9781589060470

OCCASIONAL PAPER
Malaysia:
From Crisis to Recovery
Kanitta Meesook, Il Houng Lee, Olin Liu,Yougesh Khatri,

Natalia Tamirisa, Michael Moore, and Mark H. Krysl
INTERNATIONAL MONETARY FUND

Washington DC
2001
©International Monetary Fund. Not for Redistribution

© 2001 International Monetary Fund
Production: IMF Graphics Section

Figures: Theodore F. Peters, Jr.
Typesetting: Alicia Etchebarne-Bourdin
Cataloging-in-Publication Data
Malaysia: from crisis to recovery / Kanitta Meesook . . . [et al.]—Washington,

D.C.: International Monetary Fund, 2001.
p. cm..—(Occasional paper, ISSN 0251-6365; no. 207)
Includes bibliographical references.
ISBN 1-58906-047-4
1. Malaysia—Economic policy. 2. Malaysia—Economic conditions.
3. Fiscal policy—Malaysia. 4. Finance—Malaysia. I. Kanitta Meesook.
II. International Monetary Fund. III. Occasional paper (International Mone-
tary Fund); no. 207
HC445.5.M36 2001
Price: US$20.00
(US$17.50 to full-time faculty members and
students at universities and colleges)
Please send orders to:

International Monetary Fund, Publication Services
700 19th Street, N.W., Washington, D.C. 20431, U.S.A.
Tel.: (202) 623-7430 Telefax: (202) 623-7201
E-mail: publications@imf.org
Internet: http://www.imf.org
recycled paper

Contents
Preface VII
I Overview 1
Olin Liu
II Comparative Review of Policies and Performance, 1997-2000 3

Kanitta Meesook
Initial Conditions and Economic Structure 3
Policy Responses and Performance 7
Time Frame of Malaysia's Response to the Crisis 10
Appendix. Graphical Overview of Indicators During Precrisis,
Crisis, and Recovery 17
III Potential Output and Inflation 28
Il Houng Lee and Yougesh Khatri
Introduction 28
Measuring Potential Output 28
Estimating the Output Gap 29
Determinants of Inflation 30
Appendix. Approaches to Measuring Potential Output 33
References 38
IV Challenges to Fiscal Management 39
Olin Liu
Introduction and Recent Developments 39
Fiscal Policy Rules and Recent Experience 42
Enhancing Fiscal Policy Responsiveness to Business Cycles 42
Conclusions 45
Appendix I. Off-Budget and Quasi-Fiscal Operations 45
Appendix II. Fiscal Measurement: Methodology Note 47
References 48
V Capital Controls in Response to the Asian Crisis 50

Natalia Tamirisa
Background 50
Design and Enforcement of Capital Controls 54
Capital Controls and Economic Performance 55
Policy Considerations 58
Appendix. A Preliminary Empirical Analysis of Malaysia's
Capital Controls 66
References 69

VI Financial Sector Issues 71
Mark H. Krysl and Michael Moore
Background 71
Recent Financial Sector Performance 72
Financial Sector Restructuring: Comparative Perspective 74
Toward a More Resilient Banking Sector 77
VII Corporate Performance and Reform 83

Yougesh Khatri
Key Features of the Corporate Sector 83
Corporate Sector and Financial Crises 87
Corporate Performance 89
Progress with Corporate Reforms 96
Summary and Conclusions 97
References 98
Boxes
4.1. Fiscal Policy in Malaysia, 1997-2000 41
4.2. Fiscal Policy Responses to the Business Cycle 44
5.1. Singapore: Reconciling Policy on the Internationalization
of the Singapore Dollar and Financial Development
Objectives 62
5.2. Korea: Gradual Liberalization with a Focus on Control of
Won Lending to Nonresidents 63
5.3. Thailand: Relatively Liberal Policy in Normal Times,
Temporary Capital Controls to Limit Currency Speculation 64
5.4. Japan: Toward the Internationalization of the Yen 65
6.1. Danaharta: Asset Management and Recovery 73
6.2. Danamodal: Bank Recapitalization 77
6.3. Corporate Debt Restructuring Committee 77
7.1. The Insider and Outsider Systems of Corporate Governance 85
Figures
2.1. Exchange Rate Developments 3
2.2. Bank Lending 4
2.3. Domestic Investment in Selected Asian Countries 6
2.4. Investment and Real GDP Growth 6
2.5. Market Capitalization in Selected Asian Countries 7
2.6. Private Sector Credit in Selected Asian Countries 7
2.7. Cumulative and Net Portfolio Flows 9
3.1. Comparing Estimates of Output Gap (Adj. SP) and Without (SP)
the Interpolated Point in 1998 30
3.2. Inflation, 1992-2001 33
3.3. Inflation, 1995-2001 33
4.1. Fiscal Indicators 39
4.2. Fiscal Budget and Actual Balance 40
5.1. The Evolution of Capital Controls 54
5.2. Sovereign Bond Spreads of Selected Asian Countries 55
5.3. Net Portfolio Flows 56
5.4. Volume of Interbank Spot and Swap Transactions in Kuala
Lumpur Foreign Exchange Market 58
5.5. Selected Asian Countries: Stock Market Composite Indices 58
5.6. Average Daily Trading Volume at the KLOFFE and the COMMEX 59

5.7. Capital Controls 66
7.1. Selected Asian Countries: Capital Market Overview 84
7.2. Malaysian Corporate Sector: Total Assets and Expenditures
Relative to Total Revenue 94
7.3. Indices of Efficiency, Exogenous Change and TFP 95
7.4. Malaysian Corporate Sector: Performance Indicators 95
Appendix Figures
A.2.1. Selected Asian Countries: Precrisis Macroeconomic Indicators 19
A.2.2. Selected Asian Countries: Precrisis Financial Indicators 20
A.2.3. Selected Asian Countries: Economic Indicators During the
Crisis Period 21
A.2.4. Selected Asian Countries: Economic Developments 22
A.2.5. Selected Asian Countries: Financial Market Indicators
During the Crisis Period 23
A.2.6. Selected Asian Countries: Economic Performance During the
Recovery Stage 24
A.2.7. Selected Asian Countries: Financial Indicators During the
Recovery Stage 25
A.2.8. Selected Asian Countries: Fiscal Indicators 26
A.2.9. Selected Asian Countries: Monetary Indicators 27
Tables
2.1. Asian Crisis Countries: Selected Structural Indicators, 1996-2000 5
2.2. Major Economic Policy Measures During Emergence of
Crisis and Early Stage of Contagion 11
2.3. Major Economic Policy Measures, Initial Policy Responses,
and Continued Turmoil 12
2.4. Major Economic Policy Measures, Exchange and Capital
Controls, and Aftermath 14
3.1. Estimates of Output Gap, 1996-2001 31
3.2. F and SC Statistics for Sequential Reduction from
VAR(4) to VAR(1) 32
3.3. Cointegration Analysis of Price Data, 1991Q2 to 2000Q1 32
3.4. Result of Inflation Estimation by OLS, 1992Q2-1999Q4 33
4.1. Quasi-Fiscal Contingent Liabilities 47
5.1. Key Changes in Capital Account Regulations 51
5.2. Sovereign Credit Ratings 55
5.3. Net Foreign Direct Investment in Selected Asian Countries 57
5.4. A Summary of Investment Regulations for Entry into and Exit
from Selected Markets, end-1998 60
5.5. Taxation of Capital Gains on Portfolio Investment in Selected
Emerging Markets, 1999 60
5.6. A Summary of Controls on the International Use of Domestic
Currency, end-1998 61
5.7. Data 67
5.8. Unit Root Tests 67
5.9. A Cointegration Analysis 68
5.10. A Conditional Error Correction Model of Portfolio Investment 68
6.1. The Banking System (Depository Financial Institutions) 72
6.2. The Trend in Nonperforming Loans 73
6.3. Asset Quality Indicators 74
6.4. Financial Sector Restructuring in Malaysia, Korea, and Thailand 75
6.5. Selected Financial Indicators of the Asian Crisis Countries 76
6.6. Changes in Prudential Standards in Malaysia, Korea, and Thailand 78

6.7. Proposed Banking Groups 79
6.8. Steps to Improve Prudential Supervision and Regulation in
Malaysia, Korea, and Thailand 81
7.1. A Comparison of Legal Protection, Accounting Standards,
Institutions, and Enforcement in Various Countries 86
7.2. Financial Ratios for Listed Nonfinancial Companies in
Malaysia 90
7.3. Operational Performance of Publicly Traded Corporations and
Share of Distressed Corporations in Selected Asian Countries 91
7.4. Debt/Equity Ratios in Selected Economies 91
7.5. A Comparison of Stock Market Returns and Risk in
Selected Economies 92
The following symbols have been used throughout this paper:

. . . to indicate that data are not available;
— to indicate that the figure is zero or less than half the final digit shown, or that the item
does not exist;
between years or months (e.g., 1998-99 or January-June) to indicate the years or
months covered, including the beginning and ending years or months;
/ between years (e.g., 1998/99) to indicate a fiscal (financial) year.
"Billion" means a thousand million.
Minor discrepancies between constituent figures and totals are due to rounding.
The term "country," as used in this paper, does not in all cases refer to a territorial entity that
is a state as understood by international law and practice; the term also covers some territorial
entities that are not states, but for which statistical data are maintained and provided interna-
tionally on a separate and independent basis.

Preface
This occasional paper draws on staff background studies prepared for the 2000 and
2001 consultations between the International Monetary Fund and Malaysia. The
paper takes stock of economic developments since the crisis and investigates various
aspects of economic performance, policy, and reform that contributed to the strong
recovery. The individual studies include the estimation of potential output and the im-
plications for inflation; fiscal policy and key challenges for fiscal management; capi-
tal controls, their effects, and medium-term policy considerations; and performance
and reform in the financial and corporate sectors.
The paper is the product of a team effort led by Kanitta Meesook. The team of au-
thors comprised Il Houng Lee, Olin Liu, Yougesh Khatri, Natalia Tamirisa, Michael
Moore, and Mark Krysl. Also central to the production of this paper was the research
assistance of Janice Lee, and the coordination and assistance of Margaret Tan and
Lisa Vassou.
The authors would like to thank the Malaysian authorities for their excellent coop-
eration and support during policy discussions and technical meetings, and for the ex-
tensive provision of data and background materials. The authors would also like to
thank—without implication—Kalpana Kochhar for reviewing drafts of the entire
paper at an early stage, Mark O'Brien for reviewing Section VI, and Jenifer Piesse for
contributing to the empirical work in Section VII. The authors are also indebted to the
External Relations Department and to Gail Berre who edited the paper and coordi-
nated its production and publication.
The opinions expressed are solely those of the authors and do not necessarily re-
flect the views of the International Monetary Fund, the Executive Directors, or the
Malaysian authorities.
Except where otherwise indicated, the paper reflects information available through
end-2000.

This page intentionally left blank

I Overview
Olin Liu
A s an emerging market economy, Malaysia is

clearly a success story. During the past three
decades, the Malaysian government has imple-
robust. However, there were also signs of stress as
exports decelerated and a large current account
deficit developed in the context of a gradual appreci-
mented a number of medium- to long-term develop- ation of the effective exchange rate. While the in-
ment plans, starting with the 20-year New Economic vestment-led growth strategy was successful in rais-
Policy—a development plan that strived for greater ing output and income, investment quality had
economic well-being for the ethnic Malays, or bu- deteriorated. This eventually led to major balance
miputras—and followed by the National Develop- sheet weaknesses in the banking and corporate sec-
ment Policy in the early 1990s. More recently, the tors, exposing the economy to the contagion of the
Third Outline Perspective Plan provides the general Asian crisis. Similar to its neighbors, Malaysia went
thrust of Malaysia's development strategy for the pe- through a currency crisis and a banking crisis, but its
riod 2001-10. low level of external debt spared it from an external
Malaysia's development plans were implemented debt crisis.
effectively. From 1970 to the mid-1990s, the coun- Malaysia's economic vulnerabilities stepped up
try's investment ratio was among the highest in the significantly from early 1997 through the period fol-
region, resulting in a dramatic shift in the structure lowing the onset of the crisis in mid-1997, as market
of the economy from agriculture and mining to a confidence increasingly diminished along with the
growing reliance on manufacturing. Liberalization rest of the region. Large portfolio outflows took
measures were introduced across the board that place, and equity and property values declined sub-
helped improve competitiveness and productivity. stantially. The ringgit came under tremendous pres-
Much of the investment went into electronics and sure. As currency traders took speculative positions
other export-oriented industries, while a large por- in the offshore ringgit market in anticipation of a
tion also went into nontradable sectors including large devaluation, the offshore ringgit interest rates
capital-intensive infrastructure and the real estate increased markedly relative to domestic rates. This
sector. heightened upward pressure on domestic interest
The development plans were initially financed by rates, intensified outflows of ringgit funds, and exac-
public funds. As a result, by the early 1980s, growth erbated banks' liquidity problems and overall finan-
was accompanied by increased budget deficits and cial distress. The Malaysian corporate sector experi-
an unsustainable level of public debt. The authorities enced significant loss of wealth as a result of sharp
took measures to reduce the government deficit, and falls in the value of real estate and equities used as
successfully restored fiscal prudence. Structural ad- bank collateral. Corporate incomes and cash flows
justments were also undertaken, including an open also declined, leaving some corporations unable to
trade and payments system that helped expand service their debt.
rapidly Malaysia's export base. Sustained high The initial response of the authorities was to hike
growth during the 1980s brought significant im- interest rates and tighten fiscal policy in an attempt
provements in living standards and social cohesion. to anchor market confidence in the financial system.
Economic diversification, coupled with deregulation In early 1998, fiscal policy was revised to a more ex-
and liberalization of the financial system, also pansionary stance. This policy mix proved to be in-
helped transform the country into a middle-income sufficient to correct external imbalances and bring
emerging market by the end of that decade. about the needed economic adjustment. The conta-
Malaysia's strong economic performance contin- gion effects of the crisis and the associated eco-
ued during the 1990s prior to the crisis. Real output nomic contraction were far worse than anticipated.
growth averaged81/2percent a year; unemployment Domestic imbalances quickly emerged as growth
was below 3 percent; prices and the exchange rate rates slowed and then turned sharply negative in
remained stable; and international reserves were early 1998. Market confidence faltered amid adverse

regional developments and uncertainties. Anticipa- appreciation of the ringgit, particularly during late
tion of further devaluation of the ringgit heightened. March and early April 2001, leading in turn to short-
By the summer of 1998, the stock market had fallen term capital outflows and reserve losses. These de-
to its lowest level in recent history. velopments, at a time when the economy was al-
In September 1998, the Malaysian authorities ready being hit hard by the global slowdown, has
launched a policy package designed to insulate mon- adversely affected market confidence. Nevertheless,
etary policy from external volatility. Measures in- Malaysia's external vulnerability is relatively well
cluded an exchange rate pegged to the U.S. dollar contained: the current account continues to maintain
and selected exchange and capital controls, comple- a large surplus; short-term external debt is low; and
mented by a fiscal stimulus package that stepped up reserves have remained adequate. Progress in finan-
capital spending. These measures permitted the sub- cial sector restructuring has also improved the ca-
sequent lowering of interest rates. The authorities pacity of banks to manage risks.
also pursued fundamental reforms in the financial Looking ahead, the issue is how Malaysia can bet-
and corporate sectors, including a bank consolida- ter protect itself from future shocks and avoid an-
tion program and an upgrading of prudential regula- other crisis while it seeks to regain its position as
tion and supervision in line with international best one of the fastest growing economies in the world.
practices. To these ends, its strategy should include continued
Malaysia's recovery in 1999-2000 was among the structural reforms to achieve healthy balance sheets
strongest of the Asian crisis economies, led by buoy- of the banking and corporate sectors; further deregu-
ant world demand for electronics and supported by lation to promote competition and efficiency; and
accommodating macroeconomic policies. The exter- consistent macroeconomic policies to maintain fi-
nal current account turned into large surpluses, al- nancial stability and sustainable fiscal and external
lowing a buildup of international reserves. Unem- positions.
ployment declined, and inflation remained low. The The sections that follow review policy issues and
strong growth and a gradual easing of capital controls aspects of economic management that have been as-
helped improve investor confidence. The recovery sociated with Malaysia's progress from a major cri-
was also accompanied by reduced vulnerability sis to a strong recovery, and their implications for
of the financial system. Although operational restruc- the future. Section II presents a comparative review
turing of the corporate sector has been somewhat of the country's policies and performance during
slow, much progress was achieved with corporate 1997-2000; Section III describes the study and esti-
debt restructuring. mation of potential output and inflation dynamics;
Since the latter part of 2000, however, downside Section IV reviews fiscal policy management; Sec-
risks for Malaysia have increased. Heavy depen- tion V discusses the capital controls introduced in
dence on electronic exports made Malaysia highly September 1998 and their impact; Section VI sum-
sensitive to the global slowdown in information marizes various aspects of financial sector restruc-
technology. Sharp depreciations of the yen and other turing; and Section VII reviews developments in
regional currencies have resulted in a large effective corporate sector reforms.

II Comparative Review of Policies and
Performance, 1997-2000
Kanitta Meesook
M alaysia has received much attention since Sep-

tember 1998 when, in response to a deteriorat-
ing economic situation emanating from the Asian
couraged the use of hedging instruments. From the
early 1990s, this allowed large current account
deficits—brought about in part by the appreciation
crisis, it introduced capital and exchange controls. of the real effective exchange rate (Figure 2.1)—to
Also, Prime Minister Mahathir Mohamad rebuked be sustained by short-term capital inflows, includ-
policy advice from the IMF that Malaysia had fol- ing through the stock market. Rapid credit growth
lowed up to then.1 Initially there was concern that led to excessive investment and resource misallo-
these controls might be used to avoid needed policy cation, fueling asset price bubbles in property and
adjustment, and investors and market analysts re- stock markets. As the financial system and capital
acted negatively. Market assessment turned more markets were liberalized, the important aspects of
positive, however, as it became clear that Malaysia's supervision and regulation remained weak, allow-
macroeconomic policies were not out of line, that ing extensive connected lending and obscuring the
the undervalued pegged exchange rate was con- true scale of financial and corporate problems. Fur-
tributing to the rapid recovery of exports and output, thermore, an active—but largely unregulated—off-
and that financial sector reforms were being vigor- shore market for the ringgit exposed the currency
ously pursued. market to information asymmetries and excessive
This section reviews the developments from the risk taking by private investors. All these elements
emergence of the Asian crisis to end-2000, compar- contributed to Malaysia's susceptibility to conta-
ing Malaysia's initial conditions, policy responses, gion by the crisis.
and recent performance with those of the other crisis
countries.
Figure 2 . 1 . Exchange Rate Developments

Initial Conditions and Economic
Structure
3.0 106
Malaysia's economic structure and performance Real effective exchange rate
were relatively strong prior to the crisis (see Appen- (right scale, 1995 = 100)
2.9
dix Figure A.2.1). Real GDP grew rapidly, while in-
flation was low. At the same time, fiscal policy was
prudent, domestic savings were high, international 2.8
reserves were robust, and external debt was well
managed. The legal and regulatory frameworks were 2.7
also comparatively well developed.
Underlying these positive features, however, 2.6
were signs of overheating and structural vulnerabil- Malaysian ringgit/U.S. dollar
ity, broadly similar to the other crisis countries 2.5 (left scale)
(Appendix Figure A.2.2). A predictably stable ex-
change rate encouraged foreign borrowing and dis- 2.4 1991 93 94 95 96 97
1Deputy Prime Minister and Minister of Finance, Anwar Sources: IMF Information Notice System and Asia and Pacific
Ibrahim, was Malaysia's key contact with the IMF at that time. Department database.
He was removed from office the day after the capital and ex-
change controls were introduced.

II COMPARATIVE REVIEW OF POLICIES AND PERFORMANCE
After Thailand was forced to float the baht in July not require a large official financing package or debt
1997, Malaysia's exchange and stock markets came rescheduling.2 In addition to overall financial pru-
under severe pressure. Large capital outflows that had dence, the credibility of the financial system bene-
started in April that year continued, inducing a rapid fited from a widespread bank restructuring under-
adjustment in the current account and a significant taken in the 1980s. These strong features protected
depreciation of the ringgit along with other regional the country's external position and may have less-
currencies (Appendix Figure A.2.3). Arbitrage be- ened the extent of the overall output decline. The
tween domestic and offshore ringgit markets became magnitude of domestic unemployment and potential
more active; in anticipation of further weakening of social disruption was moderated further by a signifi-
the currency, the premium on offshore ringgit interest cant migrant workforce that acted as a buffer.
rates increased sharply, exacerbating outflows of
ringgit funds and the domestic liquidity shortage.
Capital Account Regime and Reserve
Moreover, domestic demand collapsed because of
negative wealth effects from falling equity and prop- Management
erty prices and uncertain economic prospects (Ap- Malaysia's initial low level of short-term external
pendix Figure A.2.4). External demand also de- debt enabled it to maintain foreign reserves at a rea-
clined, consistent with financial difficulties and sonably high level, and this contributed to relatively
falling demand in other crisis countries. As a result robust external and domestic confidence early on in
of the ringgit's depreciation, higher borrowing costs, the crisis. As a consequence of financial vigilance
declines in equity price, and the drop in demand, exercised through prudential regulation of capital
corporate balance sheets deteriorated further. Thus, movements,3 the exposure of the financial and cor-
the financial system came under increasing stress porate systems was contained. Management of for-
(Figure 2.2), reflecting in part high corporate lever- eign exchange reserves has been cautious, entailing
age, and signs of bank runs emerged in late 1997. As no forward sales or other off-balance-sheet liabili-
the market recognized that the asset quality of both ties, so that all reserves have been usable.
the banks and corporations was worse than it ap- The capital account was nevertheless liberal in
peared, confidence eroded further. ways that left Malaysia vulnerable to contagion.
Malaysia's relative strengths going into the crisis Like the other countries, portfolio flows were free of
nonetheless helped contain the severity of its impact. restrictions; in Malaysia, these were the main chan-
Although Malaysia suffered serious currency and fi- nel for capital outflows in mid- and late 1997. Fur-
nancial crises, it was spared from an external debt thermore, cross-border activities in ringgit were
crisis. Thus, unlike the other countries, Malaysia did treated liberally, permitting the use of the currency
in trade and financial transactions with nonresidents,
and in offshore trading of securities listed on local
exchanges. As a consequence, an offshore ringgit
Figure 2.2. Bank Lending
market developed and became more active than
those in other affected countries.
30 Openness and Structure of Exports

Malaysia is the most open among the crisis coun-
25
Nonperforming loans of the banking system tries (Table 2.1), and this feature was beneficial in
20
(in percent of total loans) leading output recovery in early 1999 even as do-
mestic demand remained relatively weak. However,
15 the country's export base has become more depen-
dent on electronic and electrical products, whose
10 share in total exports reached more than 60 percent
Credit to private sector
(four-quarter percent change)
that year. Thus, like the other countries, Malaysia
5 could become more vulnerable to adverse world de-
5 2
1997 1998 1999 2000 Malaysia was the only one of the five Asian countries severely
affected by the crisis that did not resort to use of IMF resources.
3
Source: CEIC Data Company Limited. Such regulation included limits on banks' net foreign currency
open position and monitoring of liabilities of domestic corpora-
tions to ensure their foreign exchange earning potential.

Initial Conditions and Economic Structure
Table 2.1.Asian Crisis Countries: Selected Structural Indicators, 1996-2000

(In percent, unless otherwise indicated)
Malaysia Korea Thailand Philippines Indonesia
External
Precrisis
Openness, sum of exports and imports as percent of GDP, 1996 148 54 69 63 42
Share of exports to Japan and other crisis countries, 1996 23 21 25 27 38
Export growth in U.S. dollars, second half of 1995-second half
of 1996 1 -3 -5 16 6
Postcrisis
Openness, sum of exports and imports as percent of GDP, 2000 186 70 93 98 64
Share of exports to Japan and other crisis countries, second
half of 1999-first half of 2000 22 19 23 24 34
Export growth in U.S. dollars, 1999-2000 13 22 11 15 19
Banking system
Precrisis
Nonperforming loans as share of total loans, end-19971 6 6 23 5 7
Credit to private sector annual growth, average of 1993-96 21 20 31 40 22
Credit to private sector, in percent of GDP, 1996 141 150 135 52 59
Loan-to-deposit ratio (M2-based), end-1997 1.0 1.6 1.4 1.0 1.2
Postcrisis
Nonperforming loans as share of total loans, end-20001, 2 15 7 18 15 26
Impaired assets as shares of interest-earning assets, end-20003 22 23 38 23 61
Credit to private sector annual growth, 20004 6 -2 1 8 17
Credit to private sector, in percent of GDP, 20004 139 133 125 46 23
Loan-to-deposit ratio (M2-based), end-20005 0.9 1.2 0.9 0.7 0.4
Corporate sector
Precrisis
Corporate debt/equity ratios, 19976 88 425 440 174 334
Market capitalization in percent of GDP, 1996 317 28 56 98 41
Stock price changes, 1990-96 145 -6 36 385 53
Postcrisis
Corporate debt/equity ratios, 20006 114 235 280 186 277
Market capitalization in percent of GDP, 2000 131 35 26 79 21
Stock price changes, mid-1997-end-2000 -37 -32 -49 -41 -43
Sources: IMF, International Financial Statistics;World Economic Outlook; IMF staff estimates and CEIC database.
1Definitions of nonperforming loans vary among different countries, and caution should be applied in making any comparison.
2
Data cover only loans that remain in the banking system, that is, not including nonperforming loans transferred to asset management agencies. Korea
refers t o end-September.
3
Data include nonperforming loans transferred t o asset management agencies.
4
Unadjusted for transfers of nonperforming loans t o asset management agencies.Thailand data refer t o end-October.
5
Korea and Philippines refer to end-November, and Indonesia refers to end-August.
6
Nonfinancial companies. Malaysia,Thailand,and Indonesia are listed companies only.
velopments in the period ahead, especially in regard investment ratio for Malaysia was the highest among
to the technology sector. the crisis countries (Figure 2.3). This strategy was
successful in generating one of the highest precrisis
Investment-Based Growth Strategy
growth rates and elevating the income and wealth of
the bumiputras.
For two decades prior to the crisis, Malaysia pur- Although much of the investment went into elec-
sued a policy strategy aimed at raising economic tronics and other export-oriented industries, a large
growth and eradicating poverty by promoting invest- portion went into capital-intensive infrastructure and
ment in priority sectors and equipping the bumipu- the real estate sector. Investment quality, measured
tras with capital and opportunities to engage in vari- by the incremental capital output ratio and estimated
ous economic activities. For a number of years, the total factor productivity growth, eroded significantly

Figure 2.3. Domestic Investment in Figure 2.4. Investment and Real GDP
Selected Asian Countries Growth
(In percent of GDP)
50- 15
50-
40-
Investment/GDP
(four-quarter moving average,
30 - left scale)
Real GDP growth

20- (four-quarter percent change,
right scale)
10 —15
1992 93 94 95 96 97 98 99 2000
5
1994 95 96 97 98 99 2000 01
Source: CEIC Data Company Limited.
Source: IMF, World Economic Outlook.
between 1990-97.4 As the Malaysian economy un- Stock market capitalization in Malaysia grew to an
derwent an adjustment in response to the crisis, pri- extremely high level prior to the crisis, reflecting
vate investment declined sharply in 1998 and has re- both the fast expansion of the capital market and lib-
mained subdued (Figure 2.4). This adjustment eral capital account regime. At the same time, the
dampened short-term domestic demand during the level of corporate bank (short-term) indebtedness
recovery process, but reallocation from overinvested was among the highest of the crisis countries (Figure
sectors would be desirable to improve the invest- 2.6). As the crisis emerged, sharp declines in the
ment efficiency and achieve higher total factor pro- stock market led to relatively greater losses of wealth
ductivity growth over the long term. In this connec- in Malaysia than in other countries. This had negative
tion, the planned move to a knowledge-based implications for corporate ability to roll over or ser-
economy (K-economy) is intended both to raise the vice debt, because the value of stockholdings used as
productivity of investment and to diversify exports, bank collateral fell abruptly and income streams from
although its effects will not be immediate. them diminished. Multiple leveraging—made possi-
ble by cross-holding structures—exacerbated the
negative impact on corporate wealth and cash flows,
Capital Market Development and consequently on the extent of financial distress.
Beginning in the early 1980s, a rapidly growing This may explain, in part, the relative worsening of
capital market emerged as a major funding source for Malaysia's corporate performance in 1997-98. Low
higher investment (Figure 2.5). Its development was rates of return on capital, stemming from overinvest-
supported by a number of measures to strengthen, ment in the past, have likely contributed as well.
broaden, and deepen the market, which improved the
regulatory framework, as well as the trading and set-
tlement infrastructure. Legal and Regulatory Framework
Malaysia's well-developed legal and regulatory
frameworks, which provided reasonable protection
4
For total factor productivity growth estimates, see Michael to creditors and noncontrolling shareholders, permit-
Meow-Chung Yap, 2000, "Potential Output of the Malaysian ted relatively speedy but orderly settlement of prob-
Economy: Evaluating the Production Function Approach," lem loans early on. Furthermore, priority had been
paper presented at the Malaysian Institute of Economic Re- given to strengthening and modernizing prudential
search, November (Kuala Lumpur). The incremental capital out-
put ratio reversed in 1998-99, which was attributed largely to safeguards even before the crisis. This, compared
falling investment. with the weaker systems (including the absence of

Policy Responses and Performance
Figure 2.5. Market Capitalization in Figure 2.6. Private Sector Credit in

Selected Asian Countries Selected Asian Countries
(In percent of GDP) (In percent of GDP)
180
160 Korea Malaysia
140
Thailand1
120
100
80
Indonesia
60
40
Philippines
20
0
1994 95 96 97 98 99 2000 1994 95 96 97 98 99 2000
Source: CEIC Data Company Limited. Source: CEIC Data Company Limited.
1
Figure for 2000 refers t o October.
enforceable bankruptcy laws) and lack of supervi- labor was concerned, retrenchment was limited, as
sory personnel in some other countries, helped main- employers implemented other practices (also used in
tain confidence in and facilitate the recovery process other crisis countries), such as part-time, flexi-time,
of Malaysia's economy. and pay cuts, before resorting to layoffs. These "labor
Still, as in the other countries, close connections hoarding" practices were reversed as the economy re-
between large corporations and the Malaysian gov- covered. Importantly, progress in poverty reduction
ernment masked the extent of deterioration in the over the past two decades has helped preserve overall
asset quality of financial and corporate systems be- social and political stability.
fore the crisis, and this situation appears to have con-
tinued even after the crisis. Implicit government
guarantees associated with privatized infrastructure Policy Responses and Performance
projects undertaken prior to 1998 may have deferred
debt and operational restructuring by the concerned The capital controls and the ringgit peg introduced
corporations even as they became financially illiq- in September 1998 were key policy measures that
uid. The government also influenced bank lending to distinguished Malaysia from the rest of the crisis
certain projects that it deemed of social importance. countries.6 At that time, it was difficult to tell
These practices could have potentially adverse im- whether financial instability in the region was likely
plications for future claims on the government.5 to intensify or abate. Some measure of stability had
by then begun to return to the regional financial mar-
Social and Political Cohesion
kets (Appendix Figure A.2.5). From late August
1998, local currencies and stock markets had
The bumiputra policy, the presence of temporary broadly calmed down, and forward exchange rate
migrant workers, and smooth relations between labor forecasts showed expectations that the ringgit—as
unions and employers help explain the low unem- well as the won and the baht—would stabilize.
ployment in Malaysia even at the depth of the crisis. Short-term interest rates had declined to levels
Cost reductions in sectors facing output contraction below 10 percent. Sovereign bond spreads had
were brought about through nonrenewal of contracts peaked for Thailand, Korea, and the Philippines. In
for migrant workers. However, as far as domestic
6
See, below, the time frame of Malaysia's policies in response
5
See Section III for further discussions of fiscal issues. to the crisis and economic performance that ensued.

both Thailand and Malaysia, the offshore swap dif- There were also delays in implementing fiscal
ferential was trending down; nevertheless, the differ- stimulus measures in Malaysia. Fiscal restraint was
ential was in a high range and financial market pres- pursued at the early stage of the crisis, consistent
sures also remained high. with the policy recommendations of the IMF. Once
Given the relative stability experienced by the re- fiscal expansion was adopted, the delays arose from
gion, Malaysia's approach seems to have made little reliance on increased capital expenditures, which
difference to the monetary policy outcomes or produced slower effects than higher current spending
structural reforms. Crisis countries that did not in- or the tax cuts applied in some other countries. This
troduce these measures have been just as able to approach in turn reflects the policy of maintaining an
carry out accommodating monetary policy and operating surplus each year. The impact on private
maintain stable exchange rates. Boosted by rising domestic demand of the 1998-99 fiscal impulse—
export demand for electronic goods, especially from estimated to be the largest among the crisis coun-
the United States, economic recovery has been tries—was not fully felt until late 1999, reflecting
broadly similar across the Asian crisis countries the lags in carrying out capital spending. This may
(Appendix Figure A.2.6). Malaysia entered the cri- explain the lag in the recovery. Fiscal analysis is
sis later, however, and the resumption of domestic complicated, however, by the use of off-budget pri-
demand also came later. All countries embarked on vatized infrastructure projects to stimulate the econ-
structural reforms, but Malaysia's progress in finan- omy, the magnitude of which is not transparent.
cial restructuring with active involvement of the In the early stage of the crisis, Malaysia did not
government has been more rapid. Malaysia's pru- defend domestic currency as aggressively as the
dential regulation and supervision, already reason- other countries. Nominal and real interest rates were
ably well developed prior to the crisis, also seems to raised, but reached lower peaks in Malaysia, and
be superior, contributing to a relatively robust finan- very high interest rates (reaching 35 percent) per-
cial system that could help cushion Malaysia's vul- sisted for only a few days. The situation contrasted
nerability against future crises. with those in Korea and Thailand, where high inter-
The imposition of capital controls led to higher est rates were maintained for some months without
costs of external borrowing for Malaysia, but only interruption. The difference reflects two factors in-
temporarily. Bolstered by the pegged exchange rate herent in the initial conditions of the three countries.
at an undervalued level, the controls provided a safe- First, Malaysia's financing gap, if any, was modest,
guard to policymakers and domestic market partici- and the maintenance of high interest rates to attract
pants at a time of financial volatility. Overall, the ad- capital inflows was less critical. Second, the
verse impact of capital control measures was Malaysian corporate sector relied more on domestic
contained (Appendix Figure A.2.7).7 borrowing than external borrowing; therefore,
Malaysia's strategy helped mitigate the impact of the
crisis on the sector's financial distress, whereas the
Fiscal and Monetary Policies
strategy used in the other countries helped in the sta-
Macroeconomic policy responses were broadly bilization of the exchange rates that was needed to
similar in all the crisis countries. Sharp hikes in inter- contain external debt-service burden.
est rates and fiscal tightening aimed at restoring con- As monetary policy eased, real interest rates in
fidence were implemented initially. As the severity of Malaysia were brought down to a level slightly lower
output declines became evident, the strategy shifted than in the other crisis countries (Appendix Figure
to fiscal expansion supported by accommodating A.2.9), but the interest rate structure in Malaysia re-
monetary policies. The magnitude, timing, and de- mained rigid (as in Thailand). Slow credit growth was
gree of flexibility of policy responses in Malaysia, observed in all countries and does not appear to have
however, were different from the other countries to constrained the recovery because demand for large
some extent. Later implementation of macroeco- investments was small. Interest rate rigidities, in par-
nomic stimuli, consistent with the delayed output ticular, ceilings on banks' lending rates, could have
weakening and combined with rigidities in both the nevertheless hindered productive—but higher risk-
budget system and interest rate structure, may have activities and prevented faster output growth.
led to relatively slow responses in domestic demand. To counter the slow credit extension, Malaysia set
In 2000-01, Malaysia continued to pursue fiscal ex- nonbinding, minimum targets for loan growth, gave
pansion, whereas the other countries had moved to a tax incentives, and made use of directed lending to
more neutral stance (Appendix Figure A.2.8). target specific sectors. In the environment of sub-
dued domestic demand and excess liquidity that pre-
vailed through 2000, these measures appeared inef-
7
See Section V for further discussions of the capital controls fective. Government involvement in negotiations to
and their impact. restructure bank debt of some privatized infrastruc-

Policy Responses and Performance
ture projects, for example through the Corporate

Debt Restructuring Committee, may have boosted Figure 2.7. Cumulative and Net Portfolio
credit extension, although it could also have implica- Flows
tions for future claims on the government. (In billions of U.S. dollars)
Capital Controls and Exchange Rate Regime 1.5

The effect of the September 1998 controls in cur-
tailing speculative capital outflows appeared benign
because much of this capital had already left
Malaysia (Figure 2.7). Crisis countries that did not
adopt capital controls were able to pursue a low in-
terest rate policy and to keep exchange rates rela-
tively stable as the surplus from the current account
began to build up and capital outflows subsided.
These countries also made some progress with fi-
nancial and corporate sector restructuring. Never-
theless, the policies provided safety measures at a
time of foreign exchange market instability. Domes-
tic market participants reacted positively to the -2.5
1997 1998 1999 2000
fixed exchange rate regime, although it had exposed
them to exchange risks by discouraging hedging ac- Source: Malaysian authorities.
tivities. The effects of control measures on raising
funding costs for Malaysia and limiting capital in-
flows were reversed soon after the controls were
eased. mance of the financial systems.8 This may be ex-
The magnitude of the ringgit undervaluation ap- plained to some extent by the more centralized ap-
pears to have been roughly similar to those of the proach they adopted that permitted actions to be
other currencies. The undervalued ringgit benefited taken early and in a comprehensive manner, and, in
the Malaysian export sector, but it may have also the case of Malaysia, by its better starting conditions
discouraged imports and associated investment in as well. A more active involvement by the govern-
other sectors and contributed to the slow pickup in ment, however, implies that it could be taking higher
private investment overall. Additionally, market risks on contingent liabilities involved in the restruc-
expectations of an appreciation, which prevailed turing process and on implicit guarantees of the fu-
from early 1999 to mid-2000, are likely to have in- ture health of the financial system and of corpora-
duced short-term speculative inflows that were tions affected by these actions. Risks would
subsequently followed by outflows and reserve diminish, to the extent that ongoing reforms of the
losses. regulatory and supervisory frameworks promote
greater market discipline and better-managed finan-
Financial Sector Reforms cial institutions. The various aspects of Malaysia's
financial and corporate sector reforms can be sum-
Policy responses to strengthen the financial sys- marized as follows:
tem have also been broadly similar among the crisis
countries. To prevent its collapse, all countries gave • The government announced a guarantee of bank
a blanket deposit guarantee and provided liquidity deposits early in the crisis (January 1998) that
support early into the crisis. Subsequently, each was credible in view of the country's legal and
country established some form of asset-management regulatory framework, and the country did not
strategy to address nonperforming loan problems, suffer the bank runs experienced in Indonesia,
gave high priority to the upgrading of supervisory Korea, and Thailand.
and regulatory standards to international best prac- • Danamodal, the recapitalization agency, suc-
tices, and sought to recapitalize their financial insti- ceeded early in restoring the capital levels of do-
tutions based on those norms. In the process, bank mestic banks in Malaysia, whereas the capital
closures or merger programs were undertaken to es- standards had not fully been met in some of the
tablish stronger financial systems. other crisis countries that employed a more de-
Overall, the progress achieved by Malaysia and
Korea compares favorably in some aspects to the 8
See Section VI for more details on financial sector develop-
other countries, contributing to the improved perfor- ments and reform measures.

