Hamada Koichi 2009 Towards Monetary and Financial Integration in East Asia

Koichi Hamada, Beate Reszat and Ulrich Volz 2009
Chapter 11 The Peterson Institute for International Economics 2009

All rights reserved. No part of this publication may be reproduced, stored
in a retrieval system or transmitted in any form or by any means, electronic,
mechanical or photocopying, recording, or otherwise without the prior
permission of the publisher.
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Edward Elgar Publishing, Inc.
William Pratt House
9 Dewey Court
Northampton
Massachusetts 01060
USA
A catalogue record for this book is available from the British Library
Library of Congress Control Number: 2009922771
ISBN 978 1 84844 363 1

Printed and bound in Great Britain by MPG Books Group, UK
Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Downloaded from Elgar Online at 05/14/2016 12:51:27AM
via National Graduate Institute for Policy Studies, Japan
Contents
List of contributors
Foreword by Richard N. Cooper
Acknowledgements
vii
ix
xi
Introduction: prospects for monetary and financial integration in

East Asiadreams and dilemmas
Koichi Hamada, Beate Reszat, and Ulrich Volz
PART I
THE POLITICAL ECONOMY OF REGIONAL

INTEGRATION
1 The ASEAN economic community and the European experience

Michael G. Plummer and Reid W. Click
2 The political economy of European economic and monetary
union negotiations and implications for East Asia
Heungchong Kim
3 International political conflicts and economic integration
Koichi Hamada and Inpyo Lee
PART II
4
6
7
13
41
61
DEVELOPING BOND MARKETS IN EAST ASIA
Learning by doing in market reform: lessons from a regional

bond fund
Guonan Ma and Eli Remolona
Currency denomination in Asian bond markets
Eiji Ogawa and Junko Shimizu
PART III
87
104
EXCHANGE RATE POLICIES IN EAST ASIA:

THE ROLE OF THE DOLLAR
East Asias role in the revived Bretton Woods system

Michael P. Dooley, David Folkerts-Landau, and Peter Garber
Current account surpluses and conflicted virtue in East Asia:
China and Japan under the dollar standard
Ronald McKinnon and Gunther Schnabl
141
159

vi
Contents
PART IV
8
9
Three cases for monetary integration in East Asia

Ulrich Volz
Monetary and exchange rate policy coordination in
ASEAN11
William H. Branson and Conor N. Healy
PART V
10
11
12
14
222
267
290
304
THE ROLE OF CHINA
The illusion of precision and the role of the renminbi in

regional integration
Yin-Wong Cheung, Menzie D. Chinn, and Eiji Fujii
Institutional and structural problems of Chinas foreign
exchange market and the RMBs role in East Asia
Zhang Jikang and Liang Yuanyuan
Index
195
CURRENCY BASKETS FOR EAST ASIA?
Capital markets and exchange rate stabilization in East Asia:

diversifying risk based on currency baskets
Gunther Schnabl
Asian currency baskets
John Williamson
The role of an Asian currency unit
Masahiro Kawai
PART VI
13
TOWARD MONETARY INTEGRATION IN EAST

ASIA?
325
357
387

Contributors
William H. Branson, John Foster Dulles Professor of International
Affairs and Professor of Economics and International Affairs, emeritus,
Department of Economics, Princeton University, Princeton, USA.
Yin-Wong Cheung, Professor of Economics, Economics Department,
University of California, Santa Cruz, USA.
Menzie D. Chinn, Professor of Public Affairs and Economics, Robert M.
La Follette School of Public Affairs, University of Wisconsin at Madison,
USA.
Reid W. Click, Associate Professor of International Business, George
Washington University School of Business, Washington, DC, USA.
Richard N. Cooper, Maurits C. Boas Professor of International Economics,
Harvard University, Cambridge, MA, USA.
Michael P. Dooley, Professor of Economics, Economics Department,
University of California, Santa Cruz, USA.
David Folkerts-Landau, Managing Director and Global Head of Research,
Deutsche Bank AG, London, UK.
Eiji Fujii, Associate Professor of Economics, Graduate School of Systems
and Information Engineering, University of Tsukuba, Japan.
Peter Garber, Global Strategist, Deutsche Bank AG, London, UK.
Koichi Hamada, Tuntex Professor of Economics, Department of Economics
and Economic Growth Center, Yale University, New Haven, USA.
Conor N. Healy, Woodrow Wilson School, Princeton University, Princeton,
USA.
Masahiro Kawai, Dean, Asian Development Bank Institute, Tokyo, Japan.
Heungchong Kim, Research Fellow and Head of the Europe Team, Korea
Institute for International Economic Policy, Seoul, Korea.
Inpyo Lee, Professor of Economics, Ewha University, Seoul, Korea;
Visiting Fellow, Economic Growth Center, Yale University, New Haven,
USA.
vii

viii
Contributors
Liang Yuanyuan, Research Associate, Center for International Finance

Studies, Fudan University, Shanghai, China.
Guonan Ma, Senior Economist, Representative Office for Asia and the
Pacific, Bank for International Settlement, Hong Kong.
Ronald McKinnon, William D. Eberle Professor of International Economics,
Department of Economics, Stanford University, USA.
Eiji Ogawa, Professor of Economics, Graduate School of Commerce
and Management, Hitotsubashi University; Deputy Director of the EU
Institute in Japan, Tokyo, Japan.
Michael G. Plummer, Professor of Economics, Paul H. Nitze School of
Advanced International Studies, Johns Hopkins University, Bologna
Center, Italy.
Eli Remolona, Deputy Chief Representative of the Office for Asia and the
Pacific, Bank for International Settlements, Hong Kong.
Beate Reszat, Economist, Hamburg Ministry of Science and Research,
Hamburg, Germany.
Gunther Schnabl, Professor of Economics, Faculty of Economics and
Business Administration, University of Leipzig, Germany.
Junko Shimizu, Associate Professor, Faculty of Economics, Meikai
University, Japan.
Ulrich Volz, Senior Economist, German Development Institute, Bonn,
Germany.
John Williamson, Senior Fellow, Peterson Institute for International
Economics, Washington, DC, USA.
Zhang Jikang, Professor of Economics, Director of the Center for
European Studies, and Deputy Director of the Center for International
Finance Studies, Fudan University, Shanghai, China.

Foreword
Richard N. Cooper
There is a vague but palpable dissatisfaction with existing international
monetary arrangements. These entail few formal rules with regard to how
countries should manage their exchange rates, and little coherent guidance
from the economics profession beyond an apparent consensus that floating
a countrys currency is better than fixing italbeit with many reservations
about free floating for any but the largest of economies.
Against this intellectual background, Europeans defied many academic
reservations and in 1999 created a common currency among 11, later 16,
members of the European Union. The resulting euro has now had a decade
of experience, without obviously disastrous results. There have been some
strains, particularly, as was predicted and feared, regarding Italy, but also
a respectable record of achievement in terms of continued growth with
modest inflation.
At the same time, there have been large and disturbing movements in
exchange rates among the major currencies. During the past two decades
the yen has ranged from 85 yen to the dollar in 1995 to 148 in 2005, a
range of 75 percent, while during the decade of its existence the euro has
ranged by over 90 percent, between USD 0.83 and USD 1.60 per euro.
Throughout the period, inflation was relatively low in all three regions.
What justification is there for such wide swings among major currencies? What are the implications for international trade and especially for
investment in tradable goods? And what are the implications for decisions
regarding exchange rate policy for other countries, especially developing
countries with imperfect capital markets?
It is against this general background that the countries of East Asia,
one by one, have managed their exchange rates. But could they do better
by cooperating among themselves in monetary and exchange rate management, even to the point of adopting a common currency, as some
Europeans have done? This volume usefully explores these issues.
By wide agreement, a common currency is not in prospect anytime
soonif ever. There are too many political and institutional hurdles, and
even the economic gains are not clear. It is worth recalling, however, that
it took Europe over four decades between the conception of its economic
ix

Foreword
community as reflected in the Rome Treaty of 1957 and the creation of

the euro in 1999. Many forms of economic and financial cooperation are
possible short of adoption of a common currency, some of which would
facilitate an eventual move toward that end, should it prove possible and
desirable.
The first steps have been taken toward mutual financial support among
East Asian economies in times of stress, and toward the creation of an
Asian bond market in normal times. China must develop a functioning
foreign exchange market and move to full convertibility of the yuan if
it is to play a central role in regional financial cooperation. The most
vexing issue is management of exchange rates. Economic interdependence
within East Asia is growing rapidly, even while commercial interactions
with countries outside the region continue to grow. Countries thus face
the challenging task of framing exchange rate policy with one eye on the
regioncontaining both customers and competitorsand one eye on the
rest of the world.
The chapters of this book cover all of these challenging issues, and more.
They offer a worthy exploration of a complex set of increasingly important
issues and help deepen our understanding of a new financial architecture
that might be emerging in East Asia.

Acknowledgements
During parts of the work on this book Beate Reszat was Head of
Research on International Financial Markets at the Hamburg Institute
of International Economics. Ulrich Volz was a Fox International Fellow
and Max Kade Scholar at Yale University and a DekaBank Fellow at the
Hamburg Institute of International Economics.
Ulrich Volz would like to thank the Fox International Fellowship at
Yale University, the Max Kade Foundation in New York, and DekaBank
in Frankfurt for financial support during the work on this project.
The editors would like to express their gratitude to all of the contributors to this book. We would very much like to thank Andrei Popovici, our
research assistant at Yale, for his scrupulous proofreading and valuable
help with the editing of the manuscript. We also thank Edward Elgar
Publishing for the professional and speedy handling of the manuscript.
Special thanks go to Richard Cooper for writing the foreword.
Sadly, during the work on this book we lost two dear friends and colleagues, both of whom contributed chapters. Zhang Jikang of Fudan
University died very unexpectedly and tragically on February 18, 2006.
William Branson of Princeton University passed away on August 15, 2006,
after a long illness. It is to the memories of Jikang and Bill that we dedicate
this book.
xi


Introduction: prospects for monetary

and financial integration in East Asia
dreams and dilemmas
Koichi Hamada, Beate Reszat, and Ulrich Volz
The Asian financial crisis of 199798 fundamentally changed East Asias
perspective on economic regionalism and sparked great political interest
in monetary and financial cooperation and integration in the region. The
crisis revealed the fragility of the regions prevailing exchange rate arrangements and highlighted the need for a strengthening of the regional financial
architecture. Since the crisis, there has been a proliferation of proposals for
fostering East Asian monetary and financial integration.
Regional cooperation in the field of money and finance already takes
place through the Association of Southeast Asian Nations (ASEAN)
surveillance process, the Chiang Mai Initiative for the creation of bilateral
short-term financing facilities by the ASEAN13 countries (ASEAN plus
China, Japan, and Korea), and various initiatives to foster the development of regional security markets. There is also a serious debate about
establishing a regional exchange rate arrangement or even a common
East Asian currency in the longer term. In December 2005, the Asian
Development Bank announced its intention to create an Asian currency
unit (ACU) comprising a basket of East Asian currencies akin to the
European currency unit (ECU), which evolved into the euro. Although
this initiative has not come to fruition, a regional currency unit is now
being studied by the ASEAN13 countries. Particularly remarkable is the
involvement of China, which has demonstrated an increasing willingness
to accept responsibility for the economic progress of the region as a whole.
The ASEAN countries, as well as Japan and Korea, have also displayed
ample interest in fostering regional cooperation.
Deeper integration, however, is still a ways away. Countries participating in more ambitious regional monetary and financial initiatives would
need to develop a far-reaching consensus about economic policy preferences. Moreover, national authorities would need to be willing to subordinate national policy goals, at least at times, to a common goal of regional
1

Towards monetary and financial integration in East Asia
integration and stability. It does not seem that such a policy consensus
and integrationist spirit has sufficiently developed in East Asia as yet.
Integration efforts remain strained by tensions between major players in
East Asia, most notably China and Japan, both of which tend to claim
a leadership role in economic cooperation, and each of which is short of
having full trust in the other.
The aim of this book is to critically analyze these developments, to
delineate the conditions for successful integration, and to examine the
forms under which regional monetary and financial cooperation may be
favorable options for the East Asian countries. It assesses the steps already
taken toward financial integration and brings forward different proposals
for future exchange rate arrangements in East Asia. In particular, the book
evaluates the economic and politico-economic arguments and conditions
for monetary and financial integration in East Asia and explores how and
to what extent the countries of the region can integrate despite their heterogeneity and their underlying political tensions. Drawing on the European
experiences, this book analyzes the economic logic of monetary and financial integration in East Asia and its political feasibility.
The book has three unifying themes. First, it highlights the barriers
and dilemmas that East Asian countries face in pursuing monetary and
financial integration. Despite a common interest in regional cooperation,
East Asian countries are confronted with multiple problems that jeopardize a process of closer integration and that could turn their dream into
pie in the sky. If we limit our attention to the purely economic dimension, in the short run East Asian countries encounter incentive problems,
domestic as well as international, that complicate the process of attaining
the ideal state of cooperation which would spell increased welfare for all.
The dilemma is further intensified in the short run if we consider political
elements. The need to satisfy domestic constituencies limits governments
capability to participate in joint international actions. At the same time,
historical disputes, conflicting national strategic interests, and rivalry for
economic and political leadership in the region impede cooperation. In
the intermediate and in the long run, however, the situation may turn
out to be different. East Asian countries may understand that the cost of
constraints on monetary policy by engaging in regional monetary integration is somewhat overestimated. If they widen both their time horizon
and their criteria for appraising their self-interest, they will understand
the long-run merit of being freed from the de facto dollar standard and
factor in the geopolitical peace dividends that will accrue from regional
cooperation.
Second, since the European monetary union is the prime exemplar of a
multilateral currency union, it is crucial to analyze its lessons for successful

Introduction
monetary cooperation and financial integration in East Asia. European

countries experienced many difficulties on their way to monetary unification, with the currency crisis of 199293 being the most critical moment.
It is important to see how European countries overcame their political
and economic differences and how the motivation for peace and prosperity overrode longstanding national antipathies. The common agency
approach to the creation of European monetary integration helped to
balance the interests of all countries involved. For instance, the creation of
the ECU circumvented the political difficulties associated with selecting an
existing currency to be Europes lead currency (even though the German
mark, in practice, fulfilled this role) and can be seen as providing the historical precedent for the launch of the ACU. At the same time, this volume
examines economic and political differences between Europe and East Asia
and discusses the unique obstacles to a process of East Asian integration.
Finally, East Asian integration needs to be seen in the context of the
global financial system. The Asian financial crisis, which made the instability of the international financial markets all too manifest to East Asian
policymakers, triggered a desire for a greater level of East Asian policy
coordination. Regional cooperation, it can be argued, provides a way to
insulate East Asia from global shocks and reduces economic dependency
on the United States and Europe. In turn, East Asian financial integration may contribute to the stability of the world monetary system itself.
Moreover, it would increase East Asias presence and influence on the
world stage and give it a stronger voice in the reforming of the international
financial architecture.
The book is divided into six parts. Part I focuses on the political
economy of regional integration. While European monetary integration is
frequently considered a blueprint for East Asian integration, the relationships between East Asian countries are very different from those between
European Union nations. One major distinction lies in East Asian nations
conflicting strategic interests. In particular, the rivalries between China
and Japan pose great questions about the success of efforts at integration.
Nevertheless, the global contextincluding the prevalence of financial
shocks, the posture of the IMF, and the policies of the United Stateshas
opened the door to monetary and financial regionalism in East Asia.
The first chapter by Michael Plummer and Reid Click provides an
overview and assessment of economic cooperation within ASEAN, the
only supranational body promoting economic integration in the region.
Drawing on lessons from the European integration experience, Plummer
and Click elaborate the particular characteristics of the Southeast Asian
integration process. The ASEAN countries, which aim to create an
ASEAN Economic Community by 2020, have also been eager to cooperate

with China, Japan, and Korea under the ASEAN13 framework on monetary and financial matters. It is commonly asserted that this group of
countries is extremely heterogeneous, in terms of economic development as
well as in terms of economic and political systems. However, the European
Union, often presented as a genuinely homogeneous bloc, is also composed
of a very diverse set of countries, with conflicting national views and interests being the rule rather than the exception. In Chapter 2, Heungchong
Kim analyzes the roles and policy positions of the various actors in the
European monetary integration process and draws implications for East
Asian countries. East Asian cooperation, he asserts, will require a good
dose of pragmatism, a common vision, and a long breath.
In Chapter 3, Koichi Hamada and Inpyo Lee draw attention to the relationship between international political conflicts and economic integration
and examine the incentive structure needed for East Asian integration to
progress. They argue that the benefits of regional monetary and financial
integration may not only derive from the standard arguments highlighted
in the economics literature. According to Hamada and Lee, purely economic incentives may not be strong enough to convince nations to join
a monetary union. Just as they were in the case of the European Unions
decision to create an economic and monetary union, security reasons may
turn out to be a critical motive for fostering monetary integration in East
Asia.
Part II is devoted to the very topical matter of developing bond markets
in East Asia, presently the most ambitious venture in the area of finance in
the region. Financial market development can be seen both as a precondition for monetary integration and as a desired outcome of monetary integration. Moreover, financial cooperation, which does not demand as much
commitment as exchange rate cooperation, can be viewed as a testing of the
waters before moves are made toward monetary integration. In Chapter
4, Guonan Ma and Eli Remolona discuss the rationale for developing a
regional securities market and how cooperation in this area has developed
so far. In particular, they examine one of the most important regional
initiatives to promote regional bond market development, the Asian Bond
Fund II, which was established by 11 central banks in East Asia and the
Pacific in 2005. They show how working together to create a regional
index bond fund and eight single-market funds helped the involved central
banks to identify and overcome market impediments in local currency
bond markets.
In Chapter 5, Eiji Ogawa and Junko Shimizu turn to currency denomination in Asian bond markets. They estimate the risks of various currency
basket denominated bonds and show that bonds issued in a currency basket
comprised of East Asian currencies would, in general, reduce the foreign

Introduction
borrowing costs for bond issuers in all East Asian countries. Especially in
the case of bond issuers in dollar-pegging countries, the foreign borrowing
costs of issuing the currency basket denominated bonds would be lower
than those of issuing home currency denominated bonds.
The two contributions in Part III investigate the special role of the US
dollar in East Asia. Before the Asian crisis, all East Asian countries except
Japan pegged their currencies to the US dollar, either formally or informally. These (soft) pegs made them vulnerable to dollaryen fluctuations,
which contributed to the crisis. Despite numerous recommendations to
freely float their currencies in order to avoid future misalignments and
a recurrence of exchange rate crises, East Asian countries have in part
returned to their pre-crisis exchange rate policies. In Chapter 6, Michael
Dooley, David Folkerts-Landau, and Peter Garber explain why most
East Asian countries continue with their dollar pegs and develop what
has become known as the Bretton Woods II view. They argue that the
current international monetary systemin which a macroeconomically
important periphery, emerging East Asia, fixes an undervalued exchange
rate to the US dollar in order to promote an export-driven development
policyis analogous to that of the Bretton Woods period. So far, the
US has willingly played the role of center country again, running balance
of payments deficits and serving as the global financial intermediary.
Dooley, Folkerts-Landau, and Garber claim that the system will continue for years more because of the incentive that East Asian countries,
especially China, have to absorb a vast pool of underemployed labor into
the industrial system.
Ronald McKinnon and Gunther Schnabl have a similar take on the
dollars role in East Asia. In Chapter 7, they explain why continuing to
peg to the dollar is entirely rational from the East Asian perspective and
why the East Asian dollar standard, as they call it, is likely to continue.
Rather than undervaluing their currencies in order to promote exports, as
maintained in the preceding chapter, McKinnon and Schnabl argue that
East Asian governments, in particular China, are trapped into maintaining
soft dollar pegs. Because most East Asian economies have transformed
themselves from dollar debtors into dollar creditors, they face what
McKinnon and Schnabl call conflicted virtue, pressure to appreciate
their currencies that could lead to a deflationary spiral and zero interest
liquidity trap.
In Part IV, the chapters by Ulrich Volz and by William Branson and
Conor Healy argue that the region would be better off pursuing a path of
monetary integration instead of relying on the external dollar anchor. In
Chapter 8, Ulrich Volz puts forward three cases for monetary integration
in East Asia. Besides the traditional trade argument that he underscores

by estimating the effect of exchange rate stability on trade in East Asia, he

also argues that monetary integration could help overcome the regions
problems with original sin and conflicted virtue. Moreover, he questions the conventional wisdom that monetary integration always goes
along with a loss of monetary policy autonomy and argues that regional
monetary integration could actually increase monetary policy influence for
most smaller economies in the region. Volz reasons that the current situation under the East Asian dollar standard already severely limits policy
autonomy in most countries and that moving toward regional exchange
rate cooperation would increase the degree of policy freedom enjoyed by
the region as a whole.
William Branson and Conor Healy also speak out in favor of explicit
regional monetary and exchange rate policy coordination in Chapter 9.
Branson and Healy show that the trade structure of ASEAN and China,
in terms both of the geographic distribution of imports and exports and
of the commodity structure of trade, is consistent with the adoption of a
common currency basket for region-wide exchange rate stabilization. Since
East Asian countries trade patterns are quite similar and their policies
are already implicitly coordinated through their (soft) pegs to the dollar,
their real effective exchange rates tend to move together. This means that
ASEAN and China are already moving toward integration in practical
effect. Branson and Healy maintain that an explicit move toward exchange
rate coordination, which would require macro policy coordination, would
also support regional surveillance and the development of regional bond
markets. Moreover, it would facilitate reserve-sharing under the Chiang
Mai Initiative. This would, in turn, release reserves that could be invested
in a new Asian Development Fund that could be used to support growth
and poverty reduction.
Part V is dedicated to one particular form of exchange rate arrangement
that has received considerable attention since the Asian crisiscurrency
baskets. In Chapter 10, Gunther Schnabl discusses the rationale for currency
baskets and presents original evidence of a move toward basket strategies
in East Asia. He argues that because of underdeveloped capital markets
and a very limited international role of domestic currencies, the East Asian
countries (with the exception of Japan) are likely to continue exchange rate
stabilization and the accumulation of international reserves. Yet expectations of a further depreciation of the dollar may trigger a broader orientation of exchange rate policies toward basket strategies. While the direction
of trade would suggest a substantial weight for the Japanese yen in East
Asian countries optimal currency baskets, the role of the euro is enhanced
by high expectations of its long-term stability. Schnabls estimations of
the basket structures of East Asian countries suggest growing euro and

