in new bonds (Federal
Reserve Board, Flow of Funds Accounts of the United States, June 2012, Table
L.104, p. 63).
7. The terms “passive,” “active,” “ex ante,” and “ex post” have no legal meaning;
they are descriptive terms used in the chapter to help understand the nature of
the way states interact with and regulate local government borrowing.
8. Chapter 9 of the bankruptcy code is for all local governments, as defined here,
not just municipal (city) governments. Only 27 states allow their local governments to access the federal bankruptcy courts, and, of those 27, only 18 states
allow local governments access without explicit state permission. For more on
Chapter 9, see chapter 8 by De Angelis and Tian (2013) in this volume; Liu and
Waibel (2009); and McConnell and Picker (1993).
9. Liu and Waibel (2009), and chapter 8 by De Angelis and Tian (2013) in this
volume, with sources from Laughlin (2005) and Spiotto (2008). For recent
development in states’ authorization, see Spiotto (2012). Whether more or less
than half the states have active systems depends on how the nine states that
allow their local governments access to Chapter 9 under limited conditions are
counted.
10. Chapter 8 by De Angelis and Tian (2013) in this volume, with sources from the
American Bankruptcy Institute; the website of United States Courts, http://www
.uscourts.gov; “Bankruptcy Basics” (2006) by the Administrative Office of
the United States Courts and the American Bankruptcy Institute; the 2007
U.S. Census; http://www.abiworld.org; and Public Access to Court Electronic
Records, http://www.pacer.gov.
11. Since the Civil War and the Fourteenth Amendment, the national government
and federal courts have exercised control over more aspects of state and local
governments, for example, in the area of compliance with civil rights laws. But
relationships between state and local governments remain an area governed
almost exclusively by state constitutions and state laws.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
12. See Briffault 1990.
13. See Williams (1986) for a summary of the theories about the nature of local
governments.
14. Dillon’s Rule; Judge John F. Dillon of the Iowa Supreme Court, first authored
the rule in his Commentaries on the Law of Municipal Corporations, shortly after
the Civil War.
15. See, Hunter v. City of Pittsburgh (1907), a Supreme Court case; and Briffault 2008.
16. States have increased and decreased local autonomy in constitutions over time,
as we will show in general terms in the next section.
17. These constitutional provisions and laws have parallels in provisions and laws
that require general incorporation for business enterprises, banks, churches, or
any type of corporate group.
18. Section 1, Article 13, Ohio Constitution, 1851.
19. Briffault and Reynolds 2004; Briffault 2008.
20. It is impossible to give a clear date that divides the two periods, because states
varied so much in the way they interacted with and regulated local governments.
21. Teaford 1975, 79.
22. Wallis 2005.
23. See Wallis and Weingast 2008.
24. The table is taken from Wallis (2000), which also discusses the relative importance of national, state, and local fiscal activity over the entire history of the
United States.
25. For the history and development of bond counsels, see Maco (2001).
26. These are similar to a corporate prospectus but are usually called an “official
statement.” The official statement and additional information provided by the
issuer over the life of most bonds are now freely available to the public at www
.emma.msrb.org and from various commercial information vendors so that
investors may better evaluate credit risk on an ongoing basis. It is important
to note that, unlike its authority over offerings of securities by corporate entities, the U.S. Securities and Exchange Commission currently lacks authority to
mandate the contents or format of municipal disclosure documents. As a result,
the financial statements included are often quite stale and sometimes do not
adhere to Generally Accepted Accounting Principles, and it is difficult to compare one bond to another due to the lack of a uniform format or the inclusion
of uniform information in official statements for comparable bonds.
27. Such as a hotel-motel occupancy tax being used to pay bonds issued to repair
roads.
28. Today, about two-thirds of subnational debt in the United States is revenue
bonds—bonds issued collateralized by the revenue streams of the project that
the bond is to finance. The first revenue bond was issued in 1885 by Wheeling,
West Virginia, to finance a water and gas plant. The revenue bonds became
mainstream in the 1970s in the United States. (Marlin and Mysak 1991, supra
note 45, at 18, at 62)
583
584
Until Debt Do Us Part
29. As noted above, bonds for such private purposes were often held invalid under
Dillon’s Rule, and the issuer was excused from repayment.
30. Wallis 2005.
31. In addition to bond counsel representing the interests of investors, there are other
counsel involved in the debt-issuing process. Underwriter’s counsel ensures
that the issuer’s financial condition and plans and other matters that are important for an investor to know are accurately disclosed. http://www.publicbonds
.org.
32. The two federal acts draw from Maco (2001).
33. Unless otherwise noted, this section draws mainly from Maco (2001).
34. McConnell and Picker 1993.
35. For more on Chapter 9, its origins, framework, and applications, see McConnell
and Picker (1993), chapter 8 by De Angelis and Tian (2013) in this volume, and
Liu and Waibel (2009).
36. 1980–2008 data are from American Bankruptcy Institute; 2009 data are from
the website of United States Courts, http://www.uscourts.gov.
37. Maco 2001.
38. The Exchange Act requires MSRB rules for broker-dealers and municipal advisors to be designed to prevent fraudulent and manipulative acts and practices,
to promote just and equitable principles of trade, to foster cooperation and
coordination with other persons engaged in specified market support activities,
to remove impediments to and perfect the mechanism of a free and open market in municipal securities, and, in general, to protect investors and the public
interest. See section 15B(b)(2) of the Exchange Act for a complete description
of the scope of the MSRB’s authority to promulgate rules.
39. For example, through its Electronic Municipal Market Access system, the MSRB
currently provides information on more than 1.5 million state and local bond
issues to the public at no charge. See http://www.emma.msrb.org.
40. Municipal accounting is not uniform. Each state determines what accounting
standards they and their local governments will use. Not all use Governmental
Accounting Standards Board standards (Haines 2009).
41. See Briffault 1990.
42. See Williams (1986) for a summary of the theories about the nature of local
governments; see Hunter v. City of Pittsburgh; and Briffault (2008).
43. Compiling a good chronology of when states initiated active insolvency systems is beyond the scope of this chapter; however, it appears that there was
roughly a 40-year lag after North Carolina began monitoring before another
state adopted comparable institutions.
44. Cahill and James 1992; CRCM 2000.
45. Mackey 1993.
46. Mackey 1993.
47. Rodriguez-Tejedo and Wallis 2010, 2012.
48. Mackey 1993.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
49. Kloha, Weissert, and Kleine 2005. Coe found that New Mexico and Kentucky
also actively monitor, and they have been included as monitored, even though
Kloha, Weissert, and Kleine did not.
50. Honadle 2003. The numbers add up to more than 50, because several states
reported more than one definition. One of the problems with collecting information from phone surveys is that Honadle reports that North Carolina had no
definition of a fiscal crisis when, in fact, North Carolina has been actively monitoring and intervening in local finances since the 1930s. We reclassified North
Carolina in the table.
51. The information in the table was current as of 2003–05, when the various studies were undertaken.
52. Carter 1995; Coe 2008; Kimhi 2008; Wall 1984.
53. Coe 2008.
54. Only Florida had more municipal defaults than North Carolina (Laskey and
Hall 2004).
55. Laskey and Hall 2004; Wall 1984.
56. The city defaulted on US$15 million in short-term bond anticipation notes,
held primarily by six banks. The crisis of Cleveland was characteristic of local
difficulties in resolving fiscal problems. Tax increases were subject to popular referendum, and, in the 1970s, the city had a lower per capita tax burden
than many other major cities, such as Chicago, Detroit, and Philadelphia. For
more, see Ruggeri (2008) and Pennsylvania Economy League, Eastern Division
(1991).
57. Cahill and James 1991. From the passage of Act 47 in 1987 to May 2010, 26
municipalities were declared fiscally distressed under Act 47 and received state
assistance. Department of Community & Economic Development, http://www
.newpa.com, accurate as of May 2010.
58. Cannon 1993. According to estimates, the city’s deficit is 260 times larger than
the entire revolving fund of Act 47.
59. North Carolina Constitution, Article VII.
60. North Carolina General Statutes, Chapter 159, §159–36; Carter 1995.
61. Laskey and Hall 2004.
62. Marino, Woodell, and Thomas 2000.
63. The Division also provides technical assistance to municipal issuers before any
applications are presented to LGC for approval. See Moore 2008.
64. See Moore 2008.
65. Coe 2007.
66. In fiscal year 2007/08, 358 letters were sent to local units regarding the issues
of overall fiscal health such as fund balance level, compliance with adopted
budget, working capital level in water and/or sewer funds, and compliance
with various statutory requirements (“Annual Report 2008,” State Treasurer of
North Carolina).
67. Pennsylvania Constitution, Article 3, Section 31.
585
586
Until Debt Do Us Part
68. Pennsylvania Economy League, Eastern Division (1991); see the case study of
this article
69. Section 123(a) of Act 47. Certain state funds will be withheld until municipalities provide the DCED with all required data. The indicators include a deficit
over a three-year period, with a deficit of 1 percent or more in each of the previous fiscal years, expenditures exceeding revenues for three or more years, debt
service default, and payroll arrears of 30 days. See Section 201 of Act 47.
70. Section 221 of Act 47. The coordinator cannot be the elected or appointed official or employee of the municipality.
71. 564 A.2d 1015 (Pa. Commw. Ct 1989) as cited by Cannon (1993).
72. Cannon 1993.
73. According to ACT 47, the state gives a municipality a loan and grant only
when the municipality adopts the financial recovery plan created by the stateappointed coordinator. If the municipality refuses to adopt the plan, it has
to come up with its own plan, which will be subject to DCED’s examination. DCED’s determination of the plan’s inability to address problems will
trigger punishment by the state. The state will withhold certain state funds,
including grants, loans, entitlements, and payment from the state or any of
its agencies.
74. Cannon 1993.
75. Cannon 1993.
76. Section 3 of Article XVIII of the Ohio Constitution.
77. Article XVIII of the Ohio Constitution. The article was created by constitutional amendment in 1912.
78. The City of Niles, Ohio, became the first city in the fiscal emergency program,
followed by the City of Cleveland. From 1979 to November 1, 2010, the Financial Planning and Supervision Commission aided 59 fiscally distressed municipalities. For the six conditions, see Chapter 118 of the Ohio Revised Code.
79. Chapter 118 of the Ohio Revised Code. From the amendment of the law to
November 1, 2010, 22 local governments were in the fiscal watch program.