COMPARATIVE REVIEW OF POLICIES AND PERFORMANCE
centralized process. The agency's mandate in- major step to enhance the soundness of the fi-
spired confidence that all domestic financial in- nancial system, especially if financial innova-
stitutions in Malaysia would be recapitalized to tions are to be increasingly embraced by the
the required standards and that necessary opera- country's banking system.
tional restructuring would be imposed through
the exercise of control over their managements. Corporate Reforms
• The special legal power vested in Danaharta, the Progress in corporate sector reforms in Malaysia
centralized asset management company, helped has been mixed.10 Institutional frameworks have
ensure that banks were left with a manageable been upgraded, especially standards for corporate
level of problem loans and that acquired assets governance. The promotion of active bond markets
were rehabilitated; this contrasts with the more is expected to help gradually rebalance the maturity
decentralized nonperforming loan workout in profile of corporate debt and diversify system risks.
Thailand. The use of independent auditors to de- However, like in the other crisis countries, improved
termine the value of assets acquired by Dana- corporate balance sheets have largely been a result
harta avoided the subsidies required in Indonesia of economic recovery and debt restructuring,
and in Korea. The emphasis placed on the man- whereas progress in operational restructuring has
agement of assets implies their disposal at a been slow. Furthermore, a few incidents of poor
more measured pace, unlike the quick disposal governance despite the upgraded standards have
of acquired assets by Danaharta's counterparts in raised doubts about the government's commitment
other crisis countries. The agency succeeded in to adhere to best practices and protect minority
the early closure of its window for acquiring shareholders. Such incidents have in turn under-
nonperforming loans and achieved a relatively mined confidence and stock market performance,
high recovery rate of disposed loans, but the risk and weakened the ability of corporations to raise eq-
of its becoming a nonperforming loan ware- uity to support the restructuring process.
house is greater under the current strategy unless
assets are upgraded and sold before long.
• The bank consolidation program, tightly coordi- Time Frame of Malaysia's Response
nated by Bank Negara Malaysia, aimed at creat-
ing larger and more efficient financial institu- to the Crisis
tions ready to compete globally as new foreign Initial Policy Response
competition is allowed. This contrasts with the
strategy elsewhere, which entails bank closures From the time the Thai baht was floated in July
and takeovers of existing banks—including by 1997 until February 1998, Malaysia's policies (Table
foreign investors—thereby imposing market dis- 2.2) broadly aimed at addressing external imbal-
cipline and requiring financial institutions to be- ances, alleviating pressures on the ringgit, and
come internationally competitive rapidly.9 restoring market confidence. For a few days in July
Malaysia's approach has been less disruptive, 1997, the authorities intervened in the exchange
but the government's reluctance to close banks market and raised interbank interest rates sharply,
could create moral hazard. Limited foreign eq- but they quickly let the ringgit depreciate and low-
uity participation in local banks could also slow ered the interest rate to near precrisis levels. In Au-
the diffusion of technology transfers and the gust 1997, administrative measures were introduced
adoption of international best practices. to regulate foreign exchange swap transactions11 and
restrict trading of blue chip stocks in the Kuala
• Malaysia moved ahead of the other crisis coun- Lumpur stock exchange.
tries to strengthen prudential supervision and As output expansion continued to be brisk through
regulation. Bank Negara Malaysia's move to- end-1997, macroeconomic policies during this
ward internal risk management by banks, includ- period showed considerable restraint. Fiscal policy
ing for cross-border transactions, and toward was tightened through sharp spending cuts. Steps
risk-based and consolidated supervision is a were introduced to reduce loan growth through lend-
ing targets and directives to banks to cut credit for
9
The share of foreign-owned banks in Malaysia was relatively
high before the crisis, although their activities are more restricted 10
See Section VII for further analysis of the corporate sector
than local banks. The approach to not allow foreign takeovers of and recent reform efforts.
banks may partly explain the low level of foreign direct invest- 11
In May 1997, Thailand imposed partial restrictions on off-
ment flows into the country in 1998-2000, relative to recent his- shore activities in order to stabilize the foreign exchange market
tory, in comparison with such flows into the other crisis countries. and stem speculative attacks on the baht.

Time Frame of Malaysia's Response to the Crisis
Table 2.2, Major Economic Policy Measures During Emergence of Crisis and Early
Stage of Contagion
Date/Background Policy Objectives Specific Measures
March 1997: Kuala Lumpur stock Monetary:To slow credit growth • Limits on lending to the real estate sector and for the
exchange price began to fall purchase of stocks and shares
July 1997:Thai baht devalued Monetary:To defend the ringgit • Interbank rate raised, reaching 35 percent for a couple
of weeks, then lowered to around 10 percent
August 1997: Ringgit depreciation Monetary:To limit interest rate • Bank Negara Malaysia abandoned ringgit defense
continued;Thailand's program hikes and break links between • Limits imposed on ringgit swap transactions with
with IMF domestic and offshore rates nonresidents that were non trade related
Stock market:To stabilize and • Ban introduced on trading of blue-chip stocks
support Kuala Lumpur stock comprising Kuala Lumpur stock exchange composite
exchange price index
• Announcement of a large, publicly funded stock support
scheme
September 1997: Prime minister's Fiscal: To strengthen fiscal and • Government spending cut for 1997, and implementation
statement that currency trading external accounts of large privatized projects deferred
should be illegal
October 1997: Indonesia's Fiscal: To continue cautious plan • 1998 budget announced, targeting a surplus of21/2
program with IMF percent of GDP (corresponding to a real growth rate of
7 percent)
Monetary:To limit credit growth • Credit plan introduced to limit loan growth gradually to
15 percent by end-1998, but with loans to specified
sectors encouraged
• Base lending rates allowed to rise in response to higher
interbank rates
November 1997: Further fall in Corporate:To rescue Renong • Proposed takeover of Renong by United Engineers
stock prices which was in financial trouble Malaysia announced
December 1997: Evidence of Fiscal: To maintain broad fiscal • Major cuts (18 percent) in government spending plan
slowing activity; Korea's target for 1998 for 1998, offsetting lower revenue projections and
program with IMF corresponding to revised output growth of 4-5 percent
Monetary:To support the ringgit • Bank Negara Malaysia intervened in foreign exchange
interbank market
January 1998: Signs of bank runs Financial:To provide stability • National Economic Action Council set up as
and flight to quality consultative body to deal with the Asian crisis
• Government announced a guarantee of deposits at all
financial institutions
• Tighter restrictions imposed on stockbroking
companies including those on gearing ratios, margin
financing, and total exposure
February 1998 Corporate:To strengthen • High Level Finance Committee on Corporate

corporate governance Governance established
Monetary:To rebalance • Bank Negara Malaysia intervention rate raised to 11

macropolicy percent from 10 percent; statutory reserve requirement
cut to 10 percent from131/2percent
• Bank Negara Malaysia set guidelines to ensure credit to

priority bumiputra, housing, and small business sectors
• Announcement that five financial institutions were in

need of recapitalization, but no specific plans provided

Table 2.3. Major Economic Policy Measures, Initial Policy Responses, and Continued Turmoil
March 1998: Adoption of Fiscal and monetary:To rebalance • Revised fiscal plan implemented for 1998 aimed at a
macrostabilization and financial macroeconomic policy smaller surplus of1/2percent of GDP, allowing the
sector package reinstatement of previously cut social expenditures (of
1/2—1 percent of GDP) to protect most vulnerable
groups
• Credit growth of 12-15 percent for 1998 reiterated
Financial:To safeguard banking • Strategy adopted for strengthening regulatory and
system soundness prudential framework, disclosure standards, and
monetary management
• Announcement of mergers of finance companies
•Announcement of planned improvement of monetary
management, with timetable
May 1998: Political and economic Monetary:To provide banks with • Bank Negara Malaysia began to provide daily reports on
turmoil in Indonesia flexibility in liquidity management its money market operations, including its forecasts of
financial system cash flows, its liquidity operations, and
money market tender results
June 1998: Increasing signs of policy Financial:To lessen the burden in • Danaharta established to purchase the nonperforming
divergence between prime dealing with nonperforming loans loans of financial institutions
minister and deputy prime
minister
July 1998: Evidence of output Fiscal and monetary:To provide • National Economic Recovery Plan issued
collapse; equity markets in strong stimulus • Additional budgetary and off-budgetary expenditures
advanced countries began to introduced, aimed at achieving federal government
decline deficit of 21/2percent of GDP
• Statutory reserve requirement reduced to 8 percent
from 10 percent
Financial and corporate:To • Danamodal established as a vehicle to recapitalize banks

strengthen capital base of financial • Corporate Debt Restructuring Committee established
institutions and resolve bad to encourage informal arrangement for debt
debt problems restructuring
August 1998: Russia unilaterally Monetary:To stimulate economy • Bank Negara Malaysia intervention rate reduced in
announced debt-service further and to move gradually to three steps to 91/2percent from 11 percent
moratorium followed by prudential-based system • Prudential-based liquidity framework introduced,
emergency measures; yen vs. intended to replace liquid asset requirements by January
dollar rate plunged; long-term 2000
capital management trouble
publicized
financing purchases of real estate and securities. To cluded the upgrading of capital adequacy, prudential
forestall bank runs and "flight to quality," the gov- guidelines, and disclosure standards for banking in-
ernment announced a full deposit guarantee in Janu- stitutions, as well as a merger program for finance
ary 1998. companies. These were followed by a comprehen-
Additional policy measures were put into place sive approach to financial and corporate restructur-
between March and August 1998 (Table 2.3). In re- ing through the establishment of the asset manage-
sponse to signs of output slowdown, macroeconomic ment company, Danaharta, in June, and of the
policy was rebalanced in March when fiscal policy recapitalization agency, Danamodal, and the Corpo-
was relaxed, whereas the credit policy remained rate Debt Restructuring Committee in July. A new
tight relative to the previous trend. Also in March, liquidity framework for banking institutions was
the authorities adopted a strategy to safeguard the also introduced in July, requiring banks to manage
soundness of the financial system. Key measures in- assets and liabilities prudently and efficiently.

Time Frame of Malaysia's Response to the Crisis
Economic Performance Following As output collapse became evident, macroeco-

Initial Policies nomic policy became more supportive of economic
revival. A fiscal stimulus package was launched in
Although the financial reform measures would be- August 1998, entailing a significant increase in capi-
come important to the subsequent improvement of tal spending and tax reduction, and monetary policy
market sentiment, the overall policy package failed became successively more accommodating.
to restore confidence right away. Capital outflows
continued from early 1997 to mid-1998, estimated at
over $10 billion and representing over one-third of Exchange and Capital Controls
end-1996 reserves. The spread between onshore and In September 1998, Malaysia adopted an unortho-
offshore interest rates widened, suggesting heavy dox approach that involved exchange and capital
speculation in the currency market. By mid-1998, controls (Table 2.4) The ringgit was pegged at
the ringgit had depreciated by 40 percent, and the RM 3.8 per U.S. dollar, restrictions were imposed on
stock market had declined by 75 percent from pre- the repatriation of portfolio capital, and offshore
crisis levels. In addition, output contraction was ringgit activities were prohibited. The strategy was
more severe than earlier anticipated, weakening fur- designed to insulate monetary policy from external
ther the financial and corporate sectors. Plagued by volatility and facilitate a low interest rate policy, to
increasing nonperforming loans and liquidity con- contain speculative capital movements, to provide
straints, bank lending was reduced beyond the credit certainty to market participants, and to give
plan, which aggravated financial problems in the Malaysia some breathing space needed to carry out
corporate sector and further brought down investor structural adjustment.
confidence.
The market welcomed the financial reform mea-
The crisis intensified, in part because of market sures and expansionary macroeconomic policies, but
qualms about the region, but Malaysia's own politi- the capital controls had an immediate negative im-
cal uncertainty and insufficient or unclear policies pact. Rating agencies downgraded Malaysia, sover-
also contributed to the malaise. Investor confi- eign bond spreads increased relative to those of
dence—already fragile—was shaken by the an- Korea and Thailand, and Malaysia was removed
nouncement in late 1997 of a United Engineers from major investment indices. This was in part due
Malaysia-Renong deal, which lacked transparency to investors' uncertainty about the coverage of the
and may have bent stock exchange rules. controls and the potential impact on foreign direct
From early 1998, the political conflict between investment, even though the latter was not to be af-
the prime minister and the deputy prime minister be- fected by the controls. Controls on portfolio capital
came more visible and intense, creating doubts about outflows were replaced by a price-based exit levy in
Malaysia's resolve to maintain its policy commit- February 1999, which was eased and simplified in
ment. In February 1998, it was announced that five September 1999 and February 2001 and finally
financial institutions were in need of recapitaliza- eliminated in May 2001.
tion, but no specific plans for their restructuring was
given, sparking renewed concern about financial
system vulnerability. The exchange rate and stock Recovery Stage and Recent Performance
markets stabilized briefly around the time that the The negative impact of capital control measures
macroeconomic stabilization and financial sector was nevertheless contained. Regional prospects,
package was put in place, but instability returned which had strengthened just prior to the introduc-
during April-August 1998 following the political tion of capital controls, helped bolster confidence
turmoil in Indonesia, the fall of the Japanese yen, and aided in the recovery process of the crisis
fears of a devaluation of the Chinese renminbi, and countries, including Malaysia. The ringgit turned
the financial trouble in Russia. Moreover, given the out, ex post, to have been undervalued, providing a
difficulty in recognizing the sharp output slowdown, boost to exports and an incentive for retaining
implementation of countercyclical macroeconomic funds in the country. Malaysian equities once again
policies came with considerable delay.12 Fiscal pol- became fully liquid following the replacement
icy remained relatively tight even after its relaxation of quantitative restrictions on their outflows by
in March, and reliance on quantitative monetary tar- price-based controls. These equities were rein-
gets created problems, as credit was cut from firms eluded in major investment indices, and portfolio
that were potentially viable. capital returned.
Thus, along with the rest of Asia, Malaysia staged a
recovery beginning in 1999: financial sector perfor-
12
Macroeconomic restraint at this stage of the crisis was con- mance strengthened, investor confidence improved,
sistent with the policy recommendations of the IMF. real output rebounded rapidly, and the inflation rate

Table 2.4. Major Economic Policy Measures, Exchange and Capital Controls, and Aftermath
September 1998: Deputy prime Exchange and capital controls:To • Ringgit pegged at RM 3.8 per U.S. dollar
minister Anwar removed and deter speculation on the ringgit • Controls on transfers of funds from ringgit-
arrested; Bank Negara Malaysia and gain monetary independence denominated accounts for nonresidents not physically
governor and deputy governor present in Malaysia, in effect imposing a one-year holding
resigned; Malaysia removed period for repatriation of portfolio capital, with
from major investment indices retroactive effect
causing its sovereign bond
spreads to rise as those for • Prohibition of offshore ringgit transactions
other crisis countries fell Monetary:To further ease and • Reduction of Bank Negara Malaysia intervention rate to
encourage credit growth 8 percent from 91/2percent; modification of formula
to link base lending rate to policy rate; reduction in
statutory reserve requirement in two steps to 4
percent from 8 percent; reduction in liquid asset
ratio requirement to 15 percent from 17 percent
• Individual banks' minimum loan growth target set at 8
percent; relaxation of restrictions introduced in early
1997 on bank lending to purchase property and shares
• Prudential regulations regarding nonperforming loans
and disclosure requirements relaxed
October 1998-August 1999 Fiscal and monetary: To stimulate • Announcement of 1999 federal government budget
economic activity aimed at achieving deficit of 51/2percent of GDP
• Successive reductions of Bank Negara Malaysia
intervention rate to 71/2percent (10/98), to 7 percent
(11/98), to 61/2percent (4/99), to 6 percent (5/99), and to
5.5 percent (8/99)
January 1999 Monetary:To contain investment in • New lending for residential development above a
selected activities certain threshold was restricted by Bank Negara
Malaysia
January-March 1999 Corporate:To adopt best practices • Various measures to enhance disclosure for listed
companies and to restrict the number of directorships
held by an individual
• Report of High Level Finance Committee on Corporate
Governance released to the public
February 1999: After the easing Capital controls:To preempt • Easing of some controls, including replacement of 12-
of capital controls, spreads exodus of capital and reengage month holding period for repatriation of portfolio
narrowed and plans to include foreign investors capital by system of a graduated exit levy
Malaysia once again in
investment indices
announced
April 1999 Financial:To strengthen supervision • Measures included intensified supervision, performance
and regulation reviews for bank CEOs, elimination of two-tier bank
classification, and disallowing of bank lending to
controlling shareholders
September 1999 Capital controls:To provide • Graduated exit levy on portfolio investment simplified
further relaxation and replaced by a flat 10 percent levy on repatriation of
profits
• Some relaxation in swap and forward transactions for
the purpose of share purchases
November 1999: Malaysia Financial:To strengthen • Risk-based supervision initiated, with consultation
reincluded in Dow Jones/IFC management and supervision between Bank Negara Malaysia and banks regarding
investment indices minimum guidelines on credit risk-management
practices

T i m e Frame of Malaysia's Response to the Crisis
Table 2.4 (conclude
February 2000 Capital controls:To resolve • Resolution on Central Limit Order Book that froze
pending issue investors' Kuala Lumpur Stock Exchange shares traded
in Singapore as from 9/98
April 2000: Malaysia reincluded Capital markets:To develop • Announcement of consolidation of stockbroking
in Morgan Stanley Capital companies, to 15 by end-2000 from 63 (deadline later
International investment index extended)
(May) • National Bond Market Committee established to
provide policy direction and rationalize regulatory
framework for bond market development
October 2000 Fiscal:To put in place another • Announcement of 2001 budget aimed at achieving
stimulus budget federal government deficit of 41/2percent
Capital controls:To provide further • Proposal to remove 10 percent exit levy on portfolio
easing capital profits repatriated after one year (effective 2/01)
fell. A large external current account surplus and re- of Danaharta, Danamodal, and the Corporate Debt
duced capital outflows allowed a buildup of interna- Restructuring Committee, which have allowed
tional reserves that continued until mid-2000. Al- banks to be rid of nonperforming loans and to
though the real effective rate of the ringgit has strengthen their capital base, and have allowed cor-
appreciated in recent months, it remained broadly sta- porate debt to be removed. There are indications that
ble throughout 1999-2000, as exchange rates of trad- Malaysia has moved ahead of other crisis countries
ing partners that had also been affected by the crisis in respect to formulation of prudential regulation,
began to stabilize at about the same time as the peg- resolution of nonperforming loans, restoration of
ging of the ringgit. Indications are that the ringgit has capital adequacy, and implementation of a bank con-
moved close to its fair value by March-April 2001. solidation program, all of which have helped rein-
Reflecting better market sentiment, the Kuala force investor confidence that the economy was un-
Lumpur stock exchange price index surged, after dergoing fundamental adjustment.
having bottomed out, also around September 1998. There are indications, however, that the perfor-
As in the other crisis countries, the index has contin- mance of Malaysian corporations, which had been
ued to fluctuate along with stock prices in major more profitable and under less financial stress prior
markets, especially in the United States, which have to the crisis than that of their counterparts in Korea
softened significantly since early 2000. Spreads for and Thailand, deteriorated to a greater extent during
Malaysian sovereign bonds declined from their peak the crisis and has recently fared no better than the
in September 1998, even if they have edged up in re- others. Similar to the other countries, operational re-
cent months. structuring of the corporate sector has been slow
Bank balance sheets have improved, reflecting and, reportedly, has not adhered fully to best corpo-
both the economic turnaround and the achievements rate governance practices.


Appendix. Graphical Overview of Indicators
During Precrisis, Crisis, and Recovery


Figure A.2.1. Selected Asian Countries: Precrisis Macroeconomic Indicators
Malaysia's growth rate was very high. Inflation was low.

14 20
Real GDP Growth Inflation
12 — (In percent) (Percent change)
Philippines
Indonesia
Korea
4
Korea
Philippines
2 Thailand 4
Thailand
2 1991 93 95 97 1991 93 95 97
4
Fiscal positions strengthened over the period ... . . . with savings rates reaching high levels.
45
Budget Balance National Savings
(In percent of GDP) (In percent of GDP)
Korea
35
Thailand
Indonesia
Philippines
15
1991 95 97 1991 93 95 97
Short-term exposure was small . . . . . . and reserve cover was adequate.

160 800 Reserves
Short-Term External Debt
140 - (In percent of reserve) 700
700 _ (In months of imports of goods
Thailand and services)
120 (left scale) - 600 Thailand 6
Philippines Korea
100- (right scale) right scale) -500 5
4
Indonesia - 400
- 300
Philippines Korea
0 97
1991 93 95 97 1991 93 95
Sources: IMF, World Economic Outlook; International Financial Statistics; and Asia and Pacific Department databases.

COMPARATIVE REVIEW OF POLICIES AND PERFORMANCE
Figure A.2.2. Selected Asian Countries: Precrisis Financial Indicators
Malaysia's current account deficit, . . . and supported by capital flows.

however, was high . . .
Current Account
20
Total Capital Inflows
(In percent of GDP) Korea (In percent of GDP)
0 15
Indonesia
Philippines
Korea
-2 10
Philippines
-4 5
Malaysia
-6 0
Malaysia Thailand
Thailand
-8 -5
-10 -5 -10
1991 93 95 97 1991 93 95 97
This facilitated high credit growth . . . . . . that financed fast-rising investment . . .

60 60
Credit t o Private Sector Gross Domestic Investment
(Percent change) (In percent of GDP)
50 50
Thailand
Malaysia
40 Thailand
Philippines 40
30 \
Korea
30
Indonesia
20
Korea
Malaysia
10 20
Indonesia
0 10
1991 93 95 97 1991 93 95 97
. . . and drove up prices in the property market . . . . . . and pushed the stock market to unsustainable
levels.
160
Property Price Indicators1 Stock Market Price Index
150 - (1994 = 100) __ (1991 = 100)
140-
130-
120-
Korea Thailand
110
100
Indonesia
90
80
70
1991 93 95 97 1991 95 97
Sources: IMF, World Economic Outlook; International Financial Statistics, and Asia and Pacific Department databases; CEIC Data Company Limited;
Bloomberg; and Robert Deckle and Kenneth Kletzer 2001, Domestic Bank Regulation and Financial Crises:Theory and Empirical Evidence from East Asia,
IMF Working Paper 01/63 (Washington: International Monetary Fund).
1House price index for Malaysia, real estate price for Thailand, and construction value index for Indonesia. Data are not available for Philippines.

Figure A.2.3. Selected Asian Countries: Economic indicators During the Crisis Period
To defend the external position,

As capital outflows continued, reserves Malaysia's interbank rates were raised
declined. but only for a short period.
40 90
Gross Reserves Short-Term Interest Rates
50 - (In billions of U.S. dollars) 35 - (Percent per annum, end of period)
45 Korea
40
Thailand
1997 1998 1997 1998
The ringgit depreciated substantially . . . . . . and stock market prices plunged.

130
Real Effective Exchange Rates Stock Market Price Index
120 (January 1997 = 100) (January 1997 = 100)
Malaysia
Thailand Philippine
60
50 Indonesia
40
30
20
1997 1998 1997 1998
Sources: IMF, Information Notice System and Asia and Pacific Department databases.

Figure A.2.4. Selected Asian Countries: Economic Developments
Domestic demand in Malaysia began to Output collapsed in the first quarter of 1998 and
weaken in the second half of 1997, and continued down through that year.
remained subdued until late 1999.
40 25
Real GDP Growth
Real Domestic Demand (Four-quarter percent change) 20
3 0 - (Percent change)
15
Korea
10
5
Philippines - 5
Thailand
--10
Indonesia
--15
--20
--25
1997 98 99 2000 1997 98 99 2000
The weak domestic demand was consistent Credit growth also declined sharply, beyond credit
with tight fiscal policy in effect until late 1998. plans, in response to the weakening corporate
sector and rising nonperforming loans.
70 60 160
Fiscal Stimulus1 C r e d i t t o Private Sector
60 (Cumulative deficit from first quarter 5 0 — (Four-quarter percent change)
5 0 - of 1997; in percent of GDP) -
40 - Philippines
40 (left scale)
Philippines
30
20
Thailand Korea
10
Malaysia
-10
-20
-30
1997 98 99 2000 1997 99 2000
Sources: IMF, Asia and Pacific Department databases; and CEIC Data Company Limited.
1Data are not available for Indonesia.

Figure A.2.5. Selected Asian Countries: Financial Market Indicators During the Crisis Period
Expectations of ringgit depreciation peaked in Sovereign bond spreads for Malaysia were negatively
April 1998, but continued to be volatile thereafter. affected by capital controls and peaked in
By November 1998, the market clearly viewed the September 1998, one or two months after those
ringgit as under valued. of other countries.
2500
Consensus Forecast of the Three-Month Sovereign Bond Spreads
120 - Forward Exchange Rates (In basis points)
(Percent change, depreciation, +) 2000
100 Indonesia
Indonesia
80 1500
60 Korea
Malaysia
Thailand Thailand
-40 1997 1998

-500
1997 1998
Offshore swap differentials for both Malaysia . . . and, after Malaysia introduced capital controls,
and Thailand were already trending down from the differential for Thailand declined further to
July 1998 . . . close to zero.
90 45
Swap Premiums Swap Spread
80 - (One month forward) - (One month forward) 40
70 35
60
Thailand offshore
50
Malaysia onshore
Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jun. Jul. Aug. Sep. Oct. Nov. Dec
1998
Sources: Data provided by country authorities; Consensus Economics Inc., Asia Pacific Consensus Forecasts; and IMF staff estimates.
1Export-Import Bank of Korea global bond spread is used as a proxy from September 1997 through March 1998.

Figure A.2.6. Selected Asian Countries: Economic Performance During the Recovery Stage
Beginning in 1999, Malaysia staged a rapid Similar to other crisis countries, the inflation
recovery. rate fell.
50 16 120
Industrial Production Inflation
40 - (Four-quarter percent change) 14 - (Four-quarter percent change)
1998 2000
A large current account surplus, combined . . . allowed reserves to build up.

with reduced capital outflows, . . .
30 10
Current Account Reserves
(In percent of GDP) — (In months of imports of goods and services) 9
25
8
20 Thailand
7
15 6
Indonesia
Korea
5
10
4
Malaysia
5
3
Indonesia Korea
0 Philippines 2
5 1998 99 2000 1998 99 2000
Sources: IMF, International Financial Statistics; and Asia and Pacific Department databases.

Figure A.2.7. Selected Asian Countries: Financial Indicators During the Recovery Stage
The stock market performance in all crisis

Real effective exchange rate for Malaysia countries improved, peaked in early 2000, then
remained broadly stable throughout 1999-2000. softened in line with the U.S. equity market.
120 160
Real Effective Exchange Rates Stock Market Price Index
(July 1997 = 100) Philippines
(July 1997 = 100)
140
\ Korea
120
Indonesia
Philippines
100
80
60
Malaysia
40
Thailand
20
20
99 1997 98 99
0
1997 98 2000 01 2000 01
Spreads for Malaysia declined and stayed low,

close to those for Korea and Thailand.
2000
Sovereign Bond Spreads
(In basis points, end of period)
1600
1200
800
400
1997 99 2000 01
Sources: IMF, Information Notice System; Bloomberg; and Asia and Pacific Department databases.
1Export-Import Bank of Korea global bond spread is used as a proxy from September 1997 through March 1998.
2Sovereign bond issued September 19, 1990 that matured on September 27, 2000.
3Sovereign bond issued May 26, 1999 and maturing on June 1, 2009.

Figure A.2.8. Selected Asian Countries: Fiscal Indicators
Fiscal stimulus in Malaysia began in 1998 and Malaysia's cumulative fiscal impulse during 1997-
continues into 2001. 2001 was the largest among crisis countries.
8- 14
Budget Balance Fiscal Impulse 1,2
6 - (In percent of GDP) (Cumulative from 1997, in percent of GDP)
10
Thailand Malaysia
2
Indonesia Korea
0
-2
-2
1995 99 2000 2001 1997 98 99 2000 2001
Prel. Proj. Prel. Proj.
Fiscal expansion in Malaysia relied largely on . . . whereas increased current spending was used
capital expenditure, which experienced delays, . . . more in some other countries, . . .
10 24
Capital Expenditure1,2 Current Expenditure1,2
9 - (In percent of GDP) (In percent of GDP) 22
8
20
Thailand Malaysia Malaysia
18
7
16
6 Philippines
Indonesia 14
Thailand
5 Korea
12
Korea
4 10
Indonesia
3 8
Philippines
2 6
1995 97 99 2000 2001 1995 97 99 2000 2001
. . . and revenue acted as an automatic stabilizer Government debt remained at a manageable

to some extent. level.
28 140
Revenue Central Government Debt1,2
(In percent of GDP)
2 6 - (In percent of GDP) 120
24 100
Korea Indonesia
22
80
20 Philippines
60
18 Thailand Malaysia
40
16
Thailand
Indonesia
14 20
Korea
12 0
1995 97 99 2000 2001 1995 97 99 2000 2001
Sources: IMF, World Economic Outlook; International Financial Statistics; and Asia and Pacific Department databases.
1For Indonesia, 1995 through 1999 are on a fiscal year basis (fiscal year ending March). Year 2000 contains data for three quarters. Figure for 2001
is on a calendar year basis.
2
Thailand is on a fiscal year basis (fiscal year ending September).

Figure A.2.9. Selected Asian Countries: Monetary Indicators
Interest rates in Malaysia were kept lower than in . . . both in nominal and real terms, mitigating
other countries for the greater part of the crisis, . . . corporate financial distress.
40 100 30 40
1
Nominal Interest Rates Real Interest Rates 1
(Percent per annum, end of period) _ (Percent per annum, end of period)
30
10
Malaysia
(left scale)
0 Thailand
-30
0 - Malaysia
(left scale) Thailand (left scale) -40
Indonesia (right scale) (left scale)
-10 1998 1999 2000
-5
1998 1999 2000
-50
But, similar to other countries, private credit

growth in Malaysia remained subdued and did
not meet the authorities' target for 1999 or 2000.
40 160
Private Sector C r e d i t G r o w t h 1
(Twelve-month percent change)
-10
1998 1999 2000
Sources: International Financial Statistics; and Asia and Pacific Department databases.
Temporary capital controls introduced in September 1998.

Ill Potential Output and Inflation
Il Houng Lee and Yougesh Khatri
T wo groups of methods may be used to estimate

potential output. One group is based on statisti-
cal techniques that attempt to decompose a time se-
cent (year-on-year) during 2000 if the gap were to
close.2
ries into the permanent and cyclical components; the
other group is based on estimating structural rela- Measuring Potential Output
tionships, which in turn are usually obtained from
economic theory. Potential output is estimated for The concept of potential output is not well defined
Malaysia using the cubic spline-smoothing method and is difficult to measure. The literature broadly
where the result indicates that the output gap would suggests two definitions (see Scacciavillani and
close toward end-2000 or early 2001. Thereafter, in- Swagel, 1999). The first arises from the assumption
flation is forecasted using the estimated output gap that the business cycle results mainly from move-
as one of its determinants. ments in aggregate demand in relation to a slowly
moving level of aggregate supply. This occasions
(potentially substantial) swings during which there
Introduction are overutilized/underutilized resources.
The second definition follows the neoclassical tra-
Potential output is an important input to macro- dition, where potential output is assumed to be driven
economic policy design. For practical purposes, po- by exogenous productivity shocks to aggregate sup-
tential output can be defined as the maximum output ply that determine long-run growth and, to a large ex-
an economy can sustain without generating an in- tent, short-term fluctuation in output over the busi-
crease in inflation. The output gap is simply the dif- ness cycle.3 According to the latter approach, output
ference between actual and potential output. In the fluctuates around its potential level but generally
short run, estimates of the output gap provide a key without wide or prolonged divergence.
indicator of inflationary pressures. This section esti- Potential output is an unobservable variable, mak-
mates potential output for Malaysia using various ing it difficult to measure. In practice, there are a
approaches applied to data through end-1999; then plethora of techniques, each with advantages and
inflation is forecast using the estimated output gap drawbacks, and no single methodology dominates.
as one of its determinants. The different approaches These techniques could be classified broadly into
used are briefly outlined,1 an estimate of potential two groups.
output in Malaysia and the implied output gap are The first group of methods is purely statistical
provided, followed by an inflation equation esti- and attempts to decompose a time series into perma-
mated using the output gap. nent and cyclical components. This category would
To summarize, of the several approaches avail- include filtering methods (moving averages, the
able, the potential output estimated using the cubic Hodrick-Prescott (HP) filter, the Beveridge-Nelson
spline-smoothing method appears to provide method, Kernel estimators, and spline-smoothing
the most plausible result, ex post, in terms of the
implied output gap. Estimates suggest the gap is 2
expected to close in late 2000 or early 2001, imply- Discussions below on the problems of the structural break in
1998 and the use of end-period data in 1999 (which is still a point
ing that inflationary pressures could start building in the recovery path) highlight the tentative nature of this estima-
up in the near future. The estimated inflation equation, carried out in April 2000. The results should thus be viewed
tion confirms that inflation would rise above 3 per- as indicative. The latest data together with the expected slow-
down in 2001 suggest that the gap may now persist into 2002.
3
In such a case, business cycles are not necessarily driven by
shortfalls or excesses of aggregate demand, but rather by rational
agents reacting to productivity shocks by writing off old invest-
1
A more technical exposition is contained in the Appendix. ments and reinvesting in new opportunities.

Estimating the Output Gap
techniques) and unobservable component methods technique, namely the cubic spline-smoothing method
(both univariate and multivariate approaches). Most of (CSSM), was chosen to estimate the potential output
these approaches derive the permanent component as for Malaysia. The main reason for selecting this ap-
a trend by minimizing the distance between the points proach was its simplicity, especially relating to data
on the trend and the actual value at prespecified inter- requirements. A production function approach was
vals, while penalizing frequent changes of the second also tried, but it did not indicate an improvement over
moments of the estimated trend in order to ensure results obtained from filtering techniques. The CSSM
some degree of smoothness of the estimated trend. Al- was preferred over other filtering techniques as it ren-
though these approaches are attractive, in that they re- dered the most plausible potential output path in terms
quire considerably less data than other methods, they of the implied output gap, especially in relation to ac-
become ill-defined at the beginning and end of sam- tual inflation.6
ples, or if there is a structural break in the data. There are two weaknesses that need to be ad-
The second group of methods is based on estimat- dressed using any of the smoothing methods: the
ing structural relationships, including production structural break in the GDP series, due to the sharp
functions, multivariate systems of equations, struc- decline in output in 1998; and the end-period prob-
tural vector autoregressions (VARs), and "demand- lems. Given the structural break in 1998, smoothing
side" models. The production function methodology methods tend to underweight the GDP growth in
represents the middle ground between a full-scale 1997, implying a positive output gap, even though
structural model and the various univariate ap- there was no evidence of overheating (e.g., no infla-
proaches (such as filters and unobservable compo- tionary pressures were noted in 1997). Furthermore,
nents). Even then, this approach is considerably the end-point of actual data, i.e., end-1999, which is
more demanding in terms of data requirements than still a point in the recovery path, tends to bring down
filtering approaches, and the nature of inputs data— the estimated potential output such that the esti-
particularly the capital stock calculations—implies mated output gaps in 1998-99 are small. In view of
significant potential measurement errors in inputs.4 the magnitude of output decline in 1998 with no ap-
The demand-side approach relates output directly parent physical damage or dramatic shift in the
to measures of spare capacity in the economy or structure of demand that would make existing capi-
supply-side measures. The structural VAR approach tal redundant on a permanent basis, the low output
combines aggregate supply (and supply shocks) and gap during the postcrisis period does not appear to
cyclical fluctuations with changes in aggregate de- be consistent with the actual unutilized productive
mand using a structural VAR with restrictions im- capacity of the economy.
posed on the long-run effects of impacts on output To address these issues, an interpolated point was
and unemployment. These long-run restrictions are used for 1998 (instead of actual), and the sample pe-
used to identify structural supply and demand riod was extended to 2002 using the 1998 IMF
shocks by allowing supply shocks to have a perma- World Economic Outlook (WEO) forecast.7 Annual
nent effect on output, while demand shocks are as- data were used for the period covering 1970-2002.
sumed to have only a temporary effect on output.5 The estimated output gap, adjusted for these two
weaknesses (Adj. SP), is compared with the gap esti-
mated without the interpolated point (Figure 3.1) in
Estimating the Output Gap 1998 using the CSSM (SP), the HP filter (with val-
ues of X=7 (HP7) and X=100 (HP100)), the Kernel
Of the approaches outlined above (and described smoothing method (KN), and the production func-
more fully in the Appendix), a univariate filtering tion (PF). As is shown in Table 3.1, except for the
adjusted CSSM, other smoothing methods tend to
4
overstate the degree of overheating during the pre-
At the IMF, the production function approach has been applied
mainly to industrial countries (see De Masi, 1997). Relatively few
crisis period while understating the output gap dur-
empirical studies attempt to estimate potential output for developing the postcrisis period. The degree of over- and un-
ing countries, mainly due to a lack of reliable data. Also, the con- derestimation is made worse by the extension of the
cept of potential output may be less relevant when a large fraction end-period through 2003.
of output relates to primary commodities, whose production is
supply-determined, or where there are large migration-related
flows of labor and ongoing structural change (associated with the
"catch-up" phase of development).
5 6
In multivariate unobserved component models or structural The X was determined according to the generalized cross-
VARs, the relationship between inflation and output implicit in validation criteria.
7
the definition of potential output can be imposed during the esti- The CSSM was first used to estimate the 1998 value, which
mation of potential output (Dupasquier, Guay, and St. Amant, was omitted from the series as missing and interpolated, then the
1999; Apel and Jansson, 1999; Scacciavillani and Swagel, 1999; CSSM was used on the new series with the interpolated 1998
and Cerra and Saxena, 2000). value to obtain potential output.

POTENTIAL OUTPUT AND INFLATION
gether with an explicit equation for purchasing

Figure 3.1. Comparing Estimates of power parity (PPP) (e.g., Jonsson, 1999). In other
Output Gap With (Adj. SP) and Without studies, the external sector disequilibrium pressure
(SP) the Interpolated Point in 1998 is measured using the difference between actual and
(Changes in Log (Index)) estimated equilibrium exchange rates (Lim and
Papi, 1997). Single-equation models include some
form of expectation-augmented Phillips curve (Raz-
zak, 1995; and Stock and Watson, 1999). A more
formal structural approach involves various combi-
nations of the above, where, for example, Chhibber
and others (1989) introduce PPP (for traded goods),
markup (for nontraded goods), and allowances
made for controlled prices.
The approach adopted in this section is eclectic
and tries to capture the key sources of inflationary
pressures in Malaysia. The pressure exerted on
prices by excess domestic demand or supply
shocks (through the output gap) is dampened to the
extent that some of the pressure is released through
adjustments in imports. Changes in imports as a
-0.06
1970 75 85 90 95 2000
percent of GDP, therefore, would play a role in de-
termining inflation. Furthermore, the PPP condition
is introduced to accommodate the impact of foreign
prices and the exchange rate on domestic inflation.
Specifically:
p = f(gap, img, ex, wp), (1)
The output gap, estimated using the production where p = domestic price; gap = output gap; img =
function, does not indicate an improvement over re- imports as a share of GDP; ex = nominal effective
sults obtained from filtering techniques. This reflects exchange rate; and wp = foreign price.
in part the fact that smoothing techniques were used The estimation methodology is the familiar maxi-
to derive the potential labor force (i.e., at a nonaccel- mum likelihood cointegration procedure by
erating inflation rate of unemployment (NAIRU) Johansen. The long-run relationship is estimated
consistent level) and the trend total factor productiv- using the VAR on p, img, ex, and wp. The short-run
ity (TFP) growth. This implies shifting the problems dynamics are estimated using an error-correction
associated with estimation of trend GDP one layer mechanism (ECM), in which we introduce the out-
down to estimation of trend labor force and total fac- put gap, and inflation is forecast 12 months ahead
tor productivity growth.8 (through 2001Q1). Inflation is projected to reach 3 1/2
percent to 4 percent (year-on-year) by the first quar-
ter of 2001, although, given the short observation
period, the result obtained needs to be interpreted
Determinants of Inflation with caution.
Studies have used various approaches to model-
ing inflation, depending on the structure of the Data
country and objectives of the analysis. Approaches
from the supply side include markup models, where Data are quarterly, covering 1991 (Ql) through
the general domestic price level is estimated as a 2000 (Ql), although the ECM is estimated for
markup over total unit costs, including labor costs, 1991-99.
import prices, and energy prices (see de Brouwer • p: domestic price is defined as the logarithm of
and Ericsson, 1998). Another approach centers the consumer price index. As there are no de-
around a money market equation, sometimes to- tailed breakdowns of the consumer price index
for the specified period, no adjustments are
made for transitory fluctuations or rigidities that
8
Bank Negara Malaysia's estimate of potential output in early are controlled or influenced by the government.
2000, based on a production function, found that the output gap
was narrowing and potential output was back on its growth path
• img: the ratio is defined as merchandise imports
(Bank Negara Malaysia, 1999). to GDP.