Introduction
yen weights in the currency baskets of Indonesia, Korea, the Philippines,

Singapore, Taiwan, and Thailand, while the dollar remains the dominant
anchor currency for China, Hong Kong, and Malaysia.
In Chapter 11, John Williamson, the originator of the currency basket
proposal for East Asia, demonstrates how pegging to a currency basket
consisting of the dollar, the yen, and the euro would contribute to the stability of the nominal effective exchange rates of the East Asian countries.
Pegging to a common basket of major international currencies would not
only stabilize the effective exchange rate against the shocks that would be
imposed on a countrys macroeconomy by fluctuations in the value of these
currencies (which can be highly destabilizing, as shown during the run-up
to the Asian crisis) but also yield the benefit of increased regional stability. Williamsons results indicate that, for countries already pegging to the
dollar, pegging to a basket could result in significantly less instability than
the current policy. In case of a desire to realize those benefits but also to
avoid the risk of pegging, Williamson suggests adopting a managed float
guided by the use of a common basket as numeraire.
Such a numeraire is the topic of Masahiro Kawais contribution in
Chapter 12. Kawai, however, proposes a currency basket different from
the one Williamson has in mind. Rather than a basket of international
currencies, Kawai envisages an Asian currency unit (ACU) consisting of
East Asian currencies. The first idea behind the ACU is to monitor the
collective movement of regional currencies against key external currencies
(such as the US dollar and the euro) as well as each component currencys
movement against the ACU benchmark. Second, the ACU could be useful
for developing new tradable instruments, such as futures or ACU denominated bonds. Third, Kawai argues that at some point in the future the ACU
could play a role similar to that of the European ECU within a regional
monetary system. The appeal of the ACU is that it provides a venue for
initiating regional exchange rate cooperation without instantly requiring
far-reaching commitments and without requiring the politically difficult
task of selecting a single regional lead currency.
The two final chapters of the book investigate the special role of the
Chinese currency in the region. In Chapter 13, Yin-Wong Cheung, Menzie
Chinn, and Eiji Fujii take a fresh look at the renminbi debate that has
been one of the hottest issues in international economics in the last few
years. They show that the evidence of renminbi undervaluation may not
be as strong as it appears, demonstrating that, depending on the estimation method used, conclusions may range from slight overvaluation to
undervaluation. Even in the case of undervaluation, the results are not
significant in the statistical sense. Cheung, Chinn, and Fujii also note that
the fact that China is playing an important economic role in East Asia

and that it has established a complex production and trade network with
the neighboring economies complicates the calculation of the equilibrium
exchange rate. Thus, a change of Chinese exchange rate policy in response
to pressure from foreign countries and short-run considerations may have
undesirable effects not only on the Chinese economy but also on the region
as a whole.
The final chapter by Zhang Jikang and Liang Yuanyuan examines
the institutional and structural problems with Chinas foreign exchange
market and derives implications for the renminbis role in regional monetary integration. While China has made considerable progress in foreign
exchange market reform and has thereby met the preconditions for gradually introducing more flexibility to its exchange regime, Zhang and Liang
show that the foreign exchange market is still hampered by structural and
institutional problems such as low liquidity, high market concentration,
limited transaction instruments, distorted market supply and demand, and
passive intervention by the Peoples Bank of China. Along with the lack
of deep and liquid primary and secondary financial markets, these factors
will continue to limit the renminbis role in East Asia. For the time being,
therefore, the renminbi is far from being prepared to take on the role of a
regional anchor currency.
Certainly, there are many more related topics that deserve our attention, but we believe that this collection of chapters covers some of the
most momentous issues. This volume gathers together the cast of authors
best qualified, we may be allowed to say, to tackle these questions and to
explore the path to monetary and financial integration.
Many scholars writing on East Asian integration emphasize that it took
Western European countries almost half a century to create a common
market and establish a monetary union and that at least the same time
horizon might be realistic for East Asia. We are also fully aware of the
numerous problems and obstacleseconomic and politicalthat lie
ahead on the way to East Asian monetary and financial integration. And,
though we may not see an East Asian monetary union within the next
decade, isnt it amazing to observe the momentum gained and the speed
with which East Asian countries have been pushing forward toward integration? Who would have seriously believed in a common East Asian bond
market or even an ACU just a decade ago?
The dynamics of East Asian integration should by no means be underestimated. The economic potential of the region and the shared desire
since the Asian crisis to reduce dependency on Western financial markets
will make it increasingly unlikely that the countries of emerging East Asia
will continue to link their currencies to the US dollar. To draw an analogy
to Europe, it is worthwhile to recall that the main impetus for European

Introduction
monetary integration was the collapse of the Bretton Woods system. If we

have a Bretton Woods II system today, it will be interesting to see how long
the dollar will be able to provide stability to East Asia and if and when the
region will emancipate itself from the dollar the way Western Europe did
some 30 years ago.


1.
The ASEAN economic community

and the European experience
Michael G. Plummer and Reid W. Click1
1.1
INTRODUCTION
In November 2002, it was proposed at the Association of Southeast Asian

Nations (ASEAN) Heads of Government meeting in Phnom Penh that the
region should consider the possibility of creating an ASEAN Economic
Community (AEC) by 2020, a timeframe that was later shortened to
2015. The name is evocative, for an Economic Community immediately brings to mind the European experience. In fact, when the Asian
Pacific Economic Cooperation (APEC) was re-inventing itself, it was
proposed that the words behind the organizations acronym be replaced
with Asia-Pacific Economic Community. This idea was rejected explicitly for fear that it would give the impression that APEC was intending
to move in the direction of the EC model, which was thought to be too
controversial.
That the ASEAN Heads of Government should consider an Economic
Community, even with the baggage the term brings, is in some sense
nothing new. ASEAN has always studied carefully European economic
integration and seen it as a sort of role model, though certainly to be
adapted in the Southeast Asian development context. In this chapter, we
consider what lessons the European experience might hold for ASEAN, as
well as extend some suggestionsbased in part on the EU experienceas
to how ASEAN might evolve into an AEC. We begin with a brief contextual consideration of ASEAN intra-regional economic interaction, in
particular with respect to trade, with some comparison to the early years
of the EU. This is followed in Section 1.3 by a review of the evolution
of ASEAN economic integration, culminating in the AEC. Section 1.4
then proceeds to delineate some salient lessons of the EU experience for
ASEAN. Finally, we give some suggestions as to how the AEC might
proceed in Section 1.5.
13

14
1.2
The political economy of regional integration
INTRA-REGIONAL ASEAN CONTEXT
As of 2004, intra-regional trade in ASEAN came to somewhat less than

one fourth of total trade (Table 1.1). At the individual country level, intraregional trade as a percentage of total trade was highest for Myanmar
(almost one half) and Laos (one third). In terms of the value of intra-regional
trade, Singapore is number one (followed by Malaysia), though it should
be noted that Singapores share is particularly high because of the fact
that it engages considerably in intra-regional entrepot trade. The ASEAN
trade share was lowest for Cambodia (8 percent) and Vietnam (13 percent).
Outside of the region, approximately 14 percent of ASEAN exports were
destined for the EU market, less than the 16 percent accounted for by the
United States but greater than the 12 percent going to Japan. The EU is
the most important single market for four ASEAN countries (Singapore,
Laos, Myanmar, and Vietnam), the United States was the largest market
for three countries (Malaysia, Thailand, and Cambodia), and Japan was
the largest market for another three countries (Indonesia, the Philippines,
and Brunei). The Triad (the United States, Japan, and the EU) also
dominate foreign direct investment (FDI) flows to the region.
Hence, regional economic integration in ASEAN has to be appreciated
in the context of a regional organization whose most important economic
partners lie outside the region. As will be discussed below, this important
fact has been a key reason why ASEAN economic integration has been
mainly geared toward open regionalism; the cost of an inward-looking
approach to regionalism, or Fortress ASEAN, would be far too high.
Regionalism in developing countries that have focused on creating fortresses has generally failed (e.g., the Latin American Free Trade Area). It
would be a disaster in ASEANs case.
Economic integration in Europe, therefore, stands in contrast to the
ASEAN integration process. As we argue below, the EU2 in its earlier
stages of economic integration had fairly high levels of trade protection.
This was reinforced in the 1950s by the European Payments Union, which
discriminated in favor of intra-EU imports. These policies implied considerable costs in terms of lost welfare, and certain institutional arrangements that emerged from the EU integration process (e.g., the Common
Agricultural Policy, or CAP) continue to be expensive to the EU. However,
in its early stages the share of intra-regional EU trade and investment in
economic interaction was far higher than is the case within ASEAN today.
For example, in 1958, the first year of the implementation of the European
customs union, intra-EU trade was about two-fifths of total trade.3 By the
time the customs union was complete, it came to over 50 percent.4 Today,
it stands at about 60 percent.

15

71 550
126 510
39 680
179 674
96 245
4 511
518 170
2 589
540
3 161
25 850
550 310
17.1
24.0
15.5
22.0
18.1
17.2
20.5
3.0
19.4
44.6
11.4
20.2
18.2
25.1
17.2
24.3
22.0
17.2
22.6
7.5
32.8
45.3
13.0
22.2
6.4
6.7
6.7
8.6
7.4
4.5
7.4
1.1
2.1
5.9
13.6
12.2
World ASEAN- ASEAN- CH (%)

(USD mil) 6 (%)
10 (%)
22.3
10.1
20.1
6.4
14.0
38.1
12.2
3.5
1.3
5.2
13.6
12.2
JA (%)
ASEAN exports to selected partners: 2004
55.6
51.3
54.8
53.2
50.4
73.9
52.9
12.5
36.5
58.5
39.5
52.1
APT
(%)
37.8
44.8
35.0
50.7
38.5
35.8
43.9
9.1
35.3
66.2
26.4
43.0
DA
(%)
2.9
3.6
1.3
4.2
2.9
14.1
3.5
0.1
0.1
0.4
7.2
3.6
CER
(%)
12.3
18.8
18.2
13.0
16.1
8.6
15.2
55.9
0.6
0.0
20.1
15.5
US (%)
13.3
19.9
19.2
13.7
17.7
8.7
16.3
60.0
1.9
0.6
21.5
16.6
NAFTA
(%)
12.2
11.8
16.4
13.7
14.3
2.6
13.3
25.5
27.0
15.5
22.5
13.8
EU (%)
Source:
IMF, Direction of Trade Statistics, CD-Rom (April 2000).
Notes: CH 5 China; JA 5 Japan; APT 5 ASEAN-10 plus China, Korea, Japan, and Hong Kong; DA 5 Developing Asia, i.e. all of Asia except
for Japan; CER 5 AustraliaNew Zealand Closer Economic Relations Trade Agreement; NAFTA 5 North American Free Trade Agreement.
Indonesia
Malaysia
Philippines
Singapore
Thailand
Brunei
ASEAN-6
Cambodia
Laos
Myanmar
Vietnam
ASEAN-10
Source
Table 1.1
16
In addition, it is relevant to note that intra-ASEAN trade has been essentially market-driven, rather than being the result of policy-driven discrimination in favor of intra-regional economic interaction. Again, this distinguishes
ASEAN from the EU. Nevertheless, while one could argue that the change
from 37 percent to 50 percent in the case of the EU was in part a result of
discrimination, it is important to consider the relevant historical context:
in the 1950s and early 1960s, Europe was still emerging from the devastation of World War II. Although it would be difficult to assess the economic
reconstruction effect (complications with devising an anti-monde are
many), no doubt the growth over this period was the result of a normalization process, during which time Europe grew rapidly relative to the rest
of the world. Certainly, given the size, wealth, and distance of European
economies, a gravity model would predict high levels of intra-regional trade
even in the absence of EU discrimination. This is not the case for ASEAN.
Even if ASEAN trades much more than one would predict given the gravity
variables noted above (see, for example, Asian Development Bank 2002 and
Frankel 1997), the economic characteristics of ASEAN member countries
would suggest that, while by some definitions it is a natural economic bloc,
its most important trade and investment partners will continue to lie outside
the region, at least in the medium (and probably also the long) run.
In sum, in terms of real integration ASEANs situation is quite different
from that of the EU. Intra-regional trade and investment are considerably
less in the case of ASEAN. The differences in the levels of economic development in ASEAN, whose member countries include among the poorest
and the richest developing economies in the world, are far more significant
than was the case in the EU in its early years, when each EU member
country was a developed economy, or at least was so by the end of the
1950s. One should keep this in mind when trying to draw any lessons from
the EU for ASEAN. Still, we would argue that the EU process has much
to teach ASEAN, both in terms of positive and negative lessons. After all,
no two development experiences are going to be the same. But this does
not suggest that history has nothing to teach us.
1.3
EVOLUTION OF ASEAN ECONOMIC

INTEGRATION IN A REGIONAL CONTEXT
There have been many excellent surveys of regional economic integration

in Asia (e.g., Kawai 2005, Naya 2002, Asian Development Bank 2002).5
ASEAN tends to stand at the core of Asian integration, at least from an
institutional perspective. We provide in this section a brief contextual
review of the evolution of ASEAN in a regional context.

The ASEAN economic community and the European experience
17
Briefly, we would first suggest several factors influencing the regionalism trend in East Asia that stem directly from the Asian financial crisis,
including: (1) the obvious contagion relationships, which demonstrated
the presence of policy externalities across countries in ASEAN and the
newly industrializing economies; (2) major disappointment with the US
reaction to the Asian crisis, which left the feeling of being in it alone
together; (3) disappointing progress in APEC in achieving closer trade
and financial cooperation, as well as development assistance cooperation (ECOTECH); (4) Japans offer to create an Asian Monetary Fund
during the Asian crisis (opposed by the IMF and the United States), which
gave the impression that Japan wanted to be pro-active in the region; (5)
arguably, Chinas decision not to devalue during this period, which may
have also created a sense of solidarity; (6) the New Miyazawa Plan,
launched in October 1998, which dedicated USD 30 billion to help spur
recovery in East Asia (and has been deemed highly successful);6 and (7) the
discrediting of policies promulgated by the IMF to solve the crisis, which
gave greater credibility to the Asian approach.
Hence, the Asian crisis itself set the stage for serious and durable East
Asian regionalism. There are many other internal and external forces
at work that have expedited the process, such as the rise of regionalism
globally and its potential negative effects on the East Asian region; the
successful examples of the Single Market Program in Europe (discussed at
length below) and, eventually, monetary union; general pessimism regarding what can be achieved at the WTO in light of failure to move forward
at the Seattle and Cancun WTO ministerials; and the potential to tap the
inherent benefits of FTAs.
Table 1.2 gives a chronology of arguably the most important Asian initiatives, with a focus on ASEAN and ASEAN Plus Three (ASEAN13),
that is, ASEAN, Japan, China, and South Korea. As many early agreements in ASEANs history were mainly political and token in nature,7 its
first major initiative was the Asian Free Trade Area (AFTA) in 1992. With
the exception of the JapanSingapore FTA (JapanSingapore Economic
Partnership Agreement, or JSEPA), which began implementation over 10
years later, AFTA is the only example of cooperation in Asia that is similar
in concept to the North American Free Trade Agreement (NAFTA).
However, in true ASEAN fashion, rather than overly commit to regional
integration in sensitive areas, the specifics of AFTA were purposefully
left somewhat ambiguous, with the agreement basically committing the
ASEAN members to free trade in a 15-year timeframe. Also, the definition
of free trade was somewhat loose, as it included tariffs in the range of
05 percent, rather than the traditional zero percent.8 After the original
agreement, ASEAN broadened the scope of goods covered by AFTA, and

18
Table 1.2
Chronology of Asian integration: ASEAN and ASEAN13
Main points: ASEAN
ASEAN Year
Summit
1st Bali 1976

ASEAN Concord
Established ASEAN
Secretariat
Treaty of Amity:
Mutual respect
for independence,
sovereignty,
equality, territorial
integrity and identity
of nations, i.e.
noninterference
Establishment of
Zone of Peace,
Freedom, and
Neutrality
2nd
1977
ASEAN Industrial
Kuala
Project
Preferential Trading Lumpur
Agreement (PTA)
3rd
1987
PTA accelerated
Manila
ASEAN Industrial
Joint Venture (AIJV)
accelerated and
made more flexible
ASEAN Free Trade 4th
1992
Area (AFTA)
Singapore
Common Effective
Preferential Tariff
(CEPT)
5th
1995
Bangkok
Proposal for
1st
1996
ASEAN Vision
informal
2020
Jakarta
1997
2nd
ASEAN 2020
presented, a broad informal
Kuala
long-term vision
Lumpur
for ASEAN
in 2020 (with
ASEAN Economic
Community in mind)
ASEAN13 Main points: ASEAN13

Summits
1st Kuala
Lumpur
1st meeting of
ASEAN13

Table 1.2
19
(continued)
Main points: ASEAN
ASEAN Year
Summit
1998
Hanoi Plan of Action 6th
Hanoi
adopted to move
towards Vision 2020:
AFTA advanced
to 2002, 90% intraregional trade subject
to 05% tariff
ASEAN Investment
Area (AIA): goal
investment liberalization within ASEAN
by 2010, outside
ASEAN by 2020
ASEAN Surveillance
Process
Eminent Persons
Group (EPG)
proposed to come up
with plan for ASEAN
Vision 2020
EPG develops plan for 3rd
1999
Vision 2020:
informal
Financial cooperation Manila
proposed over concern
that ASEAN was not
effective in responding
to Asian crisis
AFTA to be sped up
AIA to be accelerated
To respond to surge of
China, need to become
more competitive,
attract investment,
implement faster
integration, and
promote IT
Adopted Initiative for 4th
2000
ASEAN Integration
informal
(IAI):
Singapore
Framework for more
developed ASEAN

Summits
2nd Hanoi
East Asian Vision

Group (EAVG)
proposed by Kim
Dae Jung, President
of Korea, to look
into East Asian
integration
East Asian Study

Group (EASG) to
consider
3rd Manila
4th
Singapore

20
Table 1.2
(continued)
Main points: ASEAN
ASEAN Year
Summit

Summits
members to assist
those less developed
members in need
Focus on factors
needed to enhance
competitiveness
for new economy:
education, skills
development, and
work training
Discussed
7th
challenges facing
Brunei
ASEAN: declining
FDI and erosion of
competitiveness.
Created a roadmap
to achieve ASEAN
integration by 2020
Proposed going
beyond AFTA and
AIA by deepening
market liberalization
for both trade and
investment
2001
5th Brunei
EAFTA and
agreement to hold East
Asian Summit
Two big ideas
proposed:
(1) Development of
institutional link
between Southeast
Asia and East Asia
(2) Study group for
merit of an East
Asian Free Trade
Area (EAFTA)
and investment
area
Financial cooperation
begun, e.g., Chiang
Mai Initiative (CMI)
of May 2000. By
March 2006, bilateral
swap arrangements
under the CMI came
to USD 71.5 billion
Expert group study on
ASEANChina FTA
proposed
EAVG
recommendation for
EAFTA endorsed
but overshadowed by
ChinaASEAN Free
Trade Agreement
proposal within
10 years, with the
adoption of the Early
Harvest Provision to
speed up FTA
Prompted by
ChinaASEAN FTA
proposal, Prime

Table 1.2
(continued)
Main points: ASEAN
AEC end goal of

Vision 2020
ASEAN Year
Summit
8th
Phnom
Penh
2002
9th Bali 2003
Vientiane Action
Plan
Australia attends
for the first time
Source:
21
10th
2004
Vientiane

Summits
Minister Koizumi
proposed Japan
ASEAN Economic
Partnership in
reaction to China
ASEAN proposal
JapanSingapore
Agreement for a
New Age Partnership
signed January 2002
and enforced summer
2002
6th Phnom EASG
Penh
recommendations
of deepening and
broadening of East
Asian integration
adopted. ASEAN13
Asian Bond Markets
Initiative (ABMI)
endorsed by ASEAN13
Deputies Meeting in
December 2002 in
Chiang Mai
9th Bali
At the ASEAN13
Finance Ministers
Meeting on August
7, 2003, support
is extended to the
ABMI as well as to
other means of closer
financial cooperation
10th
China speeds up FTA
Vientiane
with ASEAN from 2015
to 2010
Adapted from Naya and Plummer (2005).