80. Beckett-Camarata 2004; Coe 2007.
81. Coe 2008.
82. Chapter 118 of the Ohio Revised Code.
83. Since the inception of Ohio’s fiscal watch program in 1996 to November 2010,
of 22 local governments in the fiscal watch program, 12 were removed from
fiscal watch status; the condition of 5 deteriorated, and they entered the fiscal emergency program (http://www.auditor.state.oh.us). Twelve municipalities were assisted and graduated from the program with fiscal health and an
improved accounting and reporting system.
84. Auditor of State, accurate as of November 1, 2010. Calculated based on 35 local
entities whose fiscal emergency status has been terminated.
85. Taylor 2009.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
86. From 1979 to November 2010, 11 of 58 local units that had been in the fiscal emergency program were put in the program by the Auditor of State. From
1990 to November 2010, 4 of 40 local units that had been in the fiscal emergency program were put in the program by the Auditor of State (http://www
.auditor.state.oh.us).
87. Grants were excluded because of problems in modeling the endogenous determination of grants among national, state, and local governments.
88. The “fisc” is the combination of a government’s fiscal activity and includes revenues, expenditures, and debts.
89. Decentralizing expenditure and borrowing power will need to be sequenced
with decentralizing revenue flexibility. Subnational fiscal and debt sustainability is not separable from the intergovernmental fiscal system.
Bibliography
Alter, David, Julie Berman, and Patricia McGuigan. 1992. “Administrative Factors in
Rating Local Debt: Case Studies in the Southeast Region.” Moody’s Municipal
Issues 9: 7–9, 12.
Bailey, Robert W. 1984. The Crisis Regime: The MAC, the EFCB, and the Political
Impact of the New York City Financial Crisis. Albany, New York: State University
of New York Press.
Beckett-Camarata, Jane. 2004. “Identifying and Coping with Fiscal Emergencies in
Ohio Local Governments.” International Journal of Public Administration 27
(8–9): 615–30.
Berman, David R. 1995. “Takeovers of Local Governments: An Overview and Evaluation of State Policies.” Publius 25 (3): 55–70.
Briffault, Richard. 1990. “Our Localism: Part I—The Structure of Local Government
Law.” Columbia Law Review 90 (1): 7–8.
———. 2008. “The Central Place of State and Local Governments in American
Federalism.” In The Handbook of Municipal Bonds, ed. Sylvan G. Feldstein and
Frank J. Fabozzi, 3–23. Hoboken: John Wiley & Sons, Inc.
Briffault, Richard, and Laurie Reynolds. 2004. Cases and Materials on State
and Local Government and Local Government Law. 6th ed. St. Paul, MN:
Thomson West.
Cahill, Anthony G., and Joseph A. James. 1991. Coping with Fiscal Distress in Pennsylvania’s Local Governments: A Program Evaluation of Act 47. Harrisburg:
Pennsylvania State University, Department of Public Administration, April.
———. 1992. “Responding to Municipal Fiscal Distress: An Emerging Issue for
State Governments in the 1990s.” Public Administration Review 52 (1): 88–94.
Cahill, Anthony G., Joseph A. James, Jean E. Lavigne, and Ann Stacey. 1994. “State
Government Responses to Municipal Fiscal Distress: A Brave New World for
587
588
Until Debt Do Us Part
State-Local Intergovernmental Relations.” Public Productivity and Management
Review 17 (3) (Spring): 253–64.
Cannon, Drew P. 1993. “An Analysis of Pennsylvania’s Legislative Programs for
Financially Distressed Municipalities and the Reaction of Municipal Labor
Unions.” 98 Dickinson Law Review 28 (Winter): 281–305.
Canuto, Otaviano, and Lili Liu. 2010a. “Subnational Debt Finance and the Global
Crisis.” Economic Premise 13, Poverty Reduction and Economic Management
Network, World Bank, Washington, DC, May.
———. 2010b. “Subnational Debt Finance: Make It Sustainable.” In The Day After
Tomorrow: A Handbook on the Future of Economic Policy in the Developing
World, ed. Otaviano Canuto and Marcelo Giugale, 219–37. Washington, DC:
World Bank.
Carter, Lee K. 1995. “State Oversight of Local Government Finance.” In State and
Local Government Relations in North Carolina, ed. Charles D. Liner. Chapel Hill,
NC: Institute of Government, the University of North Carolina at Chapel Hill.
Coe, Charles K. 2007. “Preventing Local Government Fiscal Crises: The North Carolina Approach.” Public Budgeting & Finance 27 (3): 39–49.
———. 2008. “Preventing Local Government Fiscal Crises: Emerging Best Practices.” Public Administration Review 68 (4): 759–67.
CRCM (Citizens Research Council of Michigan). 2000. “Avoiding Local Government Financial Crisis: The Role of State Oversight.” Report 329, Citizens
Research Council of Michigan, Lansing, Michigan.
De Angelis, Michael, and Xiaowei Tian. 2013. “United States: Municipal Bankruptcy—Utilization, Avoidance, and Impact.” In Until Debt Do Us Part: Subnational Debt, Insolvency, and Markets, ed. Otaviano Canuto and Lili Liu, 311–52.
Washington, DC: World Bank.
Federal Reserve Board. 2012. Flow of Funds Accounts of the United States. Washington, DC, June.
Haines, Martha. 2009. “Regulations of Subnational Securities Market.” Paper presented at the Zhuhai International Conference on Subnational Debt Management, Zhuhai, China, September 21.
Hennessey, Jessica. 2009. “Endogenous Institutional Change: The Transformation of
the State-Local Relationship in the United States.” Ph.D. thesis, University of
Maryland.
Honadle, Beth W. 2003. “The States’ Role in U.S. Local Government Fiscal Crises: A
Theoretical Model and Results of a National Survey.” International Journal of
Public Administration 26 (13): 1431–72.
Hren, Henry, Michael Morelli, and Lois Briggs. 1997. “Missed Opportunity: Urban
Fiscal Crises and Financial Control Boards.” Harvard Law Review 110: 733–50.
Jeweler, Robin. 2007. Municipal Reorganization: Chapter 9 of the U.S. Bankruptcy
Code. Washington, DC: Congressional Research Service.
Kimhi, Omer. 2008. “Reviving Cities: Legal Remedies to Municipal Financial Crises.”
Boston University Law Review 88: 633.
Caveat Creditor: State Systems of Local Government Borrowing in the United States
Kloha, Philip, Carol Weissert, and Robert Kleine. 2005. “Someone to Watch Over
Me: State Monitoring of Local Fiscal Conditions.” American Review of Public
Administration 35 (3): 236–55.
Laskey, Amy, and Rebecca Hall. 2004. State of North Carolina Local Government
Commission. New York: Fitch’s Investors Service.
Laughlin, Alexander. 2005. Municipal Insolvencies: A Primer on the Treatment of
Municipalities under the Bankruptcy Code. Wiley Rein & Fielding LLP, Washington, DC. http://www.abiworld.org/pdfs/municipal_primer.pdf.
Liu, Lili, and Michael Waibel. 2009. “Subnational Governance and Insolvency:
Cross-Country Experiences.” In Does Decentralization Enhance Service Delivery
and Poverty Reduction?, ed. E. Ahmad and G. Brosio, 333–75. Northampton,
MA: Edward Elgar Publishing.
Liu, Lili, and Steven Webb. 2011. “Laws for Fiscal Responsibility for Subnational
Discipline: International Experience.” Policy Research Working Paper 5587,
World Bank, Washington, DC.
Mackey, Scott R. 1993. “State Programs to Assist Distressed Local Governments.”
Legislative Finance Paper 86, National Conference of State Legislatures, Denver, Colorado, March.
Maco, Paul S. 2001. “Building a Strong Subnational Debt Market: A Regulator’s Perspective.” Richmond Journal of Global Law & Business 2 (1): 1–31.
Marino, Richard, Colleen Woodell, and LaVerne Thomas. 2000. “North Carolina
Local Government Commission Contributes to Higher Ratings.” Standard &
Poor’s, New York.
Marlin, George J., and Joe Mysak. 1991. The Guidebook to Municipal Bonds. New
York: The Bond Buyer.
McConnell, Michael, and Randal Picker. 1993. “When Cities Go Broke: A Conceptual Introduction to Municipal Bankruptcy.” University of Chicago Law Review
60 (2): 425–35.
Moore, Richard H. 2008. “The State Treasurer’s Annual Report to the People of
North Carolina 2008.” North Carolina Department of State Treasurer, Raleigh.
National Bureau of Economic Research/Maryland Constitution Project. www.state
constitutions.umd.edu.
Pennsylvania Economy League, Eastern Division. 1991. “Fiscal Crises and Remedies:
A Comparative Study of Philadelphia and Other Large Jurisdictions, with
Emphasis on Fiscal Supervision Agencies.” Report 596, Philadelphia, April.
Raphael, Richard J., Eric Friedland, Amy R. Laskey, and Amy S. Doppelt. 2010. “The
Perils of Considering Municipal Bankruptcy.” Fitch Ratings, Public Finance,
January 27.
Ruggeri, Amanda. 2008. “Three Decades after Cleveland Defaulted on Its Debts,
Cities Face Recession Budget Woes.” U.S. News & World Report, December 15.
Rodriguez-Tejedo, Isabel, and John Joseph Wallis. 2010. “Lessons for California
from the History of Fiscal Constitutions.” California Journal of Politics and
Policy 2 (3): 1–19.
589
590
Until Debt Do Us Part
———. 2012. “Fiscal Institutions and Fiscal Crises.” In When States Go Broke: The
Origins, Context, and Solutions for the American States in Fiscal Crisis, ed. Peter
Conti-Brown and David A. Skeel, Jr., 9–39. New York: Cambridge University
Press.
Shalala, Donna E., and Carol Bellamy. 1976. “A State Saves a City: The New York
Case.” Duke Law Journal 1976: 1119.
Spiotto, James E. 2008. “Chapter 9: The Last Resort for Financially Distressed
Municipalities.” In The Handbook of Municipal Bonds, ed. Sylvan G. Feldstein
and Frank J. Fabozzi, 145–90. Hoboken: John Wiley & Sons, Inc.
———. 2012. Primer on Municipal Debt Adjustment—Chapter 9: The Last Resort for
Financially Distressed Municipalities. Chapman and Cutler LLP, Chicago.
Taylor, Mary. 2009. Fiscal Indicators: A Proactive Approach to Local Government
Financial Assistance. http://www.auditor.state.oh.us/conferences/fiscaldistress/
handouts.
Teaford, Jon C. 1975. The Municipal Revolution in America: Origins of Modern American Government, 1650–1825. Chicago: University of Chicago Press.
United States Census of Governments. 2007. Data taken from online sources at
http://www.census.gov/govs/cog.