Determinants of Inflation
Table 3.1. Estimates of Output Gap, 1996-2001
Adj. SP SP HP7 HP100 KN PF
1996 0.04 1.62 4.17 7.45 0.47 5.47

1997 0.97 4.93 6.69 9.10 4.45 5.49
1998 -9.39 -4.25 -4.70 -3.87 -3.98 -6.21
1999 -4.94 -1.32 -3.06 -3.47 -0.26 -4.31
2000 -0.86 0.23 -1.36 -2.31 0.05
2001 0.15 0.20 -0.59 -1.45 -0.18
• ex: the nominal effective exchange rate index is Table 3.3 reports the result of VAR(1) on g, m,
defined as the log of the nominal effective ex- img, ex, and wp. As indicated by the Johansen maxi-
change rate and is derived from trading partners mal eigenvalue (Xmax) and trace eigenvalue (ftrace)
weighted by their relative trade share (IMF, statistics, the result rejects the null hypothesis of no
IFS). cointegration in favor of at least one cointegrating
• wp: foreign price is defined as the log of an relationship. The rows of the B matrix in Table 3.3
index derived as a composite of price indices of can be interpreted as long-run parameters and the el-
the United States (30 percent), Japan (20 per- ements in the matrix a as adjustment coefficients
cent), Singapore (20 percent), Germany (15 per- (see Charemza and Deadman, 1992).
cent), and the United Kingdom (15 percent) The first row of B, which is the estimated cointe-
(IMF, IFS). grating vector, represents a long-run relationship
between the variables estimated and can be ex-
• gap: output gap is defined as actual output di- pressed as:
vided by potential output. The potential output is
obtained using spline-smoothing allowing for p = constant - 0.0214 img - 0.1029 ex
the structural break during parts of 1998 and
1999 and using the same approach as elaborated + 1.8302 wp (2)
above, but using quarterly data. Hence, gap < 1
implies excess capacity. All of the estimated coefficients have the right
signs and explain that Malaysia's price index de-
Integration and Cointegration clines with larger imports as a percent of GDP and
with an appreciation of the nominal effective ex-
Statistical tests indicate that all variables are inte- change rate, but increases with higher world prices.
grated of order two or lower. In particular, according Having established a long-run relationship, a single
to the Augmented Dickey-Fuller (ADF) statistic, p equation model is used to assess the short-term be-
appears to be I(1), and the other three variables ap- havior of the price level.
pear to be I(2). However, their estimated coefficients
are numerically much less than unity, i.e., the coeffi-
cients of Aimg, Aex, and Awp are (-0.14=1-1.14), The Short-Run Dynamics of Inflation
(-0.04=1-0.96), and (0.33 = 1-0.67), respectively. An unrestricted error-correction model is used to
Thus, all four variables are treated below as if they examine the dynamics of inflation in the short run. It
are I(1).9 incorporates the long-run relationship obtained
The Johansen maximum likelihood estimation above as follows:
procedure for finite-order VARs is used to obtain the
long-run relationship between integrated variables. Apt =
As there was no a priori information on the lag order
of the VAR, a fourth-order VAR was used and sim- a2St + a2Dt + vt, (3)
plified to a first-order VAR on the basis of the results
presented in Table 3.2, below. where i = (0, 1, 2, 3, 4) and j = (l, 2, 3); "res" is ob-
tained from p-p(est), where p(est) is obtained from
(2) above; S denotes the seasonal dummies; and D is
9
See de Brouwer and Ericsson (1998) for a similar approach. a dummy variable to capture the structural break

POTENTIAL OUTPUT AND INFLATION
Table 3.2. F and SC Statistics for Sequential Reduction from

VAR(4) to VAR(1)
Null Hypothesis Maintained Hypothesis 1

System Log Likelihood SC3 VAR(4) VAR(3) VAR(2)
VAR(4) 68 639.5 -32.60

VAR(3) 52 622.2 -33.26 0.988
(0.489)
VAR(2) 36 599.1 -33.54 1.409 1.873
(0.143) (0.047)
VAR(I) 20 583.5 -34.30 1.488 1.755 1.445
(0.086) (0.030) (0.151)
1The numbers represent F-statistics for testing the null hypothesis against the maintained hypothesis, and the tail
probability associated with that value of the F-statistics (in parentheses).
2
Number of unrestricted parameters.
3
Schwartz criterion.
Table 3.3. Cointegration Analysis of Price Data, 1991Q2 to 2000Q11
Ho:rank=p Amax 95% Atrace 95%

2 3
P == 0 30.0 27.1 54.9 47.2
P <= 1 15.8 21.0 24.9 29.7
P <= 2 6.3 14.1 9.1 15.4
P <= 3 2.9 3.8 2.9 3.8
Standardized P eigenvectors
P img ex wp
1.000 0.021 0.102 -1.829
-1.443 1.000 -1.044 -1.713
-2.747 -10.621 1.000 78.069
-0.739 0.569 0.217 1.000
Standardized a coefficients
P -0.236 0.016 0.001 0.009
img -0.554 -0.051 0.011 -0.186
ex 1.609 0.164 -0.008 -0.108
wp 0.166 0.001 0.001 0.004
1PcFiml 9.0 for Windows was used to estimate the VAR.

2Significant at the 95 percent level.
3Significant at the 99 percent level.
during 1998. Again, using OLS, the order of lags are t_3, Aext.-2,
reduced in sequence, and the following variables t-l, and rest_
were retained:10
Using the estimated coefficients as reported in
Table 3.4, inflation is forecast for the 12 months
10 through 2001Q1. The projection indicates an upward
Although not reported here, the F and SC statistics were used
to test the null hypothesis during the sequence of eliminating trend in inflation, reaching31/2percent to 4 percent
variables. The latter are selected according to the lowest t-values during 2000Q1. The result is presented in Figures
at each stage of the OLS estimations. 3.2 and 3.3, below.

Table 3.4. Result of Inflation Estimation by OLS, 1992Q2-1999Q41
The present sample is: 1992 (2) to 1999 (4)
Variable Coefficient Std. error t-value t-prob Part R2

Constant 0.046055 0.027674 1.664 0.1096 0.1075
Dpt-2 -0.26606 0.14359 -1.853 0.0768 0.1299
Dimg t _ 3 0.028043 0.016735 1.676 0.1073 0.1088
Dex t _ 2 -0.018984 0.015673 -1.211 0.2381 0.0600
Dwpt_3 0.77974 0.24717 3.155 0.0044 0.3020
Dgapt -0.11003 0.047982 -2.293 0.0313 0.1861
Gapt-1 -0.039224 0.027506 -1.426 0.1673 0.0812
Rest-1 -0.016748 0.083422 -0.201 0.8426 0.0017
R2 = 0.690712 F(7,23) = 7.3378 [0.0001] \sigma = 0.00348159 D W = 2.53
RSS = 0.0002787939991 for 8 variables and 31 observations
1PcGivewas used to estimate the equation;"D" represents first difference.
Figure 3.2. Inflation, 1992-2001 Figure 3.3. Inflation, 1995-2001

(Log (Index)) (Log (Index), changes)
4.9 0.06
0.05
0.04
0.03
0.02
0.01
1992 94 96 98 2000 1995 96 97 98 99 2000
Appendix. Approaches to Measuring backs are highlighted. Detailed descriptions of the

Potential Output methodologies can be found in the studies cited.
Potential output is defined as the maximum output

an economy can sustain without generating an in- Statistical Filters and Smoothing Methods
crease in inflation. There are two groups of methods Simple Trends
that may be used to estimate the potential output.
One group is based on statistical techniques that at- The simplest method of estimating potential out-
tempt to decompose a time series into permanent put is the use of a linear time trend. This can be re-
and cyclical components; the other group is based on fined by using spline trends to allow for structural
estimating structural relationships based on eco- breaks or by including a polynomial in the trend
nomic theory. Below, the various methodologies are term to allow for nonlinearity (for an example, see
outlined and some of their main attributes and draw- Bayoumi and others, 1999).

Ill POTENTIAL OUTPUT AND INFLATION
The main attraction of using a simple time trend mines the "window" of data used to calculate
or variants thereof is its simplicity. Since the publi- the trend; the larger the smoothing parameter,
cation of an influential paper by Nelson and Plosser the broader this window of data used and the
(1982), however, which suggested that output series smoother the trend.
are best characterized as integrated series, there has • The HP filter can induce spurious cyclically in
been increasing recognition that measuring the per- the smoothed series when the series are inte-
manent component of output, i.e., potential output, grated or nearly integrated13 (see Harvey and
with any degree of accuracy is a difficult task. In Jaeger, 1993; and Cogley and Nason, 1995).
particular, the existence of a stochastic permanent
component implies that potential output cannot be • Probably the most serious drawback of filtering
treated as a deterministic trend (Dupasquier, Guay, methods for policymaking purposes is that they
and St.Amant, 1999).11 become ill defined at the beginning and end of
samples; see, for example, Baxter and King
(1995), who recommend discarding three years
The Hodrick-Prescott (HP) and Other Filters of quarterly data at both ends of a sample. This
Filters and smoothing methodologies can be as is a significant drawback for those policymak-
simple as a moving average. More complex smooth- ers wishing to use the measure of the current
ing techniques include the HP filter, which chooses output gap and thus focus on the most recent
trend output, y*, such that, for a given parameter X observations.14
(which determines the degree of smoothness), the • The HP and other such filtering methods also
sum of squared deviations of y from y* plus X times tend to neglect structural breaks and regime
the sum of the rate of change of y* is minimized. shifts (Scacciavillani and Swagel, 1999).
Mathematically, trend output, y*, is derived for a
given X by:
The Coe-McDermott Method
T-\ Coe and McDermott (1997) use a nonparametric
(4) (Kernel) smoothing technique, essentially similar to
t=0
the HP filter but where the degree of smoothness (in
the case of the Kernel estimator, determined by the
where A2 indicates twice differencing. The larger the bandwidth (h), similar to the HP X in determining
parameter X, the more weight given to smoothness the "window" of data used for the smoothing) is de-
(determined by the second term) versus fit (deter- termined by the data, hence addressing one of the
mined by the first term). For a discussion of the sta- major problems associated with the HP filter. Details
tistical properties of the HP filter, see Cogley and of the methodology can be found in the appendix to
Nason (1995), Harvey and Jaeger (1993), and Soder- Coe and McDermott (1997), and a general discus-
lind (1994). sion of nonparametric techniques is provided by
• The HP and other such filters are attractive be- Hardle (1990). The method, however, is still likely
cause considerably less data are required than to share the other problems of the HP filter.
for other methods such as the production func-
tion, unobservable variables, or structural VARs, Running Median Smoothing (RMS)
discussed below. Use of these filters, however,
has certain disadvantages. A simple form of the RMS filter uses a running
window on the data, the smoothed value in each pe-
• The resultant measure of output depends criti- riod being set equal to the median of the values in
cally on the choice of the smoothing parameter, the window. The method can be extended to apply
X.12The smoothing parameter, in effect, deter- multiple passes and to use different sizes of data
11
The time trend was traditionally included in equations ex- monthly data, X might be set to 7 and 126400, respectively (see
plaining output as a measure of exogenous technical change. Microfit 4 Manual). For annual data, setting A=100 has the effect
With growing awareness of the importance of the time series of removing output cycles from the data with frequencies less
properties of variables being modeled (and particularly of the re- than eight years (Scacciavillani and Swagel, 1999).
13
gression residuals in the case of cointegrating regressions), the Guay and St. Amant (1996) find that the HP filter performs
use of the time trend has diminished. Some studies use direct poorly in identifying cyclical components of time series that have
measures of factors affecting technical change, such as research a spectrum or pseudo-spectrum with Granger's typical shape,
and development or human capital proxies, to account for techni- which is that of most macroeconomic time series.
cal change (see, for example, Adams and Coe, 1990). 14
In practice, we can use forecasts for a number of periods past
12
The choice of A, depends on the frequency of the data. For the end of the sample, which should improve the reliability of the
quarterly data, HP set A = 1600. Correspondingly, for annual and potential output measure toward the end of the sample period.

windows and observation weights. See Scacciavil- Wavelets Filters
lani and Swagel (1999) for more details and an ap-
plication of this filter. The wavelets methodology provides a "de-noising"
The RMS has the advantage of removing the ef- approach to extracting a series for potential output
fects of outliers that are not close to the particular that does not rely on arbitrary assumption regarding
smoothed value. It also allows for the possibility of the regularity of fluctuations. Rather, the approach
structural change because the window of data that is maps the observed data into more general functional
being smoothed shifts. spaces, the orthogonal bases of which are called
"wavelets." The appendix to Scacciavillani and
Swagel (1999) provides an overview of the wavelets
Cubic Spline-Smoothing Method (CSSM) theory and a technical discussion of how the filter
works, while the main paper provides an application
Cubic spline smoothing is popular in curve-fitting of the wavelets filter to deriving potential output,
applications and for interpolating data (e.g., quar- which is compared with the potential output from a
terly data from observed annual data). The cubic number of other approaches.
spline function partitions the data into N "knots." In
this case, the knots are simply periods. The knots
may be all the same length or may differ (e.g., at the Unobservable Components Methods
beginning and/or end of the period). A cubic spline
function, gi(Z), is then defined for ith knot, and each The unobservable components methods provide
cubic spline function is constrained so that where the a means of estimating unobservable variables, such
knots meet, the function values, slopes, and second as NAIRU and potential output, from the observed
derivatives are equal (to ensure a smooth and contin- data on output, inflation, and unemployment. The
uous curve). As in the HP filter, the cubic spline explicit relationships between the observed and un-
measure of potential output requires the specifica- observed variables are specified in what is called
tion of X, which provides the tradeoff between "state space" form in the measurement or observa-
smoothness and fit. The potential output measure, tion equation, which is a general way of represent-
g(Z) is then derived from the following optimization ing dynamic systems in which the observed vari-
problem: ables are specified in terms of the unobservable
variables or state variables. A separate system of
equations, the transition equation, specifies the au-
toregressive processes assumed to generate the
Min X ( (5) state variables. The unobservable variables can
t=0
then be estimated using what is known as the
Kalman filter.16 The unobservable components
The choice of X determines the polynomial order model has been developed for a univariate case
of the general solution. The choice of X can be deter- (decomposing output into a stochastic trend and a
mined by the data in a similar way to Coe and cyclical component) and multivariate case (relating
McDermott's approach by choosing X according to inflation, the output gap, and the unemployment
the generalized cross-validation (GCV) criteria, gap to derive NAIRU and a measure of potential
which essentially chooses X to minimize the out-of- output). For multivariate applications and further
sample forecast error on average.15 details of the methodology, see Apel and Jansson
The cubic spline approach is a smoothing method- (1999) and Cerra and Saxena (2000). The latter
ology which, like the other filtering methods, has the study discusses extensions, such as the inclusion of
advantage of a minimal data requirement. The knots common permanent components and the possibility
provide a means to allow for structural breaks and of asymmetric growth rates via the use of a latent
minimize the impact of outliers on the estimate of Markov-switching state variable.
potential output. The variable knot width allows for
the possibility of increasing or decreasing the num- This method has the very attractive feature of al-
ber of observations in the first and last knots. Thus, lowing the explicit specification of the relationships
reducing the size of the last knot might improve the between output, inflation, and unemployment, thus
measure of potential output and the output gap for providing theory-consistent estimates of potential
the most recent period. The spline-smoothing esti-
mator can be interpreted as a variable bandwidth
kernel estimator. 16
For a given set of starting values and model parameters, the
Kalman filter generates a sequence of optimal conditional predic-
tions of the observable variables. The prediction errors are then
15
For a fuller discussion of the use of cubic splines, the data de- used in a maximum likelihood routine to find the optimal set
termined choice of X, and an empirical application, see Khatri and of parameters and corresponding estimates of the unobservable
Solomou(1996). components.

output and the output gap.17 The requirement of an To derive potential output from the production
explicit specification of these relationships, how- function specification, the inputs and TFP have to be
ever, means that the measures of potential output set at their trend levels, consistent with full employ-
and output gap are contingent on this specification. ment and full capacity utilization. Labor input con-
Also, the method requires explicit assumptions on sistent with NAIRU (L*) could be derived by multi-
the form of the data-generating process for the ob- plying the labor force by (1-NAIRU). The trend
servable variables. For the univariate approach, component of TFP is also required and can be de-
Quah (1992) has shown that "without additional ad rived in a number of ways, including using smooth-
hoc restrictions, those characterizations are coming techniques described above.
pletely uninformative for the relative importance of Extensions to the basic methodology include the
the underlying permanent and transitory compo- refinement of inputs through quality-adjusting capi-
nents." The multivariate representations have fol- tal and labor; the use of more flexible functional
lowed, partly in response to this criticism, but they forms for specifying the production function; and
still maintain the assumption that the permanent the explicit modeling of productivity by including
component of output behaves like a random walk, technology-determining variables such as research
while in reality the dynamics of output are likely to and development, education, and spillovers (see
be much more complex (see Dupasquier, Guay, and Adams and Coe, 1990).
St. Amant, 1999). Furthermore, the results of the un- The production function approach has the attrac-
observable components method are often sensitive tive feature of determining potential output in a
to the initial "guesses" for the parameters. framework that can explicitly account for the contri-
butions of capital, labor, and TFP to output growth.
The production function, however, runs into a num-
Methods Employing Structural Relationships
ber of difficult conceptual and data problems, well
The Production Function Approach documented in the literature (see Griliches and
Mairesse, 1995, and references therein).
The production function methodology represents
the middle ground between a full-scale structural • The most basic criticism stems from the fact that
model and the various univariate approaches, such the production function may not be identified
as filters and unobservable components (De Masi, because of simultaneity problems.19
1997). The methodology in its simplest form in-
volves the estimation of a production function, most • The production function specification typically
often using the Cobb-Douglas form with two inputs, relies on an overly simplistic and probably re-
capital (K) and labor (L), and constant returns to strictive representation of the technology.
scale. Thus, the production function is specified as: • The method requires the estimation of NAIRU
lnYt = At + alnK + (1 - a)lnL, or some NAIRU-consistent level of labor (L*).
(6)
If smoothing techniques are used to derive (L*)
where At is the level of total factor productivity and for the trend TFP growth, then the problems
(TFP) and a and (1-a) are the output share of cap- of trend estimation for GDP have shifted to trend
ital and labor, respectively. The production function estimation of inputs (and TFP).
can either be estimated (e.g., as a cointegrating equa-
tion), or the value of a can be imposed (guided by • Smoothing techniques for L* and TFP also have
historical data on capital's or labor's share in out- the problem of unreliable endpoints, which will
put).18 The level of TFP can be derived residually thus affect the reliability of production function-
(Solow's residual) from (6). based estimates of potential output at endpoints,
and, importantly, for the most recent observation.
• The production function approach is consider-
ably more demanding in terms of data than fil-
17
Apel and Jansson (1999) argue that, although most econo- tering approaches, and the nature of the inputs
mists would agree that there is a close relationship between the data—particularly the capital stock calcula-
output gap and the unemployment gap and between these gaps
and the development of inflation, most studies disregard at least
one of these two conditions. They propose a multivariate unob- possibly even if using a system cointegration test (see Scacciavil-
servable component model that includes the first (Okun's law) lani and Swagel, 1999). Estimates of a might be obtained from es-
and the second (Phillips curve) relationship in the estimation of timating the production function in first differences.
19
potential output and NAIRU. If producers choose inputs to maximize profits after observ-
18 ing output and input prices, then the production function distur-
If the production function is specified as a cointegrating regres-
sion, and output, labor, and capital have unit roots, then the resid- bances will feed through into the choice of variable inputs; thus
ual-based TFP, if nonstationary, implies that a lack of a cointegrat- the exogeneity assumption required for estimating the production
ing regression will be obtained if using the Engle-Granger test, and function will not hold, and OLS estimates will be biased.

tions—implies significant potential for measure- employs a structural VAR in the sense that there are
ment errors in inputs. identifying restrictions imposed on the long-run ef-
fects of impacts on output and unemployment. These
• The production function is also likely to suffer long-run restrictions are used to identify structural
from omitted variable bias. supply and demand shocks by allowing supply
shocks to have a permanent effect on output, while
Demand-Side Models demand shocks have only a temporary effect on out-
put. The method has been extended, using the same
Bayoumi and others (1999) suggest that there are idea of imposing long-run restrictions for the purpose
two basic methods of estimating the output gap: de- of identification, to multivariate VARs and to the use
mand-side methods, which relate the output gap di- of alternative variables (see King and others, 1991;
rectly to measures of spare capacity in the economy Bayoumi and Eichengreen, 1992; Scacciavillani and
and supply-side measures, such as the production Swagel, 1999; and Cerra and Saxena, 2000).
function approach. Some researchers have combined
The approach is appealing, as it derives an esti-
these methods in simultaneous equation models, such
mate of potential output that employs a clear theoret-
as Adams and Coe (1990) or Blanchard and Quah
ical basis for the restrictions that identify permanent
(1993). Bayoumi estimates demand-side measures of
and transitory shocks. This method has several
the output gap by using a series of measures of slack
advantages.
in the economy—namely, the unemployment rate,
the ratio of job seekers to job offers, capacity utiliza- • It is not unduly restrictive in the dynamics im-
tion rates, a combination of these measures, and an posed on the permanent shocks that affect poten-
inverted Phillips curve. Measures of potential output tial output. Dupasquier, Guay, and St. Amant
and the output gap are derived from regressions of (1999) compare the structural VAR approach
the log of real GDP on the slack variable (or vari- with two alternative multivariate methodologies,
ables) together with a polynomial in the time trend, namely the Cochrane and multivariate Beveridge-
and regression of a standard Phillips curve, again in- Nelson approaches to deriving potential output,
cluding polynomial time trend, that is inverted so that and find that the dynamics of permanent shocks
output is on the right-hand side of the equation. are more complex than the random walk assumed
The proposed demand-side method is straightfor- by the other approaches.
ward and intuitively appealing. However, it does not • The approach allows the dynamics of permanent
take into account the time series properties of the shocks to be included in potential output. This is
variables. In particular, as mentioned above, output particularly appealing because one perverse im-
series tend to be integrated, difference stationary, and plication of defining potential output as a random
not trend stationary. Thus the use of a time trend, walk with drift is that, when the immediate effect
whether linear or polynomial, is not an appropriate of a permanent positive shock is smaller than the
means to isolate the stochastic permanent component long-run effect, the output gap (observed out-
of output (i.e., potential output) or detrend output to put—potential output) is negative until the full ef-
derive the output gap. Furthermore, while the output fect of the positive shock has fed through.
series is likely to be integrated, economic theory
would tend to suggest that the series for unemploy- • The potential output gap measures derived are
ment, capacity utilization rates, and the vacancy ratio not subject to end-sample biases or increased
are unlikely to be integrated in the long run. Thus, a uncertainty.
simple linear regression may be spurious, in the The structural VAR method also has certain
sense of Granger and Newbold (1974). disadvantages.
• The approach is limited in its ability to identify
Structural Vector Autoregressions (VARs)
different types of shocks (at most, there can be
the same number of types as variables used in
The structural VAR approaches that have been em- the VAR).
ployed recently to estimate potential output (see • In most applications, the method assumes uncor-
Scacciavillani and Swagel, 1999; Dupasquier, Guay, related supply and demand shocks. Theory pro-
and St. Amant, 1999; and Cerra and Saxena, 2000) vides numerous instances where shocks have
follow from Blanchard and Quah (1993). The general varying demand and supply characteristics
method combines aspects of both the Keynesian and (Cerra and Saxena, 2000). 20
neoclassical traditions in that it associates potential
output with aggregate supply and supply shocks, and
cyclical/transitory fluctuations with changes in ag- 20
For example, a technology shock may affect supply but may
gregate demand. The Blanchard and Quah method simultaneously increase demand through wealth effects.

Finally, although the approach is not demanding Dupasquier, Chantal, Alain Guay, and Pierre St. Amant,
in terms of data, the method is less straightfor- 1999, "A Survey of Alternative Methodologies for
ward to apply than many of the other approaches, Estimating Potential Output and the Output Gap,"
as it requires some nontrivial programming. Journal of Macroeconomics, Vol. 21, pp. 577-95.
Granger, Clive, and Paul Newbold, 1974, "Spurious Re-
gressions in Econometrics," Journal of Econometrics,
Vol. 2 (July), pp. 111-20.
References Griliches, Zvi, and Jacques Mairesse, 1995, "Production
Adams, Charles, and David Coe, 1990, "A Systems Ap- Functions: The Search for Identification," NBER
proach to Estimating the Natural Rate of Unemploy- Working Paper No. 5067 (Cambridge, Massachusetts:
ment and Potential Output for the United States," National Bureau of Economic Research).
Staff Papers, International Monetary Fund, Vol. 37 Guay, Alain, and Pierre St. Amant, 1996, "Do Mechanical
(February), pp. 232-93. Filters Provide a Good Approximation of Business
Apel, Michael, and Per Jansson, 1999, "A Theory- Cycles?" Technical Report No. 78 (Ottawa: Bank of
Consistent System Approach for Estimating Potential Canada).
Output and the NAIRU," Economics Letters, Vol. 64 Hardle, Wolfgang, 1990, Applied Nonparametric Regres-
(September), pp. 271-75. sion (New York: Cambridge University Press).
Bank Negara Malaysia, 1999, Annual Report (Kuala Harvey, Andrew C., and Albert Jaeger, 1993, "Detrending,
Lumpur). Stylized Facts and the Business Cycle," Journal of
Baxter, Marianne, and Robert G. King, 1995, "Measuring Applied Econometrics, Vol. 8 (July-September),
Business Cycles: Approximate Band-Pass Filters for pp. 231-47.
Economic Time Series," NBER Working Paper No. Hodrick, Robert J., and Edward C. Prescott, 1997, "Post-
5022 (Cambridge, Massachusetts: National Bureau of war U.S. Business Cycles: An Empirical Investiga-
Economic Research). tion," Journal of Money, Credit, and Banking, Vol. 29
Bayoumi, Tamim, and Barry Eichengreen, 1992, "Is There (February), pp. 1-16.
a Conflict Between EC Enlargement and European Jonsson Gunnar, 1999, "Inflation, Money Demand, and
Monetary Unification?" NBER Working Paper No. Purchasing Power Parity in South Africa," IMF
3950 (Cambridge, Massachusetts: National Bureau of Working Paper 99/122 (Washington: International
Economic Research). Monetary Fund).
Bayoumi, Tamim, and others, 1999, Japan—Selected Is- Khatri, Yongesh, and Solomons Solomou, 1996, "Climate
sues (Washington: International Monetary Fund). and Fluctuations in Agricultural Output, 1867-1913,"
Blanchard, Olivier, and Danny Quah, 1993, "The Dynamic University of Cambridge, DAE Working Paper No.
Effects of Aggregate Demand and Supply Distur- 9617 (Cambridge, Massachusetts).
bances," American Economic Review, Vol. 83 (June), King, Robert, and others, 1991, "Stochastic Trends and
pp. 653-58. Economic Fluctuations," American Economic
Cerra, Valerie, and Sweta C. Saxena, 2000, "Alternative Review, Vol. 81 (September), pp. 819-40.
Methods of Estimating Potential Output and the Out- Lim, Cheng H., and Laura Papi, 1997, "An Econometric
put Gap: An Application to Sweden," IMF Working Analysis of the Determinants of Inflation in Turkey,"
Paper 00/59 (Washington: International Monetary IMF Working Paper 97/170 (Washington: Interna-
Fund). tional Monetary Fund).
Charemza, Wojciech W., and Derek F. Deadman, 1992, Nelson, Charles, and Charles Plosser, 1982, "Trends and
New Directions in Econometric Practice: General to Random Walks in Macroeconomic Time Series:
Specific Modeling, Cointegration and Vector Autore- Some Evidence and Implications," Journal of Mone-
gression (Brookfield, Vermont: Edward Elgar). tary Economics, Vol. 10 (September), pp. 139-62.
Chhibber, Ajay, and others, 1989, "Inflation, Price Con- Quah, Danny, 1992, "The Relative Importance of Perma-
trols, and Fiscal Adjustment in Zimbabwe," World nent and Transitory Components: Identification and
Bank Policy, Planning, and Research Working Paper Some Theoretical Bounds," Econometrica, Vol. 60
No. 192 (Washington: World Bank). (January), pp. 107-18.
Coe, David, and John McDermott, 1997, "Does the Gap Razzak, Weshah, 1995, "The Inflation-Output Trade-Off:
Model Work in Asia?" Staff Papers, International Is the Phillips Curve Symmetric? Evidence from New
Monetary Fund, Vol. 44 (March), pp. 59-80. Zealand," Reserve Bank of New Zealand Discussion
Cogley, Timothy, and James Nason, 1995, "Effects of the Paper G95/7 (January), pp. 1-17.
Hodrick-Prescott Filter on Trend and Difference Sta- Scacciavillani, Fabio, and Phillip Swagel, 1999, "Mea-
tionary Time Series: Implications for Business Cycle sures of Potential Output: An Application to Israel,"
Research," Journal of Economic Dynamics and Con- IMF Working Paper 99/96 (Washington: International
trol, Vol. 19 (January-February), pp. 253-78. Monetary Fund).
de Brouwer, Gordon, and Neil Ericsson, 1998, "Modeling Soderlind, Paul, 1994, "Cyclical Properties of a Real Busi-
Inflation in Australia," Journal of Business and Eco- ness Cycle Model," Journal ofApplied Econometrics,
nomic Statistics, Vol. 16 (October), pp. 433-49. Vol. 9 (December), pp. 113-22.
De Masi, Paula, 1997, "IMF Estimates of Output: Theory Stock, James, and Mark Watson, 1999, "Forecasting Infla-
and Practice," IMF Working Paper 97/177 (Washing- tion," Journal of Monetary Economics, Vol. 44 (Octo-
ton: International Monetary Fund). ber), pp. 293-335.

IV Challenges to Fiscal Management
Olin Liu
A lesson from the Asian crisis is that the effec-

tiveness of Malaysia's fiscal policy in demand
management may benefit from greater flexibility.
government played an important role in mobilizing
public and private resources to finance its develop-
ment programs, including through off-budget and
While Malaysia's tradition of fiscal prudence pro- quasi-fiscal activities. In support of the development
vided a buffer against adverse shocks during the strategy, fiscal consolidation was conducted mainly
crisis, structural rigidities in public finances ren- through the tightening of the operating budget that
dered fiscal policy less responsive to the changing led to current surpluses, which in turn helped to
economic environment. A key challenge to fiscal moderate overheating pressures on the economy
management is to make it flexible enough to func- arising from accelerated money and credit growth,
tion as an effective countercyclical tool within the together with a surge in capital inflows.
framework dedicated to maintaining long-term fis- In the context of the consolidation, Malaysia's fis-
cal sustainability. cal conservatism was reflected in the rule of disal-
Remedies for improving fiscal flexibility could in- lowing an "operating deficit" in the annual budget to
clude extension of the fiscal planning horizon to help strengthen the financial position of the federal
cover the course of the business cycle, to be sup- government. While revenue performance deter-
ported by well-defined safeguards and escape mined the level of recurrent spending, the capacity
clauses (for fiscal rules); more realistic economic to raise nonbank financing determined the size of
forecasts and revenue projections as the basis for capital spending, thereby minimizing budgetary fi-
setting fiscal targets; a structural fiscal balance con- nancing from the banking system. This rule, together
cept; and greater use of automatic stabilizers and with a tradition of conservative revenue projections,
contingency measures in the budget.
Introduction and Recent Figure 4.1. Fiscal Indicators

Developments (In percent of GDP)
During the 1980s, Malaysia experienced large fis-

cal deficits and high public debt-to-GDP ratios. Fis- 120
cal consolidation, begun in the mid-1980s, aimed at Federal government debt
(left scale)
reducing imbalances with the support of strong gov- 100
ernment commitment to a rule-based fiscal policy, Federal government
privatization, and reorientation of capital spending fiscal balance
80 -5
to basic infrastructure development and other pro- (right scale)
ductive sectors. During the period of fiscal consoli-

dation (1992-97), fiscal policy was geared more to- 60 -10
ward achieving the medium-term development
objectives and was used, to a lesser extent, as a tool 40 -15
for demand management (Figure 4.1).
Public programs were formulated within the 20 -20
framework of five-year development plans aimed at
promoting economic growth over the medium term
through capacity expansion; improvements in pro- 1981 83 85 87 89 91 93 95 97 99
ductivity; and adequate provision of public infra-
structure, utilities, and other modern services. The

IV CHALLENGES TO FISCAL MANAGEMENT
ened market confidence. Against this background,

Figure 4.2. Fiscal Budget and Actual the government readjusted its policy mix toward an
Balance expansionary policy and has maintained an expan-
(In percent of GDP) sionary stance since mid-1998 aimed at reactivating
economic growth and, thereby, ensuring socioeco-
4 4
nomic stability. A fiscal stimulus package was intro-
duced that included a social safety net to mitigate the
Actual effects of economic adjustment on the most vulnera-
2 - 2 ble groups, as well as large increases in capital
spending, especially for infrastructure and other so-
cial sector projects.
The countercyclical fiscal policy has been deliv-
ered largely through discretionary measures. The fis-
cal stimulus package of 1998-2001 focused on se-
lected capital projects designed to ensure maximum
effect on economic growth, minimum leakage to im-
ports, large export potential, and a short gestation
period in terms of providing value-added to the
economy. Large projects were restructured into
-8
1990 92 94 96 98 2000
smaller segments to facilitate financing. Beyond the
budget, the government also provided fiscal impulse
through off-budget operations, facilitated by the
quasi-fiscal institutions. Existing credit schemes
were expanded, and new specialized funds were set
up to ensure continued access to credit at reasonable
led to large operating—and overall—surpluses each cost for priority sectors and projects.
year during 1990-97 (Figure 4.2). The strengthened The impact of the expansionary fiscal policy was
fiscal position served the country well during the en- initially weaker than expected and came with a sig-
suing crisis, in that it provided substantial scope for nificant lag (1998-99). Spending cuts implemented
a countercyclical fiscal policy and funding for finan- earlier could not be reversed easily. The fiscal rule
cial sector restructuring. of disallowing an operating deficit also limited the
At the outset of the crisis, Malaysia tightened scope of using fiscal policy as a tool for demand
macroeconomic policies to reduce vulnerability. management. In 1998 and 1999, the fiscal impulse,
The policy stance, together with a preemptive re- estimated at about 2 percent and 3 percent of GDP,
form of the financial sector, was intended to help respectively, was well below the budget targets (by
contain inflation and the current account deficit, re- 2 percent and 1 percent of GDP, respectively),
duce credit growth, and dampen the pressure on do- which explained in part the slow turnaround in do-
mestic demand. It was also expected to instill salu- mestic demand (Box 4.1). Fiscal stimulus was en-
tary effects on finances and market confidence. hanced in 2000 through both revenue and expendi-
Eventually, the 1998 budget was revised to reflect a ture measures. The budget deficit of 4.3 percent
rebalancing of policies toward a smaller surplus.1 of GDP, and possibly the lagged effects from pro-
In anticipation of declining revenue in line with an jects initiated under previous years' budgets, pro-
economic downturn, the surplus was to result from vided a strong impetus to the recovery in domestic
lower current outlays and deferred infrastructure demand.
projects. These measures were also expected to Despite the larger fiscal deficit of the government,
help reduce imports and contain the high leverage the overall financial position of the public sector was
of corporations. contained by the strengthened balance sheets of the
Notwithstanding these policy initiatives, market nonfinancial public enterprises and benefited largely
confidence faltered markedly against broader adverse from banking and corporate restructuring, a signifi-
developments in the region, including concerns about cant turnaround in economic activities, and high oil
financial system vulnerability. Economic activities prices.
also slowed much faster than anticipated, which A key challenge is to ensure that fiscal policy in
proved to be a significant factor that further damp- Malaysia can contribute to effective demand man-
agement. This would require greater flexibility
in fiscal policy to respond to domestic and external
1This degree of fiscal easing, which was supported by the IMF, macroeconomic shocks in an appropriate and
turned out to be insufficient. timely way, while maintaining fiscal prudence to

Introduction and Recent Developments
Box 4.1. Fiscal Policy in Malaysia, 1997-2000
Following the initial stage of tighter fiscal policy at Fiscal policy in Malaysia shifted to an expansionary
the outset of the crisis, the government rebalanced its stance in 1999 in support of economic recovery. The
approach through policy easing in response to a sharp fiscal stimulus provided through the budget and off-
decline in domestic demand that became evident by budget operations helped to offset in part a continued
mid-1998. Targeted budget surpluses were reduced, decline in private demand. The fiscal impulse, includ-
some previously announced cuts in social expenditures ing off-budget stimulus and estimated at about 31/2per-
were reinstated, and additional capital spending was cent of GDP, was less than the budget target of 41/4per-
implemented. Off-budget spending, most notably cent of GDP and was backloaded to the second half of
through privatized infrastructure projects, also pro- the year. On a cumulative basis, the fiscal impulse in
vided additional fiscal impetus. Malaysia during 1997—99 became larger than in the
However, the countercyclical fiscal policy was other Asian crisis countries (see figure, below).1
adopted later in Malaysia than in other Asian crisis
countries. Fiscal policies, which had been contrac-
tionary at the eve of the crisis, were eased in Korea Cumulative Fiscal Impulse of
and Thailand and subsequently shifted to expansion- Selected Asian Countries
ary stances in late 1997 (in Thailand) and early 1998 (In percent of GDP)
(in Korea). In Malaysia, by contrast, the fiscal stance
remained significantly contractionary throughout
most of 1997 (i.e., a negative fiscal impulse of 1.3
percent of GDP); this was consistent with the advice
of the IMF at that time. The shift in fiscal policy took — Malaysia
place in mid-1998. Even then, the actual fiscal im- — Thailand
pulse was smaller than the budgetary target (by about • Philippines
2 percent of GDP on average during 1998—99), owing • Korea
in part to traditionally conservative revenue projec-
tions in the budget and insufficient flexibility in fiscal
management (see figure, below).
Overall Fiscal Balances of Selected

Asian Countries
(In percent of GOP)
-2
1997 99 2000
1Thefiscal impulse for Thailand is calculated on a fiscal year

basis (October - September).
Still, the recovery in domestic demand in Malaysia

remained weak in 1999, which prompted another fiscal
stimulus in 2000. The budget deficit reached 4.3 per-
cent of GDP. The economy responded with an impres-
sive real growth of81/2percent. Domestic demand also
mounted a strong recovery, growing by more than 15
percent.
1995 96 97 98 99 2000
1Ingeneral, the fiscal impulse is used for analyzing the fis-
cal stance under a stable economic environment; therefore,
this information should be interpreted with care.
ensure medium-term fiscal sustainability and bility in the context of a medium-term fiscal strat-
growth prospects. The remainder of this section re- egy, and that explores operational guidelines to im-
views policy options that would provide such flexi- prove policy effectiveness.