22
the period of implementation has been shortened such that AFTA was
technically in full effect at the beginning of 2004 for the original ASEAN
countries and Brunei, though there are transitional periods for products
on the temporary exclusion lists (e.g., sensitive products such as rice and
automobiles in some cases) and some country-specific implementation
problems in certain areas. The original target for full implementation
for the newer ASEAN members was 2006 for Vietnam, 2008 for Laos
and Myanmar, and 2010 for Cambodia. Recently, ASEAN decided to
speed up the process such that AFTA would be fully completed by 2007.
ASEAN has also made important strides in the area of investment cooperation, such as in the form of ASEAN one-stop investment centers
and the ASEAN Investment Area (AIA).9 These efforts at industrial
cooperation have been designed with essentially the same goal in mind as
AFTA, namely, to reduce transaction costs associated with intra-regional
economic interaction.
As was noted above, in November 2002 the ASEAN Heads of
Government proposed that the region should consider the possibility of
creating an ASEAN Economic Community by 2020.10 This explicitly put
the European experience front and center in terms of design, though clearly
the ASEAN leaders had in mind an economic community with ASEAN
characteristics. The ASEAN leaders actually agreed, at the Bali ASEAN
Summit in October 2003, to create a region in which goods, services,
capital, and skilled labor would flow freely, though the details remain to
be worked out. We offer our own recommendations in this regard, colored
by the EU experience, in the penultimate section.
The reasons behind the decision to create the AEC are many, including:
(1) the desire to create a post-AFTA agenda that would be comprehensive; (2) the perceived need to deepen economic integration in ASEAN
in light of the new international commercial environment, especially the
dominance of FTAs; (3) given (2), the possibility that bilateral FTAs could
actually jeopardize ASEAN integration since all member states were free
to pursue their own commercial-policy agenda; and (4) the recognition
since the Asian crisis that cooperation in the real and financial sectors
must be extended concomitantly, and that free flows of skilled labor will
be necessary to do this.11
In addition to an ebb in progress related to the APEC Bogor Vision
of open trade and investment, there have been several events that have
shifted the ASEAN focus to its East Asian neighbors. First, even with
the successful APEC Summits at Blake Island and Bogor, the East Asian
Economic Grouping (EAEG) concept, proposed as an East Asian trade
bloc by Prime Minister Mahathir in December 1990, never faded away.12
On the contrary, it began to grow in substance. Strangely, the initiative

23
came from ASEANs effort to expand economic cooperation with the EU,
as the EUs desire to deal with all of East Asia led ASEAN to ask China,
South Korea, and Japan to participate. The first AsiaEurope Meeting
(ASEM) was held in Bangkok in March 1996, and officials from ASEAN
and the rest of East Asia met with EU representativesa format that was
regularized and has continued twice a year since. Even though the initial
impetus for these meetings was economic cooperation with the EU, the
significance for East Asian regionalism lies in the fact that these meetings
brought officials from ASEAN, China, South Korea, and Japan together
to discuss issues of economic cooperation. In 1997, these meetings culminated in an informal summit of the ASEAN13 heads of state in Kuala
Lumpur.
The original Miyazawa Plan was initiated by Japan during the Asian
crisis to create an Asian Monetary Fund to supplement the IMF. It was
opposed by the IMF and the United States but eventually led to the establishment of currency swap arrangements among East Asian countries
(basically bilateral swaps between Japan and individual countries) during
the annual meeting of the Asian Development Bank in May 2000, the
Chiang Mai Agreement (CMI).
The CMI is actually a milestone in the financial cooperation of the region
in that it constitutes the first step in the process of financial and monetary
cooperation initiatives, which have become an increasing priority of the
ASEAN13 leaders. Since the CMI, myriad proposals for cooperation
have been put on the table, and a few have actually emerged as promising
new features of Asian regional cooperation. For example, the ASEAN13
Bond Market Initiative (ABMI), which was endorsed by ASEAN officials
at the ASEAN13 deputies meeting in December 2002, endeavors to create
more efficient bond markets in Asia and to reduce the risks associated with
raising international capital.13 A complementary initiative was started by
the Executives Meeting of East Asia-Pacific Central Banks (EMEAP),
which launched the Asian Bond Funds I and II in June 2003 and June 2006,
respectively (see Ma and Remolona, Chapter 4 in this volume).
APECs lack of influence in the Asian financial crisis has served to solidify East Asias move in favor of an ASEAN13 approach. The current spate
of agreements, however, have not been extended to the entire ASEAN13,
but rather have come more from ASEAN to individual countries. For
example, the completion of the ChinaASEAN joint FTA study in the
summer of 2001 prompted Japan to quickly initiate a study of its own with
ASEAN. One month later, at the 2001 ASEAN13 meeting in November,
ASEAN and China announced their intention to negotiate a free trade area
within 10 years (the agreement was formalized in a framework agreement
in December 2004).

24
1.4
LESSONS FROM THE EU
In trying to glean EU lessons for the AEC, we might begin with several
caveats regarding the differences between the subjective environments
facing the EU (EEC) in the 1950s and those facing ASEAN today:
The institutional environment facing ASEAN in the first decade of the

21st century is much different than that facing the EU of the 1950s.
European integration was clearly pushed by both memories of a devastating war and emerging Cold War concerns. The political and social motivations for economic integration were, thus, far different than those driving
ASEAN initiatives today, though, it should be added, ASEAN has been
instrumental in keeping Southeast Asia a peaceful region, an important
contribution that is often underestimated. The European Good is interpreted much differently in Europe than the ASEAN Good in ASEAN;
this puts considerable limitations on institutional development at many
levels. Importantly, it reduces the possibility of the relinquishing of power
by ASEAN member nations to supranational organizations. Besides, such
institutional development is difficult in the ASEAN context anyway, given
that: (1) nation-state formation is much younger than was the case in the
European context, and in some countries this still requires a strong priority; (2) divergences in socio-political institutions are far greater than they
were in the European context, especially as in some European countries
these institutions were being created anew after the war; (3) it is not clear
that European institution-building has been particularly successful in all
areas, though it would receive high marks for economic-related matters
(though this, too, is a testable hypothesis); and (4) these European institutions are quite expensive, while ASEAN government budgets are much
smaller (though, fortunately, ASEAN would not have to employ an army
of translators, as the EU does).
That said, it is important to note that the notion of the ASEAN Good,
though viewed differently in the ASEAN context, has been changing over
the past 10 years. For instance, 10 years ago, few in the region (or the rest
of the world) knew what ASEAN was; today it is well known.
The international economic environment is far different today than it

was in the 1950s.
First, the contemporary global marketplace is extremely open relative

to the past. This is true because of extensive reductions in trade barriers
internationally, as a result of the GATT/WTO rounds as well as unilateral

25
liberalization and huge increases in international capital flows (including

FDI), which have increasingly been knitting an integrated global marketplace. Hence, the costs of using regional integration as a form of fortress, that is, to maximize trade diversion, are consequently much higher
than they were in the past. Second, regionalism has grown by leaps and
bounds recently; trade groupings reported to the WTO come to well over
200, with a majority being established after 1995. Some of these groupings
include ASEANs most important trading partners and could potentially
isolate ASEAN, as well as forcing it to pay the costs of trade diversion.
These trends further underscore the need for the AEC to be open as well
as for the organization to be engaged in the regionalism movement. The
more integrated the ASEAN marketplace is, the easier this will be. These
considerations were far less important in the European context.
ASEAN features far greater diversity in terms of economic

development.
We mentioned in section 1.2 that ASEAN is far more diverse in terms of

levels of economic development than was the case of the EU in its earliest phases, when all countries were developed. While the expansion of the
EU to include 10 Central and Eastern European countries in May 2004
significantly increased diversity within the EU, the region is still dominated
by developed countries and is far more symmetric than ASEAN, which
features developed countries, dynamic Asian economies, middle-income
developing countries, and least-developed countries. As of 2004, the coefficient of variation (standard deviation divided by the mean) on per capita
income levels across ASEAN member countries came to 1.62, compared
to 0.65 for the EU.14 Hence, the divergence within ASEAN is far greater
than that of the EU, and the countries are far poorer. This suggests that
matters related to the speed of implementation of the AEC, and even the
ability of ASEAN to be completely inclusive for all member states, will be
complicated and difficult. Phased 10-X strategies, which is what AFTA
in effect embraces, may be not only desirable but necessary.
ASEAN countries are far more open than was the case of Europe in
the 1950s.
ASEAN countries are (economically) small and very open relative to the
EU of the 1950s (and even with respect to most EU countries today), with
the exception of a few of the transitional CMLV (Cambodia, Myanmar,
Laos, Vietnam) countries. ASEAN countries are closely integrated with
international markets through international trade as well as multinational

26
networks. Not only is this a reality, but it is also a policy focus for ASEAN
governments. As noted in section 1.2, intra-regional trade and investment
in ASEAN is far less important than was the case with the EU and will
likely continue to be so in the future. This is another reason why one would
expect the AEC to embrace openness much more than the EU might have.
In addition, even as an integrated market, ASEAN countries together still
could not influence international terms of trade (the AEC would still be
relatively small), suggesting that the optimal Common External Tariff
would be zero. This was not the case with the EU.
Having noted these caveats, we can delineate at least three major lessons
that can be drawn from the real-side integration experience of the EU.
First, we might begin with a negative lesson: ASEAN should avoid some
of the pitfalls of inward-looking discrimination from which the EU continues to suffer (especially in agriculture) but which would be potentially
catastrophic in the context of the ASEAN countries. Intra-ASEAN trade
is only about one-fourth its global trade (compared to two-thirds in the
case of the EU) and ASEAN member states are highly integrated globally.
Hence, any real-side economic cooperation needs to be outward-looking.
In fact, this approach is exactly what the ASEAN leaders ostensibly have
in mind, that is, using ASEAN as a means of going global. Some scholars
have noted that AFTA is actually more of an investment agreement than
a trade agreement; free trade reduces intra-regional transaction costs and
presents to multinational corporations a vertically integrated market.
The AEC should never lose this vision, even when, as in the European
case, compromises may have to be made. The EU countries are developed,
high-income countries that together form a large economic space. They were
able to push economic integration behind relatively protected markets, in
the context of an international economy that was still fairly closed. Today,
the GATT/WTO has opened up markets considerably and most of the
world, the EU and ASEAN included, have internationalized extensively. It
could be argued that such a protected approach was not necessary to begin
with and should have been avoided (the CAP has been, by many measures,
a disaster); however, the cost of an inward-looking approach has increased
exponentially. It is not a viable option for the AEC.
Second, and partly related to the first, the European experience teaches us
that tradeinvestment links matter and that these relationships are shaping
in large part the economic structure of the ASEAN economies. While the
transitional ASEAN countries are still in early stages of the economic
development process, the original ASEAN countries have experienced
tremendous changes in their productive structures in general and trade in
particular. Primary-based exports (roughly estimated as SITC 04) have

27
fallen in all original ASEAN economies.15 Of the original ASEAN countries only Thailand continues to have a large agricultural-export base (it is,
for example, the largest exporter of rice in the world), but it, too, is falling
in importance. Energy (SITC 3) continues to be important to Indonesia
and Malaysia, with the former being at present a marginal oil importer.
The big change throughout the region has been the impressivein some
cases, spectacularincreases in the share of SITC 7, that is, electronics and
transport equipment (for ASEAN this means mainly electronics). Over the
1990s, the share of SITC 7 increased in all ASEAN countries. Indeed, in
most countries it is the largest export sector; it constituted 58 percent, 41
percent, 72 percent, and 68 percent of total exports in Malaysia, Thailand,
the Philippines, and Singapore, respectively.16
While economic reform has played an important role in this process of
structural adjustment, so has foreign investment. Tamamura (2002) uses
inputoutput analysis to capture the FDIexport link in East Asia, as well
as to decompose the effect of external demand (by country) on production,
using electric/electronics as a case study. He finds that, for 1995 (his latest
year), in every (original) ASEAN country, external demand induced more
production than domestic demand except (marginally) Indonesia, where,
however, domestic demand fell in relative importance from 87 percent to 52
percent. Most countries followed a similar pattern of internationalization of
electronics production. The most extreme case among the ASEAN countries
was Malaysia, where domestic demand induced only 6 percent production.
Next, it is noteworthy that most of the directives that led to the creation
of a tightly integrated market for FDI in Europe came with the Single
European Act, which commenced in 198687 and essentially created what
is mostly a common market by 1994. The European experience teaches us
that accomplishing such a feat goes well beyond mere national treatment/
most-favored-nation treatment in the regional marketplace: economic
cooperation needs to reduce myriad transaction costs associated with FDI,
including those related to the labor market, different product standards,
and the like. The AEC will have to focus per force on many of these areas.
A third lesson relates to how the EU has been able to gain from intraregional trade liberalization, though, as noted above, this could have been
better organized to minimize trade diversion. The customs union played
an important role in building a regional market; the Single European Act,
by creating a Common External Commercial Policy, was able to do much
more by keeping real-side transaction costs within the EU to a minimum
and producing a truly regional marketplace, resulting in a more efficient
division of labor in most markets.
It should be stressed, however, that the AEC should be concerned not
merely with increasing intra-regional but rather with increasing global

28
economic interaction more generally, of which the ASEAN market is

only one partin fact, a part that can be used as an international springboard. Trade and investment integration policies in ASEAN should be
expected to achieve the same general results as they did in the EU case,
but this increased interaction might actually manifest itself in a different
way, given the facts that ASEAN countries are so diverse and that most
are still developing countries. To reiterate: the AEC should be a means of
increasing economic prosperity and the social good rather than focusing
on, say, increases in (sometimes, misleading) indicators such as shares of
intra-regional trade and investment. A successful integration program
could theoretically lead to a decrease in regional integration, as measured
by trade and investment shares, for example.17
A final point would regard the European experience with respect to
financial and monetary cooperation and integration, though in part this
goes beyond the traditional interpretation of the AEC (discussed below).
ASEAN member countries have considered the formation of an ASEAN
bond market, though problems related to liquidity, potential market
depth, and the like have led ASEAN to think more in terms of an Asian
bond market. This will be a long process.18 Nevertheless, it is worthwhile
to consider the European experience as well, given that empirical studies
(e.g., Frankel and Rose 1998) have shown that monetary integration has
strong effects on trade and investment flows.
In the past, just about every regional economic integration program
focused in the beginning almost exclusively on the real side of the economy.
Financial integration was always treated as something separate, to be taken
up at a later date. In many ways, this is less true for European integration,
though the point is debatable. While the European Payments Union was
a financial arrangement, it was only ad hoc and was quickly phased out
as soon as European currencies became convertible. This was just as the
Treaty of Rome actually began implementation. The EC did publish the
Werner Report, which mapped out a plan for monetary union at a time of
great turbulence in the Bretton Woods System (1968), and after the Bretton
Woods System collapsed it tried to create the (short-lived) European
Snake and, eventually, the European Monetary System, which expanded
the Snake in March 1979. These attempts at exchange-rate cooperation
were important because the customs union plus needed stable exchange
rates in order to run well. Such cooperation was especially necessary for
the CAP: the main goal of the CAP was to stabilize farmer incomes, and
flexible exchange rates put this at risk, as a country with a depreciating currency had an advantage over an appreciating-currency country, which was
incompatible with the acquis communitaire. Hence, the EC had to develop
a green exchange rate system, called monetary compensation amounts

29
(MCAs), which prevented this adverse structural change from happening. However, this system was very expensive: Pomfret (1997) suggests that
the MCAs constituted over 15 percent of the CAPs huge budget.
Nevertheless, European capital markets tended to be substantially
segmented until the implementation of the Single European Act was
fairly advanced. There had been early attempts to create a single banking
market as far back as 1972, 15 years after the Treaty of Rome, and in 1977
the European Council established the First Banking Directive (which did
very little to integrate the markets19), but these and other attempts only
marginally integrated the regional markets until the Single European Act
initiatives. Today, the European banking system is far more integrated, but
some aspects of finance continue to be among the few areas in which the
single market is still incomplete. Capital controls were removed as part of
the Single European Act program.
In sum, even in the case of the EU, financial integration did not keep
pace with integration in the real sector. The tendency seems to be to let
financial issues wait, but experience shows that this is an unwise policy.
The Asian crisis might also be seen in this light. Prior to the Asian crisis,
APEC, for example, all but ignored financial and monetary cooperation,
and ASEAN itself did little. In creating the AEC, therefore, ASEAN
leaders would do well to focus on financial issues in tandem with real-sector
integration.
Regarding EU lessons in monetary cooperation, we must again underscore that comparisons are difficult, as relative economic-divergence
problems continue to be critical. Nevertheless, even the EU is a diverse
group, especially if one considers regions rather than countries. Moreover,
ASEANs needs in economic cooperation are obviously quite different
than those of the EU. While ASEAN integration may be popular in the
region, it is less popular than in Europe, particularly among government
leaders. In addition, various EU states had perennial macroeconomic
(especially, fiscal) problems; economic and monetary union allowed these
member states to implement necessary austerity measures in the name of
European integration. The result, after a very long process, has been convergence in terms of interest rates, inflation, and other monetary variables.
Yet, the credibility of most of the original ASEAN countries in terms of
monetary and fiscal policies is actually quite high, especially for developing
countries: inflation tends to be quite low in the original ASEAN countries,
and most countries had either budget surpluses or essentially balanced
budgets prior to the Asian crisis. Today, most have large current-account
surpluses. Nevertheless, there continue to be widely divergent interest-rate
spreads within ASEAN; convergence could have a major impact on development in certain member countries (discussed below).20

30
Hence, neither political nor political-economy dynamics, which were

favorable in the case of most Eurozone countries, could be considered
as important in the case of ASEAN. Nevertheless, in the aftermath of
the Asian crisis, things are changing. It has become clear to ASEAN
leaders that there exist policy externalities; some sort of restrictions
on the conduct of monetary and fiscal policy could not only improve the
macroeconomic environment in the ASEAN countries but also promote
regional economic stability. Moreover, the possibility of competitive
exchange-rate devaluations could be damaging to the implementation
of the AEC. Political arguments for wanting to be part of Europe for
European countries would be replaced in the ASEAN context by a fear of
repeating the economic disaster of the Asian crisis. Such cooperation could
be formally arranged inside or outside of the AEC framework, without any
pretension to the initiatives leading to monetary union.
Based on the EU experience, closer financial and monetary cooperation
in ASEAN could have the following benefits: (1) the necessary Maastrichttype agreements (e.g., restrictions on budget deficits, government debt,
inflation, even foreign-currency exposure of the banking system), perhaps
interpreted more liberally than in the EU context, that would go along with
such cooperation would create a more stable macroeconomic environment
in the region, thereby producing significant positive policy externalities; (2)
as monetary policy would likely be driven by the most credible country/
countries, less credible countries would be able to import credibility,
much as, for example, Italy was able to import German monetary credibility; (3) interest-rate spreads would converge, making it easier to price
risk at the regional level and lower the cost of capital; and (4) the harmonization of rules, accounting standards, and regulatory frameworks that
might accompany regional integration as part of the AEC and in associated financial initiatives would render the region more attractive to foreign
investors, as well as stimulate intra-regional capital flows.21 It would also
make cooperation and even institutional integration of ASEAN equity and
fixed-income markets easier, something that has happened partially in the
EU (e.g., smaller stock markets have integrated while the larger markets
continue to function separately).
The process of financial and monetary cooperation is complicated, and
effective integration demands a steady pace of progress, rather than abrupt
changes, which can actually be counterproductive. The EU process of
financial integration and exchange-rate cooperation, leading up to eventual monetary union, is instructive. The European Currency Unit (ECU)
was a basket of the currencies of the member countries of the EC, weighted
in line with each countrys GDP and foreign trade (and therefore subject
to change periodically). It was introduced in 1979 as part of the European