United States House of Representatives, Committee on Government Reform and
Oversight, Subcommittee on the District of Columbia. 1995. “Actions Taken by
Five Cities to Restore Their Financial Health.” Washington, DC, March 2.
Wall, Susan. 1984. “State Assistance to Local Governments: N.C. Local Government
Commission.” North Carolina Insight (June): 6–8.
Wallis, John Joseph. 2000. “American Government Finance in the Long Run: 1790 to
1990.” Journal of Economic Perspectives 14 (1): 61–82.
———. 2005. “Constitutions, Corporations, and Corruption: American States and
Constitutional Change, 1842 to 1852.” Journal of Economic History 65 (1): 211–56.
Wallis, John Joseph, and Barry R. Weingast. 2008. “Dysfunctional or Optimal Institutions: State Debt Limitations, the Structure of State and Local Governments,
and the Finance of American Infrastructure.” In Fiscal Challenges: An Interdisciplinary Approach to Budget Policy, ed. Elizabeth Garrett, Elizabeth A. Grady,
and Howell E. Jackson. New York: Cambridge University Press.
Williams, Joan C. 1986. “The Constitutional Vulnerability of American Local Government: The Politics of City Status in American Law.” Wisconsin Law Review
1: 83–153.
Index
Boxes, figures, notes, and tables are indicated by b, f, n, and t following the page number.
A
accounting, 251, 295. See also Public
Accountants
Acharya, S., 118
Act 47. See Municipalities Financial
Recovery Act
Act I of French decentralization of 1983
(France), 224
Acto Legislativo No. 04 of 2007 on
subnational governments
(Colombia), 198
Act on the Status of Budget Subjects of
2008 (Hungary), 297, 305n42
Act Relating to Cities and TownsProviding Financial Stability of
2010 (Rhode Island), 341–42n53
Advice Notice on Pilots Working
on Resolving the Debt for
Rural Nine-Year Compulsory
Education (Chinese State
Council), 99
Afonso, José R., 72n2, 73n4
Agreement Supervisory Committee,
202–3
Agri-Agra requirements, 439, 451n62
agrobonds, 464
Ambac, 316, 341n41
Annual Budget Law (Brazil), 52b
anticipated revenue credit operations
(AROs), 39, 46
aportaciones (earmarked transfers), 148,
170t
Argentina, sovereign debt in, 2
ARI (average regular income), 425
AROs (anticipated revenue credit
operations), 39, 46
arrears, 285, 465
asset liquidation, 284
Auditor General (South Africa), 511
Australia, government securities and,
116
automatic stays, 318–19, 331, 333,
342n58
average regular income (ARI),
425
591
592
Index
B
bailouts. See also debt write-offs
extraordinary transfers and, 166–67
“hidden,” 157–58
moral hazards and, 158, 267, 301, 371
soft budget constraints and, 266–67
Washington Public Power Supply
System, 363–65, 371
balance-of-payments crises, 37
Bangko Sentral ng Pilipinas, 425, 447n23
Bankers Association of the Philippines,
449n49
bankruptcy. See insolvency
Bankruptcy Code (U.S.), 7, 311, 313,
314, 332, 340n25, 559
Bankruptcy Courts (U.S.), 314, 320–21
Banobras, 157
Barbosa Filho, Fernando de Holanda, 65
Basel II Revised International Capital
Framework, 295, 307n52
best interests of creditors tests, 320
Blanco, Fernando, 65
BLGF. See Bureau of Local Government
Finance
Blueprint on Education Reform and
Development (China), 105n6
bonds
agrobonds, 464
automatic stays and, 318–19
bond counsels, 556–57, 558, 579
boom in, 277
debt adjustment and, 285
“double barreled,” 449n45
Eurobonds, 460, 472
global financial crisis of 2008–09
and, 272
market development for, 465–79,
522–23
municipal defaults and, 353–75. See
also Washington Public Power
Supply System
pool financing and, 523–24
“power bonds,” 130
provincial, 381–82, 390–93, 396
referendums, 549, 556–57
revenue, 319, 365–66, 370, 435, 443,
557–58, 583n28
subnational, 384, 385f, 389–94,
395–96t, 397–98, 408, 458–65, 540
void ab initio, 560–61
Bonneville Power Administration
(BPA), 357, 358, 359–60, 362,
373n12
borrowing. See subnational debt
Bose, D., 116
BOT (build-operate-transfer) financing,
452n72
Boueri, Rogério, 76n, 76n34
BPA. See Bonneville Power
Administration
Bradley-Burns Act (California), 344n94
Brazil, 10–11, 33–79
debt renegotiation, 41–53, 71
debt distress and, 41–42
fiscal consolidation and, 47–51
Fiscal Responsibility Law and,
51–53, 52b, 71–72, 74n19
macroeconomic stabilization,
42–47, 51, 71
debt restructuring in, 6, 9
fiscal balance and, 44–45, 46–47f
fiscal federalism, 33, 35–40
economic and political crises, 36–37
redemocratization, 37–40
fiscal performance, 53–65
deficit and debt trend, 54–57, 55t,
56f, 57t
municipal revenues and
expenditures, 61–65, 62f, 63t, 64f
state revenues and expenditures,
57–61, 58t, 59f, 61t, 62f
global financial crisis of 2008–09,
65–70, 71
fiscal sustainability and, 68–70
states and municipalities, 66–68
Index
overview, 33–35
Special Settlement and Custody
System (SELIC) rate and, 44,
45f, 46, 69, 73n9
subnational debt in, 2, 8
tax collection and, 38, 38f
“bridge loans,” 40
Brown, Kenneth Willy, xx, 495
Budget Code (Russia)
adoption of, 466, 467, 489
debt definition in, 488
debt management and, 471–75, 472t
federal loans and, 481
guarantees and, 482
subnational debt and, 22, 456–57
Budget Guidelines Law (Brazil), 52b
Budget Law of 1994 (China)
balanced budgets and, 383
borrowing and, 81, 94, 381
supply side demands and,
405, 406
Urban Development and Investment
Corporation and, 384
budgets
emergency, 282–84
hard constraints on, 26, 418
reform of, 300
rules and procedures for, 230–35,
231b, 232t, 235f
soft constraints on, 96, 103,
266–67
subnational governments and, 239
build-operate-transfer (BOT)
financing, 452, 452n72
Bureau of Local Government Finance
(BLGF, Philippines), 425, 426,
428, 430, 435, 447nn22–23
C
Cai, Hongbin, 403
Caisse d’Epargne, 241
California Constitution, 324, 344n94
Canada
government securities and, 116
subnational bonds and, 540
Canuto, Otaviano, xix, 1, 25, 70, 146, 436
capital grants, 60
capital investment funding, 269–70
capital markets, private credit and,
436–39. See also China,
subnational market access in
Cardoso, Fernando Henrique, 51–52
cash flow analyses, 318
Caticlan-Boracay Jetty Port
(Philippines), 437
Census of Governments (U.S.), 571,
582n5
Central Bank of Brazil, 36,
74–75nn23–24
Central Bank of Russia, 465, 481, 485
CGCT. See General Code of Territorial
Entities
Chapter 7 bankruptcy (U.S.), 316
Chapter 9 bankruptcy. See United
States, Chapter 9 bankruptcy in
Chapter 11 bankruptcy (U.S.), 314, 316
China, People’s Republic of. See also
China, rural legacy school debt
in; China, subnational market
access in
bonds and, 76–77n45, 540
subnational debt in, 447n26
China, rural legacy school debt in, 9–10,
11, 81–107
as informal practices, 386–87
moral hazard and, 82–83, 96–99, 102–3
overview, 81–83
restructuring of, 99–103
revenue assignment for education,
83–94, 85t, 89t, 94t
Compulsory Education Law, 84–85
government distribution of, 85–88
rural compulsory education,
88–94, 90–91t
significant features of, 94–96, 95t
593
594
Index
China, subnational market access in,
19–20, 379–416
borrowing frameworks prior to 2009,
382–90
informal practices, 386–87
new financing system, 387–90,
387t
onlending, 383, 410n9
revenues and expenditures, 384,
384f
subnational bonds, 384, 385f,
389–90
Urban Development and
Investment Corporation and,
383–85
expansion since 2009, 390–97
developments since 2010, 393–96
institutional reforms, 396–97, 406,
408
provincial bonds, 390–94, 395–96t
experience and implications of,
397–407
demand for subnational debt,
399–403, 407
financing gap, 402–3, 402t, 407
growth expectations, 401, 401f
subnational bonds, 397–98
supply side of debt instruments,
403–7, 406f
urbanization, 399–400, 399f, 407–8
land asset-based financing, 385–86,
400–401, 400f, 411n17
overview, 379–82
China Index Academy, 411n17
Citibank, 264
Cleveland, Ohio, financial crisis in, 566,
569
Coe, Charles K., 562, 564, 565, 585n49
Cohen, Jeffrey, 347n126
collateral
debt, 271
land, 529
Collor, Fernando Affonso, 41
Collor Plan II, 41, 73n5
collusion schemes, 192–93, 212
Colombia, 9, 14, 179–219
legal aspects of, 199–205, 200f, 205t
overview, 179–81
procedures for, 205–12
effects of, 209–12, 210t, 211f
implementation of, 205–9, 206t,
207f
subnational debt and, 189–99
borrowing trends, 189–93, 190f,
191–92t
subnational borrowing, 193–99,
194–96t
subnational governments and,
181–88, 182f, 212–13
decentralization, 182–86, 184f
responsibilities and resources,
185–88, 186–88t
Committee for the Reform of
Territorial Entities (Comité
pour la réforme des collectivités
territoriales, France), 253
companies limited by shares (Ltd.), 286
Complementary Law No. 101 of 2000.