Fiscal Policy Rules and Recent on discretionary measures, which can exert pressure
Experience on the implementation capacity. Procedural bottle-
necks, insufficient financing, and long implementa-
Malaysia follows an explicit fiscal policy rule that tion periods for new projects also reduce the timeli-
disallows an operating deficit in any given year. This ness of the intended fiscal impact.
rule aims at making a credible commitment to long- The government has made various adjustments to
term fiscal sustainability by applying discipline to overcome these constraints during the course of the
annual budgets. Other fiscal rules that have a longer crisis. Monitoring of both revenue and expenditure
time horizon are also at work and are predetermined in particular was stepped up through monthly meet-
in the medium-term framework of Malaysia's five- ings of the high-level Cash Flow Committee. In
year development plan to ensure broad consistency order to speed up spending, ministries were given
with the rule on the operating balance. These are the flexibility to move funds within the same eco-
nomic categories. Contingency reserves were used
• a balanced budget rule under the seventh five- for capital spending with the approval of the finance
year development plan (1996-2000), revised in minister. The midterm review of the seventh five-
mid-1998 in the context of the fiscal stimulus year development plan (completed in mid-1998) was
plan, that targets an overall federal government particularly timely in redirecting fiscal policy to-
deficit of no more than 6 percent of GNP by the ward stimulating the economy in the wake of a sharp
end of the plan period; contraction.
• a spending rule that places an aggregate ceiling The government also used off-budget operations
on capital expenditure during the seventh five- and quasi-fiscal activities to reinforce the fiscal im-
year plan period; and petus. A specialized fund was established to provide
• a borrowing rule that sets a ceiling on the total financing to "privatized" infrastructure projects.
outstanding stock of federal government debt During 1997-99, a total of 16 large projects (mostly
during the seventh five-year plan, with a subceil- for infrastructure development) were selected by the
ing on external federal debt. government for implementation through the
Malaysia Infrastructure Development Bank. Addi-
Overall, the rules are well defined as to the target tional loans were obtained mostly from quasi-
variables and compliance criteria, which are ap- government entities and bilateral external borrowing
proved by the parliament and are broadly enforce- (under the Miyazawa initiative) with explicit guar-
able. Internal consistencies of the fiscal rules with antees from the government (Appendix I).
other macroeconomic policies—such as low infla-
tion, sustainable government debt, and noninflation-
ary financing—are safeguarded by the medium-term Enhancing Fiscal Policy Responsiveness
policy framework. Compliance with fiscal rules is
enforced mostly through administrative means, al- to Business Cycles
though deviation from the rules requires parliamen- Fiscal policy can be effective and countercyclical
tary approval. There is no financial penalty for if the rules are so designed and institutional arrange-
nonobservance. Nevertheless, a violation of the rules ments are in place to enforce them. The key issue in
could entail loss of credibility of the government in designing fiscal rules is the balance between the
the eyes of parliament and financial markets, as well short-term need for growth and employment and the
as the public. long-term desire for fiscal sustainability. To achieve
Notwithstanding fiscal discipline, the rules pre- an appropriate balance, flexibility is critical in the
clude the use of cyclical indicators and escape formulation and application of these rules, while
clauses, thereby constraining fiscal policy from be- clearly defined guidelines and escape clauses will
coming an effective countercyclical tool. Further- safeguard the credibility of the authorities in pursu-
more, the rules are assessed against fiscal targets that ing fiscal prudence over the medium term.
do not cover off-budget and quasi-fiscal operations,
which lead to nontransparency in accounting and fis-
cal forecasting, as well as difficulty in assessing the Formulation of Fiscal Rules
total impact of fiscal policy and in estimating future Flexibility can be incorporated into fiscal rules by
government obligations. the following methods:
While effective in imposing fiscal discipline,
these rules have proved to be procyclical during the • Expanding the horizon for budget formulation,
recent economic downturn. Conservative forecasting including the application of fiscal rules, to cover
of tax revenues constrains the effectiveness of fiscal the course of a business cycle would provide the
planning. Capital spending on projects tends to rely economy with improved shock-absorptive ca-

Enhancing Fiscal Policy Responsiveness to Business Cycles
pacity. For instance, the fiscal rule on the annual triggered during budget execution if actual bud-
operating budget could be modified to allow an get performance deviates significantly from the
operating deficit during an economic downturn planned path.
while observing the balance (on average) over
the course of a business cycle. This would re-
quire a medium-term fiscal framework for plan- Greater Flexibility in Fiscal Policy
ning, analysis, and forecasting (Box 4.2). Management
• Setting fiscal targets on the basis of a structural Fiscal consolidation over the medium term will
balance concept would allow revenue and ex- require a shift from fiscal stimulus to a significant
penditure to adjust automatically for deviations scaling back of development expenditure. It is a
in output. Given the authorities' commitment to challenge to withdraw the stimulus without major
fiscal consolidation over the medium term, there disruption to economic activities. Capacity for pol-
is merit to considering the use of the aggregate icy analysis and forecasting will need to be strength-
budget deficit as a target variable, supplemented ened, taking into consideration the linkages between
by a subceiling on the operating budget. The fiscal and monetary policies, desired rate of growth,
pace of fiscal consolidation should follow a inflation, and other macroeconomic variables that
carefully mapped convergence plan to a struc- influence budget preparation and execution. Shifting
tural balance. gears with minimal disruption to the economy could
involve several initiatives.
• Enhancing the role of built-in stabilizers would
help to strengthen the countercyclical influence • Developing a medium-term fiscal framework as
of fiscal policy while limiting the use of discre- an integral part of the macroeconomic frame-
tionary measures.2 More realistic revenue pro- work. This fiscal framework should include an
jections can also facilitate assessment of the ef- analysis of the budget sensitivity to exogenous
fects of built-in stabilizers. On the expenditure shocks and help identify major fiscal vulnerabil-
side, the introduction of a well-targeted social ities. It should also incorporate forward-looking
safety net and provision of unemployment bene- estimates on resource constraints based on actual
fits could greatly improve the responsiveness of and projected macroeconomic variables. Cycli-
fiscal policy. cal indicators should be identified to reflect busi-
ness cycle behavior. Contingency measures
• Establishing safeguards and escape clauses (e.g., increases in current and capital expendi-
would allow fiscal policy to respond to large un- tures or tax measures) could be formulated in the
foreseen shocks. The choices of cyclical indica- context of this medium-term framework, which
tors and conditions that would invoke the escape would be regularly updated and applied on an
clauses need to be defined in advance and as pre- annual rolling basis.
cisely as possible. The criteria should be trans-
parent so that the timing for triggering the es- • Assessing the fiscal policy stance based on con-
cape clauses are known to the public. Choices of solidated public accounts. Provision of fully
cyclical indicators should be sensitive to the consolidated details of off-budget and quasi-
business cycle (such as the level of private in- fiscal activities for both revenue and expendi-
vestment) and should be able to be monitored. tures would greatly enhance transparency and
To be effective, the coverage of the budget allow the full impact of fiscal policy to be ana-
should include all off-budget and quasi-fiscal lyzed. The netting of certain expenditures should
operations. be replaced by transactions on a gross basis.
• Introducing contingency measures during the • Improving the quality and reliability of revenue
budget process, either to add stimulus or with- forecasts by adopting a more systematic,
draw it as required, could include the elimination model-based approach. A simple structural
(or imposition) of a surtax and introduction of a model can be developed to cover as many dif-
stabilization fund. An across-the-board increase ferent components of revenue as feasible, par-
(or cut) in capital spending, although effective, ticularly the major taxes. In this connection,
should be used only as a last resort. These and there are two basic forecasting methods: the ef-
other measures, specified in advance, can be fective tax rate method and the revenue elastic-
ity method. Time series on various taxes, par-
ticularly income taxes for individuals and
2
Although Malaysia recently took measures to enhance the sta- corporations, need to be updated with a view to
bilizers on the revenue side by shifting to current year income tax developing a microsimulation model for in-
assessment, stabilizers on the expenditure side remain absent. come tax forecasting purposes.

Box 4.2. Fiscal Policy Responses to the Business Cycle
There are two schools of thought. In line with the oped a tendency to underestimate revenues in the bud-
Keynesian theory, one group argues that the formulation get. The procyclical nature of the fiscal policy was re-
of fiscal policy should be responsive and countercycli- versed in 1998-99 through the provision of ad hoc
cal. Hence, one would observe a positive correlation be- countercyclical fiscal stimulus (see figure, lower right-
tween tax revenues and output, but a negative correlation hand side).
between government spending and output. During an
economic slowdown, the government should lower taxes
and increase spending to stimulate the economy. The
counterargument inspired by Barro (1979), however, Real GDP and Real Private
says that fiscal policy should be neutral over the business Consumption
cycle and respond only to unanticipated shocks that af- (1991 = 100)
fect the government's budget constraint.
A study by Vegh and Talvi (2000) finds that fiscal
policy in Malaysia follows neither the countercyclical
nor the neutral-response views. In fact, based on a sam-
ple of 56 countries (1977-94), the study shows that in
developing countries (including Malaysia) government
spending and taxes are highly procyclical. The authors
consider this as the outcome of the variability of the tax
base. They find that the tax base fluctuations along
with the business cycle are much larger in developing
countries than in the G-7 countries. This would imply
that, if the Barro prescription were followed, full tax 60 -
smoothing would result in large budget surpluses in a
boom and large budget deficits in a bust. Vegh and 40 -
Talvi argue, however, that the political economy could 20 -
make it costly to run large fiscal surpluses, owing to the
pressure to increase public spending in particular on 0
1991 92 93 94 95 96 97 98 99
government consumption and nonproductive invest-
ments. Thus, developing countries may choose pro-
cyclical fiscal policy as their optimal fiscal response to
favorable shocks in their tax base. Their argument is in
contrast with the standard explanation for procyclical
fiscal policy in developing countries that builds on the
imperfect access to international credit markets during
unfavorable times. They suggest that it is the inability Real GDP and Real Government
of the government to generate large enough surpluses Revenue
during expansion that forces it to borrow less during re- 1991 = 100)
cession (see figure, upper right-hand side).
The study suggested that in Malaysia the budget rev-
enue, government consumption, and private consump- 180
Real GDP
tion are procyclical.1 Their correlation coefficients with
output are estimated at 0.64, 0.54, and 0.77, respec- 160
tively (corresponding averages for the Asian crisis 140
countries were 0.66, 0.56, and 0.62, respectively). The
procyclical nature of fiscal policy is also found in other 120
developing countries in the sample, although the de- 100
grees of correlation are nonhomogenous. Evidence Real government expenditure
shows that in Malaysia economic expansion during 80

1991—97 led to real increases in government revenue, 60
but these grew at slower paces than real GDP. Higher
revenue exerted pressure for larger government spend- 40
ing. To minimize such pressure, the government devel-
20
0 1991 92 93 94 95 96 97 98 99
1
Estimation is based on real output with the cyclical com-
ponents in budget revenue, government consumption, and pri-
vate consumption (all in real terms).

• Adopting a forward-looking approach to fiscal budget implementation. Fiscal flexibility would also
management by incorporating fiscal monitoring be enhanced by the effective use of safeguards and
with forecasting. This would help identify possi- escape clauses, realistic revenue projections, and a
ble deviations in fiscal policy stance from the in- greater role for automatic stabilizers and contingency
tended path. Tax and expenditure policy options measures.
should be reviewed and updated on a timely Fiscal flexibility needs to go hand-in-hand with
basis, excluding windfall revenues and nonre- greater transparency. This can be achieved by requir-
curring expenditures, while taking into account ing the government to be explicit about its objec-
medium-term macroeconomic constraints. It tives, ensuring thereby that short-term fiscal policies
would also be helpful to initiate procedures to are consistent with long-term macroeconomic goals.
review regularly the sources of errors between Greater flexibility and transparency will require
actual outcome and the quarterly forecast for timely updating of forecasts that highlight the
major budgetary components. Errors could then changes in external and domestic conditions. Con-
be analyzed against possible attributes to solidation of off-budget and quasi-fiscal operations
changes in policies, revisions in macroeconomic are also important.
assumptions, and/or technical pitfalls in the fore-
casting model.
• Developing mechanisms for identifying the most Appendix 1. Off-Budget and
and least efficient and effective spending pro- Quasi-Fiscal Operations
grams, based on economic and social criteria.
Results could be disseminated to line ministries For the past few years, the Malaysian government
as potential candidates for expenditure adjust- has influenced overall public sector finances through
ments. In this connection, high value-added and off-budget and quasi-fiscal operations. These activi-
quick-yielding programs might be identified and ties are often carried out by semi-government enti-
targeted according to expenditure policy priori- ties and other public financial institutions. The
ties; this is particularly important where there is framework for these activities has developed as a re-
a need for additional spending. sult of significant government involvement in the
past that was essential for achieving the social objec-
• Allowing expenditure switching among sectors tives of rebalancing the ownership structure of the
or ministries, such as increases in social sector economy. The nature of off-budget and quasi-fiscal
expenditure or public maintenance, together operations in Malaysia is mostly related to the gov-
with reductions in less productive spending. ernment's role, including that of Bank Negara
Avoiding across-the-board increases (or reduc- Malaysia, as planner of the economy and regulator
tions) as much as possible will limit adverse eco- of the financial system. The government entrusts the
nomic and social impact and the loss of effi- central bank and other semi-government institutions
ciency. Contingent reserves may also be used, to provide loans for strategic investment plans and to
with the amount adjusted on an annual rolling finance large infrastructure projects, financial and
basis, to serve as a buffer against unexpected corporate sector restructuring, and operations of the
shocks in the future. informal social safety net.
Despite their potentially significant macroeco-
nomic and financial impact, as well as allocative ef-
Conclusions fects on the economy, off-budget and quasi-fiscal
operations are currently outside the consolidated fis-
Experience in recent years suggests that greater cal accounts of the public sector. Contingent liabili-
flexibility in fiscal management will help reduce the ties associated with these operations are nonetheless
economic and social costs of a crisis through timely moderate, based on identified information.
adjustments. This flexibility must be guided by well- Major off-budget and quasi-fiscal operations in
defined fiscal rules that aim at enhancing the counter- Malaysia include the following:3
cyclical nature of fiscal policy. Specifically, greater
flexibility can be achieved through a rules-based • provision of concessional lending for privatized
framework, with the horizon extended to the medium infrastructure projects through the operation of
term. This framework should have a well-defined fis- the Infrastructure Development Fund, a fund es-
cal target (i.e., a structurally adjusted balance). Tem-
porary deviations from the target will be permitted,
however, under clearly defined guidelines to accom- 3
In addition, the government's contingent liabilities arise from
modate unforeseen adverse shocks. Political support the provision of bank deposit guarantees, which were made ex-
is critical, as well as institutional arrangements for plicit in January 1998.

tablished in 1998 to provide off-budget fiscal ernment guarantees, and thus implicit contingent
stimulus through the implementation of large in- government liability related to certain privatized
frastructure projects; projects, is very limited.
• activities of Khazanah—an entity initially cre-
ated with the capital from privatization proceeds Privatized Infrastructure Projects
of public enterprises—that acts as an investment Most off-budget operations were associated with
arm of the government for strategic development the implementation of large privatized infrastructure
projects and other activities of national interest;4 projects. Sixty-eight such projects were implemented
• financial restructuring of banks, which com- during 1996-98, with a total cost of RM 67 billion.
menced in 1998, through the operations of Dana- Financial support from the federal government in
harta (to acquire nonperforming loans) and terms of advances for land and loans was provided
Danamodal (to recapitalize banks); for major infrastructure projects.5 Initial seed capital
was provided from the budget for these projects to
• provision of explicit government guarantees of help secure long-term commercial loans at favorable
loans extended by the Employees' Provident terms,6 with the objectives of improving operating
Fund, a public pension fund, and various state- efficiency, saving budgetary resources, reducing ad-
owned trust funds to nonfinancial public enter- ministrative burden, and promoting private sector
prises and some privatized projects; bonds is- participation in the economy. These projects involved
sued by Khazanah (up to RM 10 billion); and mainly the construction of highways, mass-transit
obligations of Danaharta; systems, sewage systems, and other utilities for so-
• provision of explicit government guarantees of cial sector development.
external loans undertaken by major nonfinancial The Infrastructure Development Fund was estab-
public enterprises that involve large projects of lished in 1998 and was aimed at assisting with the
national interest, and other bilateral and multilat- financing of those privatized infrastructure projects
eral external loans to public entities, including whose implementation was significantly delayed in
the state-owned financial institutions (e.g., the aftermath of the crisis. The Infrastructure De-
Malaysia Infrastructure Development Bank); velopment Fund had an initial capital of RM 1 bil-
lion and is managed by the Malaysia Infrastructure
• extension of subsidized loans by Bank Negara Development Bank. The government provided ex-
Malaysia for the operation of the informal social plicit guarantees for both domestic and external
safety net, including low-cost housing; and loans undertaken by the Malaysia Infrastructure
• potential government burden sharing in resolv- Development Bank, which amounted to nearly
ing the financial troubles of certain privatized RM 6 billion in 1999. Most of these loans are from
projects as well as other quasi-fiscal operations the Employees' Provident Fund (RM 4.5 billion)
associated with the activities of financial public and bilateral sources (RM 1.3 million, under the
enterprises. Miyazawa initiative).
Off-budget capital spending was stepped up in
Among the above-mentioned operations, financial 1999 to provide additional fiscal stimulus to the
information is mostly available on (i) the operations economy. The government handpicked 16 major
of Danaharta and Danamodal; (ii) explicit govern- projects (mainly in infrastructure, utilities, and so-
ment guarantees of the external loans of major non- cial services), amounting to about RM 9 billion (3
financial public enterprises undertaking large na- percent of GDP), to be financed by the Malaysia In-
tional projects; (iii) explicit government guarantees frastructure Development Bank as part of the off-
of domestic loans extended by the Employees' Prov- budget fiscal stimulus. Two-thirds have been final-
ident Fund (and other trust funds) to major nonfinan- ized, and about RM 1 billion is expected to be dis-
cial public enterprises; and (iv) Bank Negara bursed annually during 1999-2001.
Malaysia operations associated with the informal so- More recently, some of these large projects are
cial safety net. Information related to the operations facing financial difficulties that may have implica-
of Infrastructure Development Fund and Khazanah,
however, is partial, and information of implicit gov-
5
Most of the new and major projects were privatized through
the build-operate-transfer method.
4 6
As of end-1999, Khazanah's portfolio included its interests in Interest rates were in the range of 6-8 percent with an average
Tenaga (36 percent of equity); Telekom Malaysia (37 percent); grace period of 6 years and a repayment period of 15-18 years.
Malaysia Airports Holdings Bhd. (23 percent); and Putrajaya These are in line with the terms of the government loans for these
Holdings Sdn. Bhd. (40 percent). projects.

Table 4.1. Quasi-Fiscal Contingent Liabilities
(Outstanding stock at end of period, in percent of GDP)
1997 1998 1999
Federal government guaranteed loans 8.3 11.6 16.8

Domestic 5.0 7.1 8.3
Foreign 3.2 3.4 3.7
Additional identified debt guaranteed by the
government1 0.1 I.I 4.8
Guaranteed loans for the Malaysia Infrastructure
Development Bank 0.1 0.2 0.5
Guaranteed loans for Danaharta and Permodalan
Nasional Bhd. 0.9 3.6
Guaranteed loans for Petroliam Nasional Bhd. 0.7
Sources: Data provided by the Malaysian authorities; and IMF staff estimates.
1Excluding Danamodal bond issue of RM 7.7 billion (about21/2percent of GDP) as of February 15, 2000.
tions to the budget. The issue has caught the atten- Contingent Liabilities of the Government
tion of the public, who has questioned the viability
of these projects, as well as the governance issue Provisions of explicit guarantees to off-budget
related to government subsidies (both explicit or capital spending and operations of Danaharta have
implicit) in the form of budget spending or soft raised government contingent liabilities7 in recent
loans. years. Although data on these activities are not sys-
There are no explicit government guarantees as- tematically reported in a consolidated manner, some
sociated with these projects. Given the nature of information has been identified (Table 4.1).
these projects, however, the government has indi-
cated that it may consider some form of financial
support, depending on the merit of each case. Appendix II. Fiscal Measurement:
Modalities could include a third-party takeover, buy Methodology Note
back at discounts, and/or a rescue operation through
cash injection. In the event of a government Definition of Cyclically Neutral Balance,
takeover, outstanding loans of the government Fiscal Stance, and Fiscal Impulse
would be automatically converted to equity holding Cyclically neutral balance measures the fiscal
in the project. Liquidation is not likely, as these pro- position relative to a base year; it is defined as the
jects involve national infrastructure and social ser- difference between the cyclically neutral revenue
vices. The manner in which the government handles and expenditure. Cyclically neutral revenue repre-
these projects, including the extent to which best sents a constant revenue-to-nominal-GDP ratio,
practices are followed, is likely to have significant while cyclically neutral expenditure represents a
implications for public finances and development constant expenditure-to-potential-GDP ratio (ex-
programs in the future. cluding unemployment benefits), both relative to
The government is working on improving the the base year.
framework within which privatized projects are
regulated particularly with respect to operating Fiscal stance measures the difference between the
standards and quality of services, to ensure adher- actual balance and the cyclically neutral balance. A
ence to terms and conditions stipulated in the priva- positive fiscal stance (i.e., a situation where the ac-
tization agreements. One proposal is to establish tual deficit is larger than the cyclically neutral
sectoral regulatory bodies, initially to cover energy deficit) indicates that fiscal policy is adding stimulus
and gas, transportation, telecommunications, and to the economy.
water supply and sewerage systems. Under such a
framework, the emphasis will be on penalties for
7Information on contingent liabilities of the government asso-
noncompliance with the terms and conditions of ciated with the provision of implicit guarantees is yet to be com-
the agreements. piled and consolidated.

Fiscal impulse is defined as the change in the fis- AR(t)* (Output Gap)t
cal stance; it provides a sense of direction and the SR(t) =
(Output Gap)Pt-1
magnitude of the new fiscal stimulus to the econ-
omy. While the fiscal stance depends critically on where AR(t) = actual revenue in year t; Output
the neutral balance and, thus, the selection of the gap = (potential output)/(actual output); a = rev-
base year, the fiscal impulse is not sensitive to this enue elasticity (weighted average) in time t;B=
problem, which makes it a more useful indicator revenue elasticity (weighted average) in time t-\.
than the fiscal stance.
• Calculating the structural expenditure, SE(t), in
year t:
Calculation of Cyclically Neutral Balance,
Fiscal Stance, and Fiscal Impulse
SE(t) =AE(t)-UB(t)* [1-Structural Unemployment(t)],
• Selecting the base year t0; R(t0) — E(t0) is the
cyclically neutral balance in year t0; Actual Unemployment
where R(t0)= cyclically neutral revenue in year t0;
E(t0) = cyclically neutral expenditure in year t0. where AE(t) = actual expenditure in year t;
• Calculating the cyclically neutral revenue for UB(t) = unemployment benefits in year t.
current year t: • Calculating the structural balance, SB(t), in
Neutral Revenue = Y(t)*r, year t:
where Y(t) = nominal GDP in year t; r = the GDP SB(t) = SR(t) - SE(t).
share of revenue in the base year t0.
The structural balance is often measured in
• Calculating the cyclically neutral expenditure terms of potential GDP.
for current year t:
AE(t0) - UB(t0),
Neutral Expenditure = References
Y(t0)
where P(t) = potential nominal GDP in year t; Alesina, Alberto, and Tamim Bayoumi, 1996, "The Costs
Y(t0)— nominal GDP in year t0; AE(t0) = actual and Benefits of Fiscal Rules: Evidence from U.S.
expenditure in t0; UB(t0) = unemployment bene- States," NBER Working Paper No. 5614 (Cam-
fit in t0. bridge, Massachusetts: National Bureau of Economic
Calculating the fiscal stance, S(t), in year t: Research).
Backus, David, Patrick Kehoe, and Finn Kydland, 1995,
S(t) = [R(t0) - E(t0)] - [rY(t) - eP(t)], "International Business Cycles: Theory and Evi-
where R(t0) — E(t0) = the cyclically neutral bal- dence," in Frontiers of Business Cycle Research, ed.
ance in yeart0;rY(t) — eP(t) = the cyclically neu- by Thomas F. Cooley (Princeton, New Jersey: Prince-
tral balance in yeart;e — the GDP share of ex- ton University Press).
Bank Negara Malaysia, Annual Report 1999 (Kuala
penditure in the base year t0. Lumpur).
Calculating the fiscal impulse, I(f), in year t: Barro, Robert, 1979, "Second Thoughts on Keynesian
I(t) = [S(t)/Y(t)] - Economics," American Economic Review, Vol. 69
(May), pp. 54-59.
Chari, V.V., Lawrence Christiano, and Patrick Kehoe,
Definition of Structural Fiscal Balance 1995, "Policy Analysis in Business Cycle Models," in
Structural fiscal balance is the difference between Frontiers of Business Cycle Research, ed. by Thomas
the structural revenue and structural expenditure. F. Cooley (Princeton, New Jersey: Princeton Univer-
The structural revenue measures the actual revenue sity Press).
adjusted for the lagged impact in closing the output Cooley, Thomas F., and Gary Hansen, 1995, "Money and
the Business Cycle," in Frontiers of Business Cycle
gap. The structural expenditure measures the actual Research, ed. by Thomas F. Cooley (Princeton, New
expenditure adjusted for the provision of unemploy- Jersey: Princeton University Press).
ment benefits for cyclical unemployment (rather Dixit, Arinash, and Luisa Lambertini, 2000, "Fiscal Dis-
than the structural unemployment). cretion Destroys Monetary Commitment," IMF Semi-
nar Series No. 2000-26 (Washington: International
Monetary Fund), pp. 1-29.
Calculation of Structural Fiscal Balance
Economic Planning Unit, Malaysia, Mid-Term Review of
• Calculating the structural revenue, SR(t), in the Seventh Malaysia Plan, 1996-2000 (Kuala
year t: Lumpur).

Hemming, Richard, and Murray Petrie, "A Framework for Mackenzie, George A., and Peter Stella, 1996, Quasi-
Assessing Fiscal Vulnerability," IMF Working Paper Fiscal Operations of Public Financial Institutions,
00/52 (Washington: International Monetary Fund). IMF Occasional Paper No. 142 (Washington: Interna-
Kopits, George, and Steven Symansky, 1998, Fiscal Pol- tional Monetary Fund).
icy Rules, IMF Occasional Paper No. 162 (Washing- Potter, Barry, and Jack Diamond, eds., 1999, Guidelines
ton: International Monetary Fund). on Public Expenditure Management (Washington: In-
Kydland, Finn E., and Edward C. Prescott, 1997, "Rules ternational Monetary Fund).
Rather Than Discretion: The Inconsistency of Opti- Vegh, Carlos, and Ernesto Talvi, 2000, "Tax Base Vari-
mal Plans," Independent Central Banks and Eco- ability and Procyclical Fiscal Policy," NBER Work-
nomic Performance (Cheltenham, England: Elgar ing Paper No. 7499 (Cambridge, Massachusetts:
Reference Collection), pp. 3-21. National Bureau of Economic Research).

V Capital Controls in Response to the
Asian Crisis
Natalia Tamirisa
C apital controls introduced by Malaysia during

the Asian crisis have been a subject of much
debate. Contrary to the views that the controls would
ringgit, were prohibited; and exports and imports of
ringgit banknotes were restricted. Capital controls
were supported by other regulatory measures, partic-
have serious detrimental effects on the economy, ularly those on trading in Malaysian equities.2
only limited economic costs of the controls have As the economic situation stabilized, controls on
been identified. At the same time, the benefits of the portfolio outflows were eased and eventually re-
controls cannot be clearly established. Together with moved. However, controls on international transac-
the pegging of the exchange rate, the controls had tions in ringgit remain largely intact (Figure 5.1 and
been designed to enhance monetary independence, Table 5.1).
thereby facilitating economic recovery and provid- The impact of the capital controls appears to have
ing breathing space for the implementation of struc- been limited so far. Reflecting in part their easing
tural reforms. Given the return of confidence to the after less than six months, the controls had only a
region shortly after the introduction of the controls, transitory adverse effect on Malaysia's access to in-
however, it appears ex post that Malaysia's strong ternational capital markets and its position in major
fundamentals would have made a more accommo- investment indices. The capital controls appear to
dating monetary policy and economic recovery pos- have affected portfolio flows to some degree and
sible without resorting to these measures.1 may have contributed to a decline in foreign direct
It is too early to discern the longer-term effects investment, although isolating the impact of the con-
of the controls on capital flows or on the develop- trols from that of other policies and identifying a
ment of the financial system in Malaysia. If foreign proper counterfactual are particularly difficult. It
investors expect Malaysia to resort to controls on also appears that controls on international transac-
portfolio outflows in future periods of instability, tions in ringgit, which eliminated the offshore ring-
they may attach a higher risk premium to investing git market, helped to reduce speculation against the
in the country. As for the remaining controls on in- ringgit at a time of highly volatile exchange markets.
ternational transactions in ringgit, these need to The effect of the capital controls on the domestic eq-
be evaluated in the broader context of domestic fi- uity market was mixed, but foreign participation in
nancial system development and prudential risk derivatives markets declined.
management. The timing of the imposition of these capital con-
trols mitigated their short-term negative impact.
They were introduced well into the Asian crisis after
Background a substantial amount of capital had already left the
country, and thus their effects on portfolio outflows
In tandem with the pegging of the exchange rate, were limited. By then, the external environment had
the authorities introduced capital controls in Sep- improved, and market sentiment about the region
tember 1998, aimed at restricting portfolio outflows had reversed for the better. Ex post, the ringgit be-
and eliminating the offshore ringgit market (Table came undervalued, further reducing incentives for
5.1). Portfolio investors were restricted from with- capital outflows. The subsequent easing of controls
drawing funds invested in Malaysia for at least a on portfolio outflows led to the reinclusion of
year, and trading of the ringgit outside of the country Malaysia in the Morgan Stanley Capital Indices and
was prohibited. Additionally, fund transfers abroad helped generate new portfolio inflows. Improvement
became subject to approval; international borrowing
and lending in ringgit, as well as trade settlements in
2
A discussion of capital controls during September 1998-
February 1999 can be found in Kochhar and others (1999),
'See also Sections II, VI, and VII. Chapter I.

Table 5.1. Key Changes in Capital Account Regulations
Date/Type of Transaction Measure
January 17, 1994 A ceiling was placed on the net external liability position of domestic banks, excluding trade-related
Bank transactions and direct investment inflows. This was removed on January 20, 1995.
January 24, 1994 Residents were prohibited from selling the following Malaysian securities to nonresidents: banker's
Portfolio investment acceptances; negotiable instruments of deposit; Bank Negara Malaysia bills; treasury bills; government
securities (including Islamic securities) with a remaining maturity of up to one year; and Cagamas
bonds and notes (whether or not sold or traded on a discount basis) with a remaining maturity of up
to one year. This was removed on August 12, 1994.
February 7, 1994 Residents were prohibited from selling to nonresidents all forms of private debt securities (including
Portfolio investment commercial papers but excluding securities convertible into ordinary shares) with a remaining
maturity of one year or less.
The restriction on the sale of Malaysian securities to nonresidents was extended to both the initial
issue of the relevant security and the subsequent secondary market trade.
February 23, 1994 Forward transactions (on the bid side) and non-trade related swaps by commercial banks with foreign
Banking system transactions customers were prohibited to curtail the speculative activities of offshore agents seeking long
positions in ringgit. This was lifted on August 16, 1994.
August 12, 1994 Restrictions on the sale of Malaysian securities were lifted, and residents were permitted to sell to
Portfolio investment nonresidents any Malaysian securities.
December 1, 1994 Measures were implemented to control borrowing and lending activities in domestic and foreign
Borrowing and lending currency.
* Nonresident-controlled companies were allowed to obtain credit facilities, including immovable
property loans, up to RM 10 million without specific approval, provided that at least 60 percent of
their total credit facilities from banking institutions were obtained from Malaysian-owned banking
institutions. Short-term trade facilities, guarantees, and forward foreign exchange facilities were
excluded from the computation of the RM 10 million limit in December 1994, while the 60:40 rule
continued to apply to total short-term trade facilities.
* Nonresidents with valid work permits were permitted to obtain domestic borrowing to finance up
to 60 percent of the purchase price of residential property for their own accommodations.
* Residents were permitted to borrow in foreign currency up to a total of the equivalent of
RM 5 million from nonresidents and from commercial and merchant banks in Malaysia.
June 27, 1995 Corporate residents with a domestic credit facility were allowed to remit funds up to the equivalent
Portfolio investment of RM 10 million for overseas investment purposes each calendar year.
February 1, 1996 The threshold for the completion of the statistical forms for each remittance to, or receipt of funds
Payments for invisible from, nonresidents was raised to RM 100,000 or its equivalent in foreign currency from amounts
transactions exceeding RM 50,000.
August 4, 1997 Controls were imposed on banks to limit outstanding offer-side swap transactions in ringgit that were
Banking system transactions non-commercial related (i.e., forward order/spot purchases of ringgit by foreign customers) to
$2 million or its equivalent per foreign customer. Hedging requirements of foreigners for trade-related
and genuine portfolio and foreign direct investments were excluded.
August 8, 1997 A ban on short selling of the listed securities on the Kuala Lumpur stock exchange was introduced to
Stock market transactions limit speculative pressures on equity prices and exchange rates.
October 15, 1997 The quota on sales to foreigners of high-end condominiums was raised to 50 percent from 30 percent,
Real estate transactions and foreigners were allowed to acquire two units of condominiums (compared with one earlier) to
reduce some of the impending supply in the high end of the property market.
September 1, 1998 A number of selective exchange control measures were introduced, aimed specifically at eliminating
Offshore ringgit market the offshore ringgit market and restricting the supply of ringgit to speculators.
transactions * A requirement was introduced to repatriate all ringgit held offshore, including ringgit deposits in
overseas banks, by October 1, 1998; these required Bank Negara Malaysia approval thereafter. An
approval requirement was imposed to transfer funds between external accounts and for the use of
funds other than permitted purposes (i.e., the purchase of ringgit assets). Licensed offshore banks
were prohibited from trading in ringgit assets, which had been allowed up to permitted limits
previously.
* A limit was introduced on exports and imports of ringgit by resident and nonresident travelers,
effective September 1, 1998. No prior limits existed.
* Residents were prohibited from granting ringgit credit facilities to nonresident correspondent banks
and stockbroking companies. This had been subject to a limit previously.

CAPITAL CONTROLS IN RESPONSE TO THE ASIAN CRISIS
Table 5.1 (continued)
* Residents were prohibited from obtaining ringgit credit facilities from nonresidents.This had been
subject to limits previously.
*AII imports and exports were required to be settled in foreign currency.
* All purchases and sales of ringgitfinancialassets could only be effected through authorized
depository institutions.Trading in Malaysian shares on Singapore's Central Limit Order Book over-
the-counter market was prohibited de facto as a result of strict enforcement of the existing law
requiring Malaysian shares to be registered in the Kuala Lumpur stock exchange prior to trade.
September 1, 1998 A number of additional measures were introduced aimed at preventing heavy capital outflows by
Portfolio and other forms residents and nonresidents.
of investment * An approval requirement was imposed for nonresidents to convert ringgit held in external accounts
into foreign currency, except for purchases of ringgit assets, conversion of profits, dividends, interest,
and other permitted purposes. No such restrictions existed previously.There were, however, no
restrictions on conversions of ringgit funds in the external accounts of nonresidents with work
permits, embassies, high commissions, central banks, international organizations, and missions of
foreign countries in Malaysia.
* A 12-month waiting period was required for nonresidents to convert ringgit proceeds from the sale
of Malaysian securities held in external accounts.This excluded foreign direct investment flows,
repatriation of interest, dividends, fees, commissions, and rental income from portfolio investment.
No such restrictions existed previously.
* A prior approval requirement was imposed, beyond a certain limit, for all residents investing abroad
in any form. This was previously applied only to corporate residents with domestic borrowing.
* A specific limit was placed on exports of foreign currency by residents up to the amounts brought
into Malaysia for nonresidents. Previously there was no restriction on the export of foreign currency
notes and traveler's checks on the person or in the baggage of a traveler. Exports by other means
required approval, regardless of the amount.
December 12,1998 Commercial banks and finance companies were allowed to extend loans to nonresidents for the
Lending in ringgit purpose of purchasing residential, commercial, or industrial property, or office space in Malaysia for the
period from December 12, 1998 to January 12, 1999, subject to certain conditions.
January 13, 1999 Capital flows for the purpose of trading in derivatives on the commodity and monetary exchange of
Portfolio investment Malaysia and the Kuala Lumpur options and financial futures exchange were permitted for nonresidents,
without being subject to the rules governing external accounts, when transactions were conducted
through "designated external accounts" that could be created with tier-1 commercial banks in
Malaysia.1
February 15, 1999 The 12-month holding period rule for repatriation of portfolio capital was replaced with two
Portfolio investment measures:
* A graduated system of exit levy was applied on the repatriation on the principal of capital
investments—in shares, bonds, and other financial instruments, except property investments—made
prior to February 15, 1999.The levy decreased over the duration of the investment, and thus
penalized earlier repatriations; the levy was 30 percent if repatriated less than seven months after
entry, 20 percent if repatriated in seven to nine months, and 10 percent if repatriated in nine to
twelve months. No levy was imposed on the principal if repatriated after twelve months.
* A graduated exit levy was applied on the repatriation of profits from investments made after
February 15, 1999 in shares, bonds, and other financial instruments, except property investments. The
levy decreased over the duration of investment; the levy was 30 percent if repatriated in less than
twelve months after the profit was realized and 10 percent if repatriated after twelve months. No
exit levy was imposed on capital repatriation.
The aim was to preempt the potential exodus of funds in September when the holding period was set
to expire, and to encourage fresh inflows to facilitate the recovery.
February 18, 1999 The repatriation of funds relating to investments in immovable property was exempted from the exit
Portfolio investment levy regulations.
March 1, 1999 The ceiling on the import and export of ringgit for border trade with Thailand in selected areas was
Export and import of raised.
ringgit banknotes
April 5, 1999 Investors in the MESDAQ, where growth and technology shares are listed, were exempted from the
Portfolio investment exit levy introduced on February 15, 1999.

Table 5.1 (concluded)
July 8, 1999 Commercial banks were allowed to grant overdraft facilities not exceeding RM 200 million in aggregate
Lending in ringgit for intraday transactions and not exceeding RM 5 million for overnight transactions to foreign
stockbroking companies.
September 21, 1999 The two-tier levy system was replaced with a flat 10 percent levy on repatriation of profits on
Portfolio investment portfolio investment, irrespective of when the profits were repatriated. Bank Negara Malaysia
explained there were complaints by foreign fund managers that the graduated system complicated the
pricing of their portfolios and that the complex calculation of the amount of the applicable levy raised
administrative costs.
September 21, 1999 To provide foreign investors with more flexibility in managing their portfolios and risks, Bank Negara
Swap and forward Malaysia relaxed controls on lending in ringgit to foreign stockbroking companies. Commercial banks
transactions were allowed to enter into short-term currency swap arrangements with foreign stockbroking
companies to cover payment for purchases of shares on the Kuala Lumpur stock exchange and for
outright ringgit forward sale contracts with nonresidents who have a firm commitment to purchase
shares on the Kuala Lumpur stock exchange, for a maturity period not exceeding five working days
and with no rollover option.
October 4, 1999 Commercial banks and finance companies were allowed to extend loans to nonresidents for the
Lending in ringgit purpose of purchasing residential, commercial, or industrial property, or office space in Malaysia for the
period from October 29 to December 7, 1999.This was to support official housing campaigns and was
subject to certain conditions.
March 14, 2000 Original nonresident holders of securities purchased on the Central Limit Order Book were allowed
Portfolio investment to repatriate all funds arising from the sale of these securities without payment of the exit levy.
April 24, 2000 In line with the objective of promoting the development of the domestic bond market, resident
Borrowing companies in Malaysia were allowed to issue private debt securities for permitted purposes without
prior written approval from Bank Negara Malaysia. Nonresident-controlled companies raising
domestic credit facilities by way of private debt securities were exempted from the RM 19 million
limit and the 50:50 requirement for issuance of private debt securities on tender basis through the
fully automated system for tendering.
June 29, 2000 Administrative procedures were issued to facilitate the classification of proceeds from the sale of the
Portfolio investment Central Limit Order Book securities as being free from levy.
June 30, 2000 Guidelines on private debt securities were issued.
Borrowing
July 27, 2000 Residents and nonresidents were no longer required to make a declaration in the traveler's declaration
Export and import of form as long as they carry currency notes and/or traveler's checks within the permissible limits. For
currency nonresidents, the declaration was incorporated into the embarkation card issued by the Immigration
Department.
September 30, 2000 Licensed offshore banks in the Labuan international offshore financial center were allowed to invest in
Borrowing in ringgit ringgit assets and instruments in Malaysia for their own accounts only and not on behalf of their
and investment in clients.The investments could not be financed by ringgit borrowing.
ringgit assets
December 1, 2000 Foreign-owned banking institutions in Malaysia were allowed to extend up to 50 percent of the
Lending by foreign- total domestic credit facilities to nonresident-controlled companies, in the case of credit facilities
owned banks extended by resident banking institutions. This is to fulfill Malaysia's commitment under the General
Agreement on Trade and Services. Previously, foreign-owned banking institutions could only extend up
to a maximum of 40 percent funding.
December 20, 2000 Licensed commercial banks and Bank Islam Malaysia Berhad in Malaysia were allowed to extend
Lending in ringgit intraday overdraft facilities not exceeding RM 200 million in aggregate and overnight facilities not
exceeding RM 10 million to foreign stockbroking companies and foreign global custodian banks.
February 1, 2001 The exit levy on profits repatriated after one year was abolished. Portfolio profits repatriated within
Portfolio and other forms one year remained subject to the 10 percent levy.
of investment
May 2, 2001 The 10 percent exit levy was removed altogether.
Source: Information provided by the Malaysian authorities.

1The classification of tier-l and tier-2 banks is no longer applicable. As of September 21, 1999, all commercial banks in Malaysia are allowed to open
designated external accounts for nonresidents.

transactions in ringgit, cross-country experiences

Figure 5.1. The Evolution of Capital suggest that this regulation needs to be considered
Controls 1 carefully in the context of a longer-term policy ap-
proach to capital account regulation and financial de-
velopment, particularly the deepening of onshore fi-
1.0 nancial markets.
0.9 The remainder of this section discusses the design
and enforcement of the 1998 capital controls, fol-
lowed by a review of their impact on foreign invest-
ment, financial markets, and access to international
capital markets. A preliminary empirical analysis is
presented. Policy considerations relating to the exit
levy on portfolio outflows and controls on interna-
tional transactions in ringgit are assessed in a cross-
country context.
1997 99 2000 Design and Enforcement of

Capital Controls
Sources: Information provided by the Malaysian authorities; and
IMF staff estimates.
1To trace the evolution of capital controls, simple indices are
There is no evidence that the controls were circum-
constructed similarly to Tamirisa (1999); and Johnston and others vented on a large scale, and neither the nondeliverable
(1999). Indices equal the weighted sum of capital controls in place forward market nor a black market emerged.3 Incen-
in a given month normalized by the weighted sum of capital con- tives for circumvention were generally weak, because
trols introduced since 1991. Weights are assigned as follows: pro-
hibition is given the weight of 1; quantitative limit, approval
ex post undervaluation of the ringgit and improved re-
equirement, or a tax greater than ten percent, the weight of 0.5; gional and domestic economic prospects encouraged
and notification requirement or a tax less than ten percent, the investors to keep their funds in ringgit.
weight of 0.2.The indices range from 0 to 1, with higher values
indicating more extensive controls.
The design of the controls contributed to limiting
their circumvention. They were selective in that they
targeted offshore ringgit transactions and portfolio
flows, and not current account transactions nor for-
eign direct investments. Thus, there was no direct
in investors' sentiment was further enhanced by the reason for circumvention in relation to trade and di-
authorities' resolve to take advantage of the breath- rect investment transactions. At the same time, the
ing space provided by capital controls and their pur- controls covered the targeted types of flows compre-
suit of financial and corporate reforms. Finally, care- hensively whereby all key identifiable channels for
ful design and effective enforcement of the controls the leakage of ringgit offshore and for the access of
helped to focus on their intended objectives, thereby nonresidents to ringgit funds were closed. Along
lessening their adverse effects. The controls were se- with the introduction of capital controls, for exam-
lective and did not extend to payments for current ple, offshore trading of ringgit assets was prohibited
international transactions or foreign direct invest- (inducing a closure of the Singapore Central Limit
ment. Overall, there appears to be no evidence of a Order Book), large denomination ringgit notes were
large-scale circumvention, the emergence of a black demonetized, and the Companies Act was amended
market, or a nondeliverable forward market. to limit dividend payments.
It is too early, however, to conclude whether capi-
Additionally, effective enforcement of controls
tal controls will have any long-term effect on capital
helped reduce circumvention. While their introduc-
flows and financial system development in Malaysia.
tion initially led to confusion and uncertainty—in
Foreign investors may view Malaysia's recent resort
particular, in regard to outstanding contracts in the
to the exit levy, and controls on portfolio outflows
offshore market—Bank Negara Malaysia subse-
more generally, as a major reversal of its traditional
quently disseminated information on the controls to
policy and expect the authorities to repeat this in
ensure clarity. Bank Negara Malaysia succeeded in
times of instability. Accordingly, the risk premium on
foreign investment in Malaysia may rise. Considera-
tion of the potential usefulness of the remaining con-
trols over the medium term needs to take into account 3Circumvention could occur, for example, through leading and
lagging in the settlement of commercial transactions, dividend
these broader economic costs, as well as their effec- payments, intrafirm transfers, or misinvoicing of current account
tiveness. As regards the regulation of international transactions.