31
Monetary System (EMS), to be used as the benchmark for determining the

overvaluation/undervaluation of individual currencies and to serve as a unit
of account among the central banks participating in the EMS. No physical
ECU notes or coins ever circulated, so the ECU was strictly an artificial
denomination. However, certain European banks established a banking
product that allowed lenders and borrowers to carry out transactions in
ECU. At first, an ECU transaction was just a portfolio of transactions
in the separate underlying currencies; a deposit or loan in ECU typically
was recorded as separate deposits or loans in the individual currencies.
However, banks soon established a clearing mechanism for the ECU, thus
enabling the transfer of ECU without necessitating separate transactions in
each of the component currencies. This facilitated the growth of the ECU
for private commercial transactions; residents could use the ECU as a unit
of account for bank deposits, and companies could use it for invoicing sales
or maintaining their accounting records. The first ECU-denominated bond
was issued in 1981, just two years after the introduction of the currency
basket. The ECU subsequently became a significant currency denomination in the Eurobond markets, outranked only by the US dollar and the
German mark. A substantial amount of ECU-denominated bonds were
placed privately as well.
The use of the ECU in private transactions developed rapidly because
the ECU exchange rate tended to be more stable than those of its component currencies. For European investors and borrowers, a depreciation of
an individual home currency against other European currencies was offset
by an increase in the home-currency value of the ECU, so there was an
incentive to hold ECUs to diversify a portfolio. Similarly, non-European
investors and borrowers were drawn to the ECU because it was less risky
than the underlying individual currencies. In short, the ECU was an attractive alternative to single foreign currencies because it was less sensitive to
the volatility of a single currency.
On January 1, 1999, the euro replaced the ECU on a one-for-one basis
as part of the first stage of European Monetary Unification (EMU). The
fact that the ECU existed for 20 years prior to EMU suggests that the
simple introduction of a currency basket serves as a useful precursor
to closer monetary cooperation. The success of the ECU was partially
because its official status within the EMS bound the central banks of the
participating countries together. Its success was also due in part to the
fact that the private sector found a pan-European currency denomination
quite useful, and because the banking system was able to accommodate
the demand.
Certainly, the creation of an Asian Currency Unit that would enjoy
widespread use, or the eventual creation of a common currency in Asia, can

32
only be a long-run goal. However, the economics are ostensibly lining up

in favor of such an initiative, and certainly Asian governments have been
placing a higher priority on monetary and financial cooperation. Hence,
perhaps we will not all be dead when this goal is reached.
1.5
ON BUILDING THE ASEAN ECONOMIC

COMMUNITY
Given the tremendous diversity of ASEAN, how will it be able to create

its own customs union plus, even by 2015? Tariff dispersion rates across
ASEAN countries are, indeed, impressive: while ASEAN members tend to
have fairly low tariffs and non-tariff barriers relative to other developing
countries (except for the transitional ASEAN economies), they still vary
considerably across the region. Moreover, Singapore is unique: it essentially has no tariffs. Given the openness of its economy (trade accounts for
over 300 percent of GDP), Singapore cannot raise tariff rates to accept an
ASEAN Common External Tariff that is not equal to zero. The EEC did
not face this problem. Likely options here would include a complete freetrade zone in ASEAN, perhaps with some external tariff harmonization,
or a 10-X customs union, in which the Common External Tariff would
be determined through negotiations similar to those of the EEC, but which
not all ASEAN countries would join.
It is not clear exactly what form the AEC will take. Some scholars have
suggested a less ambitious approach to the AEC, including an FTA-plus
arrangement, which would include certain elements of a common market
(e.g., the free flow of capital, the free flow of skilled labor, and zero tariffs
on intra-regional trade) but would not have a Common External Tariff.
Noting that the European example teaches that without integrated external
tariffs markets continue to be segmented and key benefits of integration are
stymied, Plummer (2005) recommends that a 05 percent Common External
Tariff in an AEC at least be explored for the more developed ASEAN
countries. ASEAN might accept making exceptions in very few industries
that might be integrated later on (this was done in MERCOSURthe
Southern Common Market comprising Argentina, Paraguay, Uruguay,
and Brazilwith automobiles, yielding mixed results). While perhaps
more difficult to implement, this option would have the effect of reducing
transaction costs in the region substantially, mitigating any trade diversion potential of regional integration, increasing the ability of ASEAN to
negotiate integration accords with other trading partners, and augmenting
its clout in international organizations. It could be a critical step in turning
ASEAN into a truly open marketplace.

33
This approach is not really foreign to ideas that ASEAN leaders have
proposed in the past, such as the Philippine proposal to multilateralize
AFTA cuts. Moreover, many ASEAN countries have committed themselves to open trade and investment by the year 2020 as part of the Bogor
Vision of APEC. True, it is unclear exactly how the Bogor Vision will be
achieved, or even what it means: APEC has not completely spelled out the
details, and many ambiguities persist. However, tariffs and non-tariff barriers in ASEAN have been falling over time anyway and will continue to do
so thanks to Uruguay Round commitments, potential commitments under
Doha (if successful), and the liberal posture of the ASEAN leaders.
In this sense, the AEC could be recognized as a purely outward-oriented
endeavor. Fortress ASEAN was never an option. Why not, then, create
an essentially open region? The economic argument for protectionism is
extremely weak, as ASEAN leaders have recognized. Some might continue
to adhere to the infant-industry argument. But this argument has been
more of an excuse for protection than a true means of efficient industrialization in ASEAN and elsewhere. Even the accelerated AEC process
has provided plenty of time for any industry to go through its transition.
Besides, in order to make the infant-industry argument convincing, one
must identify financial bottlenecks that prevent firms from setting up comparative advantage industries. Given the state of financial markets in at
least the original six ASEAN countries, this is not a problem. Moreover,
this open-market solution does not mean that governments would have to
throw away their ability to foster industrialization directly, should they
desire to do so. Regardless of the merits of an active industrial policy, it
is still possible even in an open customs union. This is something that the
European experience clearly shows. Even today, almost a decade after
the completion of the Single European Act and four years after monetary
union, governments still tend to have active industrial policies, for example
through direct subsidies, special financial and tax credits, and even de facto
administrative rules. The EU has formal restrictions on these, but they are
constantly tested (e.g., the EU market in financial services is far from complete). Tariffs have always been a clumsy way to foster industrialization,
and non-tariff barriers tend to be even worse.
Of course, the transitional economies pose an important problem here.
Cambodia, for example, until recently received about 70 percent of its
government income from import-related taxes. However, it is reducing
reliance on international-trade-based taxes as part of its reform program,
and this has also been the case in the other CMLV countries. Vietnam has
made tremendous progress in its transition program and should be ready
to join AFTA in 2006. Allowing the logical progression of this reform
program to continue to 2015 will not be easy but would be quite desirable

34
from an economic development perspective. Again, 2015 is a long way

off, and much can happen between now and then; Vietnam has reinvented
itself from a non-market, closed, and state-directed economy into an
increasingly outward-looking, market-oriented economy in less time than
it will have for the AEC. It may even be possible for ASEAN to allow
for a longer-term transition period for Cambodia, Myanmar, and Laos,
especially since there remain political uncertainties in these countries.
Regarding labor flows, we note that it would be politically difficult to
adopt the Single European Act approach of (technically) free labor mobility. Moreover, this would not be necessary in the ASEAN context, at least
from the point of view of multinationals and of integrating the region with
the global marketplace. Yet, the free flow of skilled labor would be important, as would be the facilitation of visas for non-ASEAN nationals in the
context of a regional framework.
However, the process will be difficult, as it was in the European case.
Mutual recognition of professional qualifications, university and technical
education, and the like will require a great deal of work. Yet, this process
actually presents a good opportunity for the region, and especially for
the CMLV countries, to embrace best practices. It may well be that
the process will be easier for ASEAN than it was for the EU, as fewer
entrenched special interests and less general resistance to reform in this
area are present. Many would welcome this approach.
The idea of adopting best practices also extends to other areas that
were important in the Single European Act, such as product testing, technical standards, food/health-related standards, and the like. Mutual recognition will be necessary in these areas and, hence, harmonization of at least
minimum acceptable standards will have to be developed. Codes should
borrow from internationally accepted standards wherever possible.
Attracting FDI is an important priority among ASEAN leaders. The usefulness of a regional approach has been recognized from the beginning, with
the (generally, failed) attempts at industrial cooperation in the mid-1970s,
the (marginally more successful) initiatives of the late 1980s, and, finally,
the ASEAN Investment Area (AIA) in 1998. The AIA is surprisingly comprehensive; once the exclusion lists are incorporated into the mainstream,
it will have gone a long way toward creating an integrated ASEAN market,
though national policies will have to be increasingly harmonized in order
to create a truly regional market. There is no doubt that FDI will be a high
priority in the AEC and that this vision of an integrated market for FDI
will not be attainable without the transaction-costs-reducing liberalization
and facilitation initiatives under other aspects of the AEC.
Free flow of services will also be necessary, especially since services are
becoming increasingly important in the ASEAN countries, a process that

35
will continue as ASEAN countries develop. The ASEAN Framework

Agreement on Services (AFAS), which takes a GATS-plus approach,
is an important step toward creating an integrated market. Free flow of
services in the AEC will merely involve an expansion in existing commitments since the third round of AFAS negotiations, which began in 2001,
should at least in theory cover all sectors and modes of service provision
defined by the OECD, that is: (1) cross-border supply, in which a company
exports the service from home, for example by fax or email; (2) consumption abroad, in which the user of the service consumes it outside his/her
home country, for example tourism; (3) commercial presence, in which a
company directly supplies the service to foreign customers (this involves
establishment of an affiliate abroad and constitutes over three-fourths of
all trade in services); and (4) the presence of natural persons, in which case
the service-exporting country sends personnel abroad to supply services.
The AEC will ultimately have to ensure a generally open market in services,
including no policy-induced discriminatory restrictions (including trade
taxes), national treatment, mutual recognition, and the like. This was a difficult process in the EU, as some of these sectors remain quite sensitive. For
example, in the financial services area, the Single European Act stipulated
three principles for integration: (1) the observance of specific minimum
requirements; (2) the mutual recognition of member states legislation; and
(3) the prevailing of the home country principle, whereby the regulations
of the country in which business was taking place would take precedence
(rather than those of the host country).22 However, not even the Single
European Act has succeeded in fully integrating the financial services
sector; retail banking services in particular continue to be segmented and
protected on a national basis. Moreover, the Services Directive, which
would serve to create a more integrated market in EU services (particularly
in light of the EU 2004 expansion), was rejected in early 2005.
Hence, as the ASEAN Framework Agreement on Services is expanded
as part of the AEC process, it will be necessary to integrate services sectors
carefully, for the service sector is by its very nature more complicated than
the goods sector. Moreover, the AFAS progress to date has been weak,
and there is a reason for this: certain services are sensitive politically.
Most likely, it will be necessary to exclude certain sectors from complete
liberalization, but these should be kept to a minimum.
Developing appropriate institutions under which the AEC can evolve
will be necessary. While the ASEAN Secretariat has come a long way, it will
have to be enhanced drastically in order to facilitate the creation of the AEC.
It will need to have a much larger professional staff recruited from throughout the region and with a regionalrather than nationalcommitment, as
is the case in the EU. Many of the directorates of the EU could be emulated

36
in the ASEAN context. But it is our view that the bureaucracy should be
kept, to paraphrase Albert Einstein, to the minimum possible but no less
than that. The first reason for this is that the EU bureaucracy is simply too
big and expensive. In the second place, the drain on human capital in the
ASEAN context would be detrimental to other domestic policy priorities,
an important consideration for the CMLV countries in particular. Third,
at least in the first stages of creating the AEC, ASEAN could keep the
social bureaucracies, which are fairly substantial in the EU, somewhat
of a separate project. While these institutions were important in making
the EU what it is today, ASEAN, as noted above, is characterized by a very
different socio-political context. A fourth and related point relates to the
creation of a mini-state in ASEAN, as has been done in the case of the
EU, such as in developing an integrated executive, legislative, and judicial
system. Because the willingness within the EU to develop supranational
institutions is more the exception than the rule, in our view ASEAN should
try to minimize the supranational character of AEC, taking the idea of
subsidiarity to the greatest extent possible. The executive component
of ASEAN integration would have to be enhanced considerably, but this
could arguably be done by adapting and expanding current institutions. On
the other hand, the creation of some sort of judicial authority to enforce
(hitherto a bad word in ASEAN) AEC rules will be necessary. No doubt
this will be difficult; the EU continues to have its own problems (e.g., the
Alstom case in France is a good example, but there are many more). As in
the case of the EU, it would have to be an evolutionary process.
The Asian crisis has underscored the need for greater monetary and
financial cooperation, given the importance of these areas to future regional
economic development and their importance in supporting initiatives in
the real sector. This differentiates it to some degree from the EU experience, in which financial/monetary cooperation came later. Moreover, the
fact that ASEANs trade and investment links are dominated by partners
outside the grouping would suggest that the recent widening of cooperation outside of strictly ASEAN-based institutions to include China, Japan,
and South Korea gives it a much greater incentive to promote cooperation
in the area of financial and monetary integration, no doubt a reason for
the popularity of the ASEAN13 initiatives.
1.6
CONCLUDING REMARKS
In this chapter, we have tried to consider what the objectives and substance
of the AEC should be, using wherever possible appropriate lessons from
the worlds most successful example of regional economic integration,

37
the European Union. We note that while there is much that the EU can
teach ASEAN, ASEAN leaders should not underestimate the differences
between the regions and the differing historical contexts.
The EU integration experience is remarkable. It took a great deal of
time before it became a truly integrated marketabout 37 years from the
Treaty of Rome in 1957 until the implementation of the Single European
Act, which was essentially complete in 1994. Once the process was given
a big push in the mid-1980s, however, integration initiatives picked up
steam, culminating in monetary union only five years after the promulgation of the Single European Act.
At times, some leaders and experts gave up on the EU; the process certainly was familiar with crisis. In 1976, for example, France (temporarily)
slapped import tariffs on Italian wine. In the early 1980s, market segmentation increased with the use of non-tariff barriers outside the purview of the
EC, leading some to suggest that the EC was doomed to retreat. After the
September 1992 crisis in the EMS, it was very easy to be pessimistic about
the future of monetary union. There were skeptics up to the very the end.
But the EU was able to persevere because of the commitment of its
leaders and critical social elements. This is a very basic lesson: given the
fact that the AEC will have to be far more comprehensive and intrusive
in national markets than has ever been the case before, it will take strong
commitment, indeed, in order to move the process forward.
No doubt this is why there is much skepticism regarding the AEC. It
was no different in the case of AFTA: in the late 1980s, many pundits
were speculating that since the regions political exigencies had changed,
ASEAN had no future as a regional organization. Instead, the ASEAN
leaders responded by pushing forward impressively on the economic front,
and AFTA became the first major initiative in this process. Since then,
AFTA has expanded and deepened, cooperation has advanced significantly in the area of investment (AIA), liberalization of services is being
actively pursued in the ASEAN Framework Agreement on Services, other
deepening measures are being spearheaded, and horizontal integration
has expanded about as far as it can go, as ASEAN is now composed of all
10 Southeast Asian nations. While the AEC will take a much more extensive commitment, it can certainly be realized if the ASEAN leaders have
the political will to see it through.
NOTES
1.
An earlier version of this chapter was presented at the Inaugural Session of the
Regional Integration Seminar Series, Office of Regional Economic Integration, Asian
Development Bank, on November 28, 2005, in Manila, the Philippines (and was

38
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.

published in 2006 as one of its working papers), at the joint session of the American
Economic Association/American Committee on Asian Economic Studies (and is published along with the rest of the sessions papers in Vol. 17 (3) of the Journal of Asian
Economics), January 2006, as well as at the December 2006 HWWA/HWWI conference
on East Asian Monetary and Financial Integration in Hamburg, Germany. The authors
would like to acknowledge interesting inputs offered at these meetings, especially those
by Drs Kawai, Rana, and Capannelli. All remaining errors are those of the authors
alone.
We have a difficult acronym issue in this chapter: the European Union (EU) was the
successor to the European Community (EC), which integrated the various economic
communities in Western Europe, including the European Economic Community (EEC)
and the European Coal and Steel Community (ECSC). We use EU throughout this
chapter for simplicity, but we are actually referring here to the EEC.
Hiemenz et al. (1994), Table 2, p. 8.
Ibid.
This brief review of ASEAN integration borrows from Plummer (2006) and Naya and
Plummer (2005).
Kawai (2005).
For example, the Preferential Trading Agreement was a positive-list approach to trade
liberalization with small margins of preference and limited product coverage, expanded
somewhat during the 1980s but with no real impact on trade. Industrial cooperation,
such as the ASEAN Industrial Project system, never really got off the ground.
In fact, this range of tariffs probably contradicts the requirements spelled out in Article
XXIV of the GATT/WTO, but as was noted earlier ASEAN benefits from the Enabling
Clause, which has always freed it from these constraints.
A salient component of the AIA is the ASEAN Industrial Cooperation (AICO)
Scheme, which offers more in terms of tariff (05 percent) and non-tariff incentives
than the traditional industrial cooperation programs. Moreover, in September 1996 the
ASEAN countries created the ASEAN Agreement for the Promotion and Protection
of Investments, which among other things stipulates the simplification of investment
procedures and approval processes, as well as enhanced transparency and predictability
of FDI laws.
At the 11th ASEAN Summit in Kuala Lumpur, December 2005, the ASEAN leaders
discussed the possibility of expediting the AEC process to actually complete it by 2015
(with flexibility for new member countries). This was agreed upon at ASEANs annual
meeting in January 2007.
The free flow of all labor, including unskilled labor, was deemed too politically difficult
to consider in the AEC.
When proposed, the EAEG was highly controversial, as it suggested that Mahathir
wanted to create what might have been construed as a discriminatory bloc, which ran
contrary to the goals of ASEAN and, of course, APEC. At the 4th ASEAN Summit in
Singapore in 1992, it was downgraded to the more innocuous (and vague) East Asian
Economic Caucus (EAEC).
For more details on this initiative, see the ADBs Asian Bond Markets Initiative
homepage, http://aric.adb.org/asianbond/ASEAN-ABMI.htm.
Authors calculations, using data on GDP per capita from World Development
Indicators On-Line. In 2004, average per capita income in the 10 ASEAN countries and
the 25 EU countries came to USD 5271 and USD 18,984, respectively.
Data for this structural-change analysis come from Plummer (2003).
The case of the Philippines is the most dramatic and surprising: the value of SITC 7
exports increased over this period by over 100 percent, with the largest changes in SITC
723 (civil engineering and contractors plant and parts), SITC 728 (machine & specialized equipment), 736 (machine tools), 751 (office machines), and 752 (automatic data
processing machines).
This is because, for example, a successful AEC that brings in higher FDI flows from
abroada key aim of the AIAwould not only reduce intra-regional FDI but also

18.
19.
20.
21.
22.
39
could reduce intra-regional trade, if multinationals take advantage of the attractive

regional division of labor offered by ASEAN. For example, suppose that, as a result of
the AIA, a Japanese automobile multinational set up production stages in Indonesia
and Singapore, whereby it exports USD 2 billion in car components to Indonesia, adds
USD 100 million in labor-intensive value added to production in Indonesia, exports the
semi-processed product to Singapore for further USD 1 billion in processing, and then
finally exports back to Japan. This means that ASEAN intra-regional trade would have
changed at the margin by exports to Singapore from Indonesia (USD 1.1 billion) divided
by exports of Japan to Indonesia (USD 1 billion) plus imports of Japan from Singapore
(USD 2.1 billion), or 35 percent, whereas extra-regional trade would have increased by
65 percent. The point is that this could be a successful economic activity for all parties
involved, but intra-regional trade shares might fall anyway.
See, for example, Plummer and Click (2005).
Story and Walter (1997) note (p. 14) that of the EUs 9434 credit institutions at that
time, 429 were classified as foreign banks, and only 107 had a parent company based in
a member state. Governments were reluctant to grant licenses.
See, for example, Plummer and Click (2005).
For a more complete discussion of the potential benefits and costs of financial market
cooperation and integration in ASEAN, with comparisons to Europe, see Plummer and
Click (2005) and Click and Plummer (2005).
Story and Walter (1997).
REFERENCES
Asian Development Bank (2002). Asian Development Outlook 2002. Manila: Asian
Development Bank.
Click, Reid W., and Michael G. Plummer (2005). Stock Market Integration in
ASEAN. Journal of Asian Economics 16: 528.
Frankel, Jeffrey (1997). Regional Trading Blocs in the World Trading System.
Washington, DC: Institute for International Economics.
Frankel, Jeffrey, and Andrew K. Rose (1998). The Endogeneity of the Optimum
Currency Area Criteria. Economic Journal 108: 100925.
Hiemenz, Ulrich, Erich Gundlach, Rolf J. Langhammer, and Peter Nunnenkamp
(1994). Regional Integration in Europe and Its Effects on Developing Countries.
Kieler Studien 260. Tbingen: J.C.B. Mohr.
Institute of Southeast Asian Studies (2003). Concept Paper on the ASEAN
Economic Community. Manuscript, Institute of Southeast Asian Studies,
Singapore.
Kawai, Masahiro (2005). East Asian Economic Regionalism: Progress and
Challenges. Journal of Asian Economics 16: 2955.
Naya, Seiji F. (2002). The Asian Development Experience. Manila: Asian
Development Bank.
Naya, Seiji F., and Michael G. Plummer (2005). The Economics of the Enterprise for
ASEAN Initiative. Singapore: Institute of Southeast Asian Studies.
Plummer, Michael G. (2002). EU and ASEAN: Real Integration and Lessons in
Financial Cooperation. World Economy 25: 1469500.
Plummer, Michael G. (2003). Structural Change in a Globalized Asia: Macro
Trends and U.S. Policy Challenges. Journal of Asian Economics 14: 24381.
Plummer, Michael G. (2005). Creating an ASEAN Economic Community:
Lessons from the EU and Reflections on the Roadmap. In Denis Hew (ed.),

40
Roadmap to an ASEAN Economic Community. Singapore: Institute of Southeast

Asian Studies, pp. 3162.
Plummer, Michael G. (2006). Toward WinWin Regionalism in Asia: Issues and
Challenges in Forming Efficient Trade Agreements. In Working Papers on
Regional Economic Integration. Manila: Asian Development Bank.
Plummer, Michael G., and Reid W. Click (2005). Bond Market Development
and Integration in ASEAN. International Journal of Finance and Economics
10: 13342.
Pomfret, Richard (1997). The Economics of Regional Trading Arrangements.
Oxford: Oxford University Press.
Story, Jonathan, and Ingo Walter (1997). Political Economy of Financial Integration
in Europe: The Battle of the Systems. Cambridge, MA: MIT Press.
Tamamura, Chiharu (2002). Structural Changes in International Industrial
Linkages and Export Competitiveness in the Asia-Pacific Region. ASEAN
Economic Bulletin 19: 5282.