See Fiscal Responsibility Law of
2000
Compulsory Education Law of 1986
(China), 81, 82, 84–85
compulsory expenditures, 234, 255n27
Consolidated Sinking Fund (India),
129
Constitution, France, 223, 224, 232
Constitution, India, 12, 112
Constitution, U.S. See Eleventh
Amendment; Tenth
Amendment
Constitutional Amendment No. 3
of 1993 on public securities
(Brazil), 42
Constitutional Amendment No. 19 of
1998 on public administration
(Brazil), 51
Index
Constitutional Law No. 2003-276 on
decentralization (France), 224
Constitutional Reform of 1986
(Colombia), 182
Constitution of 1917 (Mexico), 152
Constitution of 1967 (Brazil), 40
Constitution of 1988 (Brazil), 38–39, 51
Constitution of 1989 (Hungary), 268
Constitution of 1991 (Colombia), 183,
186, 189
Constitution of 1996 (South Africa),
495, 496, 498–500, 499b, 509,
532
Constitution of 2004 (China), 84, 104n3
construction sector, 482
Consumer Price Index, 194
contracts
executory, 319, 327
take-or-pay, 358–59, 361
Contracts Clause (U.S. Constitution),
313, 339–40n24
Contrôle hiérarchisé de la dépense
(Multi-layered Control of
Expenditures), 238
Convención Nacional Hacendaria
(National Fiscal Convention),
151
Cooley, Thomas, 544
Cooley Doctrine, 544, 545, 561
cooperative legal structures
(établissements publics de
coopération intercommunale),
226
Corbacho, A., 141n40, 216n32
corporate charters, 549
Council for Mutual Economic
Assistance, 265
County Courts (Hungary), 280, 304n34
Cour des Comptes. See National Court
of Accounts
Craig, Jon, 146
cramdown powers, 6, 29n5
Crédit Agricole, 241
Crédit Local de France, 241
credit markets
deepening of, 520–23
diversification of, 2
Government Financial Institutions
(Philippines) and, 431–33,
438–39, 441–42
infrastructure financing and, 511–19,
532–33
Municipal Finance Management Act
of 2003 (South Africa) and, 511,
517
subnational debt markets and, 19–25,
431–45
credit ratings and analyses. See also
creditworthiness
capital risk and, 154
on debt securities, 461
fiscal transparency and, 404
global financial crisis of 2008–09
and, 482
project finance and, 439–41
subnational debt market and,
475–79, 484f
Washington Public Power Supply
System, 361–62
creditworthiness. See also credit ratings
and analyses
credit market deepening and, 522
debt size and, 474
stigma of bankruptcy and, 334
of subnational governments, 242,
242f, 250, 488, 530–31, 579
Washington Public Power Supply
System, 362–63, 371
currency risks, 277–79
D
DAF. See Fiscal Affairs Department
Dartmouth v. Woodward (1819), 548–49
DBSA. See Development Bank of
Southern Africa
595
596
Index
DCED (Department of Community
and Economic Development,
Pennsylvania), 568, 586n69
De Angelis, Michael A., xx–xxi, 311, 559
debt ceilings, 428, 446n17
debt collateral, 271
Debt Consolidation and Relief Facility
(FC, India), 128
debt instruments
composition of, 431–41
demand for, 429–31
subnational. See subnational debt
debt refinancing, 41, 506–7
debt relief, fiscal responsibility
legislations and, 127–29
debt renegotiations, 41–53, 71
debt restructuring
authority over, 5–6
design issues in, 5–8
institutional reform and, 126–33
Municipal Finance Management Act
of 2003 (South Africa), 510–11
national-government-led, 9, 10–13
rural legacy school debt in China,
81–107. See also China, rural
legacy school debt in
subnational debt in Brazil, 33–79.
See also Brazil
Tequila Crisis of 1994–95, 145,
155–58
debt stock, 272, 274–76
debt swaps, 120, 130–33
debt-to-grant financing, 82
debt trajectories, 121, 136
debt write-offs, 82–83, 95–96, 97–102,
127, 128. See also bailouts
decentralization
history and challenges of, 182–85
institutional structures and finances
and, 221, 223–26, 225t, 253–54
subnational governments and, 1, 420,
423
trends in, 25
Decentralization Act of 1982 (France),
221
Decision on Strengthening Rural
Education (Chinese State
Council), 87–88
Decision on the Reform and
Development of Primary
Education (Chinese State
Council), 104–5n5
Decision on the Reform of the
Education System of 1985
(Chinese Central Committee
Party), 84
Decree 28 of 2008 on subnational
governments (Colombia),
198–99, 212
Decrees 77 to 80 of 1987 on
municipalities (Colombia), 182
defaults of municipal bonds. See
Washington Public Power
Supply System
Defferre, Gaston, 254n6
“Defferre Law” of 1982 (France), 223,
254n6
deficits
debt write-offs and, 128
fiscal, 139nn15–16
trends and composition of state,
117–26, 117f, 123f
del Villar, Azul, xxi, 179
Departamento de Asuntos Fiscales. See
Fiscal Affairs Department
Department of Community and
Economic Development
(DCED, Pennsylvania), 568,
586n69, 586n73
Departmento Nacional de Planeación
(National Planning Department,
Colombia), 190, 197
Development Bank of China, 404
Development Bank of Southern Africa
(DBSA), 499b, 504, 521, 522,
524–25, 532–33
Index
Development Bank of the Philippines,
426, 431–32, 440, 446n3,
448n38
development charge financing, 525–27,
533
Dexia, 241
Dholakia, R. K., 139n19
Díaz-Cayeros, Alberto, 146, 157–58
Dillon, John, 544
Dillon’s Rule, 543–44, 545, 549, 561,
567, 580
Direct Administrative Controls model,
391–92
Direction Générale de la Comptabilité
Publique (DGCP, General
Directorate of Public
Accounting), 249–50
Direction Générale des Collectivités
Locales (DGCL, General
Directorate of Subnational
Entities), 249–50
Division of Revenue Act (South Africa),
523, 531
Dodd-Frank Act of 2010 (U.S.), 560
DOF. See Finance Department
Dotation Globale de Fonctionnement
(General Public Service Grant),
229
drug problems, 192
“Dynamics of Debt Accumulation
in India” (Rangarajan &
Srivastava), 132
E
early warning systems
Coe on, 564
fiscal distress and, 246, 247
fiscal monitoring and, 301, 566, 570
implementation of, 249
Municipal Finance Management Act
of 2003 (South Africa), 508–9
reforms and, 299
“economy of debt,” 39
education. See also China, rural legacy
school debt in
compulsory, 81–82, 84–85, 87–102,
387
revenue assignment for, 83–94
Eichengreen, Barry, 43
Electronic Municipal Market Access
system (MSRB), 584nn38–39
Eleventh Amendment (U.S.), 540, 543,
581–82n3
Emergency Social Fund (ESF), 43b,
44–45
Energy Northwest, 360
Energy Reserves Group v. Kansas Power
& Light (1983), 339–40n24
Enhance Reinsurance Co., 339n20
Enterprise Gazette (Hungary),
notification of creditors in, 283
Enterprise Law (Hungary), 297
Enterprise Registry (Hungary), 280
environmental fines, 283
Environmental Inspectorate (Hungary),
283
equalization grants, 470
ESF. See Emergency Social Fund
Eskom, 499b
Estonia, insolvency legislation in, 261
établissements publics (public
establishments), 222, 227
“EU Own Resource Subsidy,” 269
Eurobonds, 460, 472
European Charter of Local
Government, 267, 268
European Union (EU)
cohesion and structural funds and,
287
financial services and, 274
government securities and, 116
grants from, 269
euros, 275–77
executory contracts, 319, 327
“extra limit” operations, 40
597
598
Index
F
FCs. See Finance Commissions
Federal Fiscal Responsibility Law of
2005 (Mexico), 160–61
Federal Fund for Urban Development
(Brazil), 36
Federal Immediate Action Plan (PAI,
Brazil), 43, 43–44b, 44–45, 47
Federal Institute of Geography and
Statistics (IBGE), 74n23,
76n38
federalism. See fiscal federalism
Federal Loan Bonds (Russia), 463
Federal Program for Growth
Acceleration (Brazil), 61, 66
FEIEF. See Fund for the Stabilization
of the Federal Revenue for the
Federal Entities
FIFA (Fédération Internationale de
Football Association) World
Cup. See World Cup
Finance and Public Credit Ministry
(Ministerio de Hacienda y
Crédito Público, Colombia)
adjustment programs, 194–95
bankruptcy procedures, 180
Law 550 of 1999 on bankruptcy
proceedings, 216n31
restructured debt, 200, 208
subnational borrowing, 189, 197–98,
204
Finance Commissions (FCs, India)
on borrowing limits, 135
debt relief, 127–30, 128t
fiscal reforms, 110–11, 125–26, 133
Fiscal Reforms Facility, 140–41n31
on interest payments, 127, 140n29
policy recommendations of, 115,
139n17
on public sector undertakings, 120
revenue sharing, 112, 138n8
role of, 138n3
Finance Department (Philippines), 410,
417, 423–24
Finance Ministry (China)
compulsory education, 91
credit ratings, 404
on fiscal relations, 87
moral hazard, 82
onlending, 383, 390
provincial bonds, 391, 392–94, 398
on teacher qualifications, 105n6
Finance Ministry (India), 110–11
Finance Ministry (Russia), 460, 461,
468, 474
Financial and Fiscal Commission
(South Africa), 498
financial crisis. See global financial