Capital Controls and Economic Performance
Table 5.2. Sovereign Credit Ratings
Standard and Poor's Moody's
1998 first quarter A A2

1998 second quarter A- A2
1998 third quarter BBB- Baa3
1998 fourth quarter BBB- Baa3
1999 first quarter BBB- Baa3
1999 second quarter BBB- Baa3
1999 third quarter BBB- Baa3
1999 fourth quarter BBB Baa3
2000 first quarter BBB Baa3
2000 second quarter BBB Baa3
2000 third quarter BBB Baa3
2000 fourth quarter BBB Baa2
Sources: Standard and Poor's; Moody's; and Bloomberg (various dates).
monitoring and enforcing the controls through close outlook and ratings were upgraded, and the sover-
collaboration with commercial banks, building on eign bond spread narrowed to a level comparable to
the preexisting relationship with the banks that re- those of Korea and Thailand. In May 1999, the gov-
flected the fact that many capital account transac- ernment issued a ten-year global bond of $1 bil-
tions had already been subject to Bank Negara lion—the first issue in almost a decade—to test in-
Malaysia approval. Bank Negara Malaysia's reputa- vestors' sentiment and set a sovereign benchmark in
tion as a strict regulator may have also prevented international capital markets. The bond spread was
foreign banks from exploring ways to circumvent wider than that prevailing at the time for comparable
controls, for fear of losing their local branches. South Korean and Thai sovereign bonds by about
Capital Controls and Economic

Performance Figure 5.2. Sovereign Bond Spreads of
Access to International Capital Markets Selected Asian Countries
(In basis points, end of period)
and Short-Term Financing
The capital controls had an adverse, albeit tempo-
rary, effect on Malaysia's access to international 1800
capital markets and short-term financing. Following
1600
the introduction of the controls in September 1998,
international rating agencies downgraded Malaysia's 1400
Indonesia
credit and sovereign debt ratings (Table 5.2). Coun- 1200
try risk, as reflected in the sovereign bond spread
and international credit ratings, increased to a 1000
greater extent than the ratings of other emerging 800
markets in the region (except Indonesia), which
were negatively affected by Russia's default that had
taken place the previous month (Figure 5.2).4
Following the easing of controls and a strengthen-
ing of the domestic recovery in 1999, Malaysia's
4
The increase in the sovereign bond spread caused the govern-
ment to abandon its original plans to underwrite an international
issue for launching Danaharta in November 1998.

140 basis points.5 The favorable financing outlook in

2000, owing in part to high oil prices, prompted Figure 5.3. Net Portfolio Flows
Malaysia to proceed with the issue of its first euro- (In millions ofU.S.dollars)
denominated bond of €650 million and the refinanc-
ing of its $1.35 billion five-year sovereign syndi-
cated loan at 52 basis points over Libor, well below 2000
its original December 1998 issue price of 290 basis
points. In line with developments in its country risk,
access to short-term financing was affected by the
capital controls, but only temporarily. The decline in
short-term borrowing in 1999-2000 was due mainly
to repayments of existing loans and lower demand
for hedging and trade financing.
Capital Flows
The capital controls appear to have had a limited
impact on portfolio flows in 1999-2000. This was
largely due to the fact that a substantial amount of 1997 98 99 2000
capital—about $10.4 billion—had already left
Malaysia during 1997-98, before the controls were Source: Data provided by the Malaysian authorities.
imposed (Figure 5.3). Thus, outflows that occurred
following the easing of controls and the expiration
of the one-year holding period were relatively
small.
Portfolio inflows increased starting in mid-1999, Despite the explicit exemption of foreign direct
with the market sentiment turning bullish in re- investment flows from capital controls and a more
sponse to Bank Negara Malaysia's monetary easing, liberal policy starting in July 1998,7 foreign direct
the upgrading of Malaysia's outlook and credit rat- investment declined during 1999-2000 to less than
ings, and the improvement in the overall regional half of precrisis levels (Table 5.3). As a share of
prospects. The inflows increased further in early GDP, foreign direct investment flows fell to the lev-
2000, as the rising equity market stirred up in- els observed in other crisis countries, after exceeding
vestors' interest and political uncertainty related to them significantly before the crisis. Several factors
the November 1999 general elections had dissipated. unrelated to capital controls may have contributed to
The prospects for Malaysia's reinstatement in the this decline, including slower growth in Japan and
Morgan Stanley Capital Indices also firmed up, and Taiwan Province of China, the worsening of investor
fund managers who benchmarked against these in- sentiment during the Asian crisis, and the decline in
dices raised the share of their portfolios allocated to overall investment in Malaysia. Capital controls,
Malaysia.6 Later in 2000, however, portfolio flows however, could have exerted an indirect negative ef-
reversed, discouraged by a weakening equity market fect on foreign direct investment in Malaysia. For-
that followed the trends in the United States. All in eign investors had reportedly become increasingly
all, portfolio flows during 1999-2000 seem to have concerned about a higher risk of investing in
been driven largely by factors other than capital con- Malaysia, changes in investment regulations, delays
trols. A preliminary empirical analysis shows that and administrative costs associated with additional
the tightening of controls on outflows and on inter- verification and approval requirements for transfers
national transactions in ringgit had insignificant ef- between external accounts, and more limited hedg-
fects on portfolio investment so far (Appendix). ing opportunities.
Nevertheless, drawing any conclusions regarding
the effect of capital controls on foreign direct invest-
5
The bond carried a coupon of 8.75 percent, and was priced to
yield 8.86 percent, about 330 basis points above the 30-year
benchmark U.S. Treasury yield. It was assigned a BBB senior un-
7
secured rating by Standard and Poor's. The relatively high pre- Measures introduced in mid-1998 to encourage foreign direct
mium induced the government to reduce the size of the issue by investment included allowing total foreign ownership of manu-
half. facturing (except in specified sectors), regardless of the degree of
6
Malaysia was reinstated in the IFC and Dow Jones investment export orientation; increasing the foreign ownership share limit in
indices in November 1999 and in the Morgan Stanley Capital In- the telecommunications, stockbroking, and insurance industries;
dices in May 2000. and relaxing curbs on foreign investment in landed property.

Capital Controls and Economic Performance
Table 5.3. Net Foreign Direct Investment in Selected Asian Countries
Country 1996 1997 1998 1999 20001
(In billions of U.S. dollars)

Korea -2.3 -1.6 0.4 5.9 3.7
Malaysia 3.5 3.9 1.9 1.9 1.5
Philippines 1.3 I.I 1.6 0.9 0.5
Thailand 1.7 3.4 6.8 5.8 3.7
(As a percentage of GDP)

Korea -0.5 -0.3 0.1 1.4 0.7
Malaysia 3.5 3.9 2.6 2.4 1.7
Philippines 1.6 1.4 2.4 I.I 0.6
Thailand 0.9 2.3 6.1 4.6 2.9
Source: IMF World Economic Outlook (various issues).

1Preliminary.
ment from the limited evidence available is difficult. Equity Market

Foreign direct investment is an inherently longer-
term phenomenon, and in each country it is deter- The capital controls had a mixed effect on the eq-
mined by a gamut of factors, including the country's uity market in 1998-2000 (Figure 5.5). The introduc-
policy on foreign equity participation in domestic tion of a one-year holding period for portfolio invest-
activities and on investment abroad by residents, and ment and controls on international transactions in
strategy for—and progress with—financial and cor- ringgit (particularly on lending to nonresidents)
porate sector restructuring. caused an influx of ringgit funds into domestic equi-
ties, while at the same time curtailing short selling
and capital outflows.9 As a result, Malaysia's equity
Foreign Exchange Market market rallied, outperforming other markets in the re-
gion during September 1998-January 1999. Equity
The activity in the Kuala Lumpur interbank for- prices continued to rise in 1999, in line with the do-
eign exchange market declined dramatically during mestic and regional recovery. The market was largely
1999-2000 (Figure 5.4). Controls on international driven by local retail buying, in the face of controls
transactions in ringgit eliminated the offshore mar- on investment abroad by residents. The replacement
ket, and, together with the exchange rate peg and re- of the one-year holding period with a graduated levy
strictions against onshore position taking, the activity system in February 1999 had a mixed effect on the
in the foreign exchange market was limited to trade- equity market. While helping to improve market sen-
and investment-related transactions. Major players in timent, it disrupted activities of fund managers by
the foreign exchange market were domestic corpora- making portfolio pricing and risk management more
tions, although many of them relied on natural hedg- complicated. The unification of the levy in the fol-
ing through matching payables and receivables. The lowing September, along with the relaxation of con-
volume of interbank foreign exchange transactions trols on offers of credit and swap facilities to foreign
fell by 61 percent in 1999 and by a further 10 percent stockbrokers, apparently had a positive effect on the
in 2000, to the level prevailing in 1992. The share of market. The major boost came in early 2000 from the
swap transactions fell to about half of total transac- prospects for Malaysia to be reinstated in the Morgan
tions in 1998-2000 from 70 percent in 1997 because Stanley Composite Indices. This effect was short
the August 1997 restrictions on non-trade related lived, however. Similar to other markets in the re-
swaps had the impact of lowering trading in the swap
market. Transactions in U.S. dollars against the ring-
git continued to dominate, but their volume share in
total transactions declined to 68 percent in 2000 from 2000, from 2 percent of the total volume of transactions in 1998,
78 percent in 1998.8 possibly reflecting the requirement to settle trade transactions in
foreign currencies.
9
At the same time, new requirements on share trading, which
were introduced in tandem with capital controls, temporarily hin-
8
In contrast, transactions in U.S. dollars against the Singapore dered trading in American depository receipts for Malaysian
dollar increased sharply to 19 percent in 1999 and 22 percent in companies.

The capital controls appear to have dampened ac-

Figure 5.4. Volume of Interbank Spot and tivity in futures and options markets. Foreign partic-
Swap Transactions in Kuala Lumpur ipation in these relatively nascent and thin markets
Foreign Exchange Market declined after the controls were introduced in Sep-
(In billions of ringgit) tember 1998 and did not recover even after trading
in derivatives was exempted from restrictions on
repatriation of capital and profits in January 1999.
1400 Economic recovery and the improved market senti-
ment in other financial markets did not help lift the
1200 futures market either. As a result, the average daily
trading volume in the Kuala Lumpur Stock Ex-
1000 change Composite Index Futures on the Kuala
Lumpur Options and Financial Futures Exchange
800 (KLOFFE) declined by 44 percent in 1999 and by a
further 14 percent in 2000 (Figure 5.6). Trading in
600 three-month Kuala Lumpur Interbank Offered Rated
futures contracts on the Commodity and Monetary
400 Exchange of Malaysia (COMMEX) increased to a
200 daily average of 180 contracts in 2000 from 101
contracts in 1998. This increase, however, was
0 mainly due to improved liquidity in the underlying
1992 94 96 98 2000 cash market and the decline in interest rates, which
encouraged interest rate hedging. The reintroduction
Source: Data provided by the Malaysian authorities.
of the market-maker scheme in mid-August 1999,
which had been discontinued in July 1998, also con-
tributed to the market recovery. Trading was domi-
nated by local financial institutions, and foreign par-
ticipation declined to 1.2 percent in 1999 from 14
Figure 5.5. Selected Asian Countries: percent in 1998.
Stock Market Composite Indices
(January 1997 = 100)
Policy Considerations
The Exit Levy
A key control on capital outflows in effect from
February 1999 to May 2001 was the levy related to
portfolio capital repatriated within a year. There are
several arguments for using such a levy to manage
capital flows. It is a market-based measure and thus
is less distortionary than an administrative control. It
targets nondebt, short-term capital flows rather than
40
portfolio flows in general. In principle, the authori-
Singapore Thailand ties could vary the rate to achieve the appropriate de-
Malaysia
20 Malaysia
gree of monetary policy autonomy and to alter the
level and maturity composition of portfolio flows.
0 As a side benefit, the levy facilitates Bank Negara
1997 1998 1999 2000
Malaysia in its monitoring of short-term portfolio
Sources:Wharton Econometric Forecasting Associates, flows.
Bloomberg, and CEIC Data Company Limited. Other things being equal, however, the levy in the
form introduced by Malaysia raises the pretax return
required by foreign investors. The levy generally
cannot be offset by double taxation treaties because
gion, the Kuala Lumpur equity market turned in it is collected at the time of conversion of ringgit
losses by end-2000, in correlation with U.S. financial into foreign exchange for repatriation rather than at
markets, particularly the NASDAQ, despite favor- the time of the transaction; therefore, the full burden
able domestic developments. of paying the levy, including higher transaction

peared to be more restrictive than that in most other

Figure 5.6. Average Daily Trading Volume emerging markets, which tend to exempt portfolio
at the KLOFFE and the COMMEX1 investment from both repatriation taxes and capital
(Number of contracts) gains taxes (Table 5.5). This tendency may reflect
the emerging markets' efforts to attract foreign in-
vestors, in light of the fact that most developed
5000 countries do not provide a full tax credit or deduc-
tion for capital gains taxes.11
Controls on International Transactions in

Domestic Currency
During the Asian crisis, the controls on interna-
tional transactions in ringgit helped abate specula-
tion against the currency. In the context of postcrisis
policies, these controls—and particularly their long-
term implications—warrant examination from the
broader perspective of Malaysia's strategy on finan-
cial development and the capital account regime.
1997 1998 1999 2000
Demand for the international use of a country's
currency outside of the country depends on interna-
Source: Data provided by the Malaysian authorities.
tional invoicing practices; the level of development
1The Kuala Lumpur Options and Futures Exchange (KLOFFE). of the country's financial institutions, including the
The Commodity and Monetary Exchange of Malaysia (COMMEX). breadth and depth of its markets; the regulatory
regimes of the financial system and the capital ac-
count; and the country's political and economic sta-
costs, falls on foreign investors. Since the levy is not bility. For an emerging market like Malaysia, there is
indexed to inflation, the required pretax return is in- relatively little demand for the use of its currency as
creased further. A higher pretax return would imply a a unit of account in invoicing trade transactions and
higher cost of external financing for domestic firms, denominating financial instruments, or as a store of
thus possibly weakening stock market performance value outside Malaysia, whereby nonresidents hold
and domestic investment in the longer run. ringgit as an investment asset. Such demand em-
More generally, even if the levy applies only to anates largely from potential use of the currency
short-term portfolio profits, investors may attach a within the country, and is limited by international in-
risk premium to investment in the country. The pres- voicing practices.12 In addition, the risk preferences
ence of the levy may negatively influence investors' of foreign investors influence demand for ringgit-
sentiment and discourage portfolio inflows, and such denominated instruments.
effects are likely to be magnified if the levy is varied Regulations pertaining to nonresidents' ringgit
frequently. There are also limitations associated with accounts, especially transfers to and from them,
the exit levy, which, without accompanying macro- have important implications for both international
economic and prudential policies, is unlikely to be an trade and financial transactions in ringgit. In effect,
effective instrument for shifting the maturity compo-
sition of capital flows to those of longer term.10
The levy may place the country at some disadvan- 11
Notwithstanding the above comparison, definitive conclu-
tage compared to most other emerging markets. sions regarding the role of the levy in the cross-country context
Available data suggest that, until recently, Malaysia would require a comprehensive study of investment barriers in
the countries concerned, which is beyond the scope of this paper.
was the only middle-income country that taxed repa- In this connection, empirical evidence suggests that withholding
triation of capital gains from portfolio investment taxes on dividends are less likely to affect pretax rates of return
(Table 5.4). When the levy is examined in combina- because the developed countries tend to provide a tax credit or
tion with other taxes, Malaysia's overall taxation on deduction for such taxes (Demirguc-Kunt and Huizinga, 1995).
12
capital gains from portfolio investment still ap- Recent empirical studies have identified the following inter-
national invoicing practices: (i) trade in differentiated manufac-
tured products between developed countries tends to be invoiced
in the exporter's currency; (ii) trade between a developed and a
10
A preliminary empirical analysis shows that prudential mea- developing country tends to be invoiced in the currency of the de-
sures tended to be more effective than direct capital controls in veloped country; (iii) trade in primary products and transactions
influencing portfolio investment in Malaysia (see Appendix). For in financial instruments tend to be denominated in U.S. dollars;
a detailed discussion of effectiveness of controls on short-term and (iv) exports to the United States tend to be invoiced in U.S.
capital flows, see Ariyoshi and others (2000). dollars (Tavlas and Ozeki, 1992; and Dominguez, 1999).

Table 5.4. A Summary of Investment Regulations for Entry into and

Exit from Selected Markets, end-1998
Entry Exit
Availability of listed stocks Repatriation of interest, dividends,
Country to foreign investors profits, and capital
Argentina Free Free

Brazil Free Free
Chile Relatively free Restricted1
China Special classes of shares Free
Czech Republic Free Free
Hong Kong SAR Free Free
Hungary Free Free
India Authorized investors only Free
Indonesia Relatively free Free
Israel Free Free
Korea Free Free
Malaysia Relatively free Restricted2
Mexico Free Free
Philippines Free Free
Poland Free Free
Singapore Free Free
South Africa Free Free
Thailand Relatively free Free
Turkey Free Free
Sources: International Finance Corporation, Emerging Stock Markets Factbook; and IMF, Annual Report on Exchange
Arrangements and Exchange Restrictions.
1Unremunerated reserve requirement on debt and certain other flows. Repatriation of capital is free after one year.
2
The 12-month holding period for repatriation of portfolio capital from September 1998 to February 1999 and
an exit levy system from February 1999 t o May 2001. For more details, see Table 5.1.
Table 5.5. Taxation of Capital Gains on Portfolio Investment in Selected Emerging Markets, 19991
(In percent)
Top Equities2 Bonds

Country Marginal Rate Short Long Short Long
China 40 Exempt Exempt Exempt Exempt

Czech Republic 32 Ordinary income Exempt Ordinary income Exempt
Hong Kong SAR 15 Exempt Exempt Exempt Exempt
Hungary 40 20 20 20 20
Indonesia 30 0.13 0.13 Ordinary income Ordinary income
Korea 40 Exempt Exempt

Malaysia 32 Exempt Exempt Exempt Exempt
Mexico 35 Exempt Exempt Exempt Exempt
Philippines 35 0.53 0.53 3-30 3-30
Poland 40 Exempt Exempt Ordinary income Ordinary income
Singapore 28 Exempt Exempt Exempt Exempt

Turkey 55 Exempt Exempt Exempt Exempt
Source: IMF Fiscal Affairs Department, based on data from PricewaterhouseCoopers and DeloitteTouche,Johmatsu.
1The predominant tax treatment is indicated.
2
Listed.
3
Rate applies t o sales value.

Table 5.6. A Summary of Controls on the International Use of Domestic Currency, end-1998!
Hong Kong
Type of Control China SAR India Indonesia Japan Korea Malaysia Philippines Singapore Thailand
Controls on the settlement

of trade transactions in
domestic currency N N N2 N N Y Y Y3 N N
Controls on the import and

export of domestic currency
notes by residents and
nonresidents Y N Y Y N Y Y Y N Y
Controls on residents' granting

credit facilities in domestic
currency to nonresidents Y N Y Y N Y Y Y Y Y
Controls on residents' obtaining

credit facilities in domestic
currency from nonresidents Y N Y Y N Y Y Y Y Y
Controls on transfers of
domestic currency funds from
nonresidents' domestic
currency accounts Y N Y N N Y Y Y Y N
Controls on nonresidents'
issuing domestic currency-
denominated securities Y N Y Y N Y Y Y Y Y
Controls on residents' issuing

domestic currency-
denominated securities abroad Y N Y Y N Y Y Y Y Y
Source: \MF, Annual Report on Exchange Arrangements and Exchange Restrictions (1999).
'"Y" indicates yes,"N" indicates no.
2
Except for member countries of the Asian Clearing Union arrangement, other than Nepal.
3
Only for export proceeds.
the international use of the ringgit is linked to con- nomic development and the degree of its integra-
trols on the settlement of trade transactions in ring- tion into the world economy are associated with a
git, on ringgit credit operations, and on the issuance gradual liberalization of international transactions
of ringgit-denominated instruments. Regulations in domestic currencies.
concerning Malaysia's offshore financial center in Experiences in select Asian countries suggest that
Labuan also come into play, as they determine the regulation of international transactions in domestic
extent of "seepage" between international and do- currencies is generally influenced by considerations
mestic markets through the offshore center. related to the effectiveness of monetary control, the
Most emerging markets control international scope for nonresidents to take large positions against
transactions in domestic currencies, albeit to a dif- the domestic currency, and implications for financial
ferent degree (Table 5.6). The controls are typically development (Boxes 5.1-5.4).
effected through approval and reporting require-
ments, quantitative limits, and sometimes outright Considerations Regarding the Liberalization
prohibitions. Financial transactions tend to be regu-
lated more intensively than trade transactions. In of Controls in Malaysia
particular, many countries restrict lending to non- In Malaysia, a liberalization of controls on inter-
residents in order to constrain their ability to take national transactions in ringgit requires policy
large positions against the currency. Industrial measures to minimize the associated risks. Clearly
countries, in contrast, generally have a more liberal such a policy should be implemented at a time of
approach to international transactions in their cur- relative market stability. Furthermore, in order to
rencies, suggesting that a country's level of eco- limit disintermediation of financial activities off-
61

Box 5.1. Singapore: Reconciling Policy on the Internationalization of the Singapore Dollar and
Financial Development Objectives
Since the late 1960s, Singapore has pursued a strat- pore dollars as part of financial reforms aimed at pro-
egy of developing the city-state into an international fi- moting the development of the financial sector. (i) Off-
nancial center, while strictly limiting the international- shore banks had, since 1973, been permitted to make
ization of the Singapore dollar to maintain control over limited Singapore dollar loans to residents; the relax-
monetary conditions. In the 1990s, controls on the in- ation was in response to the argument that restrictions
ternational use of the Singapore dollar were substan- on lending to loyal customers in the domestic market
tially relaxed to promote financial development and fa- might cost the banks offshore business. With a view to
cilitate regionalization efforts of the private sector. providing more flexibility to offshore banks, the gov-
Traditionally, Singapore's strategy has been to pre- ernment raised the limit on such lending in the early
vent the internationalization of the Singapore dollar by 1990s (i.e., the maximum amount in domestic loans
separating offshore from domestic financial markets that offshore banks may have outstanding at any given
and restricting the operations of offshore units. The au- time) in several steps: to S$100 million in 1993; S$200
thorities have been concerned that, given the high de- million in 1997 (binding the latter commitment in the
gree of openness of the economy, the internationaliza- 1997 Financial Services Agreement of the World Trade
tion of the Singapore dollar will increase the exchange Organization); and further to S$300 million in 1998.
rate volatility and the probability of speculative attacks, The limit is expected to be increased further. (ii) Since
and will complicate control over monetary conditions. 1992, banks—including offshore banks—have been al-
Accordingly, banks maintain distinct operations or lowed to make Singapore dollar loans overseas through
"units," with separate books and balance sheets, for their domestic banking units for Singapore-related
transactions in the domestic and offshore markets. Do- trade, performance bonds, and for hedging (for im-
mestic banking units may engage in transactions in both ports) purposes. (iii) In August 1998, the government
foreign and local currencies, and are subject to stringent further relaxed some of the restrictions on borrowing in
regulatory requirements and the standard corporate in- Singapore dollars for use abroad; the Monetary Author-
come tax rate of 26 percent. Offshore units, also known ity of Singapore indicated that the regulation requiring
as Asian currency units, enjoy a concessionary tax rate resident banks to consult them on the overseas use of
of 10 percent and are not subject to reserve and liquidity Singapore dollar credit facilities exceeding S$5 million
requirements. Asian currency units focus on operating and on financing of trading activities in Singapore
in the Asian dollar market and are not permitted to dollar—denominated assets applies only to nonresi-
transact in Singapore dollars. Offshore banks may ac- dents. (iv) More flexible guidelines have been adopted
cept fixed deposits in Singapore dollars only above to allow subsidiaries of Singaporean companies and
S$250,000 and only from nonresidents and other finan- joint ventures between Singaporean and foreign com-
cial institutions. There is a ceiling on lending to resi- panies to borrow in Singapore dollars for regionaliza-
dents in Singapore dollars. To limit the short selling of tion projects. (v) Finally, foreign entities are now per-
the Singapore dollar and the funding of portfolio and mitted to issue Singapore dollar-denominated bonds for
property investments by nonresidents, lending to non- use outside Singapore upon consultation with the Mon-
residents is subject to quantitative limits. Until 1998, etary Authority of Singapore and as long as the Singa-
loans to residents for use outside the country required pore dollars are first converted into foreign currencies.
approval of the Monetary Authority of Singapore. In In parallel with liberalizing international transactions
sum, the authorities facilitated the participation of Asian in Singapore dollars, the government proceeded with
currency units in regional banking activities but limited other financial reforms. More competition will gradu-
in domestic banking activities. ally be allowed in the domestic retail banking market
Over the last decade, the government has eased con- and, in particular, the 40 percent limit on foreign share-
trols on offshore banks' wholesale operations in Singa- holding in local banks will be lifted. Various measures
are being introduced to promote the fund management
industry. The focus of prudential policies will largely
Sources: Cassard (1994); Bercuson (1995); and Cardarelli, shift to supervision and, in particular, to the review of
Gobat, and Lee (2000). banks' risk management systems.
shore, it is important that the onshore financial in international interest rates to domestic rates and
markets—especially foreign exchange markets— complicate control over monetary conditions. Ad-
become more liquid and gain more depth and ditionally, in order to prevent seepage between the
breadth, and that domestic financial institutions be- domestic economy and the Labuan offshore finan-
come more efficient and resilient. Monetary man- cial center, the policy of limiting ringgit transac-
agement also needs to be strengthened because the tions of Labuan banks needs to be continued, and
easing of controls on international transactions in consolidated bank supervision must be further
ringgit may speed up the transmission of changes strengthened.

Box 5.2. Korea: Gradual Liberalization with a Focus on Control of

Won Lending to Nonresidents
International transactions in won were liberalized, Economy. For residents, the purchase of short-term,
albeit cautiously and only partially, during 1987-97. In won-denominated securities abroad requires Ministry
particular, lending to nonresidents was deliberately re- of Finance and Economy approval and prior reporting.
stricted. The government intends to liberalize the con- To limit potential speculation, credit facilities in won
trols further by end-2000 to broaden risk management granted to nonresidents by institutional investors of
and investment opportunities for investors and to pro- more than W 100 million per borrower require Ministry
mote competition in the financial sector. Controls on of Finance and Economy approval.
nonresidents' won funding, however, will remain in During the Asian crisis, speculation against the won
place for the next one or two years. was limited. In contrast to the currencies of other crisis
International transactions in won are controlled countries, the won was not overvalued, and the current
mainly through approval requirements. While there ap- account deficit was relatively small and manageable.
pears to be no explicit policy against the international Nonresidents' access to won funds was restricted, and
use of the won or criteria for allowing these transac- short-selling of the currency became more costly,
tions, the authorities are clearly concerned about the owing to the initial high interest rate policy. Addition-
potentially destabilizing effects of such liberalization. ally, the foreign exchange market was relatively thin, as
Starting from 1987, controls on the international use of major Korean exporters, the chaebols, tended to hedge
the won were liberalized as part of the general liberal- in-house by matching receivables and payables. Al-
ization of the capital account, but this liberalization though there was a viable nondeliverable forward mar-
lagged behind that of transactions in foreign currency, ket for won in Singapore and Hong Kong SAR, it was
and approval requirements continued to apply to most also relatively thin, and position taking there tended to
transactions even after liberalization. The settlement of be transparent and costly. An offshore market for the
trade transactions in won is prohibited, and the export won did not emerge, in part because the authorities al-
and import of won banknotes are subject to quantitative lowed the rate to be determined in the market in line
limits and approval.1 Rules for nonresidents' won ac- with the prevailing flexible exchange rate regime.
counts in domestic banks and overseas branches are In the aftermath of the Asian crisis, the government
relatively liberal. The use of won for financial purposes continued to liberalize the capital account with the aim
is also controlled through approval requirements. In of broadening investment and risk management oppor-
particular, the issue of won securities abroad by resi- tunities available to investors and enhancing competi-
dents and domestically by nonresidents requires offi- tion in the financial sector. In April 1999, the govern-
cial approval. Nonresidents may list shares locally in ment streamlined procedures for current account
the form of depository receipts and may issue won- transactions and changed from a positive list system for
denominated bonds domestically, subject to reporting capital account transactions to a negative one. Offshore
requirements. The issue of collective investment secu- transactions in won, nonresidents' won deposits and
rities in the domestic market by nonresidents is al- trust accounts with maturities greater than one year,
lowed, provided they establish themselves in Korea and offshore and domestic issuance of won securities
and submit a prior report to the Ministry of Finance and with maturities greater than one year were decon-
trolled. Controls on international transactions were re-
laxed further during the second stage of liberalization
Sources: Johnston and others (1999); IMF (1998, 1999); by end-2000. Some controls, however, particularly
and Park (1998). those on won funding of nonresidents, are likely to re-
1Nonresidents are permitted to carry out current account main in place for the next one or two years (albeit pos-
transactions in won, provided remittances are made in foreign
currencies. For this purpose, nonresidents may open settle- sibly in a more relaxed form). The authorities are con-
ment accounts in won (free won accounts) for current account cerned about the reaction of hedge funds to the
transactions, reinsurance contracts, and investments in do- liberalization and intend to strengthen prudential super-
mestic securities. vision before continuing the reforms.
Policy considerations on this matter must also rec- strong incentives were to reemerge. In this case,
ognize that the effectiveness of controls on interna- leakages in capital controls may develop, for exam-
tional transactions in ringgit is not guaranteed in the ple through intercompany accounts, underinvoicing
future, particularly in times of instability. The suc- of exports and overinvoicing of imports, leads and
cessful experience of the 1998 controls so far is lags in transactions related to foreign trade, and
largely due to the appropriate macroeconomic policy small-scale exports of ringgit banknotes. Driven by
mix that prevailed at that time. Even wide-ranging fundamentals, such position taking against the ring-
controls, however, are likely to become inadequate git may occur illegally onshore, in the newly
in preventing both residents and nonresidents from emerged offshore market, or in the nondeliverable
taking positions against the ringgit if sufficiently forward market. To the extent that the offshore mar-

Box 5.3. Thailand: Relatively Liberal Policy in Normal Times,Temporary Capital

Controls to Limit Currency Speculation
Thailand's policy concerning international transac- allowing foreign exchange conversion for the under-
tions in baht was relatively liberal before the Asian cri- lying trade and investment transactions. Financial in-
sis. In 1997, prior to the floatation of the baht, Thailand stitutions were required to suspend baht lending to
imposed temporary capital controls, including those on nonresidents through swaps, outright forwards, and
baht lending to nonresidents, which created a two-tier sales of baht against other currencies. They were also
currency market. These controls were removed for the required to report daily on foreign exchange transac-
most part in early 1998. tions with nonresidents. Any purchase before the ma-
Controls on international transactions in baht were turity of baht-denominated bills of exchange and
substantially liberalized in line with the general open- other debt instruments required payment in U.S. dol-
ing of the capital account during 1989-92. Trade trans- lars. Foreign equity investors were prohibited from
actions were allowed to be settled in baht. Limits on repatriating funds in baht. Nonresidents were required
exports and imports of baht banknotes were relatively to use the onshore exchange rate to convert baht pro-
generous. Transfers and uses of funds from nonresi- ceeds from sales of stocks. Exempt from controls
dents' transferable baht accounts were allowed. Poli- were the underlying transactions related to current
cies concerning baht lending to nonresidents were also account operations, and foreign direct and portfolio
relatively liberal. investment.
A widening of internal and external imbalances and a The capital controls created a two-tier market with
weakening of the banking system induced a series of different exchange rates in the onshore and offshore
speculative attacks on the baht beginning in late 1996. markets. They squeezed offshore players with short
The attacks were facilitated by the relatively free capi- baht positions and forced them to liquidate these posi-
tal account, well-developed spot and swap markets, tions at a loss, while domestic holders of baht positions
and relatively liberal access of nonresidents to baht apparently maintained their positions. Large differen-
credit from domestic banks. Onshore and offshore tials emerged between onshore and offshore interest
speculation took place in the form of direct position rates, and the swap market dried out. However, the
taking in the forward market and short selling of the wide differentials created strong incentives for arbi-
currency through explicit baht credits. Investors who trage, expectations of baht depreciation persisted, and
were taking positions against the baht included com- capital outflows continued. Speculation in the offshore
mercial and investment banks, portfolio managers of market resumed, and eventually in July 1997 the au-
mutual funds, proprietary trading desks, as well as thorities were forced to float the baht.
hedge funds. In January 1998 capital controls were lifted, and the
To limit speculation against the baht and stabilize two-tier market was unified. In particular, prohibition
foreign exchange markets, Thailand imposed capital of offering baht credit facilities (including—but not
controls in May 1997. The measures sought to seg- limited to—loans, currency and interest rate swaps, op-
ment the onshore and offshore currency markets; to tions, and forwards) to nonresidents was replaced with
limit the supply of baht credit to nonresidents for po- a maximum outstanding limit of 50 million baht per
sition taking against the baht; and to raise the cost of counterparty without an underlying trade or investment
carrying positions overnight, while at the same time transaction. Easing of the controls helped improve in-
vestors' confidence and contributed to the appreciation
of the baht. The enforcement of the new measures was
Sources: Ariyoshi and others (2000); and IMF (1998). strengthened further in August 1999.
ket might offer more channels for arbitrage than the ment of asset prices, facilitate hedging and portfolio
onshore market, its presence may accelerate a specu- management, and generally promote financial mar-
lative attack; however, its absence is not a guarantee ket development. Pricing of financial instruments,
against position taking. Thus, key to preventing such that it becomes more reflective of returns and
destabilizing position taking is strong fundamentals, risks involved, will also be made easier, thereby pro-
especially consistent monetary and exchange rate moting the domestic bond market. In addition, more
policies, and effective prudential risk management. liberal regulation on credit operations and the in-
With an easing of controls on international trans- voicing of trade transactions in ringgit may facilitate
actions in ringgit, the liquidity of the foreign ex- the regionalization efforts of Malaysian companies
change market is likely to increase as speculators in the longer run.
enter the market as risk takers and counterparties of The presence of an offshore market could also im-
hedgers. The increased liquidity could facilitate prove financial intermediation, enhancing efficiency
portfolio rebalancing and unwinding of leveraged and investors' opportunities for risk management.
positions; this in turn would help smooth the adjust- Such a market tends to offer a narrower interest rate

Box 5.4. Japan:Toward the Internationalization of the Yen
Japan liberalized cross-border transactions in yen controlled. Residents were allowed to lend to and bor-
during the second half of the 1980s, in tandem with row from nonresidents in yen. Controls on the issuance
deregulating domestic financial markets and opening of yen bonds domestically by nonresidents and abroad
them to foreign participation, strengthening prudential by residents were relaxed. The market for euro-yen cer-
policies, and modernizing monetary policy manage- tificates of deposit was created, euro-yen credit opera-
ment. An offshore center was established to promote tions were liberalized, and the euro-yen market for
the international use of the yen. commercial papers was decontrolled. Foreign banks
Until the mid-1970s, Japan's policy was designed to were allowed to issue euro-yen bonds, and the with-
discourage the international use of the yen. The main holding tax on nonresidents' interest earnings on euro-
concern was that substantial yen holdings by nonresi- yen bonds issued by residents was removed. In parallel,
dents would jeopardize the Bank of Japan's control domestic financial deregulation continued, monetary
over money supply and increase exchange rate volatil- policy management was modernized, and prudential
ity.1 Capital flows and domestic financial activities policies were strengthened.
were tightly regulated. In addition to the liberalization of international trans-
In response to a slowdown in economic growth and actions in yen, the Japanese offshore market was cre-
widening budget deficits during the late 1970s-early ated in 1986 to promote the international use of the yen
1980s, Japan started to deregulate domestic financial and to attract euro-yen banking to Japan. Offshore
markets and the capital account. Interest rates were banks were allowed to accept yen deposits from non-
deregulated, and access of nonresidents to the Gensaki residents, and these deposits were freed from reserve
market for repurchase agreements on government requirements and other controls. By design, the off-
bonds and to fiscal bills was liberalized. The positive shore market was separated from onshore banking to
list system for regulating capital account transactions prevent "seepage" between them and to ensure the ef-
was replaced with the negative list system in 1980, al- fectiveness of domestic financial regulations.3 In par-
though many capital controls remained in place. ticular, residents were not allowed to participate in the
Cross-border yen transactions were decontrolled offshore market, transfers between domestic and off-
during 1984—89. Policy orientation concerning the in- shore accounts were restricted, and loans contracted in
ternationalization of the yen was changed in 1984, after the offshore market could not be used to finance do-
the Yen-Dollar Committee (a working group of Japan- mestic activities. In addition, securities transactions
ese and U.S. officials) and the Ministry of Finance ac- were restricted. Since its creation, the offshore market
knowledged the potential role of financial liberaliza- has grown rapidly, becoming a major international cen-
tion and the increasing internationalization of the ter for yen transactions.
Japanese economy and proposed a program of liberal- The deregulation of foreign exchange transactions
izing the international use of the yen.2 Subsequently, and the promotion of the internationalization of the yen
the conversion of foreign currencies into yen was de- continued in the 1990s. Regulations on securities in-
vestments and lending, as well as on residents' deposits
in foreign banks were liberalized. Withholding taxes on
Sources: Eken (1984); Tavlas and Ozeki (1992); Dominguez government papers for nonresidents were lifted in 1999
(1999); and Morsink and others (1999). as part of the "big bang" initiative.
1The loss of monetary control was expected to be less in
Japan than in other countries, however, given that the econ-
omy was relatively closed.
2 3
This was in part aimed at addressing U.S. concerns that the The model of the Japanese offshore market is similar to
closed nature of Japanese financial markets would artificially that of the international banking facilities, which were created
depress the value of the yen. in New York in 1981.
spread than the domestic market—where depositors ciency, the offshore market could be a convenient in-
can earn higher returns and borrowers can get access strument, especially for large corporations, for hold-
to cheaper financing—because it is not subject to re- ing excess corporate liquidity and obtaining short-
serve requirements and other regulations; it is a term financing for working capital needs.
wholesale market and can respond more quickly to All in all, the liberalization of controls on interna-
investors' needs. Therefore, reemergence of an off- tional transactions in ringgit needs to be approached
shore market could complement the onshore coun- cautiously as part of a well-sequenced medium-term
terpart and provide both domestic and foreign in- strategy of capital account liberalization and finan-
vestors access to diversified financial instruments at cial development. The liberalization of controls on
lower transaction costs and bid-ask spreads. This trade-related international transactions in ringgit
would allow investors to take advantage of high (e.g., on invoicing of trade transactions) could pre-
yields and safe, liquid investments. Owing to its effi- cede the liberalization of international financial

Figure 5.7. Capital Controls

(Indices)
1.0
CoControls on
Controls on
0.9 portfolio inflows
stock market
operations
operations
0.8
Controls on bank
0.7 transactions and
foreign exhange
0.6 market operations \
0.5
Controls on Controls on
0.4 international . . . . .
portfolio outflows
transactions
0.3 in ringgit
0.2
0.1
0
1991 93 95 97 99
transactions in ringgit. Benefits and risks of the re- Dickey-Fuller (ADF) tests are used to determine
laxation of controls will need to be carefully as- the order of integration for the variables in question
sessed, and proper safeguards and supporting poli- (Table 5.8.). Portfolio assets, domestic prices, and
cies to control the risks associated with liberalization the real exchange rate appear as I(1). The domestic
will need to be put in place. In this regard, a interest rate and the foreign interest rate in real terms
strengthened financial market environment, the (FRR) are found to be stationary.
maintenance of internally consistent monetary and Cointegration is tested using Johansen's maxi-
exchange rate policies, and adequate prudential sur- mum likelihood procedure with six lags (Table
veillance are particularly important. 5.9). The null of no cointegration is rejected in
favor of one cointegrating relationship. The recur-
sively estimated eigenvalue is reasonably stable
Appendix: A Preliminary Empirical over time. Individual tests on the feedback coeffi-
cients imply that the real exchange rate is weakly
Analysis of Malaysia's Capital Controls exogenous, i.e., there is no feedback from the coin-
The effect of capital controls on portfolio invest- tegrating relationship to the real exchange rate.
ment is examined in a simple error-correction Portfolio assets, domestic prices, the real exchange
model. Portfolio investment is modeled as a func- rate, and the trend are all significant in the cointe-
tion of internal and external factors, including the grating vector.
domestic and foreign interest rates and prices, the With weak exogeneity restrictions, the cointegrat-
real exchange rate, and capital controls.13 The ing vector is given by:
study covers the period from January 1991 to Octo- CI = pfa - 3.13 rex - 7.43 p - 0.01 trend.
ber 2000, and the evolution of capital controls dur-
ing this period is shown in Figure 5.7. Table 5.7 de- The restricted feedback coefficients for pfa and p are
scribes the data.14 0.03 and 0.004, respectively. The positive relation-
ship between pfa and rex and p is consistent with
economic theory.
13
The analysis draws on the existing literature on the econom- Weak exogeneity implies that the cointegrating
ics of capital flows. See, e.g., Fernandez-Arias and Montiel vector can be analyzed in a two-equation conditional
(1995); Calvo, Leiderman, and Reinhart (1996); and Cardoso and error-correction model, whereby portfolio invest-
Goldfajn (1998). For surveys on capital controls, see Eichengreen
and others (1998); and Ariyoshi and others (2000). ment and prices are modeled as changing in re-
14
Data on government expenditures and budget balance are not sponse to a disequilibrium in portfolio assets,
available on a monthly basis and were not included in the model. changes in macroeconomic variables, and capital

Table 5.7. Data
Variable Description Source
PFA Net foreign portfolio assets in Malaysia, billions Estimated based on Bank Negara Malaysia
of U.S. dollars Cash Balance-of-Payments Reporting System
and IMF International Financial Statistics
REX Real exchange rate, index Calculated based on IMF International Financial
Statistics
P Consumer price index IMF International Financial Statistics
FP U.S. consumer price index IMF International Financial Statistics
R Money market interest rate adjusted for expected Calculated based on IMF International Financial
depreciation, percent Statistics and Bloomberg
FR Eurodollar rate in London, percent IMF International Financial Statistics
CPFIN Controls on portfolio inflows, index
CPFOUT Controls on portfolio outflows, index
CRM Controls on international transactions in ringgit, Calculated based on the IMF Annual Report
index on Exchange Arrangements and Exchange
Restrictions and Malaysia's exchange control
regulations1
CBNKFX Controls on bank transactions and foreign exchange
market operations, index
CSM Controls on stock market operations, index
ASIA Dummy variable taking the value of 1 for the period of
the Asian crisis, and 0 otherwise
1For more details on the methodology for constructing the capital control indices, see the footnote to Figure 5.1.
Table 5.8. Unit Root Tests1,2
H0 pfa rex inflows per seRhad no FRR
1(0) -3.16 3 -1.96 -1.57 -5.23** -4.15**

1(1) -3.64* 3 -8.24** -8.90**
1ADF statistics are based on the highest significant lag. Constant and trend included.
2** (*) indicates significance at the 1 percent (5 percent) level. Lowercase denotes logarithm.
3
For the period from 1993(3) to 1997(1). Prior t o 1993(3) and after 1997(1) pfa also appears as I(1).
controls. An autoregressive distributed lag model is during January 1991-October 2000. The direction
reduced to the following parsimonious model and magnitude of these effects, however, varied by
through general-to-specific modeling (Table 5.10): the type of capital control. More specifically,
The model is fairly well specified. The goodness- • The regulation of portfolio inflows per se had no
of-fit, autocorrelation, and heteroscedasticity statis- significant effects on portfolio investment. The
tics are acceptable. Normality holds. The Chow tests tightening of these controls, however, tended to
imply constancy. discourage portfolio investment.
The results suggest that capital controls (their level • Controls on portfolio outflows had no significant
and/or changes in them) generally had statistically effects on portfolio investment, neither in terms
significant effects on portfolio investment in Malaysia of their level nor changes in them.