2.
The political economy of European

economic and monetary union
negotiations and implications for
East Asia
Heungchong Kim1
2.1
INTRODUCTION
The financial crisis of 199798 in East Asia provided a valuablethough

costlylesson for the East Asian countries in the importance of regional
cooperation. In the course of the developing crisis, East Asian countries
observed that neighboring countries were not at all helpful because of
the lack of cooperative facilities within the region, and they were shocked
by the posture of mere observance adopted by the advanced countries.
The East Asian countries realized that the remarkable economic growth
they had achieved over the years was based on a house of cards. After
the crisis had passed, various kinds of proposals for regional economic
cooperation, to strengthen economic stability as much as economic
growth, were discussed and developed in East Asia, including the Chiang
Mai Initiative, a burgeoning number of regional free trade agreements
(FTAs), potential East Asian bond markets, and so on.
While trying to develop proper institutions and tools for regional
economic cooperation, it is strongly recommended that East Asia take
an interest in the European economic integration process, investigate
Europes past experiences, and creatively adapt them to East Asias circumstances and needs. The creation of Economic and Monetary Union
(EMU) in Europe with the introduction of the euro is a good example
of economic and monetary integration that East Asia could follow. An
examination of the competing views among the European Union (EU)
member states on the path to monetary integration and how they were subsequently resolved provides important insights for East Asian countries
that have just launched the early stages of monetary cooperation.
This chapter attempts to prove this kind of examination. It analyzes
the process of EMU negotiations, mainly during the crucial 1988 to
41

42
1991 period, in order to derive lessons for potential East Asian monetary
cooperation and integration. It excogitates major issues from the EMU
negotiations, compares the views of major member states on each issue,
and clarifies the process of reaching agreements from the different negotiative positions. The chapter is organized as follows: section 2.2 introduces
a brief history of European monetary integration, including the European
Monetary System; section 2.3 clarifies some points regarding the formation of EMU and evaluates the current developments in monetary integration in East Asia; section 2.4 illuminates the roles played by the major
nations in the EMU negotiation process; and section 2.5 applies the roles
revealed in that process to the case of East Asia. Section 2.6 concludes the
chapter.
2.2
A BRIEF HISTORY OF EUROPEAN MONETARY

INTEGRATION
The idea of a united Europe was once merely a dream conceived of

by philosophers and visionaries such as Victor Hugo, who imagined a
peaceful United States of Europe inspired by humanistic ideals. But
the dream was unable to come to fruition because of centuries of war
that raged on the European continent. However, the terrible experience
of the two World Wars increased the determination to end hatred and
rivalry in Europe and build lasting peace between former enemies. The
architects of modern day Europe, such as Jean Monnet, Robert Schuman,
Konrad Adenauer, and Alcide de Gasperi, recognized that a process of
European integration could provide such a lasting peace in Europe and
that it could only be achieved gradually by starting with concrete economic and political issues that would have spillover effects in the creation
of supranational institutions. This was termed the functional approach
toward integration.
The motivation for establishing EMU was partly political and partly
economic in nature. One impetus was to facilitate trade between the
EU countries by eliminating currency exchange costs and uncertainties
relating to exchange rates. Another motivating factor was the belief
that greater integration between the EU countries economies would
facilitate the pursuit of common objectives such as a strong growth
rate and a high level of employment. However, in a Europe recently
ravaged by two World Wars, the ultimate objective was to promote
closer European integration in order to avoid future conflicts between
former enemies.

European EMU negotiations and implications for East Asia
43
The Bretton Woods Era and its Aftermath

Despite the fact that the idea of a common European currency began to be
discussed in the 1960s, it would take more than 30 years before that vision
would materialize. After the Second World War, countries with market
economies adopted the Bretton Woods system in order to promote currency stability. The Bretton Woods system was based on a system of fixed
but adjustable parities with the US dollar, and it worked well before it faced
severe problems in the late 1960s when the credibility of the US dollar was
repeatedly damaged. In the European currency market, the devaluation of
the French franc and the appreciation of the German mark threatened the
stability of the other currencies in the system and jeopardized the system of
common prices set up by the Common Agricultural Policy in 196869.
In light of these developments, the Barre report of 1969 called for greater
coordination of economic policies and closer monetary cooperation. The
report gave new momentum to the integration process. In December 1969,
at the Hague Summit, the European heads of state or government reached
an agreement to make economic and monetary union an official objective of European integration. Furthermore, a high-level group chaired by
Pierre Werner, then prime minister of Luxembourg, was entrusted with
the responsibility of drafting a report on how the goal of economic and
monetary union could be achieved by 1980.
The Werner Report, submitted in October 1970, envisioned the conclusion of a fully integrated economic and monetary union within 10 years
through the implementation of a three-stage plan. The end goal was to
accomplish full liberalization of capital movements, the irrevocable fixing
of parities, and the replacement of national currencies with a single currency. The report put forward the idea of establishing a European Monetary
Cooperation Fund in the first stage as a precursor to a community system
of European central banks in the final stage. Despite disagreement over
some of the key recommendations of the Werner Report, in March 1971
the member countries gave their approval to pursue monetary integration
in three stages. The first stage required the narrowing of currency fluctuation margins and was implemented according to the plan.2 Nevertheless,
new developments stalled the progress: the decision of the US government
to suspend the gold convertibility of the dollar in August 1971 and the subsequent breakdown of the Bretton Woods system had a destabilizing effect
on foreign exchange rates world-wide and also raised questions about the
parities between the European currencies. Consequently, an attempt was
made in March 1972 to provide intra-regional exchange rate stability by
forming the Snake in the Tunnel,3 but the oil crises, the weakness of the
dollar, and differences in economic policies caused members to back out of

44
the Snake until membership was reduced to Germany, the Benelux countries,4 and Denmark, with the Benelux currencies and the Danish krone
tied to the German mark.
The European Monetary System
In March 1979, President Valery Giscard dEstaing of France and
Chancellor Helmut Schmidt of Germany proposed the European Monetary
System (EMS), an institutional framework that would replace the Snake
with a new system of adjustable exchange rates. The EMS was intended
to lay the framework for the creation of closer monetary cooperation
leading to a zone of monetary stability in Europe. The EMS was built on
three pillars: (1) the Exchange Rate Mechanism (ERM), (2) the European
Currency Unit (ECU), and (3) the credit mechanisms of financing facilities. All of the member states of the European Community joined the
EMS, with the exception of the United Kingdom, which opted out of
the Exchange Rate Mechanism. In the period 197991, the EMS helped
stabilize exchange rate variability, and sustainable currency stability was
achieved thanks to the flexibility of the system and the political resolve to
attain economic convergence.
The ERM was one of the cornerstones of economic and monetary union.
The ERM was conceived in an effort to stabilize exchange rates, encourage trade within Europe, and control inflation. It arose as a result of a
compromise between France, which wanted the exchange rate band to be
defined around the ECU, and Germany, which wanted the exchange rate
band to be defined bilaterally. Thus, each participating countrys currency
was allowed to fluctuate within a band defined around each of its bilateral
central rates and another band defined around its ECU central rate.5 As
the EMU process advanced, the ability of each country to keep its currency
within its defined margins became one of the convergence criteria that had
to be met in order to be eligible to join the single currency.
To ensure that each member country had the necessary resources to
intervene in defense of the bilateral exchange rate parities, extensive
financing mechanisms were created. In April 1973, the member states of
the European Community decided to launch the European Monetary
Cooperation Fund (EMCF) in order to facilitate the establishment of an
economic and monetary union. The idea for setting up a fund was first
proposed by the heads of state or government at the Hague Summit in
December 1969, and it was later included in the recommendations of the
Werner Report. Within the EMS, the EMCF was responsible for facilitating the gradual narrowing of the margins of the community currencies
against one another, the administration of the Short-Term Monetary

45
Support Facility established as a result of an agreement between the communitys central banks, and the multilateralization of positions in the
Very Short-Term Financing (VSTF) Facility that arose from intervention
carried out by the central banks in community currencies. The establishment of the EMCF resulted in the administration by the Fund of all bilateral debts and credits held by community central banks in regards to the
VSTF facility of the Snake arrangement.
The adoption of the Single European Act (SEA) in the mid-1980s
allowed the community to fulfill its objective of removing impediments
to the free movement of goods, services, capital, and labor. However, the
Single Market Program also highlighted the existing problems of high
transaction costs related to currency conversions and uncertainties associated with exchange rate fluctuations, both of which hampered the exploitation of the full potential of the internal market. Thus, for federalists and
those adopting a functional approach toward integration, the next step
after the SEA would definitely be toward a monetary union.
EMU
The Hanover European Council set up a committee in June 1988, chaired
by Jacques Delors, then president of the European Commission, to draft
a proposal for an economic and monetary union. The Delors Report,
submitted in April 1989, proposed a three stage plan, with the execution
of Stage I bringing about closer monetary and economic cooperation and
with the deadline for implementation set for July 1, 1990 at the latest, to
coincide with the implementation of the full liberalization of capital flows
within the community. It also emphasized the need for greater coordination of economic policies, rules on the size and financing of national budget
deficits, and a new, completely independent central bank, which would be
responsible for the unions monetary policy. On the basis of the Delors
Report, in June 1989 the European Council agreed to begin Stage I on the
proposed date and reaffirmed the commitment to reach an economic and
monetary union in subsequent steps.6
In December 1989, the Strasbourg European Council called for an
Intergovernmental Conference (IGC) that would help identify the amendments that needed to be made to the Treaty in order to accomplish
economic and monetary union. The efforts of the aforementioned IGC
together with that of a second IGC on political union7 resulted in the
Treaty on European Union, which was formally adopted by the heads of
state or government at the Maastricht European Council in December
1991 and signed on February 7, 1992. The Treaty set out a timetable for
economic and monetary convergence in three successive stages.

46
During the first stage, which began on July 1, 1990, the European
Council was responsible for evaluating progress made with regard to economic and monetary convergence, and member states would be asked to
adopt certain measures to comply with certain restrictions specified by the
Treaty.8 Stage I did not precede exactly according to the plan because of
currency tensions during the period from September 1992 to August 1993.
Nonetheless, the transition to Stage II occurred according to the timeline
set forth by the Treaty. Stage II called for the completion of a convergence play9 of a majority of the member states and the establishment of
an operational framework for a common central bank with a single monetary policy. Thus, at the beginning of Stage II, the European Monetary
Institute (EMI) was established as a precursor to the European Central
Bank (ECB). The EMI was responsible for ensuring a smooth transition
to Stage III by strengthening the coordination of monetary policies of the
member states with the goal of attaining price stability, making necessary preparations for the establishment of the ECB, and overseeing the
progress of the ECU, including ensuring the smooth functioning of the
ECU clearing system.
Stage III started on January 1, 1999, and it saw the establishment of a
single monetary policy that was overseen by the ECB, which took over
responsibility for monetary policy from the national central banks. At
this stage the euro became the common currency of the 11 members of
the monetary union (Greece only joined in 2001). Furthermore, the ERM
was transformed into ERM2, the exchange rate arrangement linking
currencies of EU countries outside the monetary union to the euro. The
crowning achievement came on January 1, 2002, when euro notes and coins
were introduced in all EMU member countries. For a specified time, the
national notes and coins were legal tender alongside the euro. On March
1, 2002, the euro became the sole legal tender in the countries participating
in the monetary union.
2.3
REFLECTIONS ON THE FORMATION OF EMU

AND EAST ASIAN MONETARY INTEGRATION
As mentioned earlier, the founding fathers of modern Europe recognized

that European integration could only be achieved step by step, by starting
out with concrete economic and political issues that would have spillover
effects leading to the creation of supranational institutions. The integration
process was an evolutionary one, which initially began with real economic
cooperation between members. As countries began to realize the gains
from real economic cooperation, they came to the understanding that

47
they would also need to establish monetary cooperation with each other in
order to fully enjoy these gains.
European monetary integration can be described as a process of trial and
error. It is important to realize that monetary and financial integration was
not an explicit aim in the first postwar initiatives. After the creation of the
Bretton Woods system, European currencies were embedded in the worldwide system of fixed exchange rates, giving little incentive for European
countries to create their own active exchange rate policies (Reszat 2003).
It was the problems with, and the eventual collapse of, the Bretton Woods
system that provided an impetus for the project of monetary union, as formalized by policymakers in the Barre and Werner Reports. After the first
initiative of intra-regional exchange rate stabilization, the Snake, failed to
provide the desired sustainable currency stability, European leaders had
to conceive new ways of stabilizing exchange rates. Eventually, based on
the success of the EMS and the refinement of the process set out in the
Delors Report, the plan for implementing EMU was drafted, with timelines
created for the initiatives. However, as we have seen, the different stages of
advancement were not without problems, and compromises and exceptions
had to be made before the end goal, a common currency, could be reached.
East Asia can learn from the strategies employed in the course of the EMU
negotiation process and recognize the importance of sticking to principles
while at the same time allowing exceptions. For example, exceptions were
made in the case of the United Kingdom and Denmark, allowing them
clauses for exemption from the third stage of EMU negotiations. The protocol was necessary in order to secure the United Kingdoms agreement to
the Treaty.
During the process of Treaty ratification, speculation caused by the
rejection of the first Danish referendum in June 1992 and uncertainty surrounding the French referendum in September 1992 resulted in speculative monetary turbulence and compelled Italy and the United Kingdom
to withdraw their currencies from the ERM. In the summer of 1993, the
French franc, too, came under strong pressure. The crisis cast doubts on
the feasibility of EMU.10 Despite these setbacks, the preparations went
ahead and the set timetables were respected. The important lesson to be
learnt here is that when countries have a common goal (such as a monetary
union) they can work out their differences and overcome impediments that
threaten progress.
Consider now the case of the East Asian integration process. East Asia
has greatly benefited from the globalization of financial markets. From the
middle of the 1980s onwards, the dynamic economic growth of the region
was fuelled by huge capital inflows. However, the East Asian financial crisis
of 1997 triggered huge capital outflows from the region and caused great

48
damage to its economies in the form of a sharp depreciation of exchange

rates, steep falls in asset prices, collapses in investment and consumer confidence, and a sharp fall in income. The financial crisis acted as a wake-up
call to East Asian countries, clearly showing that their capital markets
were not adequately equipped to deal with globalized financial flows. In
the post-crisis period, there was growing awareness among the East Asian
countries of a need to collectively respond to such situations in the future
because of their growing regional integration, as evidenced by their mutual
susceptibility to contagion (Wang and Andersen 2002).
But East Asia faces several obstacles to monetary integration. One major
hindrance is the areas lack of experience with regard to regionalism. The
East Asian countries have very little experience of macroeconomic policy
coordination and have not developed any institutions to implement it.11
Also, huge gaps in financial expertise, technology, infrastructure, and
market size between the East Asian emerging markets and the international financial centers in developed countries have led to the financial
services in the region being provided mostly by foreign corporations,
often in New York or London. This could in turn act as a barrier against
sufficient expertise being developed in the region and also put constraints
on the extent to which regional identification develops as a driving force
(Wang and Andersen 2002).
Moreover, the two largest economies in the region, China and Japan,
differ considerably in their strategies. Japan, the most developed country
in the region with a strong international currency, is confronted with
skepticism by the other East Asian countries and especially China, which
is not willing to let Japan play the lead part in regional integration. Also,
as a result of Japanese commercial banks having been driven out of
the financial services markets because of prolonged financial problems,
the yen is declining in importance, thus making the role of Japan as a
leader in the creation of an East Asian economic entity seem increasingly
unlikely. China, on the other hand, despite its fast growth and expanding
trade relations with its neighbors including ASEAN, is presently lacking
deep financial markets and an international currency that would enable
it to act as the center country of East Asian monetary integration.12 All
of these factors will make it increasingly difficult to reach a consensus
on economic and financial integration in an increasingly bipolar East
Asian region. While an economic upturn in Japan and more intimate
Sino-Japanese relations might help in fostering regional cooperation,
nonetheless, the greater differences in governments, legal and regulatory
environments, levels of economic development, and local histories than
was the case with Europe suggest that the integration process will take
considerable time.

49
Finally, one of the driving forces behind European integration was the
desire for a united Europe. This idea of a common citizenship is lacking
in East Asia. In order to reap the benefits of a monetary union, countries will have to cooperate in the political arena as well, and East Asian
leaders will have to build a common framework in which decisions are
reached based on what is best for East Asia. At present, they do not show
the same ambitions that European leaders exhibited half a century ago.
Furthermore, East Asia needs to develop ways of interacting that promote
collaboration rather than competition between countries. In light of all
these obstacles, we can easily see that East Asia has a long way to go
before regional integration can be accomplished.
2.4
THE MAIN PLAYERS IN EUROPES EMU

NEGOTIATIONS
East Asia still has a lot of work to do before it can hope to establish an
economic and monetary union. As East Asia differs considerably from
Europe and cannot simply emulate its experience with monetary union,
it will have to chart its own course toward integration. We nevertheless
believe that the process of EMU in Europe can provide important lessons
and models for different types of possible coordination and cooperation. In
order to derive such implications for monetary cooperation in East Asia,
we investigate the roles of the major players in European economic and
monetary integrationGermany, France, the United Kingdom, Benelux,
and the European Commissionin negotiating EMU.
Germany
The joint initiative of Chancellor Helmut Kohl of Germany and President
Francois Mitterrand of France resulted in a great leap toward the establishment of EMU during their years in power in the 1980s and 1990s.
They were convinced that Europe could only be united on the basis of
Franco-German reconciliation, with both countries acting together as a
locomotive driving the process of European integration. This shared belief
contributed a lot to the success of EMU negotiations by preserving key
bilateral relations between Germany and France.
During his 16 years in power (198298), Kohl presided over Germanys
unification and became the driving force behind European integration.
Kohls early years in power were marked by Germanys growing pains
from unemployment and a deep national divide over the deployment of
nuclear weapons on German territory. In 1989, the collapse of the East

50
German regime and the fall of the Berlin Wall allowed Kohl to grab the
mantle of history. He became the main influence behind German unification. By skillfully balancing pressure and persuasion, he convinced
leaders in Eastern and Western Europe as well as the Soviet Union and
the US to accept a large and unified Germany, ending half a century of
Cold War division. In Kohls view, German unity and European unity
were two sides of the same coin. Tying Germany into the double framework of EU and NATO, Kohl hoped to avoid a replay of great power
rivalries.
However, the triumph of national unification was soon dampened by
a string of problems, not only because of structural problems with the
European economy but also because of the costs and consequences of
unification itself. Like a majority of European countries in the 1990s,
Germany faced increased global competition, rising welfare costs, and
high unemployment, particularly in its traditional industrial sector. No
less importantly, it also faced the overwhelming added expenses of unifying
East and West Germany. As a result of a legacy of inefficiency, the former
East German economy collapsed, hundreds of thousands of East Germans
faced unemployment, and East Germany became heavily dependent on
federal subsidies.
In a sense, Germanys EMU participation added another burden to the
economic challenge of German reunification. Within the ERM regime, the
German mark was the only currency that had never been devalued and
acquired great credibility and a reputation as a hard currency. In order
to maintain parity with the mark, other countries had to adopt domestic
monetary policies as disciplined as Germanys. This caused asymmetry
within the functioning of the ERM, with the burden of adjustment falling
disproportionately on other countries. Thus, the West German people
were generally opposed to entering into EMU because they feared the loss
of their monetary supremacy. There were also concerns that entering a
monetary union with less disciplined countries would result in a European
currency potentially weaker as compared with the mark. To soothe these
concerns, the German government, as well as the Bundesbank, demanded
many concessions from the other European countries in order to ensure a
stable European currency. For instance, the German negotiators insisted
on the independence of the common central bank and pushed through the
Stability and Growth Pact to ensure that monetary stability would not be
undermined by the fiscal recklessness of individual member countries.
Over the course of EMU negotiations, Germany had allies in the
Netherlands, Denmark, and Luxembourg (and sometimes the United
Kingdom), all of which supported the economist approach. The
economist approach to monetary integration stressed that monetary