crisis of 2008–09; Tequila Crisis
of 1994–95
Financial Fund for Educational Services
(FUNDEF), 75n31
Financial Investment Fund, 73n5
Financial Planning and Supervision
Commission (Ohio), 569
financing gaps, 402–3, 402t, 407
FINDETER INFIS, 191, 215n14
Fiscal Affairs Department
(Departamento de Asuntos
Fiscales, Colombia)
adjustment programs, 195, 209
insolvency cases, 213
Oil Saving and Stabilization Fund, 205
restructuring agreements, 200–201,
202
subnational borrowing, 198, 208
fiscal consolidation, debt renegotiation
and, 47–51
fiscal federalism
debt restructuring, 33, 35–40
subnational debt management,
146–54, 148t, 150f, 152t
fiscal incentives, 3–4, 8–9
Fiscal Offense Punishment Act of 2004
(China), 393
Index
Fiscal Reforms Facility (India),
140–41n31
Fiscal Responsibility and Budget
Management Act of 2003
(India), 12, 129
Fiscal Responsibility Law of 2000
(FRL, Brazil)
debt renegotiation, 41, 76n44
debt service ratio, 64
enactment of, 35, 71–72, 74n19
fiscal sustainability, 51–53, 52b
fiscal turnaround, 54
personnel expenditures, 60
fiscal responsibility legislations (FRLs)
adoption of, 12
debt relief, 127–29, 141n40
finance reforms, 110–11, 131–32
impact of, 116, 134
state borrowing, 114, 122
targets of, 127
fiscal risks
assessment of, 404
subnational insolvency and, 243–45,
243t, 254, 257n48
Fiscal Stability Act of 2011 (Hungary), 268
Fiscal Transparency and Responsibility
Law of 2003 (Colombia), 197
Fitch Ratings, 242, 243, 477, 491n28
Floor Area Ratios, 386, 528
Floor Space Indexes, 386, 528
Florida, debt default in, 2
Fondo de Ahorro y Estabilización
Petrolera (Oil Saving and
Stabilization Fund, Colombia),
205
Fondo de Aportaciones para los
Servicios de Salud (Fund for
Health Services, Mexico), 148
Fondo de Estabilización de los Ingresos
de las Entidades Federativas. See
Fund for the Stabilization of the
Federal Revenue for the Federal
Entities
Fondo General de Participaciones
(General Participation Fund,
Mexico), 147
Fondo Nacional de Pensiones en
las Entidades Territoriales
(National Pension Fund,
Colombia), 187, 202
Fondo Nacional de Regalías (National
Royalty Fund, Colombia), 186–87
FPE. See State Participation Fund
France, 9, 14, 15–16, 221–59
central government, 230–38
budget rules, 230–35, 231b, 232t,
235f
Prefects, 230–36, 253–54
Public Accountants, 222, 230, 231,
236–38, 253–54, 256n34
Regional Chamber of Accounts
(RCAs), 230–36, 253–54
financial distress, 245–53, 249t
fiscal risks, 243–45, 243t, 254, 257n48
institutional structures and finances,
223–30, 224f
administrative, 223
decentralization, 221, 223–26,
225t, 253–54
intermunicipal cooperation and,
226–27
subnational finances, 227–30,
228–29f
overview, 221–22
subnational debt, 238–42
borrowing framework, 238–40
creditworthiness and, 242, 242f
evolution of, 240–41, 241f
lenders and, 241–42
free-rider problems, 3, 26–27, 103
FRL. See Fiscal Responsibility Law
of 2000; fiscal responsibility
legislations
FUNDEF (Financial Fund for
Educational Services, Brazil),
75n31
599
600
Index
Fund for Basic Education (Mexico),
148
Fund for Health Services (Fondo de
Aportaciones para los Servicios de
Salud, Mexico), 148
Fund for Maintenance and
Development of Basic
Education and Teacher Training
(Brazil), 68, 75n31
Fund for Regional and Municipal
Finance Reform (Russia), 468
Fund for the Stabilization of the Federal
Revenue for the Federal Entities
(Fondo de Estabilización de
los Ingresos de las Entidades
Federativas, Mexico), 161, 162,
163, 175n33
G
Gaillard, Norbert, xxi, 221
Gamboa, Rafael, 146, 157–58
Garson, Sol, xxii, 33
General Code of Territorial Entities
(Code Général des Collectivités
Territoriales, France), 239, 247,
248, 250, 256n30
General Council (France), 224
General Directorate of Public
Accounting (Direction Générale
de la Comptabilité Publique,
DGCP), 249–50
General Directorate of Subnational
Entities (Direction Générale des
Collectivités Locales, DGCL),
249–50
General Fund obligations, 327–28,
345n104
“general incorporation acts” (U.S.), 545,
554, 580–81
General Inspection of Finance
(Inspection Generale Des
Finances, France), 237
Generally Accepted Accounting
Principles, 560, 583n26
General Participation Fund (Fondo
General de Participaciones,
Mexico), 147
General Price Index, Domestic
Availability (IGP-DI), 48, 66, 67,
68–69, 73–74n12
General Public Service Grant (Dotation
Globale de Fonctionnement,
France), 229
General System of Social Security for
Health (Colombia), 204
General Transfer System (Sistema
General de Participaciones,
Colombia), 185–86, 215n12
Germany
government securities
and, 116
subnational bonds and, 540
GFIs. See Government Financial
Institutions
Giambiagi, Fábio, 73n4
Giugale, Marcelo, 146, 153
GJMC (Greater Johannesburg
Metropolitan Council), 499b
Global Competitiveness Report
(World Economic Forum),
172–73n9
global financial crisis of 2008–09
financial distress and, 246–47
impact of, 65–71, 133–37, 271–72,
426, 457, 479–82
subnational debt and, 3, 145,
158–67, 159t, 161f, 164–65f,
479–89
“Golden Rule,” 446n3
good faith requirements, 314–15, 318
Good Governance Index, 430
Governmental Accounting Standards
Board (U.S.), 560
Government Finance Officers
Association, 560
Index
Government Financial Institutions
(GFIs, Philippines)
borrowing from, 430
credit markets and, 431–33, 438–39,
441–42
defaults and, 426
loans of, 438
Local Government Units and, 417,
419, 436–37, 445
supervision by, 428
Goyal, Rajan, 119, 146
grants
capital grants, 60
debt-to-grant financing, 82
equalization, 470
fiscal reform and, 140–41n31
rural legacy school debt and, 83, 98,
101, 103, 105–6nn12–13
subnational finance and, 269–70
Great Depression, 2, 17, 559
Greater Johannesburg Metropolitan
Council (GJMC), 499b
Greater Johannesburg Metropolitan
Municipality (South Africa),
499, 499b
gross state domestic product (GSDP),
123, 124f, 132
GTS. See General Transfer System
Guarantee Act (India), 129
Guarantee Redemption Fund (India),
129
guarantees
fiscal responsibility legislation and, 129
issuance of, 120, 139–40n21
recourse and nonrecourse, 473
subnational governments and, 481–82
“Guidelines for the State Debt Policy in
2012–14” (Russia), 486
H
Harrisburg, Pennsylvania, Chapter 9
bankruptcy in, 329–30
Helios (computer system), 238
Hernández, Fausto, 146, 157–58
Hewitt Associates, 151
holdout problems, 7, 312
home rule concept, 545–46, 546–47t,
554–55, 563–64t
Honadle, Beth W., 564–65,
585n50
Hungary, 9, 14, 16–17, 261–309
bankruptcy procedures in, 180
cramdown power in, 6
debt types and, 284–86
ex-ante regulations, 272–79
debt composition, 274–79, 275t,
276f, 278f
debt market, 272–74
goals of, 279
governments and finance,
267–72
debt collateral and, 271
finance sources, 269–71
global financial crisis of 2008–09,
271–72
structure of, 267–69
implementation, 288–93
database of events, 288, 289–91t,
305n43
insolvency causes, 288, 292–93,
292t
judicial approach to insolvency in, 8
lessons learned, 293–300,
305–6nn44–45
recent changes, 297
strengthening of, 298–300,
301
Municipal Debt Adjustment Law,
context of, 264–67
municipal sector laws, 264, 264b
other issues in, 287
overview, 261–64
procedural steps for, 280–84, 281f,
282–83b
utilities and corporations, 286–87
601
602
Index
I
IBGE (Federal Institute of Geography
and Statistics), 74n23, 76n38
ICMS. See intermunicipal
transportation and
communications services
IFIs. See International Financial
Institutions
IGP-DI. See General Price Index,
Domestic Availability
IMF (International Monetary Fund),
40, 74n18
Incentive Program for the Reduction
of Participation of the States in
Public Sector Banking, 49
Incentive Program for the States’
Restructuring and Fiscal
Adjustment (PAF), 49, 50
incentives, fiscal, 3–4, 8–9
India, 10, 12, 109–44
bank deposit rates, 119, 119–20t
borrowing regime, 111–17, 113–14b,
115–16f
debt restructuring and institutional
reform, 9, 126–33
debt relief and fiscal responsibility
legislations, 127–29, 128t
debt swap and securitization, 130–33
deficit and interest burden, 117–26,
117f, 123–24f, 136
extended debt, 121, 121f
fiscal stress in, 2
global financial crisis of 2008–09,
133–36, 137
interest rate, 132–33, 133f
overdraft facilities, 130–31, 131t
overview, 109–11
state vulnerability, 123, 125t
industrialized products, value-added
taxes on (IPI), 59, 66, 75n28
inflation, 43–44, 44b, 50, 71
infrastructure
for bond markets, 460
development of, 430–31
investments in, 379–80, 382–83,
389, 396
private financing for, 495–537.