Table 5.9. A Cointegration Analysis1
Maximum eigenvalue statistics

p=0 p <\ 1 p<2
30.620** 9.212 6.719
Standardized eigenvectors (3
pfa rex P trend
1.000 -3.277 -4.574 -0.015
-2.529 1.000 70.428 -0.194
-0.296 0.715 1.000 0.001
Standardized feedback coefficients a

pfa 0.036 0.007 0.053
rex 0.019 0.007 -0.123
P 0.005 -0.001 -0.003
Tests of weak exogeneity X2( 1)

pfa rex P
11.420** 2.429 12.445**
Tests of significance X2( 1)

pfa rex P trend
5.205* 12.089** 6.646** 8.270**
* (*) indicates significance at the 1 percent (5 percent) level. Constant, trend, and a dummy for 1991 included.
Table 5.10. A Conditional Error Correction Model of Portfolio Investment1,2
Equation for Apfat Equation for Apt

Variable Coefficient t-value Coefficient t-value
Apfat_1 0.191*** 2.610 0.020** 2.108

Apfat_2 -0.018 -1.296
Apfat_4 -0.067*** -6.245
Apfat_7 0.196*** 2.618 -0.020** -2.004
Apfot_9 0.275*** 3.539
Apt-9 -2.342*** -4.012 -0.299*** -4.029
clt_1 0.058*** 2.901 0.004 1.536
Clt-2 0.092 1.372 0.018 1.526
Clt-3 -0.119* -1.883 -0.011 -0.977
Clt-6 -0.006*** -3.514
Arex t 0.060*** 5.495
Arext-2 0.431** 2.186 0.054 1.450
Arex t _ 4 0.045*** 4.510
Arext-9 0.009 1.067
AR t 0.026* 2.455
ARt-1 -0.010*** -4.795
ARt-2 0.034*** 2.965 -0.008*** -4.600
AFRRt 1.714** -2.456
AFRRt_2 1.154* -1.746 0.212** 2.384
AFRRt-8 1.706** 2.519 -0.396*** -4.038
ASIAt_6 0.003** 2.355
ASIAt_9 0.038*** 5.012 -0.005*** -3.004
Constant 1.427*** 5.046 0.210*** 4.126
CPFINt -0.004 -1.436
CRMt 0.052*** 3.026 -0.006*** -2.738

Table 5.10 (concluded)
Equation for Apfat Equation for Apt

Variable Coefficient t-value Coefficient t-value
CBNKFXt -0.122*** -4.982 0.013*** 2.795

ACPFINt -0.087*** -5.599 0.011*** 3.481
ACSMt -0.101*** -3.679 -0.006 -1.547
Single equation tests Portmanteau = 32.27 Portmanteau = 24.10
AR 1-7 F(7,58) = 1.94 AR 1-7F(7, 58)= 1.95
Normality X2 (2) = 0.90 Normality X2 (2) = 0.51
ARCH 7 F(7,51) = 0.48 ARCH 7 F(7, 51) = 0.26
Heteroscedasticity test: Heteroscedasticity test:
X2 (55) = 55.77, F (55, 9) = 0.17 X2 (55) = 50.99, F (55, 9) = 0. 15
Number of observations = 108, 1991 (11) to 2000( 10).
Full information maximum likelihood estimation. Strong convergence.
Log-likelihood = 1172.235. LR test of over-identifying restrictions: X 2 (14) = 12.03.
Vector tests: Portmanteau = 77.07. AR 1-7 F(28, 112) = 1.12. Normality X 2 (4) = 0.94.
Heteroscedasticity test: X 2 (165) = 150.54 and F-form (165, 39) = 0.22.
1*** (**, *) indicates significance at the 1 percent (5 percent, 10 percent) level.

2
The model controls for seasonality and the following time periods: 1991, 1993(11), 1994(1), and 1995(2).
• The regulation of bank operations and foreign tion, IMF Occasional Paper No. 190 (Washington: In-
exchange transactions tended to reduce portfo- ternational Monetary Fund).
lio investment. By influencing banks' risk- Bank Negara Malaysia, 2000, Annual Report 1999 (Kuala
taking incentives, regulatory measures pertain- Lumpur).
ing to bank operations apparently affected capi- Bercuson, Kenneth, ed., 1995, Singapore: A Case Study in
Rapid Development, IMF Occasional Paper No. 119
tal account transactions. Controls on swap and (Washington: International Monetary Fund).
forward transactions with nonresidents, in turn, Calvo, Guillermo, Leonardo Leiderman, and Carmen
directly constrained hedging and portfolio man- Reinhart, 1996, "Inflows of Capital to Developing
agement opportunities of foreign investors. Countries in the 1990s," Journal of Economic Per-
• The regulation of international transactions in spectives, Vol. 10 (Spring), pp. 123-39.
ringgit had a positive, albeit a relatively small, ef- Cardarelli, Roberto, Jeanne Gobat, and Jaewoo Lee, 2000,
Singapore—Selected Issues, IMF Staff Country Re-
fect on portfolio investment. This positive effect port No. 00/96 (Washington: International Monetary
may reflect the role of these controls in abating Fund).
speculative pressures on the ringgit. Changes in Cardoso, Eliana, and Ilan Goldfajn, 1998, "Capital Flows
controls on international transactions in ringgit to Brazil: The Endogeneity of Capital Controls," Staff
had no significant effects on portfolio investment. Papers, International Monetary Fund, Vol. 45
• The tightening of controls on equity market trans- (March), pp. 161-202.
Cassard, Marcel, 1994, "The Role of Offshore Centers in
actions tended to discourage portfolio investment. International Financial Intermediation," IMF Work-
The regulation of trading in equities per se had no ing Paper 94/107 (Washington: International Mone-
significant effects on portfolio investment. tary Fund).
The results should be considered only preliminary. Cheong, Latifah M., 2000, "Evaluation of Capital Con-
Their interpretation is subject to a caveat because the trols: Financial and Economic Implications," paper
time series used in this study are relatively short to presented at the MIER National Economic Outlook
fully reflect effects of the capital controls introduced Conference, Kuala Lumpur, Malaysia (January).
Demirguc-Kunt, Asli, and Harry Huizinga, 1995, "Barri-
recently during the Asian crisis. ers to Portfolio Investments in Emerging Stock Mar-
kets," Journal of Development Economics, Vol. 47
(August), pp. 355-74.
References Dominguez, Kathryn M., 1999, "The Role of the Yen," in In-
ternational Capital Flows, ed. by Martin Feldstein, a Na-
Ariyoshi, Akira, and others, 2000, Capital Controls: tional Bureau of Economic Research conference report,
Country Experiences with Their Use and Liberaliza- pp. 133-71 (Chicago: University of Chicago Press).

Eichengreen, Barry, and others, 1998, Capital Account Issues, World Economic and Financial Surveys
Liberalization: Theoretical and Practical Aspects, (Washington: International Monetary Fund).
IMF Occasional Paper No. 172 (Washington: Interna- -, 1999, Annual Report on Exchange Arrangements
tional Monetary Fund). and Exchange Restrictions (Washington: Interna-
Eken, Sena, 1984, "Integration of Domestic and Interna- tional Monetary Fund).
tional Financial Markets: The Japanese Experience," Johnston, Barry, and others, 1999, Exchange Rate
Staff Papers, International Monetary Fund, Vol. 31 Arrangements and Currency Convertibility: Develop-
(September), pp. 499-548. ments and Issues, World Economic and Financial
Errico, Luca, and Alberto Musalem, 1999, "Offshore Surveys (Washington: International Monetary Fund).
Banking: An Analysis of Micro- and Macro-Pruden- Kochhar, Kalpana, and others, Malaysia—Selected Issues,
tial Issues," IMF Working Paper 99/5 (Washington: IMF Staff Country Report No. 99/86 (Washington:
International Monetary Fund). International Monetary Fund).
Feldstein, Martin, ed., 1999, "International Capital Morsink, James, and others, 1999, Japan—Economic and
Flows," National Bureau of Economic Research con- Policy Developments, IMF Staff Country Report
ference report (Chicago: University of Chicago No. 99/181, Revision 1 (Washington: International
Press). Monetary Fund).
Fernandez-Arias, Eduardo, and Peter J. Montiel, 1995, Park, Yung C, 1998, "Gradual Approach to Capital Ac-
"The Surge in Capital Inflows to Developing Coun- count Liberalization: The Korean Experience," a
tries: Prospects and Policy Response," World Bank paper prepared for the seminar of the International
Policy Research Working Paper No. 1473 (Washing- Monetary Fund on Capital Account Liberalization,
ton: World Bank). Washington, March.
Frankel, Jeffrey, 1984, "The Yen/Dollar Agreement: Liber- Securities Commission, Malaysia, 2000, Annual Report
alizing Japanese Capital Markets," Policy Analyses in 1999 (Kuala Lumpur).
International Economics, Vol. 9 (Washington: Insti- Tamirisa, Natalia, 1999, "Exchange and Capital Controls
tute for International Economics). as Barriers to Trade," Staff Papers, International
International Finance Corporation, 1999, Emerging Stock Monetary Fund, Vol. 46 (January), pp. 69-88.
Markets Factbook (Washington: International Fi- Tavlas, George S., and Yuzuru Ozeki, 1992, The Interna-
nance Corporation). tionalization of Currencies: An Appraisal of the
International Monetary Fund, 1998, International Capital Japanese Yen, IMF Occasional Paper No. 90 (Wash-
Markets: Developments, Prospects, and Key Policy ington: International Monetary Fund).

VI Financial Sector Issues
Mark H. Krysl and Michael Moore
Iration
n response to the Asian crisis, Malaysia undertook
a number of policy measures to curtail the deterio-
in the financial sector and assist in its recov-
in 1996 and 1997. As a result, the level of credit in re-
lation to GDP was high (160 percent), and financial
institutions were exposed to those vulnerable sectors.
ery. These actions were initiated early and, sup- There was, nevertheless, the sentiment that the bank-
ported by better domestic and regional economic ing sector in Malaysia was more sound than those in
conditions, have led to substantial improvement in the other crisis countries, because its financial institu-
the sector's performance. This section gives a brief tions had lower amounts of nonperforming loans and
background of the financial sector, discusses its per- higher capital, and there was a stronger banking cul-
formance up to end-2000, and examines key ele- ture, with a better supervisory environment, higher
ments of the reforms under way to strengthen its standards of accounting and auditing practices, and
structure and regulatory regime in the context of the superior prudential supervision.
Financial Sector Masterplan.1 Reform aspects that Three categories of financial institutions are au-
are highlighted include a bank merger program that thorized to take deposits: commercial banks, finance
aims at creating larger and more efficient domestic companies, and merchant banks (Table 6.1).2
institutions, and the upgrading of the prudential su-
pervision system. • Commercial banks engage in retail and corporate
banking, and are the only institutions authorized
to take demand deposits.3 Through subsidiaries,
commercial banks provide other financial ser-
Background vices, including merchant banking, stockbroking,
The financial crisis in Malaysia was compara- insurance, and finance company activities. For-
tively well contained, attributable in part to bank re- eign banks have operated in Malaysia prior to its
structuring efforts and the development of domestic independence in 1957. Current regulation limits
capital markets in the 1980s. The country's low for- to 30 percent new equity holdings by foreigners in
eign debt at the outset also placed it in a relatively domestically controlled banks4 and restricts exist-
good position to confront the crisis. The country's ing foreign banks from opening new branches.
traditional policies to limit short-term borrowing, • Finance companies are able to offer hire purchase
encourage foreign direct investment inflows, and lending and other types of installment credit to
rely on equity capital prevented the corporate sector consumers and small businesses, with funding
from building up excessive unhedged foreign ex- provided primarily from time and savings de-
change exposures and very high debt/equity ratios posits. Facing diminishing returns from tradi-
that were so damaging in the other crisis countries. tional business lines, finance companies went into
Like the other countries, however, a decade of riskier real estate and share purchase lending,
strong growth prior to the crisis lulled governments making them more vulnerable to an economic
and creditors, both foreign and domestic, into compla-
cency. Easy access to bank credit contributed to spec-
ulative price bubbles in the real estate and securities 2
This structure may change further following the completion of
sectors, and credit growth in the years leading to the the merger process.
3
Malaysia has two pure Islamic banks, Bank Islam and Bank
crisis rose substantially, reaching 25 percent annually Muamalat, with total assets of RM 14 billion at end-2000, repre-
senting around 2 percent of the banking system assets. They offer
Islamic bank products such as interest-free leasing, hire purchase
lending, profit sharing, and joint-venture financing. The total
1This section was prepared largely prior to the finalization of market share of Islamic banking assets increased to 6.9 percent in
the bank merger program and the issue of the Financial Sector 2000 from 5.5 percent in 1999.
Masterplan (March 2001), and includes only a brief outline of 4
Foreign banks have minority interest in ten domestic commer-
those initiatives. cial banks, three finance companies, and seven merchant banks.

FINANCIAL SECTOR ISSUES
Table 6.1. The Banking System (Depository Financial Institutions)

(In billions ofringgit;asof February 2001)
Number of
Depository
Financial Percent Percent Percent Percent
Institutions of Total Assets of Total Loans1 of Total Deposits of Total
Commercial banks 29 57 516 78 305 76 365 77

Domestic-owned banks 15 29 391 59 233 58 278 59
Foreign-owned banks 14 27 125 19 72 18 87 18
Finance companies 12 24 110 17 11 19 84 18
Merchant banks 10 20 38 6 17 4 27 6
All depository financial
institutions 51 100 664 100 398 100 475 100
Source: Bank Negara Malaysia.

1Excludes loans sold to Cagamas, the national mortgage corporation.
downturn. Because of the vulnerability, Bank Ne- • Following a pretax loss totaling RM 8.5 billion
gara Malaysia began consolidation of the finance during the 12-month period ending March 1999,
company sector early in the crisis, and the process the banking system began to generate profits in
is expected to be accomplished with the comple- line with the turnaround of the overall economy.
tion of the bank merger program. The system recorded an aggregate pretax profit
of RM 4.7 billion during 1999 and RM 9.7 bil-
Merchant banks are involved primarily in fee- lion in 2000, with major banking institutions
based activities, such as syndication of loans, turning profitable.
corporate advisory services, securities under-
writing, and portfolio management. They » The interest margin of the banking system also
can only accept time deposits greater than widened in 1999, reversing the trend of the pre-
RM 200,000. In past years, lending by merchant vious year, and leveled off in 2000. For the com-
banks grew to be a key—and at the time prof- mercial banks, the interest margin declined to a
itable—activity, but the diversion proved very low of 3.2 percent in September 1998, owing to
costly with a weakening economy. Bank Negara locked-in deposit funds in the face of more vari-
Malaysia has redirected merchant banks back to able loan rates. The margin improved as interest
the traditional fee-based activities by limiting rates began to fall, reaching 4.5 percent in the
their loan exposures. final quarter of 1999, but declined slightly to 4.3
percent by end-2000. The margin for finance
companies reached a low point of 1.7 percent in
Recent Financial Sector Performance June 1998, as deposit rates rose against a high
volume of fixed rates for hire purchase loans;
The financial sector, which suffered losses in since then, the margin grew substantially from
1998, has recovered along with the rest of the econ- the declining interest rates, to about 6.5 percent
omy. In 1999-2000, the banking system recorded in 1999 and 7 percent by end-2000.
pretax profits. The recovery can be attributed in
part to the restructuring of the banking sector that • Loan growth was flat in 1999 and picked up
took place (see below), and in part to a more stable slowly to 5.4 percent by end-2000. However,
interest rate environment that led to a sharp reduc- taking into account the large amount of loans
tion in loan loss provisions. Nonperforming loans sold to, or managed by, Danaharta,5 bad debts
were reduced to 15.3 percent of total loans at end- written off, and loans converted into private debt
2000 (Table 6.2). Although asset growth averaged securities, the increase in total bank financing
only 4 percent in 1999-2000, there is no evidence has been much larger.
of constraints on the supply of credit to the econ-
omy, as witnessed by information on loan approvals 5
See Box 6.1 for a description of Danaharta's role in the re-
and disbursements. structuring of the banking sector.

Recent Financial Sector Performance
Table 6.2.The Trend in Nonperforming Loans1
1997 1998 1999 2000

December December December March June September December
(In billions of ringgit)
Commercial banks 14.2 44.9 40.9 41.7 41.6 42.3 41.6

Finance companies 10.0 24.9 19.1 18.1 18.3 17.9 17.7
Merchant banks I.I 7.2 5.6 5.1 5.2 5.3 4.3
Total 25.2 77.0 65.5 64.9 65.0 65.5 63.6
Total including loans sold

to Danaharta2 90.0 100.9 101.3 101.7 102.3 101.3
(In percent of total loans)

Commercial banks 4.9 15.0 13.8 14.1 13.7 13.7 13.2
Finance companies 9.2 27.0 23.6 22.4 22.5 21.7 21.1
Merchant banks 4.8 32.2 29.6 29.0 29.7 30.5 24.7
Total 6.0 18.6 16.6 16.4 16.2 16.1 15.3
Total including loans sold

to Danaharta2 0.0 21.8 25.5 25.6 25.3 25.1 24.3

1Loans are classified as nonperforming if payments are overdue for three months or more; prior to January 1,
1998, this period was six months. Total loans include housing loans sold to Cagamas.
2
Loans were first sold t o Danaharta (national asset management agency) beginning in the third quarter of 1998.
Box 6.1. Danaharta: Asset Management and Recovery
Danaharta, a wholly government-owned agency, was chased loans with face values totaling about RM 20.5
established in June 1998 to acquire and manage banks' billion at market value, as determined by independent
impaired assets. It has RM 1.5 billion in capital provided auditors, and averaging 45 percent of the face values
by the Finance Ministry, and is authorized to issue up to of the loans. Financial institutions were allowed up
RM 15 billion (face value) in zero-coupon bonds. RM to five years to amortize the difference between
11.4 billion in such bonds has been issued. the book value and the sale price, thereby avoiding
Danaharta was to purchase nonperforming loans with immediate recognition of the total loss. In addition,
face values of RM 5 million or more. Financial institu- Danaharta has been managing RM 26.2 billion worth
tions seeking recapitalization from Danamodal were re- of assets owned by the government in connection
quired to sell their nonperforming loans in excess of 10 with government-assisted bank mergers.
percent of total loans to Danaharta as a precondition. Resolution has been reached for 74 percent of
Legislation has vested Danaharta with special power the RM 46.7 billion worth of nonperforming loans
over borrowers, including insulation of the agency (and acquired or managed by Danaharta, involving various
of subsequent purchasers) from undisclosed claims workout processes. For viable loans, these include
made after the initial purchase of assets by Danaharta; loan restructuring, settlement, and special administra-
the ability to appoint special administrators without tion; for nonviable loans, these include sales
having to go to court; and the power to abrogate under- of collateral, sales of business, foreclosures,
lying contracts when it forecloses on collateral. liquidation, and special administration via a bid
In late 2000, when its window for acquiring non- process. The average recovery rate of the loans was
performing loans was closed, Danaharta had pur- 66 percent.
Barring severe external shocks, Malaysian banks ing system (Table 6.3). There are, however, potential
are expected to recover at a faster pace than from the vulnerabilities that could slow the recovery. Sub-
previous recession because of less erosion of their stantial increases in interest rates could trigger fur-
asset quality, and because of the authorities' rapid ther credit problems for weaker borrowers and cre-
support for, and proactive restructuring of, the bank- ate additional nonperforming loans.

VI FINANCIAL SECTOR ISSUES
Table 6.3. Asset Quality Indicators1

(As of January 31,2001)
Total Bad-Debt
Nonperforming Nonperforming Provision to
Loans2 Loans to Loans Nonperforming Loans3
3-month 6-month 3-month 6-month 3-month 6-month
(In billions of ringgit) (In percent)

Commercial banks 40.9 32.5 13.0 10.3 57.3 66.4
Finance companies 17.9 13.5 21.2 16.0 51.9 62.1
Merchant banks 3.9 3.0 22.4 17.2 41.0 52.6
Banking system 62.7 49.0 15.1 11.8 54.8 64.3

1Including loans sold to Cagamas with full recourse.
2
Nonperforming loans are shown gross of interest in suspense. Malaysian accounting calls for the continued ac-
crual of interest on nonperforming loans with an offsetting provision to the interest-in-suspense account. The
interest-in-suspense balance is a provision equal t o the amount of interest accrued but not collected from non-
performing loans. Indicators are shown here for classifications of loans as nonperforming after payments are over-
due for three months and six months.
3
Total bad-debt provision equals the aggregate of provisions for general, specific, and interest in suspense.
Financial Sector Restructuring: financial sector indicators, as measured by nonper-

Comparative Perspective forming loans and capital adequacy ratios, have im-
proved in all countries, but those in Malaysia com-
In response to the Asian crisis, the affected coun- pare favorably (Table 6.5).
tries—despite different initial conditions and the ap-
proaches they used—adopted similar policies aimed Financial and Corporate Debt Restructuring
at improving the structure of the financial sector and
lessening its vulnerability (Table 6.4). To prevent a Malaysia's initiatives to restructure financial and
collapse of the system, all countries provided a blan- corporate debt in a coordinated way entailed the es-
ket deposit guarantee and liquidity support to finan- tablishment of Danaharta to acquire nonperforming
cial institutions. Subsequently, each country adopted loans and help banks clear their balance sheets,
some form of asset management strategy to address Danamodal to recapitalize banks (Box 6.2), and the
nonperforming loan problems, assigned high priority Corporate Debt Restructuring Committee to facili-
to upgrading supervisory and regulatory standards to tate debt workout by large borrowers (Box 6.3).6
international best practices, and sought to recapitalize This multipronged approach has proved to be a cred-
financial institutions based on those norms. In the ible plan in the restructuring of Malaysia's financial
process, bank closures or merger programs were un- sector.
dertaken to establish stronger financial systems. Also, Danaharta's broad legal mandate helped ensure
standards of corporate governance were upgraded. that nonperforming loans would be dealt with
Malaysia has achieved considerable progress in promptly, while its setup under the Companies Act
implementing these reforms in comparison to the
other crisis countries. The approach adopted by
Malaysia (and also by Korea) in resolving bad loan 6
problems and restructuring banks involved a high Korea also adopted a centralized approach to asset resolution
and bank recapitalization from the outset, with active government
degree of government involvement, which had the involvement. In contrast, Thailand aggressively liquidated the
advantages of speed and coherence, notwithstanding impaired assets of closed finance companies through a central
the possibility that it could also raise expectations of agency, but it did not permit public sector purchases of impaired
future government bailouts. Malaysia's efforts also assets from private commercial banks. Instead, each bank was en-
couraged to establish its own asset management company. Fol-
benefited from the country's relatively strong initial lowing a change in the government in 2001, however, Thailand
position, including its well-developed legal and in- announced that it would establish a centralized asset management
stitutional frameworks. Reflecting these efforts, the company.

Financial Sector Restructuring: Comparative Perspective
Table 6.4. Financial Sector Restructuring in Malaysia, Korea, and Thailand
Malaysia Korea Thailand
Initial government response

Establishment of an overarching restructuring authority Yes1 Yes Yes
Establishment of a separate bank restructuring authority Yes (Danamodal) No No
Liquidity support (in billions of U.S. dollars) 9.2 23.3 24.1
(In percent of GDP) 13 5 20
Introduction of a blanket guarantee Yes Yes Yes
2
Deposit insurance No No No
Financial distress resolutions
Bank closures 1 of 15
Elimination or dilution of current shareholder stakes of
insolvent banks Yes Yes Yes
Closure of other financial institutions 0 Over 200 57 of 91
Mergers or interventions Yes, 54 t o be merged Yes, 8 of 26 absorbed Yes, 5 banks and 13
into 10 groups by 12/00 by other banks finance companies,3
also 3 banks privatized
Bank recapitalization strategies
Public funds for recapitalization Danamodal injected $7.1 Government injected Government injected
billion into 10 institutions $36 billion into 9 about $11 billion into
commercial banks; 5 public banks
out of 6 major banks
now 90 percent
controlled by state
Majority foreign ownership of banks Control of domestic 1 announced; 4 completed;
banks not allowed; 1 pending 5 2 pending6
foreign bank share
is, however, significant 4
Instruments used to recapitalize and purchase Bonds o r cash Bonds or cash Debt-to-equity
nonperforming loans conversions
Asset resolution strategies
Establishment of a centralized asset management Yes (Danaharta) 7 Yes ( K A M C O ) 8 No, but an asset
corporation management company
is planned9
Operational autonomy of restructuring agencies Yes Yes Not applicable
Centralized asset management companies purchased Purchased assets are Assets initially pur- Not applicable
assets at subsidized prices valued by independent chased above market-
outside auditors clearing prices with
recourse. Since 1998,
purchase endeavors at
market prices
Nature of agency; restructuring or disposition Restructuring N o t clearly defined; Not applicable
mostly disposal of
assets
Eligibility of loans All financial institutions, All financial institutions Finance companies
including Labuan thus far, but also
subsidiaries of Malaysian banks subject to
banks and development intervention
financial institutions
Sources: Information provided by country authorities; Claessens and others, 1999, "Financial Restructuring in East Asia: Halfway There?" Financial Sec-
tor Discussion Paper No. 3 (Washington:World Bank); Lindgren and others, 2000,"Financial Sector Crisis and Restructuring: Lessons from Asia," IMF Oc-
casional Paper No. 188 (Washington: International Monetary Fund).
1Steeringcommittee chaired by the central bank.
2
Under consideration in Malaysia and Korea.
3
Between government-owned institutions in which the government has intervened.
4
Foreign banks are allowed to purchase up to 30 percent equity of domestic banks.
5
From 15 percent to 100 percent.
6
Approval required from Board of Investment.
7
Assets transferred; loans larger than RM 5 million and mostly loans secured by property or shares.
8
The powers and resources of a preexisting asset management company were substantially increased; worst assets were to be transferred.
9
Nonperforming loan workout is decentralized. Three banks have established private asset management companies and more are being considered.
Hybrid approach to reprivatization of banks subject to intervention is evolving.Thailand has announced that it will establish a centralized asset manage-
ment company.

Table 6.5. Selected Financial Indicators of the Asian Crisis Countries
Malaysia1 Indonesia2 Korea3 Philippines4 Thailand5
Nonperforming loans of the banking system

(as a percent of total loans)
December 1997 6.0 8.4 6.1 4.7 22.6
December 1998 18.6 48.6 7.4 10.4 45.0
December 1999 16.6 32.9 8.3 12.3 38.9
December 2000 15.3 25.6 5.6 15.1 17.9
Risk-weighted capital ratio(s) of the banking

system
December 1997 10.5 8.0 7.0 15.9 9.9
December 1998 11.8 -11.6 8.2 17.5 10.9
December 1999 12.5 -2.4 10.8 17.0 15.3
December 2000 12.4 12.7 10.3 15.8 12.4
Total outstanding bad-debt provision(s) of

the banking system (as a percent of
nonperforming loans)
December 1997 66.2 39.6 23.7 42.1
December 1998 42.4 40.4 59.6 36.4 21.8
December 1999 50.2 29.0 75.4 45.1 32.2
December 2000 53.8 84.7 43.6
Sources: Data provided by the country authorities; and IMF staff estimates.
1Nonperforming loans include those with interest in suspense and specific provisioning, and are defined as loans for which payments are overdue for
three months or more.
including Rp 254 trillion worth of nonperforming loans transferred t o the Indonesian Bank Restructuring Agency; nonperforming loans of the banking
system totaled 65.1 percent at end-March 2000.The capital ratio data are equity (as declared by banks) as a percent of total (unweighted) assets (data for
2000 relate t o October).
3
For nonperforming loans, commercial banks' data are used (based on delinquency); December 1997 data include past-due loans over six months and
bankruptcy loans; December 1998 data include past-due loans over three months and bankruptcy loans; December 1999 data include past-due loans
over three months and nonaccrual loans reflecting forward-looking criteria. Risk-weighted capital ratio for December 1999 also reflects the application
of new forward-looking criteria. Total outstanding bad-debt provisions of the banking system data show total provisions on balance sheet as a percent of
nonperforming loans.
4
Nonperforming loan data are commercial banks' data. Capital ratios of the banking system are not risk weighted, based on Basel guidelines. Regula-
tions are being drafted in order t o bring the practice into line with the guidelines, in accordance with recently passed legislation.The 2000 nonperform-
ing loan data are for September 2000.
5
Nonperforming loans include those for Thai private and state-owned banks. Capital ratios for Thailand are based on phased-in provisioning rules, as al-
lowed by Bank of Thailand, scheduled t o be fully met at end-2000.Total outstanding bad-debt provision data for December 1998 and December 1999 are
balance sheet loan-loss provisions, and do not include surplus capital in excess of the regulatory minimum.
meant that it would be managed as a private entity, accounts) meant that the process was manageable.
subject to normal auditing. The key issue in asset Furthermore, the emphasis on resolution of nonper-
purchases by Danaharta was realistic valuations to forming loans—and not simply their disposal—
ensure that it did not become a tool for indirect assisted in the restructuring of the corporate sector.
bailouts of existing shareholders, which would un- There are, nevertheless, risks of the agency becom-
dermine the incentives for private sector recapital- ing a warehouse for nonperforming loans unless as-
ization and proper governance of the agency and the sets are upgraded and sold before long.
banks. There are indications that loan valuations by The mandate of Danamodal inspired confidence
Danaharta were reasonable, although some latitude that all domestic financial institutions would be re-
was given to banks in that the losses on the sales of capitalized to the required standards, maintaining the
their assets to the agency could be amortized over safety and soundness of the banking sector. The re-
five years rather than recognized immediately. quirement that institutions seeking Danamodal's
Overall, the plan for Danaharta was well con- capital would have to sell nonperforming loans in
ceived, with nonperforming loans being taken over excess of the specified proportion to Danaharta gave
to ease bank operations. Concentration on larger banks the incentive to deal with their bad assets in a
nonperforming loans (involving only 2,000 to 3,000 timely and coherent manner. Danamodal achieved

Toward a More Resilient Banking Sector
Box 6.2. Danamodal: Bank Recapitalization
Danamodal was established in July 1998 with the Institutions requesting capital injections must submit
main objective of recapitalizing the banking system. recapitalization plans and are subject to monthly re-
Capital injections from Danamodal were destined to porting of performance against a list of targets.
enable institutions to restore their capital adequacy ra- Danamodal exercises control over management by ap-
tios to 9 percent. To fund its needs, Danamodal raised pointing at least two members to the boards of direc-
RM 10.7 billion, comprising RM 3 billion in paid-up tors, of which one is to be an executive director or
capital from Bank Negara Malaysia, and RM 7.7 bil- chairman of the board.
lion raised through the issuance of zero-coupon bonds Ten institutions received a total capital injection of
to financial institutions. RM 7.1 billion, initially in the form of tier-two subordi-
Selection of candidates for recapitalization was nated debt that, per definitive agreement, is to be con-
initially guided by Bank Negara Malaysia's watchlist, verted into equity; irredeemable, noncumulative con-
based on stress tests of banking institutions. vertible preference shares; and/or subordinated loans,
Danamodal's participation was also determined by the depending on the cash flow characteristics of the in-
nonfeasibility of market solutions, the systemic impact strument and circumstances of the banking institution.
of the failure, and the future viability and competitive Eight of the ten institutions receiving assistance have
positioning of concerned institutions. The "first loss" fully repaid their loans. Danamodal's excess funding
principle, by which original shareholders' equity is has gone unneeded because undercapitalized institu-
written down, is applied strictly to all transactions. tions were able to restore capital on their own.
Debt restructuring in Malaysia has taken a number

Box 6.3. Corporate Debt of forms. The approach taken by the Corporate Debt
Restructuring Committee Restructuring Committee was intended to minimize
losses to creditors and company shareholders
The Corporate Debt Restructuring Committee was through coordinated debt workouts that avoid plac-
established in July 1998 to help mediate voluntary ing viable companies into liquidation or receiver-
out-of-court restructuring of large debt involving a ships, and to have banking institutions play a greater
number of major creditors, following the London role in the financial rehabilitation of the corporate
Rules model. Debt restructuring under the Commit- sector.
tee is reserved for viable businesses and not those in
receivership or liquidation. Aggregate bank loans
must be RM 50 million or more, with at least three Prudential Accounting Standards
lending institutions participating, and the creditor
committees representing the interests of at least 75 Prudential accounting standards have been
percent of total debt of all creditors. brought closer to compliance with international best
The Corporate Debt Restructuring Committee has practices for all crisis countries. The valuation of
no legal status, but debt restructuring under its aus- nonperforming loans was hampered by the lack of
pices is facilitated by a joint public-private sector clear market values and continuously changing eco-
steering committee appointed by Bank Negara nomic conditions. To better support the valuation
Malaysia, which is assisted by a secretariat set up in process, all countries tightened their rules for loan
Bank Negara Malaysia. Over 70 companies applied classification, loss provisioning, income recognition,
to the Committee for workout arrangements, with and collateral valuation, and they have substantially
debts totaling RM 39.4 billion. The majority were strengthened supervisory scrutiny of compliance by
property, construction, and diversified holding com-
panies. At end-2000, 21 of these applications with bankers and auditors with these rules. In the case of
debts of RM 7.8 billion had been withdrawn or re- Malaysia, these changes took place in early 1998
jected; 42 applications with debts of RM 27.3 billion (Table 6.6).
have been completed or resolved with assistance
from Danaharta; and 12 applications with debts of
RM 12.1 billion are outstanding. Toward a More Resilient
Banking Sector
its goal of restoring the financial industry's capital As the next step in reforming the financial sector,
level to above precrisis levels. Also, no systemic the Financial Sector Masterplan was issued in March
banking failure was encountered, and the payment 2001. The plan has a long-term vision "to develop a
system functioned smoothly throughout the crisis. more resilient, competitive, and dynamic banking

Table 6.6. Changes in Prudential Standards in Malaysia, Korea, and Thailand
Date when changes took effect 1/1/98 6/30/98 3/31/98
Loan classification: days elapsed before considered 180 days, with parallel Reduced to 90 days Reduced to 90 days
past due classification of 90 days from 180 days from 360 days
for supervision purpose
Present criteria for classifying

substandard 6 months 3 months 3 months
doubtful 9 months 3 months 6 months
loss 12 months 12 months 12 months
Loan-loss provisioning (in percent)

substandard to 20 from 0 20 to 20 from 15-20
doubtful 50 to 50 from 75 to 50 from 100
loss 100 100 100
Interest accrual No change, that is, Reduced to 3 months Reduced to 3 months

accrual for 6 months maximum from 6 maximum from 6
maximum months maximum months maximum
Source: Data provided by the country authorities.
system with best practices, that supports and con- alization and technological advances; and to iden-
tributes positively to the growth of the economy tify the optimal supervisory philosophy to be
throughout the economic cycle, and has a core of adopted within the existing structure, where Bank
strong and forward-looking domestic financial insti- Negara Malaysia remains the sole regulatory au-
tutions that are more technologically driven and thority of the banking system. Two key components
ready to face the challenges of liberalization and of the plan are the bank merger program and signif-
globalization."7 The plan will be a blueprint for the icant changes to regulation and supervision, in line
financial sector for the next ten years and will focus with best practices.
on a series of best practices for the industry, as well
as initiating a process of corporate governance based The Bank Merger Program
on effective risk management. In addition, the capi-
tal market is to have a more important role in the al- For all crisis countries, the strategies for systemic
location of resources. restructuring have sought to restore the financial
The main thrust of the plan is to develop strong do- systems to soundness as soon as possible. The
mestic banking institutions that form the core of an ef- process involves the introduction of the legal, insti-
ficient, effective, and stable financial sector. They will tutional, and policy frameworks necessary for deal-
be expected to operate in an environment of emerging ing with nonviable financial institutions, strengthen-
new technological advances and more differentiated ing viable ones, and resolving value-impaired assets
and demanding consumers, and to provide a more di- in the system.
versified range of financial services. The banking sec- Malaysia has undertaken to deal with its banking
tor will not only serve a more internationally inte- problems through a comprehensive bank merger
grated and dynamic economy, it should also have a program, designed to take advantage of economies-
leading role within this economy. of-scale, to tap potential synergies, and to determine
The broad strategies for the banking sector an exit strategy for the weakest banks. In the
within the plan are to manage the process of liber- process, domestic banks were given broad flexibility
alization, including the positioning of domestic fi- to form their own merger groups, which total ten,
nancial institutions vis-a-vis foreign institutions; to each comprising a commercial bank, a finance com-
strengthen the financial sector in the wake of glob- pany, and a merchant bank (with the exception of
one group). Each group is required to have a mini-
mum capitalization of RM 2 billion by end-2001,
7
Bank Negara Malaysia, 2001, "The Financial Sector Master- implying the asset size of most banking groups to be
plan" (Kuala Lumpur), March. in excess of RM 25 billion (Table 6.7).