51
union must be the end result of a long process of economic convergence,

following the freeing and opening of markets and, not least, the complete
liberalization of capital movements (Stoltenberg 1988). These strategic
allies greatly helped the German view dominate during the planning
stages of EMU.
France
France, under the leadership of Francois Mitterrand, played a great role
in the advancement of EMU. Mitterrands strategy on EMUbalancing
domestic interests against international and European interestsreflected
those of former French presidents Charles de Gaulle, Georges Pompidou,
and Valry Giscard dEstaing. They all envisaged an EMU that would
counter the monetary hegemony of the US with a common European stance
on international monetary reform. At the same time, it would deal with
German monetary power by Europeanizing it (Dyson and Featherstone
1999). Thus, with the fall of the Berlin Wall in 1989, Mitterrand pushed
for the goal of an EMU with a single currency in order to deepen German
integration within Europe.
Some of the notable influences on Mitterrands policy on EMU were a
belief in the unique French role in Europe as the birthplace of democratic
values, the desire to be disassociated from the fiscal irresponsibility that
was linked to the political left, and memories of French humiliation at
German hands, this last factor being the single most important one in the
decision to forge closer ties with Germany to ensure peace between the
two countries. The vested interest that the French had in European integration also manifested itself in the idea that the French had ownership of
European integration. The French felt that they could play a special civilizing role in the world. While Germanys contribution to the EMU would
be economic, the French would contribute through their art, language,
and culture.
In contrast to the economist approach favored by Germany, France took
the so-called monetarist approach, as did Italy, Belgium, and the European
Commission.13 The monetarists believed that the way to achieve economic
convergence was through the establishment of a new monetary institution
that would in turn change market behavior. This argument was consistent
with the community method approach to European integration, which
stressed the role of elite socialization in new institutional structures as the
most appropriate method for building Europe (Dyson and Featherstone
1999). The theoretical premise of the monetarist argument was that external discipline could be used to promote the twin goals of domestic policy
reform and external credibility.14

52
The United Kingdom

As mentioned earlier, the Franco-German alliance formed the core of the
integration process in Europe, as it was the political will of these two countries that motivated further integration. European unification and FrancoGerman reconciliation were judged to be in both France and Germanys
best interests. British negotiators, however, held a different opinion about
the relation of European unification to vital national interests. They saw
Europe as one of several fora, rather than a central forum, through which
British interests could be pursued (Dyson and Featherstone 1999). Unlike
the French and the Germans, the British had a different perception of the
necessity of European unification for securing strategic interests.
The United Kingdom stands on the boundaries of Europe. As victors
emerging from the Second World War, the United Kingdom did not
see European integration as a necessary foundation for its security.
Furthermore, Britains special relationship with the United States caused
it to place more emphasis on transatlantic relations.
Thus, while the British government favored the single market policy,
leading to liberalization, deregulation, and the free movement of capital,
it rejected the idea of a monetary union with a single currency, a single
European central bank, and stated fiscal policy rules. The then prime minister of Britain, Margaret Thatcher, refused to participate in the Stage I
negotiations outlined by the Delores Report if there would be any binding
commitments requiring participation in the next stage and the eventual
adoption of a common currency. British negotiators like Thatcher,
Lawson, Major, and Lamont were reluctant to transfer the sovereignty of
monetary policy to a European Central Bank and accept EC rules on fiscal
policy. Thatcher was deeply suspicious that European integration was progressing in a federalist manner, with institutions such as the Commission
and the European Court of Justice increasingly infringing on national
sovereignty.
The British government proposed a Hard ECU (HECU) that would
become the common currency of the Community. It also proposed that
the central value of the HECU not be devalued against other currencies
and that over time the HECU become the dominant currency of the union.
Unsurprisingly, the British proposal had little support, as the Bundesbank
did not want to create a European currency without having an independent
European central bank oversee it.
Like the Germans, the Danish, and the Dutch, the British mostly supported the economist approach to monetary integration. They advocated
that a potential monetary union be the end result of a long process of
economic convergence among EC countries, beginning with the opening

53
of markets and capital liberalization. They also believed that a multi-speed

Europe should be a possibility, since different states would converge at
different speeds.15
Benelux
In the course of European integration and especially the process of EMU,
we cannot deny the important role of big member states such as France and
Germany. However, it would be misguided to attribute the formation of
the EMU exclusively to the Franco-German alliance or to the convergence
of the interests of the four largest countries. Rather, the construction of
the EMU was a collective, multilateral achievement, in which all member
countries, including the smaller ones, played a part (Maes and Verdun
2005). Great achievements such as approval of the Treaty on European
Union could not have been realized without support from small and
medium-sized member states, as all member countries of the European
Union enjoy the same formal rights within the union, irrespective of their
population and economic size.
In this regard, the Benelux countries occupied a peculiar position in the
process of European integration. The three countries had already formed
a customs union before the European integration process that started
after the Second World War.16 Hence, they were regarded as a model for
European integration and, as such, a testing bed for Europe-wide integration. The Benelux countries had long realized that smaller countries
(or regions) would be better off if they fused their strengths, so it was an
easily accepted premise in the Benelux countries that European integration
was the best policy for guaranteeing peace and prosperity for themselves
as well as for Europe. The Benelux union was a forerunner of economic
integration, which made its countries voices more credible and persuasive in the negotiation process. The Benelux countries were some of the
strongest supporters of monetary integration throughout the whole EMU
negotiation process.
Benelux contributed to European integration in many areas, developing many new ideas and tactics. The Spaak report, the Werner report, the
Tindermans report, and Duisenberg plan, all bear the names of big figures
of European integration from Benelux countries that greatly contributed
to the acceleration of the integration process when its momentum began to
diminish. The idea of a flexible, multi-speed Europe was originally proposed by Belgium, which preferred a realistic approach to European integration. Benelux diplomats were famous for being mediators competent
at reaching consensus among seemingly conflicting views and contributed
greatly to the success of negotiations.

54
In the area of monetary integration, however, Belgium on the one side

and the Netherlands and Luxembourg on the other side displayed competing views. While Belgium supported the monetarist approach together
with France and the European Commission and was very active in creating
the institutional background for monetary union in the Community, the
Netherlands and Luxembourg went hand in hand with Germany in their
support for the economist approach.
All in all, Belgium is said to have played a pace-setting role, as it was a
consistent demandeur (Maes and Verdun 2005). Belgiums constant concern
was to have EMU on the agenda. In the 1970s and 1980s, the Netherlands
and Luxembourg had a more restrictive view of the conditions under which
EMU was possible. Together with Germany, they also took a harder line
on the criteria for entry and were seen as gate-keepers (Maes and Verdun
2005). The Netherlands and Luxembourg, however, did not merely side
with Germany. They, together with Belgium, were keen to promote vivid
discussion among member states about monetary integration.
The European Commission
It is worthwhile to note the central role played by the European Commission
in the negotiation process, despite the fact that it is not a nation with sovereignty but an organization. The European Commission was from the start
an active participant in negotiations on the proposed monetary union.
Within the institutional framework of the European integration process,
the original treaties gave the Commission specific duties, chief among
which were that it be an advocate of the common interests of the member
states, an initiator of common positions, an objective arbiter between
national interests, and, lastly, the guardian of the treaties.
The Commission generally supported the French positions on the
process of candidacy requirements, EMU implementation, such as the
introduction of the opt-out clause, the flexibility of convergence criteria,
and fiscal federalism, as it explicitly preferred EMU progress under more
relaxed requirements to making no progress under stricter conditions.
Jacques Delors played a very important role in the relaunching of the
project of EMU. As president of the Commission, Delors was extremely
invested in the EMU, and after overcoming resistance from Thatcher he
went ahead with his goal of putting monetary union back on the ECs
agenda. The Delors Report helped design the EMU and set timelines for
subsequent negotiations. He is one of the main actors and contributors of
the formulation of EMU in Europe.
The European Commission was also important in facilitating the negotiation process, since it was an independent institution. The European

55
Commission was actively involved in the negotiation process that resulted

in the Hague Summit, which proposed the establishment of a step-by-step
plan for the achievement of EMU. It produced a number of documents
addressing shortcomings in economic integration and monetary cooperation and thus provided the background for the decisions made at the
Hague Summit. The commission created the first plan toward monetary
integration in stages, and the Werner Report drew greatly on this plan. The
Delors Report became the reference document for the Maastricht Treaty
and for the implementation of EMU over the next 10 years, which resulted
in the introduction of a single currency in 1999.
2.5
THE ROLES OF THE EAST ASIAN STATES17
Turning to East Asia and its still young attempts to foster regional monetary and financial cooperation and integration, it is still to be seen how the
process will advance and who the key players will be. One possible scenario
is that China and Japan will come to realize that despite the differences in
their political and economic strategies, working together will be in both
countries interests and crucial to developing a common political will in
East Asia. Sakakibara (2003) and Murase (2004) argue that the roles of
China and Japan in East Asias integration process would be synonymous
with those of France and Germany in Europes integration process.18
Similarly, the Kobe Research Project report19 states that [i]t is essential for
the JapanChina cooperation, as a core in East Asia, to lead the process of
economic and financial integration, as the Franco-German alliance played
a central role in the integration and cooperation process in Europe.
Despite this and similar claims, Japan and China are not yet in a position to emulate the Franco-German partnership and promote regional
economic and monetary integration. A Sino-Japanese alliance is unlikely
for the time being, as the differences between China and Japanin terms of
economic development, political systems, and regional security interests
offset the similarities. In addition, it must be kept in mind that, while Japan
is experiencing stagnating economic growth, China is experiencing high
levels of growth and could potentially become a superpower in economic,
political, and military terms. Thus, in order for any partnership between
these two countries to form, many differences would have to be resolved
and mutual benefits recognized.20
With the potential leadership roles in a process of East Asian monetary integration assigned to Japan and China, Korea could take on a
proactive role in fostering Sino-Japanese collaboration while at the same
time acting as a representative of smaller countries in the region. Korea

56
could serve as a mediator between China and Japan, helping to cultivate

a common political will. Murase (2004) likens Koreas potential role in a
process of East Asian integration to the one the Benelux countries played
in Europe. Murases argument on the role of Korea is interesting, but a
few issues need to be clarified with regard to Koreas role in the integration process. First, as was mentioned before, it is quite difficult to sort out
a common Benelux role, especially in the early stages of the negotiation
process. Frances views and strategy for EMU negotiations were more
or less duplicated by Belgium, while the Netherlands and Luxembourg
firmly stood by Germany during the negotiations. It was only in the last
stages of EMU that a common Benelux view developed that was not so
different from what France and Germany agreed upon. Second, Korea
can take a strategic position between China and Japan. As China is still
considered to be an introvert country that lacks policy harmonization
with its neighboring countries, while Japan has not yet shown regional
leadership, Korea can function as a kind of honest broker in dealing
with cooperation and integration issues. Third, it is important to note that,
unlike China or Japan, Korea maintains a strong will to promote regional
cooperation and integration. Korea is one of the countries most concerned
with North Koreas nuclear crisis and would be devastated if a war in
the Korean peninsular were to occur. Recalling that one of the strongest
impetuses toward economic integration in Europe after the Second World
War was an institutionalization of permanent peace in the region, Koreas
position may indeed mirror that of the Benelux countries in that Korea,
as a medium-sized country, has strong motives for proposing regional
integration and mediating between its two large neighbors. Korea may
be a proponent of hard currency if the negotiation process for monetary integration were started, possibly together with Japan, Singapore,
Malaysia, and Thailand.21 Korea may also play a gate-keeping role like
the Netherlands and Luxembourg did. However, it would be premature to
expect Korea to play an active role in the near future, as its role has been
rather limited in the past, while ASEAN has had one of the most active
roles in ASEAN13.
2.6
CONCLUDING REMARKS
This chapter suggests several implications for East Asian monetary

integration and cooperation. First, the European monetary integration
process has demonstrated the need for pragmatic compromises in developing frameworks for monetary cooperation. All through the process of
EMU, there were competing views among the EU member states on the

57
need for and the form of monetary integration. To ensure the continued
participation of all parties, compromises were struck and exceptions, such
as the opt-out clauses for the UK and Denmark, were made. In East Asia,
governments also need to take a pragmatic approach in working toward
a common goal. This might involve, at times, countries agreeing to decisions that only indirectly benefit their economies by helping the overall
integration process.
Second, the European experience has shown that integration can only
be achieved step by step, by starting out with concrete economic and
political measures that subsequently help develop a larger framework of
cooperation that could later be enforced by common supranational institutions. East Asian countries need to be exposed to opportunities of binding
negotiations on regional issues and to accumulate experience in resolving conflicting issues. It took many years for the countries who designed
Europes EMU to acquire negotiation skills and to reach a stage of mutual
understanding. Similarly, a gradual process of integration would help East
Asian countries to develop the mutual trust and understanding that is so
crucial for monetary cooperation.
Third, while the European monetary integration process was a process
of trial and error, it was helpful to have blueprints in the form of the Barre,
the Werner, and the Delors reports. Although the Barre and the Werner
reports had not been implemented immediately, they contributed greatly
to shaping Europes monetary integration by stirring up discussion about
the direction of integration. The Delors plan, which became the actual
blueprint for the creation of the EMU, provided binding benchmarks
and timelines for the monetary integration process. East Asian countries
should also develop an advanced blueprint for monetary cooperation and
integration. This would help clarify positions and force the participating
countries to agree on common principles for potential integration.
Fourth, a variety of policies are required in order to encourage the
possible losers of monetary integration to continue to participate in the
process. Europes Common Agricultural Policy (CAP) and Structural
Funds played a key role in this, although the original programs were not
developed for that purpose. While it is not recommended that East Asian
countries implement programs such as the CAP (and the region would
also be lacking the resources to do so), certain fiscal transfer mechanisms
would help support the common integration policies within those countries
experiencing transitional losses.
Fifth, it is recommended that some East Asian countries form a kind of
core group so as not to lose the momentum toward integration. In Europe,
Germany and France maintained key bilateral relations over the course of
the whole integration process. In East Asia, Japan and China could play

58
such a role, but, if the two countries are not yet ready, Korea is strongly
recommended to initiate trilateral relations. Korea, the uniquely divided
country in the region, would have a special interest in initiating and leading
a process of economic and monetary integration in order to secure regional
peace and prosperity.
The process of economic and monetary cooperation in East Asia has just
started and is still at a rudimentary stage. However, we need to recall that
the original six members of the European Economic Community, which
today seem to be more or less homogeneous economies, were regarded
as genuinely heterogeneous when the Community was founded some 50
years ago. In those days, the differences between the nations economic
structures caused people to believe that the integration of such heterogeneous economies was absurd. Therefore, without putting too much emphasis
on the differences between East Asian countries, it would be important
to encourage closer regional cooperation now in order to prepare the
way for solid performances in economic and monetary cooperation and
integration.
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
9.
I would like to thank Beate Reszat, Ulrich Volz, and participants of the HWWA/HWWI
conference on East Asian Monetary and Financial Integration in Hamburg, on December
1516, 2005 for their helpful comments and suggestions. The first draft of the chapter
was presented at the conference on Regional Monetary Cooperation and Coordination:
the Experience in Europe and Feasibilities in Asia, hosted by the Institute of World
Economy at Fudan University, Shanghai, on October 2526, 2004. Comments by the
participants of the conference are gratefully acknowledged. The remaining errors are
mine alone.
The first stage was launched on an experimental basis and did not require a binding commitment from its members to complete the process of economic and monetary union.
This was a mechanism for the managed floating of the European currencies: the snake
within narrow margins of fluctuation against the dollarthe tunnel.
Benelux refers to Belgium, the Netherlands, and Luxembourg.
Exchange rates were based on central rates against the ECU, the European unit of
account, which was not a numeraire based on a single currency but a weighted average,
or basket, of the currencies of the member states. A grid of bilateral rates was calculated
on the basis of these central rates, expressed in ECUs, and currency fluctuations had to
be restricted to within a margin of 2.25 percent on either side of the bilateral rates (the
Italian lira was allowed a wider margin of 6 percent).
The commitment to the formation of an economic and monetary union had already
been stated in the SEA of 1986.
Both IGCs were launched at the Rome European Council in December 1990.
These included restrictions on capital movements, constraints on the overdraft facilities available to public authorities and public undertakings, and limitations on giving
preferential access to financial institutions for public undertakings.
The four convergence criteria that are presented in Article 121(1) of the Maastricht
Treaty relate to price stability, government deficits, exchange rates, and long-term
interest rates. Each member state must satisfy all four criteria in order to be able to

10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
59
participate in the third stage of economic and monetary union. Budgetary rules were set
to become binding and member states not complying with them were likely to be penalized. During the negotiations, Denmark and the United Kingdom opted out of the third
stage of EMU.
For an analysis of the ERM crisis see Volz (2006).
ASEAN13 Financial Ministers Meetings have discussed the possibilities of macroeconomic policy coordination, but this remains in the rudimentary stages.
See also the chapter by Zhang and Liang in this volume (Chapter 14).
The monetarists had nothing to do with the school of thought of the same name that
was developed by Milton Friedman and the Chicago School.
The existence of the monetarist and economist policy advocacy groups added tension to
the Werner Group negotiations in the 1970s, and also to negotiations and debates about
the ERM. Both groups offered diverse opinions on the relationship between economic
convergence and monetary union.
Still today, the British government maintains that potential membership in the EMU
would depend on the fulfillment of five economic tests that would ensure convergence
with the Eurozone.
Belgium, the Netherlands, and Luxembourg founded Benelux on September 5, 1944.
Before the formation of Benelux, Belgium and Luxembourg established BLEU (the
BelgiumLuxembourg Economic Union) in 1921 and have run a fixed exchange rate
system ever since.
This section is developed from Section 4 in Kim and Wang (2005).
Like China and Japan, France and Germany also had a wartime legacy of animosity.
It was submitted to the fourth gathering of the finance ministers of the AsiaEurope
Meeting (ASEM Finance Ministers Meeting) held in Copenhagen in July 2002.
Searching for desirable roles of Japan and China, it would be helpful to also take note of
other relations such as the Anglo-German relations and the GermanPolish reconciliation process. Japan has shown a more or less similar strategic position to the UK in its
regional policies. GermanPolish reconciliation is also interesting to note, as a wartime
legacy between the two countries had not been resolved until the 1990s because of the
Cold War.
For an application of EMU criteria to East Asian countries, see Kim and Park (2004,
pp. 1259).
REFERENCES
Dyson, Kenneth, and Kevin Featherstone (1999). The Road to Maastricht:
Negotiating Economic and Monetary Union. Oxford: Oxford University
Press.
Kim, Heungchong, and Sung-Hoon Park (2004). The Political Economy of the
EMU Negotiation and its Implications to East Asian Monetary Integration.
KIEP Policy Analysis 0410. Korea Institute for International Economic Policy,
Seoul.
Kim, Heungchong, and Yunjong Wang (2005). Financial Integration in East
Asia: Which Role for Korea? Paper presented at a conference on The Rise of
China and Koreas Regional Policy at Ifri, Paris.
Maes, Ivo, and Amy Verdun (2005). The Role of Medium-sized Countries in
the Creation of EMU: The Cases of Belgium and the Netherlands. Journal of
Common Market Studies 43: 2748.
Murase, Tetsuji (2004). The East Asian Monetary Zone and the Roles of Japan,
China and Korea. Manuscript, Keio University, Tokyo.

60
Reszat, Beate (2003). How has the European Monetary Integration Process
Contributed to Regional Financial Market Integration? HWWA Discussion
Paper 221, HWWA, Hamburg.
Sakakibara, Eisuke (2003). Asian Cooperation and the End of Pax Americana.
In Jan Joost Teunissen and Mark Teunissen (eds), Financial Stability and
Growth in Emerging Economies: The Role of the Financial Sector. The Hague:
FONDAD.
Stoltenberg, Gerhard (1988). The Further Development of Monetary Cooperation
in Europe. Memorandum to ECOFIN Council, Bonn: Ministry of Finance.
Volz, Ulrich (2006). On the Feasibility of a Regional Exchange Rate System for
East Asia. Lessons of the 92/93 EMS Crisis. Journal of Asian Economics 17:
110727.
Wang, Seok-Dong, and Lene Andersen (2002). Regional Financial Cooperation
in East Asia: The Chiang Mai Initiative and Beyond. www.unescap.org/pdd/
publications/bulletin2002/ch8.pdf.