See also South Africa
public-private partnerships and, 442
urban, 387, 387t
urbanization and, 1, 496
Infrastructure Finance Corporation
Limited, 521
Inman, Robert P., 28–29n3
insolvency
collective enforcement and, 6–7
Colombia, 179–219. See also
Colombia
France, 221–59. See also France
Hungary, 261–309. See also Hungary
judicial vs. administrative approach
to, 7–8
public and private, 5
subnational debt and, 1–3, 14–19,
189–99, 238–42
United States, 311–51. See also United
States, Chapter 9 bankruptcy in
local government borrowing in,
539–90. See also United States,
local government borrowing in
Insolvency Act of 1936 (South Africa),
511
insolvency tests, 315
Inspection Generale Des Finances
(General Inspection of Finance,
France), 237
institutional reforms, 396–97, 406, 408,
498–505
Inter-Bank Bond Market, 405
interest burdens, 117–26, 124f, 136
interest rates, 40, 120, 132–33, 133f
intermunicipal transportation and
communications services
(ICMS), 49, 54, 59, 65, 66–67,
75n31
Index
Internal Revenue Allotment (IRA,
Philippines)
capital projects, 432
debt service offset, 450n58
finances and revenues, 421–23
Local Government Code, 420
Local Government Units, 429–30,
448n34
payments of, 425, 427–28, 436–37, 439
reliance on, 443
use of funds from, 426
International Financial Institutions
(IFIs), 423, 436, 441, 450n59
International Monetary Fund (IMF),
40, 74n18
IRA. See Internal Revenue Allotment
ISS (service taxes), 54, 61
J
Jain, R., 116
Jamet Report (2010), 230
Japan, subnational bonds and, 540
Jefferson County, Alabama, Chapter 9
bankruptcy in, 3, 328–29, 334
Johannesburg Stock Exchange, 521
joint action agencies, 356
Jókay, Károly (Charles), xxii, 261,
305n43
JPMorgan, 346–47n123
judicial embargoes, 212
judicial writs (precatórios), 42, 57, 73n8
K
Karan, N., 139n19
Kerala Ceiling on Government
Guarantee Act of 2003 (India),
129
Kleine, Robert, 562, 564, 571
Kloha, Philip, 562, 564, 571
Korobow, Adam, 146, 153
Kurlyandskaya, Galina, xxii–xxiii, 455
L
labor unions, 327
Lakshmanan, L., 116
Land Bank of the Philippines, 426,
431–32, 446n3, 448n38, 450n56
land-based financing, 385–86, 389,
400–401, 400f, 411n17, 527–30,
533
land-use exchanges, 529
Las Vegas Monorail Company, 316
“latent” bankruptcies, 288, 292, 298
Latvia, insolvency legislation in, 261
Law 12/1986 on revenue sharing
(Colombia), 183
Law 14/1983 on tax bases (Colombia),
182–83
Law XLIX on corporate insolvency
(Hungary), 263
Law 60/1993 on subnational borrowing
(Colombia), 189
Law 358/1997 on subnational
borrowing (Traffic Light Law,
Colombia), 193–94, 195,
197–99, 204, 210t, 213
Law 550 of 1999 on bankruptcy
proceedings (Colombia), 14–15,
179–81, 190, 191t, 198–201,
200f, 204–14, 206t, 216n31,
216–17n36, 217n40
Law 617/2000 on operational
expenditures (Colombia),
196–98, 199, 204, 208, 209,
210t, 213
Law 715/2001 on transfers (Colombia),
186
Law 756/2002 on royalty rates
(Colombia), 186
Law 795 and 819/2003 on subnational
borrowing (Colombia), 197–98,
199, 213
Law 1176/2007 on transfers
(Colombia), 186
603
604
Index
Law No. 2004-809 of 2004 on
decentralization (France), 224
Law No. 7976 of 1989 on debt
refinancing (Brazil), 41, 73n6
Law No. 8388 of 1991 on debt
renegotiations (Brazil), 41
Law No. 8727 of 1993 on debt
renegotiations (Brazil), 41–42
Law No. 9496 of 1997 on debt
renegotiations (Brazil), 42, 50
Law on Insolvency and Compulsory
Liquidation of 1991 (Hungary),
286
Law on Local Government of 1990
(Hungary), 16, 264, 268, 300
Law on Local Government of 2011
(Hungary), 273
Law on Local Rights and
Responsibilities (France), 240
Law on Local Self-Government of 2003
(Russia), 467–68
legacy debt. See China, rural legacy
school debt in
Leigland, James, xxiii, 353
Leite, Cristiane Kerches da Silva, 74n
Lemmen, J., 116
LGC (Local Government Commission,
U.S.), 566–67
LGUGC. See Local Government Unit
Guarantee Corporation
LGUs. See Local Government Units
limited liability companies (LLCs),
286
Liu, Lili
biographical information, xix–xx
on Brazilian presidency, 51–52
on China’s legacy debt, 81
on China’s subnational market
access, 379
on Colombia’s subnational
insolvency framework, 179
on ex-post insolvency mechanisms,
337
on fiscal responsibility legislations,
132
on France’s subnational insolvency
framework, 221
on “Golden Rule,” 446n3
on South Africa’s infrastructure
financing, 495
on special districts (Philippines), 436
on subnational debt, 1, 25, 70,
146, 417
on U.S. local government borrowing,
539
on Washington Public Power, 353
Llanto, Gilberto M., xxiii, 417
LLCs (limited liability companies), 286
loans
“bridge loans,” 40
in debt adjustment, 284–85
debt structure and, 485
global financial crisis of 2008–09 and,
480–81
maturities on, 522
short-term bank, 457–58
small savings, 114–15
state borrowing, 113–14b
water revenue, 435, 443
Local Government Code of 1991
(Philippines)
credit access, 431
decentralization, 420, 423
goals of, 442
Local Government Units, 21, 417,
424–25, 444
Local Government Commission
(LGC, U.S.), 566–67
Local Government Financial
Emergencies and Accountability
Act of 1979 (Florida), 562
Local Government Unit Guarantee
Corporation (LGUGC,
Philippines)
bonds and, 418, 437–38
creation of, 21
Index
credit ratings and, 439
debt data and, 428
guarantees by, 434
innovations and, 444
ownership of, 449n49
private credit and, 436
Local Government Units (LGUs,
Philippines)
borrowing by, 417–19, 424–25, 424t,
431, 442–43, 449n456
constraints on, 444–45
credit ratings and, 439–41
debt instruments and, 429–31
Financing Framework, 419
private credit and capital markets,
436–39
Public Financial Institutions and,
431–36, 433–34t
spending and revenues of, 420–28,
421t, 422–23f, 427–28t
Locally Own Sources Income
(Philippines), 425
Local Water Utilities Administration
(LWUA, Philippines), 431–33,
446n3
London Club debt restructuring
agreements, 465
Ltd. (companies limited by shares), 286
LWUA (Local Water Utilities
Administration, Philippines),
431–33, 446n3
M
Mackey, Scott R., 562, 565
macroeconomics
debt renegotiation and, 41–53, 71
insolvency and, 213–14
subnational debt markets and, 25–26,
168, 409, 456, 488–89
mandamus orders, 312, 330, 338n11,
342n58
mandato (mandates), 152–53
Manoel, Alvaro, xxiii–xxiv, 33
market access, subnational. See China,
subnational market access in
market-based financing, 130–33
Markets in Financial Investments
Directive of 2004 (EU),
303n25
MDF (Municipal Development Fund,
Philippines), 431–32, 446n3
MDFO. See Municipal Development
Fund Office
Member of the Executive Council
(MEC), 504, 509, 510
Merrill Lynch, 344n96
Mexican Fiscal Federalism Framework,
146
Mexican National Association of State’s
Secretaries of Treasury, 163
Mexico, 2, 10, 12–13, 145–76
debt crisis, 40
debt restructuring, 6, 8, 9, 145,
155–58
fiscal federalism, 146–54, 148t, 150f,
152t, 169–70t
global financial crisis of 2008–09,
145, 158–67, 159t, 161f,
164–65f
lessons in, 166–68
overview, 145–46
states’ local revenues, 149, 171t
Tequila Crisis, 145, 155–58, 156f,
159t, 160, 164, 164f
MFMA. See Municipal Finance
Management Act
MFPC. See Finance and Public Credit
Ministry
Milan, B., 121
minban teaching staff
(nongovernmental employees),
90–91
Mohan, R., 118
Monetary Board (Philippine National
Bank), 434, 439
605
606
Index
Moody’s, 242, 312, 354, 358,
360, 362
Mora, Mônica, xxiv, 33
moral hazards
bailouts and, 158, 267, 301, 371
Chapter 9 bankruptcies and, 315
debt repayment and, 539
debt restructuring and, 110, 127, 336
hard budget constraints and, 26
insolvency and, 294
“power bonds” and, 130
rural legacy school debt and, 82–83,
96–99, 102–3
subnational debt and, 4
Mosqueira, Edgardo, xxiv, 179
Motsoane, Tebogo, xxiv–xxv, 495
MSRB. See Municipal Securities
Rulemaking Board
Multi-layered Control of Expenditures
(Contrôle hiérarchisé de la
dépense), 238
Multiyear Plan (Brazil), 52b
Municipal Act of 1990 (Hungary), 265
municipal bond defaults. See
Washington Public Power
Supply System
Municipal Budget and Reporting
Regulations of 2009 (South
Africa), 531
Municipal Council (South Africa), 495,
498, 505, 508
Municipal Debt Adjustment Law of
1996 (Hungary)
application of, 263, 286, 295
bankers, 302n8
changes in, 297
context for, 264–67
creditors, 296
debt collateral, 271
default, 302n9
enactment of, 16, 22–23, 261
“latent” bankruptcies, 292
long-term borrowing, 299
mayors and, 306n47
municipalities, tasks of, 303n14
policy goals of, 279, 300–301
procedural phases in, 280–84, 281f,
282–83b
public services, 262, 294
Municipal Development Fund (MDF,
Philippines), 431–32, 446n3,
448n38
Municipal Development Fund Office
(MDFO, Philippines), 426, 431,
433, 440
Municipal Finance Management Act of
2003 (MFMA, South Africa)
borrowing, 506–7, 518, 522–23
context of, 501–5
credit markets, 497–98, 511, 517
debt relief and restructuring,
510–11
drafting of, 500
early warning system of, 508–9
financial distress, 507–11
financial monitoring, 511
fiscal adjustment, 509–10
goals of, 496–97
reforms of, 531
regulatory framework of, 505–6
Municipal Finance Management Bill
(South Africa), 502, 503–4
Municipal Financial Emergency
Authority (South Africa), 504
Municipal Financial Recovery Service
(South Africa), 504, 509–10,
535n43
Municipal Fiscal Powers and Functions
Act of 2007 (South Africa), 531
municipalities, definition of, 315–16,
325
Municipalities Financial Recovery Act
(Act 47, Pennsylvania), 562,
566–67, 568, 586n73
Municipalities Participation Fund
(Brazil), 66, 68
Index
Municipal Property Rates Act of 2004
(South Africa), 531
Municipal Securities Rulemaking Board
(MSRB, U.S.), 25, 366, 560, 579,
584nn38–39
Municipal Structures Act of 1998
(South Africa), 498–99, 531
Municipal Systems Act of 2000 (South
Africa), 531
N
National Bank of Economic and Social
Development (Brazil), 61, 66
National Bank of Hungary, 275, 302n5,
304n33
National Constituent Assembly (Brazil),
38
National Council of the Comptroller
(Conseil national de la
comptabilité, France), 240
National Court of Accounts (Cour des
Comptes, France), 223, 232, 234,
244, 250–51, 256n29
National Development and Reform
Commission (China), 404
National Federation of Municipal
Analysts (U.S.), 560
National Fiscal Convention
(Colombia), 151
National Housing Bank (Brazil), 37
National Monetary Council (Brazil),
36, 54, 67
National Pension Fund (Fondo
Nacional de Pensiones en
las Entidades Territoriales,
Colombia), 187, 202
National Planning Department
(Departmento Nacional de
Planeación, Colombia), 190,
197
National Renewal Alliance Party
(Brazil), 37
National Royalty Fund (Fondo Nacional
de Regalías, Colombia), 186–87
National Savings Bank (Hungary),
273–74
National Small Savings Fund (NSSF,