Table 6.7. Proposed Banking Groups

(Mergers and acquisitions completed or close to completion as at December 31, 2000)
Assets of Post-
Commercial Finance Merchant Anchor Banks Merger Assets Percentage of
Anchor Banks Banks Companies Banks (in RM billion)1 (in RM billion)1 System Assets
Maybank Pacific Bank Mayban Finance Aseambankers Malaysia 106 145 28.2
PhileoAllied Bank Kewangan Bersatu
Sime Finance
Bumiputra- Bumiputra- Commerce International 58 68 13.2

Commerce Bank Commerce Merchant Bank
Finance
RHB Bank Interfinance RHB Sakura Merchant 49 53 10.3

RHB Delta Finance Bankers
Public Bank Hock Hua Bank Public Finance Public Merchant Bankers 33 52 10.1
Advance Finance
Arab-Malaysian Arab-Malaysian Arab-Malaysian Merchant 11 41 8.0

Bank Finance Bank
Hong Leong Wah Tat Bank Hong Leong 23 40 7.8

Bank Finance Credit
Corporation
Alliance International Sabah Finance Bumiputra Merchant 15 20 3.9

(Multi-Purpose) Bank Malaysia Bolton Finance Bankers
Bank Sabah Bank Alliance Merchant Bank
Affin Bank BSN Commercial Asia Commercial Perwira Affin Merchant 23 32 6.2
Bank Finance Bankers
BSN Finance BSN Merchant Bank
Southern Bank Ban Hin Lee Bank Perdana Finance Perdana Merchant Bankers 17 25 4.9
Cempaka Finance
United Merchant
Finance
EON Bank Oriental Bank EON Finance Malaysia International 15 25 4.9

City Finance Merchant Bankers
Perkasa Finance
Others not Bank Utama MBf Finance Utama Merchant Bank 13 2.5
yet merged

1As at February 28, 2001.
The merger and acquisition phase was largely com- timely manner. In this regard, strategic plans and inte-
pleted by end-2000, but much remains to be done. gration issues for each banking group are important.
Cultural adaptation alone could present barriers to a
quick and smooth progression of the process. Selec- • Integration plans. An important segment of inte-
tion of management teams will be critical where there gration is "cultural changes," namely, blending
is no clear choice of chief executive officer or chief the cultures of two or more entities.
operating officer already in place. Bank Negara • Strategic plans. Such plans are needed to pre-
Malaysia requires that each banking group employ an vent several banking groups from concentrating
independent consulting firm to advise and recommend on the same market sectors.
individuals best qualified to manage the company.
Bank Negara Malaysia will also closely monitor the • Staffing. This issue deals not only with over-
performance of the banks' management teams to en- staffing and retraining and/or early retirement of
sure that any identified weaknesses are addressed in a excess staff, but also with keeping and develop-

FINANCIAL SECTOR ISSUES
ing key individuals to manage the "new" bank- banking institutions to have formalized risk-
ing system for the future. management systems is most important in that it
will create an environment of self-supervision
• Information technology. The integration of two for the industry and enhance risk-based supervi-
or more data systems must be compatible and ef- sion. Stronger risk-management capabilities
ficient, resulting in little or no turmoil to internal of banking institutions, in turn, will render risk-
operations and consumer interests. based supervision more effective. Implementa-
The bank merger program in Malaysia, ambitious tion of formalized risk management at the hold-
in nature and timing, presents short- and longer-term ing company level will further improve the
challenges. The reduction to 10 groups from over 60 systems.
banks (precrisis) is significant, but the distribution in • Improvements of the systems' risk management
terms of asset size will initially be disproportionate. have so far included requirements related to de-
Three of the new banking groups will each comprise rivative products management and liquidity
5 percent or less of total banking assets, less than framework (implemented), and credit risk man-
one-fifth the size of the Maybank group. Some of the agement (scheduled to be formalized by mid-
smaller banking groups may not achieve the desired 2000). Internal risk-management systems would
efficiencies or economies-of-scale, creating compet- entail guidelines for other types of risks, in con-
itive mismatches with their larger counterparts. nection with interest rates, foreign exchange,
The possibility for additional mergers in the future transactions, strategy, reputation, prices, and
could also be triggered by a liberalization of the fi- other types of risks deemed significant to the
nancial services industry to allow for more foreign oversight of each bank's risk propensity.
competition and competitive factors, that is, for
larger companies to dominate specific segments.8 • A bank-by-bank early warning system currently
This possibility needs to be watched closely because being developed by Bank Negara Malaysia will
the improved ability to compete is a key goal of the provide leading indicators for supervisory atten-
bank merger program and the Financial Sector Mas- tion and allow for preventive measures against
terplan. There is also a concern regarding the man- the catastrophic deterioration of individual insti-
agement of some of the new banking groups, tutions and the overall financial sector. The sys-
namely, depth of experience and strengths to manage tem, based on an individual bank failure model,
a more complex and larger entity in a highly com- will serve to enhance confidence in the financial
petitive environment. sector.
• Consolidated supervision is expected to play an
Prudential Supervision and Regulation important role as the bank merger program is
completed, with the new banking groups be-
Malaysia moved rapidly to strengthen the frame- coming larger and more complex, and engaging
work of prudential supervision and regulation in more varied activities. It will help contain
(Table 6.8). In particular, the requirement that banks overleveraging by financial institutions. For
establish internal systems to manage risks, includ- consolidated supervision to be operative, Bank
ing for cross-border transactions, and Bank Negara Negara Malaysia will need to have supervisory
Malaysia's move toward risk-based and consoli- mandates over financial institutions' holding
dated supervision are major steps to enhance the companies and other subsidiaries and affiliates.
soundness of the financial system, especially as the An amendment to the Bank and Financial Insti-
merger program is completed and financial innova- tutions Act would be warranted to include the
tions are embraced by banks in Malaysia. scope of permissible activities for these entities,
• Risk-based supervision—with supervisory atten- their minimum capital requirements, approval
tion for weak institutions—will allow Bank Ne- of dividends by the authorities, and approval for
gara Malaysia to focus its resources on the most the publishing of reports and appointment of
critical areas in the individual institutions, as well directors.
as on the stratification of risk areas across the fi- • Increased transparency of macro- and microeco-
nancial sector. The impending requirement for nomic data and policies will help ensure more
effective supervision. Similar to the other crisis
8
countries, Malaysia is gradually implementing
The share of banking system assets controlled by foreign policies to foster improved corporate gover-
banks increased to 19 percent in 2000 from around 15 percent in
1997. The government's share of the banking system has also in- nance and lower corporate leverage. Ongoing at-
creased over this period, through increased government equity tention is also needed to improve protection for
holding and the recapitalization process. outside investors through stronger enforcement
80
Table 6.8. Steps to Improve Prudential Supervision and Regulation in

Malaysia, Korea, and Thailand
Development of an overall plan Yes Under consideration 2 Yes3

and/or strategy for the Masterplan 1
supervision and regulation of
the industry
Steps t o develop and implement In process 4 Under consideration 5 Under consideration 6

consolidated supervision
Steps t o develop and implement Yes7 Yes8 Yes9

risk-based supervision
Requiring banks and holding Yes10 Unknown 11 Early stages of

companies t o develop and development 12
implement formal risk-
management processes
Source: Information provided by country authorities.

1The Financial Sector Masterplan was released in March 2001; it sets broad goals for the supervision and regula-
tion of the industry.
2
Through the refinancing of a World Bank loan, Korea is in the process of securing an outside consulting firm to as-
sist in developing a plan to bring its supervision and regulation function in step with current worldwide standards.
3
A complete reengineering of the supervision group is under way. As part of the reengineering process, actions
to ensure compliance with international standards and the Core Principles are under way.
4
Bank Negara Malaysia is moving toward consolidated supervision. A change to existing banking law will be re-
quired t o give Bank Negara Malaysia the authority.
5Consolidated supervision is t o be considered as part of an overall plan.
6
Through the proposed Financial Institutions Law and new supervisory policies and procedures, consolidated
supervision is being considered.
7
Risk-based supervision was in the implementation stage when the crisis emerged, and is now being reintro-
duced into the examination process.
8
Risk-based supervision will be part of an overall plan for the supervision and regulation of the financial services
industry.
9
Risk-based supervision approach is embodied in all new policies, procedures, regulations, and other supervisory
guidelines that are developed. A risk-based supervisory approach for on-site examination is in the testing stage.
10
Bank Negara Malaysia is moving the industry toward formalized risk-management processes.This is evidenced
by the implementation of mandated risk-management practices for derivatives, liquidity, and credit.
11
The consultant's plan does not specifically address this issue.
12
New regulations t o accompany the proposed Financial Institutions Law will require appropriate risk manage-
ment processes.
of disclosure requirements and shareholder and and stockbroking companies or discount houses are
creditor rights. to be encouraged, with a view to developing mer-
chant banks into full-fledged investment banks. To
Other Initiatives
facilitate this process, the legal and regulatory
framework governing the banking and securities in-
In April 2000, the authorities announced a consol- dustries (supervised respectively by Bank Negara
idation plan for the stockbroking industry to reduce Malaysia and the Securities Commission) would
the number of companies to 15 from the present 63, need to be harmonized.
in conjunction with a reduction in transaction com- The Financial Sector Masterplan envisages that a
missions, although the stipulation of the final num- deposit insurance fund be established to replace the
ber of brokers was subsequently abandoned, and the current blanket deposit guarantee by the government
schedule revised. The objective of the consolidation that has been in place since January 1998. Contribu-
is to form a group of well-capitalized universal bro- tions to the fund would be risk adjusted. Such a pre-
kers that can provide efficient and cost-effective in- mium structure will avoid the moral hazard created
termediation for investors, and are robust enough to by a blanket guarantee and create an incentive for
withstand the pressures of the stockbroking busi- prudent management in line with the move toward
ness. Furthermore, mergers between merchant banks performance-based prudential regulation.

The 1999 amendment of the Labuan Offshore Fi- cation of the financial players in the center; the de-
nancial Services Authority Act provides the authori- velopment of Islamic banking; and the development
ties with the power to supervise offshore banking ac- of capital market, e-commerce, and the ancillary
tivities in Labuan. The agency is allowed to perform activities.
on-site work in the branches where necessary and to An efficient, progressive, and comprehensive
require the submission of any information relating to Islamic financial sector is also an element of the
specific activities within its supervisory mandate. long-run vision for Malaysia's financial system. To
The amendment also provides for home country su- achieve this, efforts will be needed to enhance insti-
pervisors, including Bank Negara Malaysia, to con- tutional capacity, and to develop financial infra-
duct on-site supervision of their country's branches structure and the appropriate regulatory framework.
in Labuan. Beyond supervisory matters, the govern- It is envisaged that by 2010, Islamic banking could
ment's objectives regarding the Labuan Offshore Fi- comprise 20 percent of the banking and insurance
nancial Center include the promotion and diversifi- market share.

VII Corporate Performance and Reform
Yougesh Khatri
C orporate sector vulnerabilities and governance

issues are increasingly seen to have played key
roles in the Asian crisis. Recent economic literature
prices and a high level of new equity issues and pri-
vatizations (relative to the region). The Malaysian
corporate sector is highly concentrated both in terms
has gone so far as to place these at the center of the of ownership and control. The "insider" system of
crisis. While the debate as to the main causes of the corporate governance applies and, together with an
crisis will no doubt continue, it is clearly important to overreliance on bank financing, the sector has some
be able to identify the main corporate sector vulnera- innate vulnerabilities.
bilities and to ensure that sufficient reforms are under-
taken in order to avoid future crises stemming from
Capital Market Structure
the corporate sector and to ensure that this sector can
withstand financial crises, whatever the cause. The Malaysian capital market and underlying cor-
This section provides a brief overview of the corpo- porate sector are large by any standard. The total capi-
rate sector in Malaysia; discusses possible links be- talization of the Kuala Lumpur stock exchange, in-
tween the corporate sector and the recent crisis; exam- cluding financial and nonfinancial corporations on the
ines corporate performance before, during, and after main and second boards, amounted to RM 424 billion
the crisis; and assesses progress with corporate re- ($112 billion) or 125 percent of GDP at end-2000.
form. Various indicators provide evidence of a deteri- At its peak in 1993, market capitalization reached
oration in corporate performance before the crisis, a 360 percent of GDP. The International Finance Cor-
significant negative impact on the sector with the poration's Emerging Stock Markets Factbook 1999,
onset of the crisis, and strong signs of recovery in ranked Malaysia's market capitalization at end-1998
1999. There is clear evidence of progress with re- as the twenty-third largest in the world; its total value
forms, although this progress has been somewhat un- traded was twenty-ninth highest in the world; and, by
even: debt restructuring has proceeded well under the number of listed companies, Malaysia ranked fif-
Corporate Debt Restructuring Committee and Dana- teenth in the world.
harta; some operational restructuring has taken place The corporate sector was characterized by rapid
through the resolution by Danaharta of its nonper- growth throughout the 1990s. The number of listed
forming loans portfolio; and there have been con- companies in Malaysia increased to 795 from 285
certed efforts made to adopt the proposals of the Fi- in 1990 by the end of the decade (Figure 7.1). The
nance Committee's Report on Corporate Governance. number of listed companies grew on average by
These reforms will take time to complete and become 11.6 percent a year throughout the 1990s, and by
effective, but momentum should not be lost in 14 percent for the period 1990-96. Compared to
the face of either an improving or deteriorating the other crisis countries, this was relatively rapid:
environment. Indonesia averaged a 10 percent annual growth in
the number of listed companies throughout the
decade, but that was from a much lower base (122
Key Features of the Corporate in 1990); Korea averaged only 1 percent annual
growth, but from a much higher base (669 listed
Sector1 companies in 1990);2 Thailand averaged over 7
The Malaysian corporate sector is large and for percent annually for the decade, but 13.6 percent in
most of the last decade has been characterized by
rapid growth, mainly driven by increasing stock 2
The number of firms listed on the Korean stock exchange and
its overall capitalization may be misleading as the KOSDAQ
(similar to the NASDAQ in the United States) has been growing
rapidly, and this growth is not reflected by the stock exchange
1See also Kochhar and others (1999), Chapter IV. figures.

VII CORPORATE PERFORMANCE AND REFORM
Figure 7.1. Selected Asian Countries: Capital Market Overview
160 500 1000

Market Capitalization Number of Firms Listed
140 - (In percent of GDP) (In units)
900
800
700
600
500
1990 94 96 98 2000
700 Stock Market Index

45
Funds Raised
(End of period, 1990 = 100) (Equity in billions of U.S. dollars) 40
600
Korea 35
500 Philippines 30
400
300
200
100 Korea
Indonesia
1990 92 94 96 98 2000 1990 94 96 98 2000
Sources: IMF, World Economic Outlook; Asia and Pacific Department core database, and CEIC Data Company Limited.
the 1990-96 period. Market capitalization in privatization in the region (Harvey and Roper,
Malaysia grew by an average rate of over 40 per- 1999).3
cent in the 1990-96 period and by an average of While the amount of new equity raised was large
nearly 30 percent for the decade. This was not un- by regional comparison, Malaysia was still highly de-
usual for the region; Indonesia and the Philippines pendent on bank financing. New financial flows to
had more rapid growth rates but from a much lower corporations in the period prior to the crisis
base. (1995-97) were mainly from the domestic banking
The growth in market capitalization in Malaysia system, representing nearly 60 percent of net funds
was driven mainly by increases in stock prices, but compared to around 15 percent from equity, 11 per-
also by new equity issues and privatization. Stock cent from domestic debt markets, and 16 percent
prices increased by a factor of over 2.4 between raised through external borrowing (World Bank,
1990 and 1996, and by a factor of 5 for the Philip- 1999a). The authorities, however, are attempting to
pines, compared to between 1 and 1.5 for the other develop the domestic bond market and are encourag-
crisis countries. In the precrisis period, only Korea ing corporations to diversify their sources of funding.
raised more funds through the equity market in
absolute terms than Malaysia. Malaysia was at the
forefront of Asian privatization, and Malaysian 3
In early 1991, Malaysia released its privatization master plan,
privatizations, which accounted for a large portion which had the objective of privatizing key industries, and proceeded
with Tenaga Nasional, the national electricity company—one of the
of total new equity raised in the country, also largest privatizations to date in Asia at $1.2 billion. Between 1992
constituted around one-third of total revenue from and 1995, privatization revenue averaged 3 percent of annual GDP.
• ©International Monetary Fund. Not for Redistribution

Key Features of the Corporate Sector
Box 7.1.The Insider and Outsider Systems of Corporate Governance
Corporate governance relates to ways in which in- also play an active role in management and have the de-
vestors and owners (principals) oversee managers who cisive vote in major decisions.
run the firms (agents). There are essentially two main To illustrate the contrast between the outsider
systems of corporate governance, referred to as the in- (mainly the United Kingdom and the United States) and
sider and outsider systems. insider systems (most other countries, including conti-
The outsider system refers to the Anglo-American nental Europe and Asia), Crama and others (1999) find
system where typically ownership of firms is widely that the largest owner in the median U.K.-listed com-
dispersed and control is delegated to professional man- pany in 1998 had a stake of less than 15 percent, and
agers; the number of listed companies is large; the less than 5 percent in the median U.S.-listed company;
process of acquiring control is market oriented (i.e., compared with over 80 percent of listed nonfinancial
there is a liquid capital market with frequently traded companies in continental Europe that had shareholders
ownership and control rights); there are few interlocked with a blocking minority (at least 25 percent) and where
patterns of ownership; and there are few major control- around half the companies had one shareholder with an
ling shareholders, which are rarely associated with the absolute majority. La Porta, Lopez-de-Silanes, and
corporate sector itself. Shleifer (1998) find the average share of common stock
The insider system is characterized by the following owned by the largest three shareholders in the largest
features: there is a high concentration of ownership; the companies to be 54 percent in Malaysia, 46 percent in
corporate sector has controlling interests in itself; the Thailand, 20 percent in Korea, and 18 percent in Japan,
number of listed companies is relatively small; the capi- compared to less than 15 percent in the United King-
tal market is illiquid because controlling blocks are held dom and the United States.
by a few major shareholders, and these are held rather Both systems have relative strengths and weaknesses
than traded; there are a large number of holding or in- that are typically analyzed in a principal-agent frame-
terlocked companies acting to deter outsiders from work. Some of the main issues are summarized below
acquiring control; and major shareholders typically (mainly from Crama and others, 1999).
Ownership and Voting Power: Structures and Consequences
A: Dispersed ownership and dispersed voting power B: Dispersed ownership and concentrated voting power
(United Kingdom, United States) (Countries where a stake holder can collect proxy votes and
shareholder coalitions are allowed)
Advantages: Portfolio diversification and liquidity; takeover Advantages: Portfolio diversification and liquidity;
possibility monitoring of management
Disadvantages: Insufficient monitoring and free-riding problem Disadvantages: Violation of one-share-one-vote principle;
reduced takeover possibility
Agency conflicts: Management vs. shareholders Agency conflicts: Controlling block holders vs. small
shareholders
C: Concentrated ownership and dispersed voting power D: Concentrated ownership and concentrated voting power
(Any company with voting right restrictions) (continental Europe,Asia, and any company after a takeover)
Advantages: Protection of minority holders' rights Advantages: High monitoring incentives; more focused strategic
direction, restructuring, and long-term commitment
Disadvantages: Violation of one-share-one-vote principle; Disadvantages: Low portfolio diversification possibilities; low
low monitoring incentives; low portfolio diversification liquidity; reduced takeover possibilities
possibilities; low liquidity; higher cost of capital; reduced
takeover possibilities
Agency conflicts: Management vs. shareholders Agency conflicts: Controlling block holders vs. small
shareholders
The Asian crisis economies (which fall under Type D) are characterized by highly concentrated ownership and control, and thus the classic
agency problem (between management and shareholders) becomes irrelevant. Another agency conflict arises, however, between insiders (major
shareholders) and outsiders (small shareholders), and the main problem of corporate governance becomes how to ensure that the insiders do
not exploit (expropriate) the assets of outsiders.

Table 7.1. A Comparison of Legal Protection,Accounting Standards,

Institutions, and Enforcement in Various Countries
Shareholder Creditor Degree of Judicial Accounting

Protection 1 Protection 1 Enforcement2 Standards3
Malaysia 3.0 4 7.7 76

Korea 2.0 3 6.7 62
Thailand 3.0 3 5.9 64
Latin America 4 2.5 1 6.2 53
United States 5.0 1 9.5 71
United Kingdom 4.0 4 9.4 78
Japan 3.0 2 9.4 65
Germany 1.0 4 9.4 62
Source: La Porta, Lopez-de-Silanes, and Shleifer (1998).

1On a scale from 1 (weakest) t o 5 (strongest).
2
On a scale from 1 (lowest) to 10 (highest).
3
The higher the measure, the higher the standard.
4
Average of Argentina, Brazil, Chile, and Mexico.
Malaysia, like the other crisis countries, is charac- concentration of shareholding can lead to poor gov-
terized by the insider system of corporate governance ernance because a small group can exercise control
(Box 7.1), in which there is a high degree of owner- over a firm and pursue the objectives of the insiders
ship concentration, cross holdings, and participation at the cost of the outsiders or small shareholders
of owners in management. A few large corporations (Claessens, Djankov, and Lang, 1999; and Box
account for a significant proportion of financial assets 7.1). An interesting feature of corporate ownership
and productive capacity in the country.4 Concentra- in Malaysia was the prevalence of nominee ac-
tion also occurs at the level of stock ownership, counts. Nominee accounts at end-1997 were the
which, given the large capitalization, is in the hands largest type of shareholders in the top five share-
of relatively few institutional and corporate investors. holders of listed companies, and about half the ben-
Another layer of concentration occurs in terms of con- eficial owners of the nominee accounts were for-
trol, where "pyramiding" or cross-holding of share eigners. A recently issued set of rules by the Kuala
ownership magnifies the actual control of a few indi- Lumpur stock exchange requiring securities ac-
viduals or entities well beyond their actual level of counts to be opened in the name of the beneficial
ownership in each company. owner means that nominee accounts can be ex-
The development path of the corporate sector has pected to be phased out soon.
resulted in some innate vulnerabilities. First, the
development of the private sector under the activist
Legal and Institutional Structure
industrial policies of the government has resulted
in close ties between government and large corpo- The nonfinancial corporate sector in Malaysia is
rations. Second, the cross-holding structures can mainly governed by three acts: the Companies Act,
create incentives for double leveraging and thus 1965; the Securities Industry Act, 1983; and the Secu-
create a multiplier effect in the sensitivity of corpo- rities Commission Act, 1993.5 The Kuala Lumpur
rate wealth to changes in the equity market stock exchange's listing requirements and rules also
(Kochhar and others, 1999, Chapter IV). Third, the play an important role in regulating investors, brokers,
4
The International Finance Corporation produces an indicator 5
of this type of concentration derived from the ten largest stocks in Also relevant to the financial corporations and the broader
the IFC global indices for each country relative to the total Inter- legal and institutional environment in which corporations operate
national Finance Corporation Global market capitalization for are: the Futures Industry Act of 1993, the Banking and Financial
each country. At end-1998, this measure of concentration was Institutions Act of 1989, and the Malaysian Code on Takeovers
31.5 percent for Malaysia compared with 61.5 percent for In- and Mergers of 1987, all of which were supplemented by guide-
donesia, 37.9 percent for Korea, 55.4 percent for the Philippines, lines such as those of the Kuala Lumpur stock exchange (men-
and 45.8 percent for Thailand. tioned above) and the Foreign Investment Committee.

Corporate Sector and Financial Crises
and issuers. The Companies Act deals with the pre-in- Corporate Sector and
corporation, incorporation, operations, and duties of Financial Crises
companies and their directors, as well as the rights and
obligations of shareholders and directors. The Securi- There are two main branches of economic litera-
ties Industry Act and Securities Commission Act ture that link the corporate sector with financial
make up the legislative and regulatory frameworks of crises. The main literature on crises has been at the
Malaysia's capital markets, under the authority of the macroeconomic level focusing on macroeconomic
Ministry of Finance. The powers of the Kuala Lumpur fundamentals or self-fulfilling crises (modeled on
stock exchange were recently strengthened through bank runs), and has only recently (in a model by
amendments to the Securities Industry Act, such that Krugman) included explicitly the corporate sector as
the exchange may now take action against directors a central element. The other branch of the literature
and any person involved with its listing requirements.6 is more microeconomic based, looking at either
Malaysia has comprehensive laws relating to cor- firm-level data to investigate the role of the corpo-
porate governance, and the laws governing creditor rate sector in the crisis or focusing on the role of in-
rights are comparable to those of OECD countries stitutional factors and corporate governance.
(Table 7.1). While accounting standards are also
good, enforcement and actual practice have been
weaker. The Financial Reporting Act of 1997 was Macroeconomic Approaches
designed to address this issue by giving the force of Standard Models of Financial Crises
law to (i.e., requiring all companies to comply with)
the accounting standards approved by the national The literature on financial crises has a long his-
accounting body, the Malaysian Accounting Stan- tory and extends well beyond the scope of this sec-
dards Board. Most of the accounting standards ap- tion.9 The main prevailing theories have been classi-
proved by this board are based on international ac- fied by Eichengreen as "first generation" and
counting standards. While progress with improving "second generation" models. First generation mod-
standards and adopting international best practices els are associated with the seminal paper by Krug-
has been notable, the accuracy of financial data man (1992) and generally explain crises as a result
needs to be improved by bringing financial disclo- of a deterioration in fundamentals. The main feature
sure requirements in line with international best of second generation models (following Obstfeld,
practices. The broader legal and institutional envi- 1986) is self-fulfilling speculative attacks, although
ronment is strong, although transparency and ac- even for these some perceived weakness in funda-
countability in the public sector need more attention mentals seems to be the key trigger for the specula-
(World Bank, 1999b). tive attack.10 The development of the second genera-
There are a number of alternatives for dealing with tion models followed the breakdown of the
distressed corporations. The Companies Act allows exchange rate mechanism in 1993 and the Mexican
creditors to petition the high courts to wind up a com- crisis in 1994, as aspects of these crises were hard to
pany if that company defaults on debt payments, and reconcile with the first generation models. This was
allows debtors to petition for court protection under mainly because the decisions to abandon pegs were
Section 176 until a group of creditors (representing not related to the exhaustion of reserves in defend-
three-fourths of the outstanding debt) agree to a reor- ing the currency and, in some cases, were not easily
ganization plan.7 Companies and creditors can also explicable at all from the point of view of economic
opt for voluntary out-of-court restructuring of the fundamentals.
debt or, for companies with debt exceeding RM 50 The Asian crisis seemed to consolidate the con-
million and having more than three creditors, the vol- sensus toward the second generation models as rep-
untary out-of-court restructuring can be done through resenting more recent crises. Krugman (1999) sug-
the Corporate Debt Restructuring Committee.8 gests, however, that for the major crises in the Asian
countries, neither of these models seems to have
6
For more details on the legal and institutional environment gov-
9
erning corporations, see the forthcoming publication, Malaysia: See, for example, the survey in Berg and others (1999) and the
Corporate Governance Assessment by the World Bank. references therein.
7 10
In response to certain weaknesses identified in the Companies A crude caricature of the first generation model is one in
Act, key provisions of Section 176 were tightened after the onset which a budget deficit is financed by "printing money," which re-
of the crisis in an attempt to prevent misuse of the Act as a means sults in the eventual collapse of a fixed exchange rate. In second
to delay adjustment. generation models, the crisis results from a conflict between a
8
For a fuller discussion of the problems with enforcement of fixed exchange rate regime and the desire to pursue a more ex-
laws, autonomy of regulators, transparency in exercising existing pansionary monetary policy; investors bet that the authorities will
regulation, and confusion over jurisdictional boundaries, see let the peg go rather than compromise on another front, such as
Kochhar and others (1999), Chapter IV. employment, and this bet is self-fulfilling.

much relevance.11 The fiscal positions, by conven- • balance sheet problems: Most descriptive ac-
tional measures, were strong, and there were not re- counts of the crisis place significant emphasis on
ally the clear trade-offs between employment and the role of firms' balance sheets, but this had not
exchange rate stability (e.g., as faced by the United been featured as a central element in the crisis
Kingdom in 1992). Krugman suggests the need for a literature.14
"third generation" of crisis models and is skeptical
about the current bank-centered candidates (Corsetti, The model is characterized by multiple equilibria,
Pesenti, and Roubini, 1998; or Chang and Velasco, where a loss of confidence, for whatever reason, can
1998).12 Krugman acknowledges that these models lead to a self-fulfilling collapse; the mechanism dif-
capture some aspects of the crisis, but outlines an al- fers from the Diamond-Dybvig approach in that the
ternative candidate that emphasizes factors not for- main mechanism is the transfer problem. Very
mally included in previous models, namely, the role roughly speaking, the loss of confidence leads to the
of companies' balance sheets in determining their transfer problem, and to achieve the required current
ability to invest and the role of capital flows in af- account reversal, the country must experience a
fecting real exchange rates. large real depreciation and/or output decline, either
of which adversely affects the balance sheets of do-
mestic firms. This validates the initial loss of confi-
Krugman's Proposed Third Generation Model dence, i.e., moving from the high expected invest-
ment equilibrium to the low expected investment
Reliance on the moral hazard argument, according equilibrium. According to the model, the factors that
to Corsetti, Pesenti, and Roubini, is discounted as a make such a crisis possible (i.e., reinforce the feed-
key feature by Krugman, who argues that there was back loop between investment, real exchange rates,
ample evidence of significant investment in the Asian and balance sheets) are high leverage, low marginal
crisis countries prior to the crisis, including in direct propensity to import, and large foreign currency debt
foreign purchases of equity and real estate, which relative to exports.
clearly were not protected by any form of implicit
guarantee. Krugman does, however, accept the exis-
tence of multiple equilibria as a necessary element in Microeconomic and Institutional Approaches
modeling the crisis, although not the mechanism im-
plicit in the Diamond-Dybvig type approach. The A number of studies have used firm level data to
Krugman model incorporates three key elements: investigate (in a comparative context) the corporate
performance of the crisis countries during the 1990s,
• contagion; including Claessens, Djankov, and Lang (1998),
Pomerleano (1998), Harvey and Roper (1999), and
• the transfer problem:13 A huge change is needed Claessens, Djankov, and Xu (2000). These studies
in the current account as a counterpart to the re- provide new evidence suggesting that the causes of
versal in capital flows, which is evidently central the Asian crisis may lie in firm-based decisions.
to the crisis yet has not been explicitly included Claessens, Djankov, and Xu summarize the litera-
in previous models; and ture on the role of the corporate sector, in both per-
formance and financing, in the Asian crisis and iden-
tify four main possible links, summarized briefly
11
Note that Krugman (1999) has a somewhat different view below.
from that of Krugman (1998).
12
First, the weak corporate performance after the
Two major views have emerged in the postcrisis literature. crisis was related largely to the shocks experienced
The first view (Corsetti, Pesenti, and Roubini, 1998) suggests that
the apparent soundness of macroeconomic policy was misleading
by the Asian countries, including declines in aggre-
because a large hidden subsidy via implicit government guaran- gate demand, reversal of capital flows, sharp depre-
tees to banks and corporations led to moral-hazard lending and ciations, and increases in interest rates (Furman and
implied a hidden government deficit. Thus the apparent sound- Stiglitz, 1998). As yet, there is little empirical work
ness of the macroeconomic policy was an illusion. The second assessing the importance of aggregate shocks to cor-
view, associated with Radelet and Sachs (1998), may be charac-
terized by the idea that there was not a major problem with the porate performance. There is some evidence from
policies pursued by the crisis countries and that investments were
basically sound; at most, the countries suffered from "financial
fragility" and were thus vulnerable to self-fulfilling pessimism on 14
The role of firms' balance sheets in the crisis has most often
the part of international lenders (see models by Chang and Ve- been related to the impact of massive exchange rate depreciations
lasco, 1998). on the domestic value of the foreign currency-denominated debt
13
The transfer problem—discussed by Keynes, among others— of these firms, which had been accumulating prior to the crisis.
refers to the difficulty in transferring large quantities of capital Balance sheets were further weakened by declining sales and
from one country to other countries. The problem relates to the high interest rates. These problems in turn led to the increase in
burden of making the transfer, but also to the burden associated nonperforming loans, but this view implies that the problem was
with the resultant change in exchange rates and relative prices. not, per se, a banking system problem.

Corporate Performance
survey data, such as the results of a survey of Thai 1981). The balance sheet problem (Bernake and
industrial firms reported in Dollar and Hallward- Gertler, 1995) may exacerbate the effect of a shock:
Driemeier (2000), 15 that indicates these shocks in the presence of asymmetric information and
played an important but not exclusive role in the per- principal-agent relationships, the corporation's net
formance of these firms during the crisis. worth or wealth becomes an important determinant
Second, the poor performance of the corporate of the amount it can borrow, as assumed in the Krug-
sectors during and after the crisis reflects previous man model, rather than the prospects of the project
fundamental weaknesses (Corsetti, Pesenti, and for which the borrowing is undertaken. Thus, a de-
Roubini, 1998). This view implies that the perfor- cline in the wealth of a firm (e.g., through deprecia-
mance of firms was not adequately monitored by tion that reduces the domestic value of foreign as-
shareholders and investors, and/or firms were not sets) can reduce the credit available even for viable
subject to sufficient competition, thus poor perform- new projects.
ers or riskier firms were not forced to fully adjust and Fourth, the efficiency of debt-resolution mecha-
increase their rates of return to compensate investors nisms will determine, in part, the extent of the im-
for higher risk. This view may also imply that prof- pact of financial and other shocks. It has long been
itability was overstated by firms: thus the lack of recognized that the institutional framework is impor-
transparency, relatively weak accounting practices in tant in avoiding and resolving systemic financial
the region, and weak corporate governance may have crises and that exceptional mechanisms, such as
hidden the extent of the problems and delayed the Fund programs, may be required during periods of
onset of the crisis.16 A number of studies indicate that systemic crisis. This broad area of literature spans
ownership structure may induce risky behavior. The economic principles for optimal workouts to the im-
insider system (Box 7.1) prevails in the Asian coun- portance of creditor rights to enforce claims and
tries: the extensive links and cross-holdings of seize collateral, both in the context of domestic and
shares, particularly between corporations and banks, external borrowing, as seen in the review by La
are likely to have distorted the market allocation of Porta, Lopez-de-Silanes, and Shleifer (1999).
resources and resulted in excessive and nontranspar-
ent risk. These ownership links clearly played a sig-
nificant role in Korea and Indonesia. Government in- Corporate Performance
volvement—through direct participation in bank
ownership and through links with corporations and Financial ratios, which are commonly used to ana-
banks—is also likely to weaken the allocation of re- lyze corporate performance, have the advantage of
sources because a political dimension is introduced being simple to compile and are broadly understood,
into the allocation decisions. but they have some major drawbacks. Simple finan-
Third, aggregate and financial shocks to the financial ratios generally give a partial indication of perfor-
cial sector may result in a credit crunch, constraining mance in a particular dimension. Financial ratios can
lending to viable corporations with profitable invest- also be misleading (e.g., looking at rates of returns
ment and trading opportunities. Shocks—financial, rather than risk-adjusted rates of return). Economic
real, or regulatory—may cause a real or perceived measures based on concepts of efficiency—measured
shortage of capital for banks and lead to their curtail- relative to other firms—or total factor productivity—
ing credit for investment or trade, thus impairing the measured for a firm over time—provide more com-
performance of firms. A credit crunch may result prehensive measures of performance, but require de-
from weak financial institutions or from a change in tailed production (input-output) data. Generally, only
the regulatory and supervisory environment. In- accounting data are readily available on a consistent
creased uncertainty regarding whether and at what basis between firms and over time, and thus an analy-
price loans will be available may also result in a sis of corporate performance has generally employed
shortfall of loanable funds (Stiglitz and Weiss, a financial ratio analysis.
Below, performance is analyzed based on finan-
cial ratios, but also on approximations to the eco-
15
The survey found that 60 percent of firms said that the sub- nomic measures of efficiency and productivity using
stantial decline in domestic demand and higher input costs relat- accounting data. This provides evidence that perfor-
ing to the depreciation were the primary sources of difficulty.
Only one-third of the firms cited access to capital as a major mance in the Malaysian corporate sector, and in the
problem, although more cited the cost of capital as a problem. corporate sector of the other crisis countries, deterio-
16
Weak corporate governance and a lack of transparency are rated prior to the crisis, but deteriorated even more
the central factors in explaining the Asian crisis according to dramatically with the onset of the crisis. The evi-
Johnson and others (1998). But, as Furman and Stiglitz (1998)
point out, countries with few problems in terms of corporate dence is consistent with a number of the possible
governance and transparency have still experienced crises (e.g., links identified above, although no formal tests were
Sweden). conducted.

Table 7.2. Financial Ratios for Listed Nonfinancial Companies in Malaysia

(In percent, unless otherwise indicated)
1995 1996 1997 1998 1999
Profit (after tax)/Turnover (net profit margin) 10.5 10.1 7.5 -3.7 -I.I
Return on shareholder's funds 11.0 10.8 7.6 -4.9 -2.4
Return on assets 5.3 4.6 2.9 -1.8 -0.8
Earnings yield (equals 1/Net price-earnings ratio) 4.3 3.9 4.2 -4.4 -1.6
Net dividend yield 1.5 I.I 1.5 2.1 1.8
Current ratio (equals current assets/current
liabilities) 1.2 I.I I.I 1.0 1.0
Total debt-to-equity ratio 0.5 0.7 0.9 1.0 I.I
Number of companies included in calculating

these ratios 473 541 627 664 325
Source: Kuala Lumpur stock exchange data provided in April 2000.
Financial Indicators of Corporate an increase in leverage, thus enhancing corporate

Performance and Risk stress. An alternative explanation is suggested
below (see Corporate Performance and Links to the
Performance, as measured by the net profit mar- Crisis).
gin, seemed relatively healthy in Malaysia and did An alternative measure of corporate perfor-
not appear to diminish markedly before the crisis mance is the return on assets, which has the advan-
(Table 7.2). The net profit margin compared favor- tage of not being affected by the liability structure
ably with those of the other crisis countries, and of a firm while providing a measure of return on
prior to the crisis was only surpassed by that of In- capital.18 The return on assets in Malaysia tell a
donesia (Table 7.3).17 Unlike the other crisis coun- similar story to net profit margins, although there
tries, the Malaysian corporate sector in aggregate is a more obvious decline in the return on assets
was still profitable in 1997; thus, Malaysia appears just prior to the crisis (Table 7.2). Claessens,
to have entered the crisis later or fared better going Djankov, and Xu (2000) compare real return on as-
into it. There were, however, signs of increasing cor- sets, which they define as return on assets less the
porate distress going into the crisis, as indicated by inflation rate, in a sample of Asian countries and
the percentage of firms not able to cover interest ex- the United States and Germany. The findings re-
penses from operational cash flows. veal that the average real return on assets for the
While Malaysia appears to have fared compara- precrisis period of 1988-96 in Thailand, the Philip-
bly well going into the crisis, it did no better than pines, and Indonesia were the highest among the
the other countries during the crisis or the recovery sample of 36 countries (that report to Worldscope)
period. Lack of a differentially superior postcrisis at 9.8 percent, 7.9 percent, and 7.1 percent, respec-
performance, given the more favorable starting con- tively. Malaysia's average for the same period was
ditions, may reflect either the lag with which the not far behind at 6.3 percent—ranking eighth in the
crisis affected Malaysia or the later pickup in do- sample of 36—and greater than that of either the
mestic demand compared with the other countries United States (5.3 percent) or Germany (4.7 per-
(see Section II). Also, because Malaysia has a sig- cent). Korea had one of the lowest real return on
nificantly larger market capitalization relative to assets for the period in the sample at 3.7 percent.
GDP, the stock market declines during the crisis are Indonesia, Thailand, and Korea had declining
likely to have resulted in a larger loss of wealth and trends in real return on assets between 1990 and
1996, with Korea measured from an already low
17
base, while Malaysia had an increasing trend
The net profit margin reported by Claessens, Djankov, and between 1988 and 1993, but a declining one there-
Lang (1999) in this table differs from that derived using the Kuala
Lumpur stock exchange data in Table 7.2. This may be due to a
different sample (Table 7.2 is limited to nonfinancial corpora-
18
tions) or different definitions. Also, the net profit margin in Table The return on assets is nevertheless a partial measure of per-
7.2 is defined as earnings before interest and tax less tax relative formance because it only takes into account returns with respect
to turnover. to capital and not other factors, such as labor.