3.
International political conflicts and

economic integration
Koichi Hamada and Inpyo Lee
If goods dont cross borders, armies will.
(Attributed to Frdric Bastiat)
3.1
INTRODUCTION
In ancient times, the Far East, notably China, was a center of civilization and a theater for displays of brilliant military strategies. Now, the
Chinese economy is emerging as a dominant player in the world thanks
to its sheer size alone. The Japanese economy, with its rapid growth, was
hailed as a rising sun. South Korea, Taiwan China, Thailand, and others
achieved Asian miracles and then were disrupted by the East Asian crisis
but recovered quite successfully. Thus, East Asiapeople seldom call it the
Far East nowhas become a dramatic, intriguing, and even threatening
area.
East Asian countries were notably unconcerned about regional economic cooperation until the late 1990s. The Asian financial crisis of 1997,
however, clearly showed that the heightened degree of interdependence
in the international economy has increased the need for closer global and
regional economic cooperation.
Kuroda (2003) suggested a step-by-step plan for financial and monetary
cooperation in East Asia: (1) formation of a regional financial safety net,
(2) promotion of effective regional surveillance, (3) development of regional
bond markets, (4) exchange rate stabilization among regional currencies,
and (5) establishment of an Asian common currency. Various efforts for
the first to third stages have been taken, with some of them materialized,
but the later stages have yet to be realized.
After the proposal to create an Asian Monetary Fund was shelved,
ASEAN plus China, Japan, and Korea (ASEAN13) agreed to strengthen
the existing cooperative frameworks in the region through the Chiang
Mai Initiative (CMI) in May 2000. Regional cooperation for financial and
61

62
monetary arrangements has just started by way of the CMI, but considerable work remains to strengthen cooperation for a safety net, regional
surveillance, bond market development, exchange rate stabilization, and
a common currency.
In particular, building up the Asian bond market has been a focal
point of regional financial cooperation.1 The virtual absence of foreign
investment in local bonds in East Asia has been highlighted as a great
barrier to financial stability. That is, foreign borrowing by the banking
sector, which was used to finance domestic fixed investment, was pointed
out to have resulted in the so-called double mismatch of maturity and
currency. If so, the high risk of short-term borrowing and equity investment still leaves East Asian countries vulnerable to shocks in the global
financial market.
A true Asian bond market could contribute to the stabilization of the
regional financial system by reducing the heavy reliance on short-term
external financing of Asian companies through the banking sector. In addition, vitalization of the bond market may help reduce the risk of maturity
and currency mismatch. More importantly, successful development of the
Asian bond market will signify that ultimate economic and monetary integration in East Asia is quite promising and could accelerate the integration
process. In this regard, the East Asian economies have just taken the first
steps in a process that could eventually result in an economic and monetary
union (EMU) in Asia akin to that in Europe, although it certainly will be
a long process.
On the other hand, we see several unsettling factors in the diplomatic
relations between Japan and its neighbors, China in particular. Japans
Prime Ministers visit to the Yasukuni Shrine, threats to reduce foreign aid
from Japan, and Japans textbook description of wartime conducts generate suspicion about the future course of Japans diplomatic policy. Chinas
alleged operation of submarines within Japans territorial sea near the
Okinawa islands, its opposition to Japans permanent membership in the
General Council of the United Nations, and its benign treatment of student
protests before the Japan Embassy in Beijing all make Japan nervous. The
diplomatic relations between China and Japan appear to have been chilled
and to involve more uncertainty and even some military risks. The tension
between China and Japan casts clouds over the future of Asias political
economy, about which serious questions have already been raised, such as
those regarding North Korea.
This chapter develops economic observations on financial and monetary
integration in East Asia. Considering the delicate political situation in
East Asia at present, however, we could not help but cross the border of
disciplines and discuss the international political implications of monetary

International political conflicts and economic integration
63
and financial integration. Needless to say, economic cooperation and integration is not an issue that could be understood and discussed from purely
economic perspectives. Peace is of enormous importance for all of East
Asia and for the world.
In section 3.2, we analyze economic rationale behind the efforts to create
the Asian bond market and discuss various policy issues. We address, in
addition to the question of what the ideal state of bond market integration
would be, the question of what kinds of policy measures should be implemented to achieve a desirable financial integration in terms of the bond
market.
In section 3.3, we turn to monetary integration and discuss the costs
and benefits of joining a monetary union. The normative justification for
monetary unification or financial integration is not quite decisive from the
purely economic point of view.
In section 3.4, we explore the relationship between economic integration and political conflicts from an international political perspective. In
order to understand the issues related to economic integration, we have
to be aware of the incentives for each participating nation. It is one thing
to recommend the formation of a free trade area or common currency
area based on benefitcost analysis, but it is just as important to understand under what conditions nations are motivated to join a club. As is
understood from the political economy of monetary integration, purely
economic incentives may not exist for nations to join a monetary union.
As in the case of the European Union and the euro, security reasons may
turn out to be a critical motive for realizing a monetary union in the East
Asia.
In section 3.5, we illustrate the logic of the situation in East Asia by
examining a simple model of nested games, that is, a concurrence of an
economic game with a security game between two countries. The economic
game can be characterized as a type of the Prisoners Dilemma, while the
security game is characterized as a type of HawkDove game. If both
governments accurately know the probability of military confrontations,
their direct and indirect costs, and the link between the economic game and
the security game, then the total game can be seen as a simple game with
multiple strategies. The total outcome will not substantially distort the
interests of voters in participating nations. In addition, we point out the
factors that may distort the play of the above nested game and jeopardize
the peace and welfare of nations.

64
3.2
FINANCIAL INTEGRATION AND THE

COMMON BOND MARKET
Economic and Conceptual Issues

Before going into various policy issues regarding bond market integration,
it is helpful to review the conceptual distinctions between three basic forms
of financial intermediation: bank lending, bond financing, and equity
financing. It is important to recognize, first, that financial integration is
indeed closely related to monetary integration, but there exists a fundamental functional difference between them.2
Bank lending is characterized by the face-to-face direct relationship
between a lender and a borrower and can be described as the negotiable
form of financial intermediation. Conditions of loans are worked out
between the related parties. A prospective lender investigates the credit
conditions of prospective borrowers and screens safe borrowers from
less safe ones. After the loan is made, the lender often monitors, if not
supervises, the business of the borrower. Activities related to information gathering and monitoring are conducted on a bilateral basis between
the creditor and the debtor. Normally, the creditor is supposed to hold
the claim until the end of the term unless the borrower decides to pay
back the debt, and only occasionally is the claim transferred to a third
party. The negotiable credit contract cannot go with anonymity.
This form of transfer can exploit lender-specific and borrower-specific
information that is unavailable to the public. This is often considered a
positive aspect of relationship banking. On the other hand, this type
of transfer of funds can be associated with the crony banking of the
Japanese Zaibatsu (literally financial cliques) or South Korean Chaebol
(family-controlled business conglomerates) in the sense that one lends to
the other not because of the borrowers financial suitability but because the
lender happens to be a relative or a friend of the borrower. It is no coincidence that the main bank system grew out of the Zaibatsu relationship that
was originally based on family ties.
On the other hand, bond financing and equity financing can be characterized as the market form of financial intermediation. Because of
the absence of a direct information channel between essential lenders
and borrowers, firms are required to reveal their essential financial data.
Underwriting is important not only for the treatment of risks related to
public offers but also for the dissemination of business information of a
debtor firm to the public. Bonds are standardized and most importantly
transferable from one holder to another in the open market. Public business information is also published in a standardized form in order to be

65
able to be transmitted to anybody in the market. Bonds have more liquidity

because they can be instantaneously sold. Anonymity works in the market
form of financial intermediation.3
At the least, there are various merits to cultivating a well functioning
bond market. First, since lenders wish to minimize risks through portfolio management, the existence of new forms of investment always helps.
Therefore, adding new bond instruments to an economy dominated by
bank lending implies that the economy obtains a new form of financial
instrument. Second, for firms that are able to demonstrate their sound
financial prospects in the open market, bond financing creates a new and
profitable opportunity. This is the reason that many prominent Japanese
and Korean firms are successfully obtaining funds through public bond
offers. Third, this may be a good way to get away from crony finance.
Although a holder of bonds has a pay-off profile quantitatively similar
to that of a holder of loans in terms of the contingent pay-off schedule,
they are constructed differently in contractual and procedural terms, in
particular when delinquency or bankruptcy situations emerge. Also, the
mode of and costs related to monitoring the agency relations are different. Therefore, it is always better for an economy to be able to use both
of these alternative means of finance. Bond market development can thus
be regarded as a positive direction for economic development policy from
this point of view.4
As an application, consider what happens to various asset holders in
the case of a financial crisis taking place in a borrowing country. Equity
holders of non-performing firms will lose wealth almost immediately.
Bondholders retain claims to the remaining worth of the firm and can join
the bankruptcy procedure, but the value of bonds will depreciate. In informal expression, the claim to the firm under problematic conditions will be
handled by the asset market almost immediately and in parts.
On the other hand, if a firm is financed mainly by a bank, it takes time
for loans to be known to be delinquent. The problem of oigashi (roll-over
loan)the phenomenon observed when a bank official additionally extends
to a delinquent zombie firm in order to avoid the revelation that the firm
is near bankruptcyaggravates the situation. If the majority of investments
were financed by the market type instruments rather than by the negotiable
type instrument, the cost of financial crises would be smaller.
Political Economy Aspects
In addition to the question of what the ideal state of financial intermediation is, we must address the question of what kinds of incentives promote
financial integration in terms of bond markets. In general, unless there are

66
regulations against the mode of finance, and unless there are externalities
or increasing returns, the financial market will choose the right mode or
the right combination of modes of financial intermediation. There will be
no need for government interventions to facilitate a particular mode of
financial intermediation.
In the East Asian countries, when a government preaches the need for a
well-developed bond market, it often means the need for a domestic market
on which it can sell its public debt with a lower rate of interest. This is the
wrong reason to develop mature bond markets. By no means should the
idea of common basket bonds be the main device for solving the large debt
situation of Japan, Korea, and other countries. It does not make sense
to spend more tax money in order to facilitate the deficit financing of the
government.
Incidentally, since the essential characteristic of the bond market lies
in its universal, non-name-specific nature, in contrast to the bank loan
market, it is an oxymoron to define an Asian bond market as a bond market
By Asians, For Asians, and In Asian currencies.5 Indeed, denomination
may matter. Because of the existence of externalities in the denomination
of bonds, people are motivated to use the denominations that are already
prevalently used. This gives governments some rationale for promoting
the denomination of the bond market in local currencies. But it should be
of little significance in determining who issues and who buys a particular
brand of bonds. Such a market would not grow and flourish unless the
bonds traded there became accessible to non-Asians.
Before making any attempts to encourage the creation of a regional
bond market through financial support, governments must start dismantling unnecessary regulations that they themselves imposed, such as
regulations on the issuance of bonds and obstructions against the free
movement of capital and the acquisitions of domestic firms. It is almost a
self-defeating policy strategy to promote the Asian bond market through
cooperative actions by using taxpayers money while keeping barriers to
free capital movements across Asian countries.
Governments should also improve the accounting, legal, and corporate
governance systems to make them incentive compatible and transparent.
As Takagi (2002) maintains, the building of proper infrastructures is a prerequisite for building sound bond markets. Politically, of course, the goal of
creating a sound bond market may facilitate the building of infrastructure.
Under normal conditions, our recommendation for the creation of bond
markets is of the laissez-faire type: let the market mechanism develop any
new market. What can be the exception to this principle?
In general terms, government interventions are justified under the following conditions:

1.
2.
3.
67
Externalities exist among various financial markets or among local

markets, so the promotion of a bond market and cooperation among
many national markets may yield some positive returns. International
cooperation through the initiation of experimental bond markets may
give market participants signals for noticing Asian markets as focal
points.
The bond issue in the currency basket may well be a helpful device
for focusing the attention of market participants and for developing a
market with a greater degree of depth compared to bonds in a single currency. Here an analogy to language use applies. Although this subject
needs further empirical research, by initially promoting the common
bond market using public funds, government may possibly gain some
advantage thanks to the depth of the market in order to market their
government bond on better terms. As it is difficult to disseminate a
language to another location, this process will take a long time.
Similarly, international cooperation may be a sound device for making
use of the economies of scale involved in initiating new markets,
because a single country market may be too small.
These attempts are justified only as an analogue to the pump priming

policy that is reminiscent of the earlier Keynesian economics. Unless
externalities or increasing returns and information asymmetry are important enough and also of the nature that interventions would be the only
remedies, we do not endorse using government policy to artificially
promote debt markets.
In conclusion, increasing the variety of financial instruments is always
welcome. Therefore, the creation of bond markets through natural market
mechanisms is advised. In the mean time, deregulation and the construction of financial infrastructures are also always necessary. Excessive artificial encouragement of a particular form of bond market at the expense of
taxpayers money, however, could be costly for the regional economy as
well as individual nations economies.
3.3
MONETARY INTEGRATION OR CURRENCY

UNION6
Economic and Conceptual Aspects

In order to discuss the costs as well as benefits of joining a monetary union,
we have to identify alternative forms of monetary integration on the basis
of the degree of the strength of the binding agreement.7

68
1.
2.
3.
4.
Under a floating exchange rate regime, monetary authorities can

engage in cooperative actions to keep exchange rates within a certain
range, or a zone. Naturally, joint interventions are more effective
than unilateral interventions. As shown in Hamada (1985), the fixed
exchange rate necessitates, at least partially, exchange rate coordination by the nature of its rule, and, if not necessitated, coordination
has a positive benefit under the fixed exchange rate. Under the flexible
exchange rate, however, national monetary authorities recover the
independence of their monetary policies. The need for, as well as the
benefit from, coordination becomes almost negligible.
In its weakest form, monetary integration implies the linking of
national currencies with fixed parities accompanied by a narrowing or
vanishing band of exchange without common reserves or a common
central bank. This is what Corden (1972) called the pseudo-exchange
rate union. The coordination of economic policies, particularly
monetary policies, is needed to prevent disequilibria in the balance of
payments. There are a few sub-examples of this type of union.
Governments can announce the fixity of exchange rates. The degree
by which the public believes in the endurance of the fixed exchange
rates depends on various economic and political factors. The Bretton
Woods regime of adjustable pegs is a typical example. Participants
could change the parity if there was a fundamental disequilibrium, a
phrase that caused academic and practical controversies but was never
defined clearly by the IMF.
Or, under the gold standard and other commodity standards, a metal
or some other commodity is used as an anchor for keeping the fixed
exchange parities among currencies. Through the pricespecie flow
mechanism, monetary policy coordination is supposed to function
automatically. In this regime, however, confidence in the endurance
of parities is not absolute. Governments have the option of changing
the parity rate of currency with gold, for example.
The degree of monetary integration is enhanced by the establishment
of public confidence in the irrevocable nature of the fixed exchange
parities accompanied by full convertibility between currencies for
capital and current account transactions. This confidence normally
emerges only after a substantial transition period during which de
facto fixed exchange parities are successfully maintained or after some
kind of political unification.
Full monetary integration or unification is only realized when a
common currency issued by a single central bank circulates in the area
of the monetary union. A unified currency may prevail in an area where
multiple governments exist; this has been the situation in the euro zone

69
since the circulation of the euro. Each participating nation retains the
right to exit the union, and an exit option will be conceptually possible.
We can also think of the circulation of a unified currency under a single
government, where the common currency circulates in all regions of
the country or federation. Under a united government, discrepancy of
fiscal policy can be neglected as people believe that this currency area
will never be divided again. For example, after the German unification, Germans trusted in the irrevocable exchange rate between the
D-Mark and the O-Mark again. Historically, however, as in the cases
of the Habsburg Empire and the Soviet Union, a single currency under
a unified government could dissolve into multiple currencies under
multiple states.
Now consider the benefits and costs of attempting monetary integration. Aside from the possible strategic benefits, to be discussed later, the
benefits from monetary integration are of a microeconomic nature: users
of a common currency economize on information costs and transaction
costs. Monetary union members enjoy the benefits of increased trade and
differentiated products as a result of the reduction or even disappearance
of uncertainty about fluctuations in the exchange rates among member
currencies. This benefit is partially delivered by a pseudo-exchange union
but substantially realized only after the emergence of public confidence in
the fixity of exchange rates. Only after full monetary unification is achieved
are the transaction costs arising from currency conversion eliminated and
the consequent benefits of increased trade and tourism enjoyed.
These microeconomic benefits are closely associated with moneys function as a medium of exchange. Money economizes on the information costs
required for transactions and allows for the procurement of a stable bundle
of goods at a lower cost than under barter.8 The use of a common currency
carries intrinsic externalities as a result of its informational properties.
These benefits from economizing on transaction costs and from information spillover have the characteristics of a public good: non-rivalry (consumption by one member does not reduce the goods availability to other
members) and non-excludability (certain members cannot be excluded
from enjoying the good). These benefits of complete monetary integration
are obtained only partly through the adoption of the fixed exchange rate
regime.
A secondary benefit of monetary integration is macroeconomic.
Mundells (1968) theory of policy assignment indicates that the effect
of regionally specific real shocks may be absorbed by flexible exchange
rates. However, recent studies on regime choice show that country-specific
monetary shocks can be better managed under fixed exchange rates or

70
under highly managed exchange rates (e.g., Fukuda and Hamada 1987,
Eichengreen 1998).
Thus the primary benefits of joining a monetary union are microeconomic, while the macroeconomic benefits are limited to smoothing
monetary disturbance. In contrast, the costs, making each country
vulnerable to real country-specific shocks, are mainly macroeconomic.
While countries differ in their rates of productivity growth and in their
preferences concerning the choice between unemployment and inflation,
countries adhering to a fixed exchange rate or participating in a single
currency union sacrifice the ability to independently pursue individual
monetary policy objectives. This is particularly true when international
capital mobility is high and when wages and prices are rigid. The floating exchange rate system gives national economies the opportunity to
secure the minimum state of affairs that can be obtained independently,
that is, the opportunity to follow a maximin strategy in the interplay of
monetary policies. By joining a monetary union, a country gives up this
maximin position and must seek the mutual consensus demanded by
policy coordination.
In general, the benefits and costs of monetary integration have several
characteristics. First, in contrast to the benefits of monetary integration,
which are enjoyed collectively and have a public-good nature across
nations, the sacrifices made by joining a monetary union are made at the
level of the individual nations. This distinction will play a crucial role in
our analysis of the incentives of joining a monetary union.
Second, the benefits and costs to participating countries vary over time.
Initially, the costs of sacrificing domestic economic objectives and an independent monetary policy are large. As capital market integration proceeds,
the financing of fiscal deficits becomes easier, and, hence, these adjustment
costs become smaller. However, the common benefits of monetary integration are usually enjoyed only at a later stage when the credibility of the fixity
of exchange rates is established or when participating nations currencies are
unified into a single one. For example, the saving on the costs of currency
conversion occurs only after exchange rate union has been accomplished,
and the benefits arising from the stability of exchange rates can be reaped
only after confidence in the fixity of parities has been established. The only
way to make the time profile of the benefits lean toward the present is to
adopt a single currency. In this respect the euro zone has had apparently
considerable success. We say apparently because the assessment of the
great euro experiment should be made with caution, as will be discussed
later.
Third, the openness to trade and factor flows of a monetary union
member country has an important influence on the magnitude of the

71
benefits and costs arising from monetary integration. If an economy is

relatively open, with large import and export flows relative to GDP, the
costs of adjusting its output or employment level for balance of payments
reasons will be small (McKinnon 1963). The savings in currency conversion costs will also be proportionately larger in a more open economy.
If the economy is closed, however, these savings will be less significant,
while the costs of adjustment will be relatively large. Prior integration
of the markets for goods and services and factors of production among
member countries may increase the desirability of monetary integration
by increasing average levels of openness.9
Empirical Appraisals
First, Andrew Rose, Jeffrey Frankel, and others have conducted extensive research on the benefit of monetary union in numerous articles (e.g.,
Frankel and Rose 2002, Glick and Rose 2002, Rose and Engel 2002).
They conclude, mostly by applying the gravity model to cross-country
panel data, that having a single currency exerts a substantial influence on
the flow of trade and national income. According to their estimates, the
use of a single currency triples the amount of trade and increases national
income approximately by a half percent. The effect of the use of a single
currency is much larger than the effect of the fixity of exchange rates. The
institutional arrangement for keeping exchange parity is one thing, while
that for having a single currency is completely different. The latter gives
rise to much more confidence in the constancy of future exchange rates
and eliminates uncertainty about the future. Although Europes experiment with the euro is still young and it is too early to evaluate whether the
currency union has resulted in substantial changes, the general consensus
in the literature is that the euros effect on trade within euro zone countries
as well as on trade between euro countries and non-euro countries is positive and significant.10 It is worth noting that the level of trade of member
countries with those outside of the euro region did not shrink. Thus, the
trade diversion from external trade to internal trade seemingly did not take
place in the euro zone.
Second, Alesina and Barro (2002) and Alesina et al. (2002) show, based
on various indicators, that intra-regional trade in Europe is intense and
that there is a satellite-like relationship between the US and many of its
trading partners. In other words, one might define a European Trade
Area and a North American Trade Area. In contrast, they argue, Japan
has many trade partners all over the world, so that it would be difficult to
conceive of an entity to be called the Japan Trade Area. In their opinion,
the EU had, and NAFTA has, preconditions for a currency, but, compared