India), 114–15, 129, 139n16
National Tax Services (China), 86
National Treasury (Brazil), 40, 41, 49,
64, 74n21
National Treasury (South Africa)
budgeting practices, 531
development charges, 527
financial monitoring, 511
guarantees, 507
Municipal Financial Recovery
Service, 509
municipal treasury functions, 525
private sector investors, 501
supervisory role of, 508
treasury norms, 502
National Treasury Secretariat (STN,
Brazil), 51, 74–75nn23–24
National Valuation Authority
(Philippines), 448n33
net billing arrangements, 357, 358
net current revenue (RCL), 67, 76n40
New Central Bank Act (Philippines),
426
New Deal, 17
Nine-Year Compulsory Education plan
(China), 91, 92
nonearmarked transfers. See
participaciones
nonprofit associations as public entities,
227
North Carolina, fiscal monitoring in,
561–62, 566, 567–68, 570, 581
Nossa Caixa, 76n33
Notice on Improvement of
Administration of Compulsory
Education in Rural Areas
(Chinese State Council),
104–5n5
607
608
Index
NSSF (National Small Savings Fund,
India), 114–15, 139n16
nuclear power plants, 357–60, 364
O
OCIP (Orange County Investment
Pool), 315–16, 324–25
OECD countries, economic growth in,
134
Off-Track Betting Corporation (OTB),
316
Ohio
Constitution of 1851, 545, 568–69
fiscal monitoring in, 566, 568–70,
586n83
oil revenues, 160
Oil Saving and Stabilization Fund (Fondo
de Ahorro y Estabilización
Petrolera, Colombia), 205
oil shocks, 37
Oliveira, Joao de Carmo, 146
onlending
market access and, 382, 388
subnational governments and, 380,
381, 383
“On the Promissory Note and the
Bill of Exchange” law of 1996
(Russia), 460
OPEC (Organization of the Petroleum
Exporting Countries), 562
Orange County, California,
Chapter 9 bankruptcy in, 3,
324–26, 328, 334–35,
345n103
Orange County Investment Pool
(OCIP), 315–16, 324–25
Organization of the Petroleum Exporting
Countries (OPEC), 562
Osvath, Laszlo, 305n43
OTB (Off-Track Betting Corporation),
316
OTP (bank), 273–74
P
PACER (Public Access to Court
Electronic Records), 322,
344n89
PAF (Incentive Program for the States’
Restructuring and Fiscal
Adjustment), 49
PAI. See Federal Immediate Action Plan
participaciones (nonearmarked
transfers), 147, 152–53, 155, 157,
164–65, 169t, 172n6, 173n14
payroll taxes, 149, 172n7
Pellegrini, Josué A., 76n44
Pennsylvania
fiscal monitoring in, 566–67,
568, 570
Municipalities Financial Recovery
Act (Act 47), 562, 566–67, 568,
586n73
Pennsylvania Intergovernmental
Cooperation Authority (PICA),
567, 568
pension plans, 327, 331, 346n111
People’s Congress Standing Committee
(China), 392
Pessôa, Samuel, 65
Petersen, John E., xxv, 417
PFIs. See Private Financial Institutions
Philippine Dealing & Exchange
Corporation, 439
Philippine National Bank, 432, 448n40
Philippines, 9, 19, 21, 417–54
constraints and opportunities,
444–45
credit market, 431–45
access and innovations, 441–44
credit ratings and project finance,
439–41
private credit and capital markets,
436–39, 438t
Public Financial Institutions, 419,
431–37, 433–34t, 445
Index
debt instruments
composition of, 431–41
demand for, 429–31
finance and borrowing framework,
420–28
spending and revenues, 420–28,
421t, 422–23f, 424t, 427–28t
overview, 417–19
Philippine Valuation Standards, 430
Philippine Veterans Bank, 434, 437–38,
450n53
Philippine Water Revolving Fund
(PWRF), 440, 444, 451n624
Piancastelli, Marcelo, 76n34
PICA (Pennsylvania Intergovernmental
Cooperation Authority), 567,
568
Pineda, Emilio, 173n16
Planning Commission (India), 111, 112
Plan of Debt Adjustment (U.S. Chapter 9),
313, 319–20
bankruptcy courts, 321
debt adjustment, 315
general obligation bonds, 318
municipal authorities, 332
Orange County, California, 325–26,
344n97
rejection of, 336
Vallejo, California, 327–28
Westfall, Pennsylvania, 330
Poland
credit ratings, 478
private capital, 70
Policy Framework for Municipal
Borrowing and Financial
Emergencies (South Africa),
500, 503
pool financing, 523–24
“pork barrel” projects, 430, 442
Poterba, J., 116
“power bonds,” 130
PPPs. See public-private partnerships
Prakash, Anupam, 119, 146
Prasad, Abha, xxv–xxvi, 109, 146
Prefects (France)
ex-ante control by, 15, 238
executive power of, 223
financial distress, 249
monitoring by, 222, 230–36, 253–54
recovery plans, 248
Prichard, Alabama, Chapter 9
bankruptcy in, 331, 335
prisoners’ dilemma model, 3, 28–29n3
Private Financial Institutions (PFIs),
418, 438–39
private sector infrastructure financing.
See South Africa
“procedural debt restrictions,” 549
project finance, 439–41
property taxes, 149, 430
“proprietary property,” 5, 312, 338n12
Provisional Measure No. 1891 of 1999
on debt restructuring (Brazil),
50
PSUs. See public sector undertakings
Public Access to Court Electronic
Records (PACER), 322, 344n89
Public Accountants (France), 15, 222,
230, 231, 236–38, 253–54,
256n34
public authorities, concept of, 354–56
public establishments. See établissements
publics
Public Finance Management Act of
1999 (South Africa), 502, 507,
523, 534n37
Public Financial Institutions
(Philippines), 419, 431–37,
440–41, 442, 444, 445
public officials, 3
public-private partnerships (PPPs)
capital finance and, 497
debt financing and, 430–31
financial distress of, 245–46
fiscal risk and, 254
infrastructure and, 442, 530, 533
609
610
Index
insolvency and, 222, 293, 298–99,
301
intermunicipal cooperation and, 227
land-based financing and, 386, 528
subnational governments and, 221
public sector undertakings (PSUs), 119,
120, 122, 129, 136, 140n21
public securities, debt refinancing and,
42
public services
financial distress, 248
insolvency, 286–87, 294
Municipal Debt Adjustment Law,
262
Public Utility Districts (PUDs), 356–57,
372n5
PWRF. See Philippine Water Revolving
Fund
Q
Qiao, Baoyun, xxvi, 81, 379
quo warranto concept, 548, 556
R
Raiffeisen, 264
Rainy Day Funds (RDFs), 160–61, 162,
163, 164–65, 167
Ram Mohan, T. T., 139n19
“Ramo 28,” 169t
“Ramo 33,” 148, 170t
Rangarajan, C., xxvi, 109
Rao, H., 118
RBI. See Reserve Bank of India
RCAs. See Regional Chamber of
Accounts
RCL (net current revenue), 67, 76n40
RDFs. See Rainy Day Funds
real net revenue (RLR), 42, 46, 48,
50, 73n7
Real Plan (Brazil), 34–35, 43–44, 44b,
47, 50, 51, 71
real value units (URV), 44b
redemocratization, subnational
governments and, 37–40
refinancing of debt. See debt
refinancing
“reform communism,” 264
Regional Chamber of Accounts (RCAs,
France)
decentralization, 223
financial distress, 249
financial risk, 256n32
General Code of Territorial Entities
and, 256n30
monitoring by, 222, 230–36, 253–54
Public Accountants and, 237
recovery plans, 248
structured products, 245
subnational borrowing, 15, 240
Regional Courts of Accounts (France),
256n29
registered debt, 154, 173n17
Regulation 1248 of 2001 on debt
restructuring (Colombia), 180
“Regulation of the Budget
Administration of the 2009
Subnational Government Bond
Funds” (China), 392
Regulatory Framework for Municipal
Borrowing of 1999 (South
Africa), 522
“relationship” banking, 296
remittances, 427
renegotiation of debt. See debt
renegotiations
rentas cedidas (earmarked taxes in
Colombia), 185
Reserve Bank of India (RBI), 111, 112,
118–19, 130–31, 135, 139n12,
140n21
restructuring of debt. See debt
restructuring
revenue assignments, 83–94, 85t,
89t, 94t
Revilla, Ernesto, xxvii, 145, 146
Index
Rezende, Fernando, 73n4
Rigolon, Francisco, 73n4
Riordan, Richard, 347–48n144
risks, fiscal. See fiscal risks
RLR. See real net revenue
Road Law of 1998 (China), 383, 410n8
Romania
insolvency legislation in, 261
private capital in, 70
Royal Bank of Scotland, 241–42
royalties, 185–86
Rueben, K., 116
rural areas, financing schools in,
81–107. See also China, rural
legacy school debt in
Russian Federation, 9, 19, 21–22,
455–93
bond market (1991–2000), 458–65
as bond issuer, 76–77n45
crisis and recovery (1998–2000),
463–65, 464f
regional and municipal, 459–63,
462f
global financial crisis of 2008–09,
479–89
deepening markets, 487–88
fiscal positions and debt activities,
485–86
impact of, 479–82
public finance and debt structure,
483–85
overview, 455–58
reform, regulation, and development
(2000–08), 465–79
Budget Code, 471–75, 472t
credit ratings, 475–79, 476–77t,
484f
debt load and, 476, 477f
public finance reform, 466–69
regional bond issue, 478, 478f
results assessment, 469–71
Russian Government Treasury
Bills, 463
S
salaries of teachers, 91–94
Sarkozy, Nicolas, 253
savings rates, 405
Sberbank, 488
Schmid, Juan Pedro, xxvii, 179
school debt. See China, rural legacy
school debt in
Schuknecht, L., 116
Schwartz, G., 141n40, 216n32
SEC. See Securities and Exchange
Commission, U.S.
secondary market disclosure problems,
367–68
Securities Act of 1933 (U.S.), 366, 558
Securities and Exchange Act of 1934
(U.S.), 366, 558, 584n38
Securities and Exchange Commission
(Philippines), 425–26
Securities and Exchange Commission,
U.S. (SEC), 354, 366–68, 370,
583n26
securitization, debt swaps and, 130–33
SELIC. See Special Settlement and
Custody System
SEMs. See public-private partnerships
Senate Resolution No. 93 on credit
operations (Brazil), 40
Serra, José, 72n2
service taxes (ISS), 54, 61
sewer systems, 328–29
Sistema General de Participaciones. See
General Transfer System
situado fiscal, 182
small savings loans, 114–15
SNGs. See subnational governments
snowballing products, 244
SOC. See Superintendency of
Corporations
Société Générale (bank), 241
sociétés d’économie mixte locales. See
public-private partnerships
611
612
Index
SOEs. See state-owned enterprises
solvency. See also insolvency
defined, 138n10
state debt in India and, 109–44. See
also India
South Africa, 9, 19, 22–23, 495–537
credit markets, 511–19, 532–33
metropolitan borrowing, 512–16,
513–17f, 532
municipal borrowing, 512, 512f
secondary cities borrowing,
516–19, 518t, 532
historical context and institutional
reforms, 498–505
Constitution of 1996, 498–500,
499b
Municipal Finance Management
Act, 501–5
white paper on local government,
500–501
insolvency rules, 7
overview, 495–98
private capital in, 70
private financing leverage, 519–31, 520f
access to, 523–25
credit market deepening, 520–23
creditworthiness, 530–31
Development Bank of Southern
Africa and, 524–25, 532–33
development charges, 525–27,
533
land-based financing, 527–30, 533
pool finance, 523–24
public-private partnerships, 530,
533
treasury function capacity, 525
regulatory framework for borrowing,
505–11, 532
borrowing provisions, 506–7
debt relief and restructuring, 8,
510–11
early warning system, 508–9
financial adjustment, 509–10
financial distress provisions,
507–11
financial monitoring, 511
sovereign debt markets, 388
Soviet Union, federal structure of, 458.