Table 7.3. Operational Performance of Publicly Traded Corporations and

Share of Distressed Corporations in Selected Asian Countries
(In percent)
1999
1995 1996 1997 1998 (First half)
Net profit margin

Indonesia 12.4 13.9 -3.6 -13.3 -8.9
Korea 2.7 0.4 -0.3 -2.6 2.7
Malaysia 12.2 12.0 6.9 -2.8 1.3
Thailand 7.1 5.1 -3.6 2.2 4.8
Firms unable to cover interest

expenses from operational
cash flows
Indonesia 12.6 17.9 40.3 58.2 63.8
Korea 8.5 11.2 24.3 33.8 26.7
Malaysia 3.4 5.6 17.1 34.3 26.3
Thailand 6.7 10.4 32.6 30.4 28.3
Source: Claessens, Djankov, and Klingebiel (1999).
Table 7.4. Debt/Equity Ratios in Selected Economies

(In percent)
1990 1991 1992 1993 1994 1995 1996
Hong Kong SAR 1.8 2.0 1.8 1.8 2.3 2.0 1.6
Indonesia — 1.9 2.1 2.1 1.7 2.1 1.9
Japan 2.9 2.0 2.0 2.1 2.2 2.4 2.4
Korea 3.1 3.2 3.4 3.6 3.5 3.8 3.5
Malaysia 1.0 0.6 0.6 0.7 1.0 1.0 1.2
Philippines — 0.8 1.2 1.2 I.I 1.2 1.3
Singapore 0.9 0.9 0.9 I.I 0.9 1.0 1.0
Taiwan Province of China — 0.7 0.9 0.9 0.9 0.8 0.8
Thailand 2.2 2.0 1.8 1.9 2.1 2.2 2.4
Germany 1.6 1.6 1.5 1.5 1.5 1.5 1.5

United States 0.9 1.0 I.I I.I I.I I.I I.I
Source: Claessens, Djankov, and Xu (2000).
after through 1996. Thus, return on assets—mea- returns would normally be required to justify this in-
sured in real terms and over a longer period— creased risk. In effect, accounting returns, although
seems to indicate declining corporate performance generally high, were declining, and implicit risk-ad-
in the precrisis period. justed returns were declining even more rapidly.
A number of risk indicators worsened prior to the
crisis. Measures such as the proportion of distressed
corporates (Table 7.3), leverage (Table 7.4), and the one-third for the period 1988-96. The ratios for the same period
maturity structure of debt clearly indicate that risk were 34.1 percent in Indonesia, 43.7 percent in Korea, 52.2 per-
was increasing in the run-up to the crisis.19 Higher cent in the Philippines, and 30.9 percent in Thailand. These are
low compared to the average U.S. and German long-term-to-total
debt ratios for the same period: 55.3 percent for Germany and
19
The ratio of long-term debt to total debt was relatively low in 75.9 percent for the United States (Claessens, Djankov, and Xu,
the region and declined steadily in Malaysia to average less than 2000).

Table 7.5. A Comparison of Stock Market Returns and Risk in

Selected Economies
(During period ofJanuary 1990-December 1996)
Average Annualized Annualized Standard

Buy and Hold Monthly Return Deviation
Indonesia 11.10 3.09 103.55

Korea -47.33 -9.52 95.47
Malaysia 127.13 11.80 85.02
Philippines 118.13 9.55 109.07
Taiwan Province of China -42.89 -4.26 151.92
Thailand 29.56 3.12 111.48
United States 184.14 13.96 39.76

World 78.55 7.60 46.20
Source: Harvey and Roper, 1999.
Summary statistics on the distribution of firm- riod, while the U.S. Morgan Stanley Capital Index
level indicators of nonfinancial corporate perfor- had both a higher return and lower volatility (or risk)
mance reported by Harvey and Roper also suggest than all of the Asian countries (Table 7.5). Through-
that corporate performance in the Asian countries— out Asia, Harvey and Roper also find that returns in
but also in other emerging markets—deteriorated in individual markets deteriorated well before the onset
the run-up to the Asian crisis. Harvey and Roper re- of the crisis. Although the buy-and-hold investment
port statistics on the entire distribution of returns on strategy is a naive one, after considering alternative
equity and returns on invested capital from firm- dynamic strategies Harvey and Roper still find that
level ratios for a number of emerging markets. They Asian returns failed to outperform either U.S. or
determine that the median of these indicators had world dynamic strategies.
clearly declined in Indonesia and Thailand, and less A popular measure of liquidity, the current ratio
clearly so in Korea, the Philippines, and Malaysia.20 (see Table 7.2) indicates that liquidity in Malaysia's
More generally, the reported medians of a series of corporate sector declined slightly before and during
corporate performance indicators—namely return on the crisis. Negative profits during and immediately
equity, return on invested capital, total debt-to-com- after the crisis imply that cash reserves and other liq-
mon equity, and interest payments relative to earn- uid assets likely had declined. New borrowing for
ings before interest and tax—show a deterioration in investment declined significantly with the onset of
the main Asian and Latin American emerging mar- the crisis, and new investment was partly financed
kets between 1992 and 1996, with the clear excep- from retained earnings, further decreasing current
tion of Mexico, for which three out of four indicators assets relative to current liabilities.
improved. Leverage in the corporate sector, measured as the
A related form of return measure is that on stock ratio of total debt to equity, was rising rapidly in
market investment. Harvey and Roper estimate the Malaysia in the precrisis period, but has since
buy-and-hold returns accruing to investors between stabilized (Table 2.1 and Table 7.5). Harvey and
January 1990 and December 1996 for various coun- Roper note that the entire distribution of leverage
tries and conclude that returns in the Asian crisis ratios at the firm level shifted right or worsened
countries declined prior to the crisis and that risk- between 1992 and 1996. Reporting the distribution
adjusted returns were relatively poor in Asia. Only by quartiles, they find that in 1992, a quarter of the
Malaysia and the Philippines outperformed the firms had leveraged 2.1 percent of the common
World Morgan Stanley Capital Index over the pe- stock, half had reported a leverage ratio of 18.9 per-
cent or less, and three-quarters of the firms had
20
stated their leverage ratio was less than 50 percent.
For Indonesia, the median return on equity declined to 12.5 By 1996, the corresponding values for the first,
percent in 1996 from 15 percent in 1992, while for Thailand, it
declined more markedly to 7.7 percent from 19.4 percent over the
second, and third quartiles had increased to 12 per-
same period. cent, 58 percent, and 112 percent, respectively.

Economic Measures of Corporate Performance The Malmquist index 23 (Malmquist, 1953; and
Fare, Grosskopf, and Lovell, 1994) allows us to
Efficiency and productivity are the main eco-
jointly derive measures of technical efficiency (dis-
nomic measures of performance. Technically effi-
tance from the frontier) and exogenous change (i.e.,
cient firms use the least combinations of inputs, such
shifts of the frontier that are often attributed to tech-
as labor and capital, to produce a unit of output at a
nological change) (Figure 7.3). The derived indices of
given point in time and for a given environment,
technical efficiency and exogenous change, using the
such as technological and institutional infrastructure.
29 firms for which 1999 data were already available,
The firms using the least inputs to produce output
indicate that average efficiency declined during
define the production frontier, which relates output
1995-99, while exogenous factors, such as techno-
to inputs, or, equivalently, the efficient unit isoquant,
logical change, institutional and regulatory factors,
the efficient input combinations that can produce
and macroeconomic and international variables, dete-
one unit of output. Firms not on the production fron-
riorated somewhat but were strongly positive contrib-
tier (i.e., not on the unit isoquant) are said to be inef-
utors to total factor productivity in 1999. Figure 7.3
ficient; the level of inefficiency can be measured by
plots chained indexes of changes in average measured
the "distance" from the frontier.21 Shifts in the pro-
efficiency for firms and shifts in the frontier. Bearing
duction frontier—and, correspondingly, the unit iso-
in mind the significant reservations with respect to the
quant—can be thought of as technological change
data expressed earlier, there are indications that tech-
but will generally reflect changes in the exogenous
nical efficiency and, more strongly, total factor pro-
environment in the broad sense of all factors exoge-
ductivity, declined prior to the crisis. This provides
nous to the firm. Total factor productivity conflates
further evidence of a deterioration of fundamentals in
efficiency changes with technological change and
the corporate sector and in the environment in which
broader changes in the exogenous environment and
corporations operated prior to the crisis. The contin-
is thus a residual catch-all measure.
ued decline in efficiency in 1999 is consistent with
Ideally, to calculate efficiency economic measures the view that the benefits from any restructuring were
of inputs and outputs would be used, but as these are not likely to be manifested immediately because of
not available accounting data are used as proxies. transition costs and the dynamics of adjustment. The
Crude approximations of output (using total sales or deterioration in exogenous factors—a shift of the
turnover), capital (using total assets), and labor and frontier away from the origin—is consistent with the
other inputs (using total expenses = turnover - earn- initial deterioration in factors external to the firms,
ings before interest and tax) are derived from the such as increases in interest rates, the large real depre-
published balance sheets and income statements for ciation, and a decline in aggregate demand, because
the 29 largest nonfinancial corporates for the period these will have affected costs for all firms and thus
1995-99. 22 Dividing the two input measures (total will have shifted the efficient frontier. As these exter-
assets and total expenses) by output (turnover or nal factors improved, however, it is also possible that
total revenue), a scatter plot can be derived of the the initial pressure to improve efficiency diminished.
input combinations, of which the lower boundary
formed by the points furthest south and/or west is in- The deterioration in corporate sector performance
dicative of a unit isoquant. The larger the scatter, the is also evident from the aggregate data on listed nonfi-
less the implied relative efficiency, and thus the scat- nancial corporations. Aggregating total revenue, total
ter plots for the 1995-99 period (Figure 7.2) indicate assets, and total expenditure for the whole corporate
a worsening of relative efficiency for these large cor- sector in Malaysia (see Table 7.2 for the number of
porations between 1995 and 1998, but it is unclear firms included for each year) similar indicators of per-
from the scatter plots whether efficiency improves or formance, which combine efficiency and shift in the
worsens in 1999. efficient frontier, can be plotted at an aggregate level
over time (Figure 7.4). The aggregate indicators
clearly demonstrate a deterioration in the corporate
21
In addition, allocative efficiency requires that the firm not sector's use of total assets (i.e., an increase in the total
only be on the unit isoquant but also at the point where the budget assets/total revenue ratio) in the run-up to the crisis. In
line or surface determined by the relative input prices is tangen- 1997, performance in both dimensions is weaker, and
tial to the unit isoquant. in 1998 a major deterioration occurs in the dimension
22
The Kuala Lumpur stock exchange kindly provided account-
ing information on the top 40 firms, of which only 29—represent- of expenses relative to revenue, possibly due to a sig-
ing around 40 percent of total market capitalization—had com- nificant shift in the frontier relating to the various ex-
plete accounts through 1999. There are of course significant ogenous shocks that are discussed further, below.
problems in this analysis, not the least of which are the use of ac-
counting variables as proxies for economic variables, the problem
of comparing companies across industries, the aggregation of in- 23
The Malmquist indices of efficiency and total factor produc-
puts into only two groups, and the use of only a subset of firms. tivity reported here were estimated by Jenifer Piesse of Birkbeck
This analysis should thus be taken as indicative. College, University of London.

Figure 7.2. Malaysian Corporate Sector:Total Assets and Expenditures

Relative to Total Revenue
7 7
1995 1996
6- 6- -
C
5- 5- -
4- • - 4- -
3- -
ass
3- -
2- - 2- -
1 - - 1 - -
0
0.3 0.6 0.9 1.2 1.5 0.3 0.6 0.9 1.2 1.5
Expenditure/revenue Expenditure/revenue
7 7
1997 1998
6- - 6- -
5- - 5- -
4- - 4- -
3- 3-
2- - 2- -
1 - 1 - -
0
0.3 0.6 0.9 1.2 1.5 0.3 0.6 0.9 1.2 1.5
Expenditure/revenue Expenditure/revenue
6 - 1999
5 -
4 -
3-
2 -
1 -
0.3 0.6 0.9 1.2 1.5
Expenditure/revenue
Corporate Performance and Links deteriorate before the crisis, and there is thus support
to the Crisis for the fundamentals-based theories. The evidence,
however, is also consistent with Krugman's model if
The evidence of deterioration in corporate perfor- the deterioration in precrisis corporate performance
mance in Malaysia and within the region is consis- is considered insufficient to have warranted the ex-
tent with various aspects of the links discussed tent of the crisis. Contagion, together with the deteri-
above. What emerges from the returns measures, oration in fundamentals, could have acted as the trig-
particularly when adjusted for risk, and from the ger by causing the change in sentiment that resulted
economic performance indicators is that perfor- in a shift from high to low equilibrium and ulti-
mance in the Malaysian corporate sector did indeed mately in the full-blown crisis. This view provides

Figure 7.3. Indices of Efficiency, Exogenous F i g u r e 7.4. Malaysian C o r p o r a t e S e c t o r :

Change, and TFP Performance Indicators
(1995 = 100)
1.02
135 1.00 _ 98 •
99
0.98 _
125 0.96 - -
Exogenous change
115 0.94 - -
0.92 - -
c 0.90 - 97 -
•
0.88 - -
0) 95 96
0.86 - -
0.84
1.5 1.7 1.9 2.1 2.3 2.5
Total assets/total revenue
1995 96 97 98 99
evidence indicates that stock market returns in the

Asian countries adjusted for risk (volatility) were
well below those in other equity markets in the 1990s
(Harvey and Roper, 1999); that returns in the Asian
another possible explanation as to why Malaysia stock markets declined well before the onset of the
went into the crisis in a stronger position and yet, crisis (Harvey and Roper); and that returns on assets
with the onset of the crisis, its performance was sim- were declining in all the crisis countries in the run-up
ilar to the other crisis countries. to the crisis, and even more rapidly if adjusted for the
Relating to the first link discussed above, associ- increasing risk associated with higher leverage, in-
ating corporate performance to stocks, there is ample creasing external debt, and an increasing share of
evidence from the accounting and economic mea- short-term debt. Scatter plots and Malmquist indices
sures of corporate performance in Malaysia of a sig- show some deterioration in efficiency of the largest
nificant deterioration in performance with the onset Malaysian corporation before the crisis. Thus there is
of the crisis. Measures of net profit margins, return some evidence that appropriate market discipline on
on assets, the number of distressed firms, and firm- firms was missing, but there does not appear to be a
level and aggregate efficient measures all deterio- problem with misreporting of profitability, and ac-
rated in 1997 and significantly more so in 1998. The counting standards appear to be stronger than in other
survey evidence in Dollar and Hallward-Driemeier countries in the region. Other factors, such as the
(2000) identifies aggregate shocks as a major, but ownership structure and government involvement,
not the sole cause of the deterioration in perfor- clearly could also be contributory factors to the in-
mance of manufacturing firms in Thailand. The creased risk in the corporate sector that did not result
measure of a shift in the efficient frontier from the in a correspondingly higher required return.
Malmquist index provides another indicator of the Performance indicators say little about whether or
impact of exogenous factors in explaining a deterio- not there has been a credit crunch nor do they indi-
ration in corporate performance. cate whether the real effects of the financial and
Although not immediately evident from the other shocks were related to the efficiency of debt-
accounts-based financial ratios, there is evidence of a resolution mechanisms, the third and fourth possible
decline in corporate performance in Malaysia and the links. It appears that slow growth in the private sec-
other Asian countries before the crisis and thus the tor credit is due to a lack of demand for funds rather
second potential link.24 The accounting measures are than a lack of supply, which is consistent with the
only partial indicators and do not adjust for risk. The overinvestment story (see also Section VI). Ghosh
and Ghosh (1999) find that, for Korea and Thailand,
the binding constraint was slowing demand rather
24
Of course, this can be considered consistent with the previous than an inadequate supply of funds.
link to the extent that there was evidence of a deterioration in cor-
porate performance before the crisis, but the main decline fol-
With respect to debt resolution, the mechanisms
lowed the onset of the crisis and the various shocks. available were inadequate at the onset of the crisis,

but Malaysia took decisive and effective steps to Act. Of these, 18 percent of the applications relate
deal with the emerging problems by creating institu- to the first quarter of 2000; thus the rate of applica-
tions to deal with nonperforming loans and debt re- tions does not appear to be slowing. Firms in the
structuring, as discussed in Section VI and below. sectors described as finance, insurance, property, or
Overall, the country has fairly strong legal and insti- trading have dominated, representing nearly 40
tutional frameworks and laws relating to corporate percent of annual, as well as total, applications.
governance, and those governing creditor and share- Over 1,000 petitions to liquidate companies have
holder protection are comparable to those of OECD been submitted, and there has also been an increase
countries (La Porta, Lopez-de-Silanes, and Shleifer, in mergers and acquisitions.26
1998; Table 7.1). Of the Asian crisis countries, Korea and Malaysia
Malaysia is characterized by a high prevalence of have made the most progress in restructuring debt,
collateral-based bank lending, and thus the rapid achieved mainly out of court. By August 1999, the
asset and equity price inflation is likely to have con- two countries had completed restructuring of about
tributed to an increase in bank lending. The roles of one-third of the debt registered under their respec-
contagion, leverage (corporate balance sheets), and tive programs.
the transfer problem are clear in Malaysia, and all of
these are central to the Krugman model. The asset
Operational Restructuring
price collapse, with the reversal of capital flows and
its impact on net wealth, however, was probably a Operational restructuring of Malaysian corpora-
more important factor than the currency depreciation tions has proceeded partly through the resolution by
in the feedback loop that resulted in the crisis, but Danaharta of its acquired nonperforming loans ac-
such a collapse is not an explicit part of the Krug- complished by rehabilitation of businesses and liqui-
man model.25 dation and foreclosure of collateral, and through the
elimination of noncore businesses as part of debt-
restructuring agreements coordinated by the Corpo-
Progress with Corporate Reforms ration Debt Restructuring Committee. By end-2000,
Danaharta had appointed special administrators to
Corporate reforms can be classified as relating to oversee the management of more than 80 companies
debt restructuring, operational restructuring, and im- under its control to assist in their stabilization and
provements in corporate governance. The progress restructuring.
achieved by Malaysia with debt restructuring has Another important aspect of operational restruc-
been significant, and the proactive approach adopted turing in the crisis countries has been reductions in
by the authorities seems to have paid off. The extent labor, which have been the main source of improve-
of progress with operational restructuring is less ment in operational cash flows (Claessens, Djankov,
clear, and the adoption of measures to improve cor- and Klingebiel, 1999). Average labor shedding in
porate governance will need to be applied evenly be- publicly listed companies resulted in a 34 percent
fore their benefits become apparent. decrease in payrolls in mid-1999 compared to mid-
1997 for Korea, a 12 percent decrease for Thailand,
but only a 7 percent decrease for Malaysia. The use
Debt Restructuring of foreign workers in the latter meant that, although
Debt restructuring in Malaysia has taken a num- there was evidence of labor hoarding (Bank Negara
ber of forms. The Corporate Debt Restructuring Malaysia, 1999), the flexibility to dismiss foreign
Committee provides a platform—based on London workers helped to cushion the domestic labor force
rules—to achieve out-of-court corporate debt re- from the full impact of the crisis.
structuring (see Section VI). Restructuring is also
occurring on an out-of-court basis and outside of
the formal Corporate Debt Restructuring Commit- Corporate Governance
tee or Danaharta framework. As of end-March Significant progress has been made in imple-
2000, more than 192 companies had filed for court menting the recommendations of the Finance Com-
protection under Section 176 of the Companies mittee's Report on Corporate Governance. The re-
26
The number of company liquidations increased from 681 in
25
Kochhar and others (1999) estimate a relative wealth shock 1996 to 1,898 in 1997 (an increase of 179 percent) and to 4,800
due to changes in the stock market capitalization in Malaysia dur- in 1998 (an increase of 152 percent). Comparing liquidations
ing the financial crisis, of 155 percent of GDP, compared to the between January and September 1999 of 3,778, with those be-
impact on net wealth of the exchange rate depreciation (at its tween January and September 1998 of 3,438, we see a much
peak) of only 14 percent of GDP. smaller increase.

Summary and Conclusions
port was made public in March 1999, and the Fi- An assessment of corporate reforms shows that
nance Committee established an Implementation significant progress has been achieved with debt
Project Team, consisting of the Ministry of Finance, restructuring, and concerted efforts have been
the Securities Commission, the Registry of Compa- made toward improving corporate governance.
nies, the Kuala Lumpur stock exchange, the Federa- There is, however, less evidence of operational re-
tion of Public Limited Companies, Bank Negara structuring over and above labor shedding, and ef-
Malaysia, and Malaysian Exchange of Securities ficiency measures—adjusting for changes in ex-
Dealing and Automated Quotation. Implementation ogenous circumstances—suggest that the benefits
of the recommendations is taking the form of updat- of operational restructuring have not yet led to in-
ing of laws, regulations, and rules in line with inter- creases in firm efficiency.
national best practices; amendments to the listing While an analysis of corporate performance in
requirements of the Kuala Lumpur stock exchange; Malaysia finds a number of possible links between
and development of accreditation programs for ex- the corporate sector and the crisis, the failure of cor-
isting directors of listed companies. Malaysia also porate governance is an underlying theme. As in
led the Asia-Pacific Economic Cooperation (APEC) other Asian countries, the net worth of a firm—
finance ministers' initiative on corporate gover- rather than the profit potential of a particular under-
nance; in May 1999, a report was submitted to the taking—is an important determinant of how much a
APEC finance ministers entitled "Strengthening firm can borrow in Malaysia. Such a mechanism is
Corporate Governance in the APEC Region." Most likely to lead to an inefficient allocation of funds and
recently, a monitoring body on corporate gover- to the balance sheet problem, both of which may ex-
nance was set up, the Minority Shareholders Watch- acerbate the effects of a shock that reduces the
dog Group, to monitor corporations, provide advice wealth of a firm. Also, evidence indicates that corpo-
on best practices, and eventually to offer other ser- rations in Malaysia, and in the other Asian countries,
vices like proxy voting. While there has been a con- were increasing leverage despite their declining
certed effort to improve the corporate governance profitability, referred to as the "Asia Bet" by Harvey
environment, a few high-profile cases of poor gov- and Roper (1999). Increasingly, corporations were
ernance relating to companies with political links tapping short-term and foreign debt markets, effec-
have undermined market sentiment. Thus it will be tively betting that exchange rates would remain sta-
important to apply the new code of governance ble: they lost both bets. The close links between the
evenly. corporate sector and the financial sector and the lack
of a well-developed corporate bond market also led
to overreliance on bank lending.
Summary and Conclusions Thus, it could be argued that the system of corpo-
rate governance led to inefficient allocation of funds,
The analysis indicates that corporate performance created innate vulnerabilities in the corporate sector,
in Malaysia deteriorated notably prior to the crisis and failed to discipline managers and contain the risks
and recovered somewhat in 1999. Some accounting they assumed. The importance of good governance is
measures of performance, such as financial ratios, clear, and to achieve it requires better monitoring of
indicate a relatively stable performance immediately managers, improved information quality and avail-
prior to the crisis. Increasing risk—indicated by ability, and transparency in corporate management.
higher leverage, an increase in the proportion of dis- Adoption of the recommendations of the Finance
tressed firms, and the lower proportion of long-term Committee's Report on Corporate Governance is pro-
to total debt—meant, however, that risk-adjusted re- ceeding well and, if fully implemented and adhered to
turns were deteriorating. Stock market returns in in- in spirit, will significantly enhance corporate gover-
dividual markets in the crisis countries declined as nance in Malaysia. These reforms will take time to
well in the run-up to the crisis. Measures of effi- implement and become effective, but momentum
ciency also indicate that corporate performance should not be lost because of improving or deteriorat-
worsened in the precrisis period. The indicators ing corporate and general economic performance.
point to some role for the decline to corporate-level There is significant evidence of a deterioration in
fundamentals, such as performance and sustainabil- corporate sector performance, but it does not appear
ity, in explaining the crisis, particularly in Korea and to be commensurate with the resulting crisis. If the
Thailand. More recently, the corporate sector in decline in corporate fundamentals did not justify
Malaysia has benefited from a strong economic re- the extent of the crisis in Malaysia, then the role of
covery and lower interest rates, and the accounting contagion, the balance sheet problem, and the trans-
measures of corporate performance show significant fer problem suggest the type of crisis described by
improvement in 1999, although measures of effi- the model in Krugman (1999). The implications for
ciency have not improved. avoiding such crises are similar to those suggested

above. A more specific implication is the reduction Firms," The World Bank Research Observer, Vol. 15
in the reliance on short-term debt and external (Washington: World Bank), pp. 1-22.
debt—both are key to the feedback mechanism—to Fare, Rolf, Shawna Grosskopf, and C.A. Knox Lovell,
reduce susceptibility to sudden losses of confi- 1994, Production Frontiers (Cambridge: Cambridge
dence, resulting in self-fulfilling crises. Once such University Press).
a crisis is under way, it is clear that sufficiently Fare Rolf, and others, 1997, "Biased Technical Change
and the Malmquist Productivity Index," Scandinavian
large funds from a lender-of-last-resort could re- Journal of Economics, Vol. 99 (March), pp. 119-27.
store confidence and avoid a downward spiral and Furman, Jason, and Joseph Stiglitz, 1998, "Economic
protracted crisis. In reality, the amounts needed are Crises: Evidence and Insights from East Asia,"
very large and their availability may create moral Brookings Papers on Economic Activity, Vol. 2
hazard. (Washington: Brookings Institution), pp. 1-114.
Ghosh, Swati, and Atish Ghosh, 1999, "East Asia in the
Aftermath: Was There a Crunch?" IMF Working
Paper 99/38 (Washington: International Monetary
References Fund).
Harvey, C, and A. Roper, 1999, "The Asian Bet," in
Bank Negara Malaysia, 1999, Annual Report (Kuala Financial Markets and Development: The Crisis
Lumpur). in Emerging Financial Markets, Alison Harwood,
Berg, Andrew, and others, 1999, Anticipating Balance of Robert E. Litan, and Michael Pomerleano,
Payments Crises: The Role of Early Warning eds. (Washington: Brookings Institution Press),
Systems, IMF Occasional Paper No. 186 (Washing- pp. 29-115.
ton: International Monetary Fund). Johnson, Simon, and others, 1998, "Corporate Gover-
Bernake, Ben, and Marie Gertler, 1995, "Inside the Black nance in the Asian Financial Crisis, 1997-98," MIT,
Box: The Credit Channel of Monetary Policy Trans- mimeo (Cambridge, Massachusetts: Massachusetts
mission," Journal of Economic Perspectives, Vol.9 Institute of Technology).
(Fall), pp. 27-48. Kochhar, Kalpana, and others, 1999, Malaysia—Selected
Chang, Roberto, and Andres Velasco, 1998, "Financial Issues, IMF Staff Country Report No. 99/86 (Wash-
Crisis in Emerging Markets: A Canonical Model," ington: International Monetary Fund).
NBER Working Paper No. 6606 (Cambridge, Massa- Krugman, Paul, 1992, "A Model of Balance-of-Payments
chusetts: National Bureau of Economic Research). Crises," Currencies and Crises (Cambridge: MIT
Claessens, Stijn, Simeon Djankov, and Larry Lang, 1998, Press), pp. 61-76.
"Who Controls East Asian Corporations?" IMF Semi- , 1998, "What Happened to Asia?" paper presented
nar Series, No. 1999-6, pp. 1-40. at a conference in Japan, January (Cambridge: MIT).
, 1999, "Corporate Diversification in East Asia: Available via the Internet: http://web.mit.edu/krugman/
The Role of Ultimate Ownership and Group Affilia- www/#other.
tion," Policy Research Working Papers No. 2089 , 1999, "Balance Sheets, the Transfer Problem, and
(Washington: World Bank). Financial Crises," paper prepared for the festschrift
Claessens, Stijn, Simeon Djankov, and Daniela Klinge- volume in honor of Robert Flood, MIT, Cambridge.
biel, 1999, "Bank and Corporate Restructuring in Available via the Internet: http://web.mit. edu/
East Asia: Opportunities for Further Reform," paper krugman/www/#other.
presented at International Monetary Fund Annual La Porta, Rafael, Florencio Lopez-de-Silanes, and Andrei
Meetings seminar, Washington, September. Shleifer, 1998, "Corporate Ownership Around the
Claessens, Stijn, Simeon Djankov, and Lixine Xu, 2000, World," NBER Working Paper No. 6625 (Cambridge,
"Corporate Performance in the East Asian Financial Massachusetts: National Bureau of Economic Re-
Crisis," The World Bank Research Observer, Vol. 15, search).
pp. 23-46. , 1999, "Investor Protection: Origins, Conse-
Corsetti, Giancarlo, Paolo Pesenti, and Nouriel Roubini, quences, Reform," NBER Working Paper No. 7428
1998, "What Caused the Asian Currency and Finan- (Cambridge, Massachusetts: National Bureau of Eco-
cial Crisis? A Macroeconomic Overview," NBER nomic Research).
Working Paper No. 6833 (Cambridge, Massachusetts: Malmquist, Sten, 1953, "Index Numbers and Indifference
National Bureau of Economic Research). Surfaces," Trabajos de Estadistica, Vol. 4, pp.
Crama, Yves, and others, 1999, "Corporate Governance 209-42.
Structures, Control and Performance in European Obstfeld, Maurice, 1986, "Rational and Self-Fulfilling
Markets: A Tale of Two Systems," Center for Eco- Balance of Payments Crises," American Economic
nomic Research Discussion Paper No. 9942 (Lou- Review, Vol. 76 (March), pp. 72-81.
vain, Belgium: Catholic University of Louvain). Pomerleano, Michael, 1998, "The East Asia Crisis and
Diamond, Douglas, and Philip Dybvig, 1983, "Bank Runs, Corporate Finances: The Untold Micro Story," World
Deposit Insurance, and Liquidity," Journal of Politi- Bank Policy Research Working Paper No. 1990
cal Economy, Vol. 91 (June), pp. 401-19. (Washington: World Bank).
Dollar, David, and Mary Hallward-Driemeier, 2000, "Cri- Radelet, Steven, and Jeffrey Sachs, 1998, "The Onset of
sis, Adjustment, and Reform in Thailand's Industrial the East Asian Financial Crisis," NBER Working

Paper No. 6680 (Cambridge, Massachusetts: National , 1999a, East Asia: Recovery and Beyond (Wash-
Bureau of Economic Research). ington: World Bank).
Stiglitz, Joseph, and Andrew Weiss, 1981, "Credit Ra- , 1999b, "Malaysia Structural Policy Review: Path
tioning in Markets with Imperfect Information," to Recovery," Report No. 18647-MA, June (Washing-
American Economic Review, Vol. 71, pp. 393-410. ton: World Bank).
World Bank, 1998, East Asia: Road to Recovery (Wash- , forthcoming, Malaysia: Corporate Governance
ington: World Bank). Assessment (Washington: World Bank).

OCCASIONAL PAPERS
Recent Occasional Papers of the International Monetary Fund

207. Malaysia: From Crisis to Recovery, by Kanitta Meesook, Il Houng Lee, Olin Liu, Yougesh Khatri, Natalia
Tamirisa, Michael Moore, and Mark H. Krysl. 2001.
206. The Dominican Republic: Stabilization, Structural Reform, and Economic Growth, by Alessandro Gius-
tiniani, Werner C. Keller, and Randa E. Sab. 2001
205. Stabilization and Savings Funds for Nonrenewable Resources, by Jeffrey Davis, Rolando Ossowski,
James Daniel, and Steven Barnett. 2001.
204. Monetary Union in West Africa (ECOWAS): Is It Desirable and How Could It Be Achieved? by Paul
Masson and Catherine Pattillo. 2001.
203. Modern Banking and OTC Derivatives Markets: The Transformation of Global Finance and Its Implica-
tions for Systemic Risk, by Garry J. Schinasi, R. Sean Craig, Burkhard Drees, and Charles Kramer.
2000.
202. Adopting Inflation Targeting: Practical Issues for Emerging Market Countries, by Andrea Schaechter,
Mark R. Stone, and Mark Zelmer. 2000.
201. Developments and Challenges in the Caribbean Region, by Samuel Itam, Simon Cueva, Erik Lundback,
Janet Stotsky, and Stephen Tokarick. 2000.
200. Pension Reform in the Baltics: Issues and Prospects, by Jerald Schiff, Niko Hobdari, Axel Schimmelpfen-
nig, and Roman Zytek. 2000.
199. Ghana: Economic Development in a Democratic Environment, by Sergio Pereira Leite, Anthony Pelle-
chio, Luisa Zanforlin, Girma Begashaw, Stefania Fabrizio, and Joachim Harnack. 2000.
198. Setting Up Treasuries in the Baltics, Russia, and Other Countries of the Former Soviet Union: An Assess-
ment of IMF Technical Assistance, by Barry H. Potter and Jack Diamond. 2000.
197. Deposit Insurance: Actual and Good Practices, by Gillian G.H. Garcia. 2000.
196. Trade and Trade Policies in Eastern and Southern Africa, by a staff team led by Arvind Subramanian, with
Enrique Gelbard, Richard Harmsen, Katrin Elborgh-Woytek, and Piroska Nagy. 2000.
195. The Eastern Caribbean Currency Union—Institutions, Performance, and Policy Issues, by Frits van Beek,
Jose Roberto Rosales, Mayra Zermeno, Ruby Randall, and Jorge Shepherd. 2000.
194. Fiscal and Macroeconomic Impact of Privatization, by Jeffrey Davis, Rolando Ossowski, Thomas
Richardson, and Steven Barnett. 2000.
193. Exchange Rate Regimes in an Increasingly Integrated World Economy, by Michael Mussa, Paul Masson,
Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andy Berg. 2000.
192. Macroprudential Indicators of Financial System Soundness, by a staff team led by Owen Evans, Alfredo
M. Leone, Mahinder Gill, and Paul Hilbers. 2000.
191. Social Issues in IMF-Supported Programs, by Sanjeev Gupta, Louis Dicks-Mireaux, Ritha Khemani,
Calvin McDonald, and Marijn Verhoeven. 2000.
190. Capital Controls: Country Experiences with Their Use and Liberalization, by Akira Ariyoshi, Karl Haber-
meier, Bernard Laurens, Inci Otker-Robe, Jorge Ivan Canales Kriljenko, and Andrei Kirilenko. 2000.
189. Current Account and External Sustainability in the Baltics, Russia, and Other Countries of the Former So-
viet Union, by Donal McGettigan. 2000.
188. Financial Sector Crisis and Restructuring: Lessons from Asia, by Carl-Johan Lindgren, Tomas J.T.
Balino, Charles Enoch, Anne-Marie Guide, Marc Quintyn, and Leslie Teo. 1999.
187. Philippines: Toward Sustainable and Rapid Growth, Recent Developments and the Agenda Ahead, by
Markus Rodlauer, Prakash Loungani, Vivek Arora, Charalambos Christofides, Enrique G. De la Piedra,
Piyabha Kongsamut, Kristina Kostial, Victoria Summers, and Athanasios Vamvakidis. 2000.
186. Anticipating Balance of Payments Crises: The Role of Early Warning Systems, by Andrew Berg,
Eduardo Borensztein, Gian Maria Milesi-Ferretti, and Catherine Pattillo. 1999.
185. Oman Beyond the Oil Horizon: Policies Toward Sustainable Growth, edited by Ahsan Mansur and Volker
Treichel. 1999.

Occasional Papers
184. Growth Experience in Transition Countries, 1990-98, by Oleh Havrylyshyn, Thomas Wolf, Julian Beren-
gaut, Marta Castello-Branco, Ron van Rooden, and Valerie Mercer-Blackman. 1999.
183. Economic Reforms in Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan, by
Emine Gurgen, Harry Snoek, Jon Craig, Jimmy McHugh, Ivailo Izvorski, and Ron van Rooden. 1999.
182. Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union, by a staff team led by
Liam Ebrill and Oleh Havrylyshyn. 1999.
181. The Netherlands: Transforming a Market Economy, by C. Maxwell Watson, Bas B. Bakker, Jan Kees
Martijn, and Ioannis Halikias. 1999.
180. Revenue Implications of Trade Liberalization, by Liam Ebrill, Janet Stotsky, and Reint Gropp. 1999.
179. Disinflation in Transition: 1993-97, by Carlo Cottarelli and Peter Doyle. 1999.
178. IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, by Timothy Lane,
Atish Ghosh, Javier Hamann, Steven Phillips, Marianne Schulze-Ghattas, and Tsidi Tsikata. 1999.
177. Perspectives on Regional Unemployment in Europe, by Paolo Mauro, Eswar Prasad, and Antonio Spilim-
bergo. 1999.
176. Back to the Future: Postwar Reconstruction and Stabilization in Lebanon, edited by Sena Eken and
Thomas Helbling. 1999.
175. Macroeconomic Developments in the Baltics, Russia, and Other Countries of the Former Soviet Union,
1992-97, by Luis M. Valdivieso. 1998.
174. Impact of EMU on Selected Non-European Union Countries, by R. Feldman, K. Nashashibi, R. Nord,
P. Allum, D. Desruelle, K. Enders, R. Kahn, and H. Temprano-Arroyo. 1998.
173. The Baltic Countries: From Economic Stabilization to EU Accession, by Julian Berengaut, Augusto
Lopez-Claros, Francoise Le Gall, Dennis Jones, Richard Stern, Ann-Margret Westin, Effie Psalida,
Pietro Garibaldi. 1998.
172. Capital Account Liberalization: Theoretical and Practical Aspects, by a staff team led by Barry Eichen-
green and Michael Mussa, with Giovanni Dell'Ariccia, Enrica Detragiache, Gian Maria Milesi-Ferretti,
and Andrew Tweedie. 1998.
171. Monetary Policy in Dollarized Economies, by Tomas Balino, Adam Bennett, and Eduardo Borensztein.
1998.
170. The West African Economic and Monetary Union: Recent Developments and Policy Issues, by a staff
team led by Ernesto Hernandez-Cata and comprising Christian A. Francois, Paul Masson, Pascal Bou-
vier, Patrick Peroz, Dominique Desruelle, and Athanasios Vamvakidis. 1998.
169. Financial Sector Development in Sub-Saharan African Countries, by Hassanali Mehran, Piero Ugolini,
Jean Phillipe Briffaux, George Iden, Tonny Lybek, Stephen Swaray, and Peter Hayward. 1998.
168. Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility, by a staff team
led by Barry Eichengreen and Paul Masson with Hugh Bredenkamp, Barry Johnston, Javier Hamann,
Esteban Jadresic, and Inci Otker. 1998.
167. Exchange Rate Assessment: Extensions of the Macroeconomic Balance Approach, edited by Peter Isard
and Hamid Faruqee. 1998
166. Hedge Funds and Financial Market Dynamics, by a staff team led by Barry Eichengreen and Donald
Mathieson with Bankim Chadha, Anne Jansen, Laura Kodres, and Sunil Sharma. 1998.
165. Algeria: Stabilization and Transition to the Market, by Karim Nashashibi, Patricia Alonso-Gamo, Stefania
Bazzoni, Alain Feler, Nicole Laframboise, and Sebastian Paris Horvitz. 1998.
164. MULTIMOD Mark III: The Core Dynamic and Steady-State Model, by Douglas Laxton, Peter Isard,
Hamid Faruqee, Eswar Prasad, and Bart Turtelboom. 1998.
163. Egypt: Beyond Stabilization, Toward a Dynamic Market Economy, by a staff team led by Howard Handy.
1998.
Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMF
Publication Services.

9781589060470

Uploaded by

Copyright:

Available Formats

9781589060470

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

9781589060470

Uploaded by

Copyright:

Available Formats

OCCASIONAL PAPER

Kanitta Meesook, Il Houng Lee, Olin Liu,Yougesh Khatri,

INTERNATIONAL MONETARY FUND

©International Monetary Fund. Not for Redistribution

Production: IMF Graphics Section

Malaysia: from crisis to recovery / Kanitta Meesook . . . [et al.]—Washington,

Please send orders to:

©International Monetary Fund. Not for Redistribution

II Comparative Review of Policies and Performance, 1997-2000 3

V Capital Controls in Response to the Asian Crisis 50

©International Monetary Fund. Not for Redistribution

VII Corporate Performance and Reform 83

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

The following symbols have been used throughout this paper:

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

A s an emerging market economy, Malaysia is

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

M alaysia has received much attention since Sep-

Figure 2 . 1 . Exchange Rate Developments

©International Monetary Fund. Not for Redistribution

30 Openness and Structure of Exports

©International Monetary Fund. Not for Redistribution

Table 2.1.Asian Crisis Countries: Selected Structural Indicators, 1996-2000

Malaysia Korea Thailand Philippines Indonesia

©International Monetary Fund. Not for Redistribution

Real GDP growth

©International Monetary Fund. Not for Redistribution

Figure 2.5. Market Capitalization in Figure 2.6. Private Sector Credit in

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

ture projects, for example through the Corporate

Capital Controls and Exchange Rate Regime 1.5

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

Date/Background Policy Objectives Specific Measures

February 1998 Corporate:To strengthen • High Level Finance Committee on Corporate

Monetary:To rebalance • Bank Negara Malaysia intervention rate raised to 11

• Bank Negara Malaysia set guidelines to ensure credit to

• Announcement that five financial institutions were in

©International Monetary Fund. Not for Redistribution

Date/Background Policy Objectives Specific Measures

Financial and corporate:To • Danamodal established as a vehicle to recapitalize banks

©International Monetary Fund. Not for Redistribution

Economic Performance Following As output collapse became evident, macroeco-

©International Monetary Fund. Not for Redistribution

Date/Background Policy Objectives Specific Measures

©International Monetary Fund. Not for Redistribution

Table 2.4 (conclude

Date/Background Policy Objectives Specific Measures

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

©International Monetary Fund. Not for Redistribution

Malaysia's growth rate was very high. Inflation was low.

Short-term exposure was small . . . . . . and reserve cover was adequate.