See also Russian Federation
special districts (Philippines), 435–36
special districts (U.S.), 557
special fund doctrine, 440, 451n69,
452n74
Special Purpose Districts (U.S.),
321–22, 339n15
special purpose vehicles, 114b, 328, 329,
448n30
special revenue bonds, 319
“special revenues,” 342n59
Special Settlement and Custody System
(SELIC), 44, 45f, 46, 69, 73n9
speculation in borrowing, 250
Standard & Poor’s, 242, 254n1, 358, 360,
362, 404
Standing Technical Committee (Reserve
Bank of India), 139n12
State Administration of Taxation
(China), 86
State Audit Office (Hungary), 265, 276,
288
State Auditor (Ohio), 569–70
State Budget Act (Hungary), 300
State Council (China)
credit risk analysis and, 404
institutional reforms and, 396
onlending and, 383, 410n9
provincial bonds and, 381–82,
390, 394
Urban Development and Investment
Corporation and, 384
State Council Regulations on the
Implementation of the Tax
Sharing System (China),
86–87
Statement of Income and Expenditure,
428
Index
“State of Local Government Finances
and Financial Management”
(South African Treasury), 511
state-owned enterprises (SOEs), 48–49,
53, 389
State Participation Fund (FPE), 49, 59,
66–67
state sovereignty, 364, 366, 539–40,
545–46, 559
State Treasurer (Tresorier-Payeur
General), 237
stays of obligations. See automatic stays
“step-in-provisions,” 435, 443
stigma effects of bankruptcy, 7, 329,
332, 334
STN (National Treasury Secretariat,
Brazil), 51, 74–75nn23–24
Story, Joseph, 548–49
stress tests, 252
structured products, 244–47, 250, 251,
253, 257n53
Study of State Budgets (Reserve Bank of
India), 118–19
subnational debt
demand for, 399–403
framework for, 193–99, 194–96t,
221–22, 238–40
government markets and, 417–54.
See also Philippines
insolvency and, 1–3, 189–99, 238–42
management of, 145–76. See also
Mexico
public finances and, 455–93. See also
Russian Federation
trends in, 189–93, 190f, 191–92t
U.S. state systems for, 539–90.
See also United States, local
government borrowing in
subnational governments (SNGs)
bond markets, 460, 486
contingent fiscal risks, 243–45
creditworthiness of, 242, 242f, 250,
488, 530–31, 579
debt crisis, 33–35, 70–71
debt portfolio of, 487–88, 489–90
debt renegotiation, 41–53, 71
decentralization, 1, 420, 423
expenditure assignments of, 87–88
finances of, 269–72
financial distress, 41–42, 245–53
fiscal incentives and, 3–4, 8–9
fiscal performance, 53–65
fiscal sustainability, 68–70
guarantees, 481–82
insolvency, 5, 179–81, 199–205,
211–14, 221–22
intermunicipal cooperation, 226–27
market access, 390–97
onlending, 380, 381
redemocratization, 37–40
revenue sources of, 470
rural legacy school debt, 82–83,
94–103
structures of, 181–88, 223–30,
267–69
subnational debt, 189–99, 238–42,
382–90, 455–56
taxes, 182–83, 187–88, 187–88t, 229,
255n23, 270, 403
transfers to, 4, 185–86, 189, 198–99,
229
transparency of, 197–98, 251,
252–53
“Suggestions on Improving
Subprovincial Fiscal Relations”
(Chinese Finance Ministry), 87
“sunshine,” 298, 315
Superintendency of Banks (Colombia),
195, 202–3
Superintendency of Corporations
(Superintendencia de
Sociadades, Colombia), 14–15,
180–81, 199–200, 202–3, 213,
214n3, 215n23
Swiss francs, 275–77
Szepesi, Gábor, 305n40
613
614
Index
T
Tagaytay City Convention Center
(Philippines), 437
take-or-pay contracts, 358–59, 361
Tax Code (Russia), 22, 456, 466, 489
taxes
finance reforms, 111, 470
good faith requirements, 318
municipal collection of, 38, 38f
payroll, 149, 172n7
property, 149, 430
service tax (ISS), 54, 61
subnational governments, 182–83,
187–88, 187–88t, 229, 255n23,
270, 403, 483
tax increment financing, 443–44, 526
value-added taxes (VAT), 34, 49, 59,
66, 75n28, 86, 247, 255n23
Tax-for-Fee Reform (China), 82, 88, 90,
92, 104n4
Tax Sharing System (TSS, China),
85–86, 382
teachers’ salaries, 91–94
Tenth Amendment (U.S.), 313, 316,
336, 340n30
Tequila Crisis of 1994–95, 8, 13, 145,
155–58, 156f, 159t, 160, 164, 164f
Ter-Minassian, Teresa, 146
Tian, Xiaowei, xxvii–xxviii, 311, 539, 559
tolerable debt, 118, 139n19
Traffic Light Law. See Law 358/1997 on
subnational borrowing
transfers, financial
aportaciones (earmarked transfers),
148, 170t
decentralization and, 182–86, 184f
global financial crisis of 2008–09
and, 480
Local Government Units and, 418
participaciones (nonearmarked
transfers), 147, 152–53, 155, 157,
164–65, 169t, 172n6, 173n14
public finance reform and, 468
subnational governments and, 4,
185–86, 189, 198–99, 229
transparency
credit ratings and, 404
in insolvency proceedings, 298, 301
in land-based financing, 529
municipal bonds and, 369, 371
of subnational debt market, 153–54,
475, 476–77t
subnational governments and,
197–98, 251, 252–53
treasury management skills, 522, 525
Treisman, Daniel, 403
Tresorier-Payeur General (State
Treasurer), 237
trustees, 280, 282–84, 282b, 294
trust funds, 435, 449n44
TSS. See Tax Sharing System
U
UBS (bank), 241
UDC (Urban Development
Corporation), 373n15
UDICs. See Urban Development and
Investment Corporations
underwriters and underwriting, 363,
366–67, 394
Unicbank, 264
United States. See also United States,
Chapter 9 bankruptcy in;
United States, local government
borrowing in
bankruptcy procedures in, 180
Constitution. See Eleventh
Amendment; Tenth
Amendment
cramdown power in, 6
development rights in, 528
“proprietary property” in, 5
public services in, 248
revenue bonds in, 443
Index
special purpose vehicles in, 448n30
states’ incomes in, 174n26
subnational bond market in, 238,
410n13, 411n28
subnational debt in, 447n267
tax incremental financing in, 526
United States, Chapter 9 bankruptcy in,
9, 14, 17–18, 311–51
automatic stay and, 318–19,
342n58
bankruptcy courts and, 320–21
debt restructuring and, 559
eligibility under, 314–18
insolvency requirement, 318
municipality, definition of, 315–16,
325
state authorization, 316–17, 317t
impact of, 332–35
on financially stressed
municipalities, 334–35
pros and cons, 333–34
overview, 311–21
Plan of Debt Adjustment, 319–20,
336
approval of, 319–20
executory contracts, 319
stigma and, 7
use of, 321–31, 542, 559
selected cases, 323–31, 334–35
statistics on, 321–23, 322–23f
United States, local government
borrowing in, 9, 19, 23–25,
539–90
conceptual and constitutional issues,
543–48, 546–47t
empirical results on, 570–77, 572t,
574–76t
historical context, 548–61
from colonial period to late 19th
century, 548–55, 550–53t
government debt, 555, 556t
private and public institutions,
555–61, 579
overview, 539–43
state interventions and monitoring,
561–70, 563–64t, 578–79,
581
voter discipline and, 577–81
Unit Investment Trusts, 367
urban areas, infrastructure in, 387, 387t.
See also urbanization
Urban Development and Investment
Corporations (UDICs)
bond issuance of, 394, 408–9, 410n11,
412n43
borrowing of, 405
credit ratings of, 404
debt of, 19–20, 388, 390, 396,
408, 413n53
institutional reforms and, 396–97
land-based financing and, 386, 389
onlending and, 380, 383–85
Urban Development Corporation
(UDC), 373n15
urbanization
deepening of, 33–34, 71
infrastructure and, 1, 496
subnational debt and, 399–400, 399f,
407–8
trends in, 25
URV (real value units), 44b
utilities, 286–87, 321–22, 447n24. See
also public services
V
Vallejo, California, Chapter 9
bankruptcy in, 326–28, 335,
345n104, 346n109
valuation standards, 430, 448n33
value-added taxes (VAT), 34, 49, 59, 66,
75n28, 86, 247, 255n23
Von Hagen, J., 116
voter discipline, 577–81
voter sovereignty, 544, 559
VTB (bank), 488
615
616
Index
W
Waibel, Michael, xxviii, 146, 221, 337,
446n3
Wallis, John Joseph, xxviii, 539
Washington Public Power Supply System
(WPPSS), 3, 18–19, 353–75
bailout, failure of, 363–65, 371
contributing factors to default, 360–63
from creation to default, 354–60
background, 356–60
public authority concept, 354–56
lessons learned, 368–71
overview, 353–54
regulatory reforms and, 365–68
water revenue loans, 435, 443
ways and means advances (WMA),
114b
Webb, Steven B., xxix, 51–52, 132, 146,
153, 179
Weissert, Carol, 562, 564, 571
Westfall, Pennsylvania, Chapter 9
bankruptcy in, 330–31, 335
West Virginia, revenue bonds in,
583n28
“White Paper on Local Government”
(South Africa Government), 22,
496, 498, 500–501, 531
WMA (ways and means advances),
114b
Wolswijk, G., 116
World Bank
debt restructuring and, 74n17
on Local Government Units
(Philippines), 447n21
World Cup (FIFA), 23, 497, 512–13,
514
WPPSS. See Washington Public Power
Supply System
ECO-AUDIT
Environmental Benefits Statement
The World Bank is committed to preserving
endangered forests and natural resources.
The Office of the Publisher has chosen to
print Until Debt Do Us Part on recycled
paper with 50 percent postconsumer fiber in
accordance with the recommended standards
for paper usage set by the Green Press Initiative, a nonprofit program supporting publishers in using fiber that is not sourced from
endangered forests. For more information,
visit www.greenpressinitiative.org.
Saved:
• 27 trees
• 12 million BTUs of
total energy
• 2,327 lbs CO2 equiv. of
net greenhouse gases
• 12,621 gallons of waste
water
• 845 lbs of solid
waste