72
with the EU and the NAFTA area, the Asian region has less compelling
reasons for an Asian trade zone and, accordingly, has fewer grounds for
an Asian Monetary Union.
Third, Goto (2002) argues, on the other hand, by using the principal
component approach, that the confluence of business cycles among Asian
countries is even stronger than that among European countries and that
among NAFTA countries. Principal component analysis is useful for
measuring the degree of synchronization of multiple variables in contrast
with the conventional method of measuring the bilateral output correlation between each pair of countries (Goto and Hamada 1996). Goto
(2002) particularly emphasizes the time series observation that integration
among Asian economies has progressed in recent years. While most East
Asian variables were correlated with US variables in the past, they are now
moving together with Japanese variables, as well. Reservations remain,
however, about his findings. First, even though macroeconomic and trade
variables are moving together in Asia, as detected by principal component
analysis, the openness of Asian countries to trade is not as great as in
Europe. In Asia trade moves together, but the amount of trade is often
small. Therefore, the level test may fail. The use of the trade intensity index,
which is a powerful tool for describing the interdependence of trade flows,
may not satisfactorily indicate the closeness of trade relationships between
countries because the trade intensity index measures comovements in trade
but does not factor in the level of trade.
Political Economy Aspects
In sum, the pros of creating a monetary union between China, Korea,
Japan, and neighboring countries can be stated as follows:
The microeconomic benefits will include increased trade and differentiated
products for member countries. Moreover, trade partners outside the union will
benefit from the savings on transaction and information costs. In addition, we
can expect the macroeconomic benefits such as better management of monetary
shocks.
Meanwhile, the cons of creating a monetary union can be summarized as

follows:
Monetary union entails each member countrys loss of monetary independence.
Furthermore, the European experience has illustrated that in order to enjoy the
true benefits of monetary integration, an exchange rate union is not sufficient,
and a uniform currency should be an end goal. Are Asian countries prepared to
unify their currency systems despite institutional and cultural difference across

73
national borders? Are central banks, including the Peoples Bank of China, the
Bank of Japan, and the Bank of Korea, willing to give up their autonomy and
pool their monetary authority?
Even if we admit the findings by Frankel and Rose (2002) of a large

and positive effect of currency union on trade, it is not perfectly certain
whether this trade-flow effect dominates the cost in terms of the loss of
monetary independence effect. The cost of joining a single currency area
beyond the present fixed exchange rate regimes to the dollar could be really
serious. Inflation is not serious right now, and there is currently a worry of
worldwide deflation. We might encourage worldwide deflation by creating
an Asian single currency zone. Just as the leader country of the euro zone,
Germany, seems to be suffering from recession, might Japan not suffer
from deflationary pressure after unification?
Though we recognize the long-run benefit of creating a uniform currency in terms of trade creation, in the foreseeable future the cost in terms
of loss of monetary control seems to us more serious. Therefore, as far as
economic effects are concerned, we are skeptical about the significance of
the benefits of Asian monetary unification.
There remains the question of incentives. Under what conditions will
national governments find incentives to agree on a monetary union with
the benefits far into the future?
We strongly believe in the merit of and the need for policy coordination
among countries in which fixed exchange rates prevail (Hamada 1985).
Under flexible rates, the number of monetary instruments is sufficient for
each country to be able to conduct monetary policy regarding its price levels
and macroeconomic situations as a policy assignment problem (Feldstein
1997, 2000). On the other hand, if 12 countries join a monetary union,
then 11 policy instruments will be lost. Of course, this is true only if the
participating countries had monetary autonomy before. In Ulrich Volzs
contribution to this volume (Chapter 8), he argues that most (small) East
Asian countries do not have full monetary independence now because of
external exchange rate targets as well as weak domestic financial markets.
In particular, in the problem termed conflicted virtue by McKinnon,
countries with large international reserves as well as assets denominated
in dollars are constrained against the devaluation of their currencies. This
argument is rather persuasive, and we agree with it to a certain extent.
The loss of monetary autonomy in the case of the adoption of a common
currency is, however, definite, and a country cannot easily go back to a
floating rate without a substantial transition cost. This cost must be much
more significant than the cost of foregoing devaluation because of the large
scale of international reserves in dollars.

74
Therefore, the answer to the positive (rather than normative) question

from an economic standpoint also does not give strong support in favor of
a uniform currency.
3.4
ECONOMIC INTEGRATION AND

INTERNATIONAL POLITICAL CONFLICTS
We have seen that the justification for monetary unification or financial

integration is not decisive from the economic point of view. If we extend
the scope of the arguments from the purely economic dimension to the
political one and further to the strategic or security point of view, however,
the outcome changes significantly. A positive political economy analysis
of nations incentives for joining a currency union can tell us what may be
needed to create a currency area in East Asia.
Over the past few decades, there has been abundant literature on the
relationship between economic interdependence and political conflicts.11
Political scientists attention to the relationship between national security
and economic integration has been centered on the relationship between
trade and international conflicts. Among numerous papers on this topic,
Oneal and Russett (1997) conducted an extensive study on the relationship
between economic integration and the existence of national conflicts. They
found, in general, significant results demonstrating that the openness of the
pair (dyad) of countries will reduce the probability of conflicting situations
between them. For example, if the dyadic trade-to-GDP ratio increases by
one standard deviation, the likelihood of conflict between these countries
will decrease by about 4.7 percent.
As Mansfield and Pollins (2001) pointed out, the direction and timing
of causality between economic interdependence and political conflicts
should be examined. Gartzke et al. (2001) indicate that investment flows
are important, as well, and that incentive aspects should be clarified.
Given the serious damage potentially inflicted by wars, we consider
the gain in peace from openness as reported by Oneal and Russett
to be a substantial dividend. The process of economic and monetary
union in Europe can also be understood in the light of these strategic
considerations.
If openness to trade is a significant factor, then openness to capital
market and financial integration should also be an important factor for
peace, and the benefits of using a common currency should be even greater.
Unfortunately, empirical research is scarce concerning the relationship
between monetary unification and peace. However, we can expect that the
use of a common currency will significantly increase the amount of trade

75
between participating nations, which, in turn, would reduce the probability

of war quite substantially.
While most of the literature on the relationship between economic interdependence and political conflict considers that the causal arrow goes from
economic integration to the possibility of war, the arrow may also go in the
other direction. In their highly persuasive paper, Glick and Taylor (2005)
focus their attention on the question of how detrimental war is to trade.
They show that wars may cause a serious breakdown in trade and a consequent loss of national and global economic welfare and that the indirect
damage to trade by wars may be economic losses of a magnitude equivalent
to the direct costs of the war, such as lost human capital.
Studies such as those listed above reinforce the argument that a case
for a monetary union including security grounds may be much stronger
than the case for a monetary union on economic merits alone and that we
should develop solid relationships among nations. To this end we have
to rely on whatever means are available to foster economic ties between
countries, such as free trade areas, foreign investment liberalization, and
currency unions.
To temper our optimism about the peace effects of monetary union, we
should mention the intriguing history of the disintegration of monetary
unions, such as that of the Austro-Hungarian empire or the Commonwealth
of Independent States (CIS).12 In these examples, economic and monetary
union did not preclude political division and even conflicts. Without going
into the details of these historical experiences, we will discuss the example
of the European Union, where the political leadership apparently succeeded in overcoming the economic difficulties of monetary integration.13
The development of the European Monetary System (EMS) since its
inception in 1979 has offered a striking contemporary example of the
process of monetary integration. The EMS was founded as a flexible,
symmetric version of the Bretton-Woods system (Kenen 1992), following long-standing plans of economic and monetary union in Europe. The
Werner Report of 1970 envisaged the development of a monetary union
by 1980 through economic policy harmonization. Harmonization is a
process wherein the differences in participating nations key economic
indicators, such as the inflation rate, interest rates, and the levels of government deficits and accumulated debt, are gradually narrowed. After
frequent early adjustments to the Exchange Rate Mechanism (ERM),
there were no major realignments from 1987 until September 1992. This
period of stability contributed to the sense of inevitability in the progression toward a fixed parity system, but it proved to be the lull before
the storm of the suspensions and devaluations of member currencies in
September 1992.

76
During this period of turmoil the Deutsche Bundesbank increasingly

served as the monetary anchor of the EMS by virtue of the size of the
German economy and strength of the Deutschmark. This strength was
zealously guarded by the Bundesbank, with its strong reputation for
inflation aversion. However, the monetary discipline provided by the
Bundesbank became punitive as Germany suffered the real fiscal shock of
unification. The cost to other countries of enjoying Bundesbank credibility
soared, as high German real interest rates were transmitted throughout
the ERM. Members with below full employment felt the effects of the
squeeze on interest-sensitive sectors like housing and construction. The
growing divergence of the costs and benefits of maintaining the fixed
parity created uncertainty about the future of currency union as well as
of monetary coordination. It opened the door for speculators to bet that
the ERM link was too costly for weaker currencies to sustain. Despite
regularly reaffirmed commitments to preserve exchange rate parities, and
despite defensive purchases of weaker currencies by central banks, the
speculative attacks of September 1992 succeeded in forcing several devaluations. Thus, temporarily, confidence in the ERM bands was substantially
undermined, and the system tumbled into its most severe crisis in its then
14-year history.
Instead of giving up the process of monetary integration, EMS countries
(except the UK and Italy) continued their strategy of creating a common
currency area without the slow process of converging exchange rates.
In January 1999, the euro was introduced as the single currency, and
the European Central Bank (ECB) was established to conduct a unified
monetary policy.
The euro experience lends a new set of insights to the positive theory of
monetary integration. Contrary to conventional wisdom, monetary integration could precede political integration.
In spite of economists criticism (e.g., Feldstein 1992, 1997), and in spite
of severe speculative attacks in 1992, the momentum was strong enough
to establish the euro zone. The economic benefit of unification is clear:
the economy of transaction and information costs. More importantly, the
political will, that is, the perception of political benefits by national leaders,
must have been the driving force behind monetary unification in Europe.
Europes perseverant approach to the introduction of the euro, in spite of
the speculative attacks in 1992, could hardly be explained without taking
political and strategic considerations into account.

Table 3.1
77
Game I: Prisoners Dilemma

Country B
Country A
3.5
Free Trade
Protection
Free Trade
Protection
(4, 4)
(6, 1)
(1, 6)
(2, 2)
ECONOMIC COMPETITION AND THE ARMS

RACE AS A CONNECTED GAME
In this section, we present a simple representation of two kinds of nested

gamesan economic game and a security gameto illustrate the situation
in East Asia. There are some typical patterns in the pay-off structures of
economic games. First, the trade game is characterized by the prisoners
dilemma. Let us consider the simple case with the pattern shown in Table
3.1.
It is well known that the Nash equilibrium is the protectionprotection
combination (2, 2), because protection is the dominant strategy for both
players. Cooperation may emerge in this situation so that by a combination
of trigger strategies the cooperative equilibrium (4, 4) may be sustained.
This is known as the Folk theorem. Liberalization of the process of trade,
the creation of a free trade zone, and the deregulation of foreign investment
roughly fit into this type of strategic situation (e.g., Myerson 1991).
Alternatively, the pay-off of currency unification can be depicted as
shown in Table 3.2.
In this example with China and Japan, disagreement over the choice
of currency for a common currency union results in the prevailing of the
status quo. If the two countries agree on a currency, they benefit from the
formation of a currency union, but the country whose currency is chosen
will benefit more than the other. The Nash solution is either (3, 2) or (2,
3). A mixed strategy may be realized through the selection of a currency
basket, but technical problems remain with currency basket use.14
On the other hand, the security or strategic game has the nature of a
chicken game or a HawkDove game. If both countries take Dove strategies, both share the dividend of peace. If one country takes Hawk and the
other takes Dove, then the Hawk country will gain. If both countries take
Hawk, the world is subject to at least the possibility of severe destruction.
The pay-off appears as shown in Table 3.3.
(1, 6) and (6, 1) are the Nash equilibria. Once a country takes the Hawk
position, the other country would not take the Hawk because that would

78
Table 3.2
Game II: Matching Pennies or the Battle of the Sexes

Japan
China
Table 3.3
Yuan
Yen
Yuan
Yen
(3, 2)
(1, 1)
(0, 0)
(2, 3)
Game III: HawkDove game

Country B
Country A
Dove
Hawk
Dove
Hawk
(4, 4)
(6, 1)
(1, 6)
(10, 10)
lead to mutual destruction or at least the possibility of mutual destruction,

as represented by (10, 10). By itself, the repeated game version of this
HawkDove game has a sub-game perfect equilibrium that maintains the
pay-off close to the cooperative solution (4, 4) provided that the players
discount rate is sufficiently small. The trouble is that it also has a sub-game
perfect equilibrium that maintains the pay-off close to (10, 10). The
evolution of cooperation (Axelrod 1981) is not automatically guaranteed
(Myerson 1991, p. 323ff).
In the real world, those games are connected. As the reviewed literature
has shown, successful completion of currency union is likely to increase
the benefits of trade; the successful cooperation in Game II will improve
the cooperative pay-off of Game I. The most serious problem is that the
strategic game and economic games are nested or connected. The strategic
game of HawkDove is played along with the trade game of the Prisoners
Dilemma and/or the currency unification game of the Battle of the Sexes.
And the pay-off matrices are interrelated. If these two games were played
by a coherent and rational government, the nested or connected nature
would cause few problems. The problem is that it is hard to find such
coherent and rational governments.
One can explain the connected nature of the games as follows. Cooperative
plays in economic games, either trade or currency, will increase the opportunity cost of the war and deter the onset of the war, changing (10, 10)
into a more severe outcome, such as (15, 15). On the other hand, more
peaceful plays of the strategic game (Game III) will create a sense of
assurance about future economic relations and enhance the pay-off of the

79
cooperation solution of the Prisoners Dilemma as well as the pay-off of

the choice of currency game.15 Cooperation in one sphere will reinforce the
possibility of cooperation in the other sphere. A winwin situation will be
realized as a virtuous circle.
Mistrust over military matters will result in difficulty in economic
matters, and vice versa. Here, vicious circles may develop. In particular,
the recent strategic or security-related mini-confrontations may imply that
China and Japan are engaging in a pecking order game in order to assume
a leadership role in the choosing of a currency to be at the center of possible monetary integration. It is also understandable that Japan tries to
avoid the tug-of-war conflicts of leadership in monetary integration and
that it encourages bond market integration that is more consistent with the
concept of a multiple currency basket, a kind of mixed strategy solution.
Even if these two kinds of games, economic and strategic, are played
with full consistency by two rational governments, the possibility of the
emergence of an undesirable, or even destructive, equilibrium still cannot
be completely ignored. Myerson (1991) shows that a near-destructive state
can be sustained as a sub-game perfect equilibrium by trigger strategies in
a repeated HawkDove game.
The situation becomes more precarious when we consider the
following:
1.
2.
3.
4.
Military conflicts often take place not as a result of governments

deliberate decisions, but because of accidental events, such as minor
cross-border conflicts. Accidental factors may change the pay-off
matrix, or the perception of it, drastically. This will increase the risk
of uncalculated war.
Psychologists find that people have the tendency to assign a smaller
probability to events that they dislike, a phenomenon called cognitive dissonance (Festinger 1957). Governments tend to underestimate
the probability of war and the resulting casualties triggered by their
hawkish attitudes.
Governments and voters may underestimate the cost of war. Even if
they ever take into account the war expenditure, destruction of military capital, the grave human and health losses, and the destruction
of physical capital resulting from a war, they may not fully take into
account the economic loss caused by deteriorated trade relationships
after the war.
If a nations government is a coherent entity or an effective common
agent of the voters, then the benefitcost assessment of war as well the
benefitcost assessment of economic relations will be taken account in
a balanced fashion. The reality is that economic bureaus and security

80
bureaus are operated rather independently of each other. The repercussion to ones pay-off matrix by the others is often neglected by the
presence of red tape and sectionalism.
Historically, East Asia has enjoyed a common cultural heritage. Many
nationals can communicate with each other, even though they do not speak
foreign languages, by writing Chinese characters. Japan used coins made
in China in the 16th century, and the yen itself is allegedly named after the
Hong Kong dollar. Confuciuss teaching has long been a guiding principle
of government in the East Asian countries. There are many focal elements
that would facilitate coordination in the Asian basin.
When Myerson (1991) introduces the possibility of good as well as
grim equilibrium, he notices the necessity of the cooperative attitude.
Qualitatively, the more cooperative equilibria seem to involve a kind of
reciprocal linkage (e.g., expect me tomorrow to do what you do today),
whereas the more belligerent equilibria seem to involve a kind of extrapolative linkage (e.g., expect me tomorrow to do what I do today) (Myerson
1991, p. 331). The first cooperative attitude is similar to Confuciuss
Golden Rule, expressed in the Analects: Do not do to others what you
would not wish others do to you. Diplomacy and clever politics must go
beyond the difficulty pointed out by game theorists and endeavor to foster
mutual trust to prevent the destructive consequences of a war.
3.6
CONCLUSION
On the grounds of economic reasoning alone, the case for financial integration and, in particular, for monetary integration is not decisive. East Asian
nations calculus of participation does not give strong motivation for the
creation of a common bond market, an integrated financial market, or a
uniform currency area. Creating a common bond market may seem less
painful and involves only a limited degree of commitment, unlike exchange
rate coordination, which involves a certain degree of macro coordination
and possibly the pooling of reserves. However, the true functioning of a
common bond market still requires, we believe, the development of credit
market infrastructures in most East Asian countries.
Once we introduce strategic or security considerations, however, the situation changes drastically. This chapter is a tentative survey of the ongoing
work in the field of economics and political science and an attempt to
explore the importance of security considerations in the issue of monetary
cooperation. It is meaningful to compare economic costs and benefits. But,
at the same time, if we consider the tremendous human and material costs

81
generated by possible internal conflicts in the region, we need to consider

seriously the mutual causality between economic integration and international conflicts.
As an impetus for European countries to agree to implement first
economic union and then monetary union, the repentance over the past
military conflicts in Europe must have played an important role. This must
be an important reason that Germany, an economic leader country whose
central bank played a dominant role, agreed to give up some of its monetary hegemony in Europe.16 It may be the time to push forward economic
and monetary integration in East Asia, not just in order to realize economic
gains but also to prevent unfortunate military tragedies such as those seen
during World War II. The following statement by Wim Duisenberg (2005),
the founding President of the European Central Bank, well illustrates the
intention for peace in Europe:
European economic integrationin all its aspectsreflects the desire to integrate Europe politically, which, at least in my view, implies that European
economic and monetary integration is irreversible. This has been true from the
outset of the European integration process in 1952, when six countries established the European Coal and Steel Community (ECSC). The aim of the ECSC
was explicitly politicalto remove control of the two most important raw
materials for the production of heavy weaponry from states that had just fought
the bloodiest war in history. This drive for peace remained, quietly, the key
motivation behind further steps toward Europes economic integration, which
was regarded as the vehicle for achieving political integration. [] In particular,
Germany under Chancellor Helmut Kohl often linked monetary integration
with the objective of political union. Kohl once referred to the single currencys
success as a matter of war and peace.
NOTES
1.
The Asia Cooperation Dialogue (ACD), in its first meeting held in June 2002, agreed to
set up a working group on financial cooperation in order to set guidelines for the development of Asian bond markets. The second ACD meeting of June 2003 adopted the
Chiang Mai Declaration on the Asian Bond Market and the Asia Bond Fund Initiative.
In June 2003, the Executives Meeting of East Asia-Pacific Central Banks (EMEAP)
agreed to establish an Asian Bond Fund (ABF) of USD 1 billion, which was invested in
US dollar denominated government bonds issued by Asian countries. In June 2006, the
EMEAP central banks launched ABF2, which has invested USD 2 billion of EMEAP
central bank reserves in local currency denominated sovereign and quasi-sovereign
issues (see Ma and Remolona, Chapter 4 in this volume). In March 2003, the financial
ministers of ASEAN13 agreed to create working groups to promote the Asian Bond
Markets Initiative (ABMI) for the development of regional bond markets. In April
2003, the APEC convened its first meeting for the development of the Asian bond
market and emphasized closer cooperation to achieve better supervision and information sharing. For details, see Ryou et al. (2005).

82
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.

For the conceptual issues, see Ryou et al. (2005).
See Cargill and Royama (1988) for the notions of negotiable forms and market forms
of financial intermediation.
For the contingent nature of financial instruments and its implications, see Allen and
Gale (2000).
Ito (2003) defines Asian bonds as bonds issued by Asian institutions (government, corporations, financial institutions), denominated in an Asian currency, and sold, traded,
and settled in an Asian financial center (Tokyo, Singapore, Hong Kong). Investors are
expected to be mainly from regional economies (in particular, Japan, Singapore, and
Hong Kong), but others are also welcome.
Hamada (2004) contains a more detailed account of the conceptual issues as well as the
political economy aspects of economic and monetary integration.
The term monetary integration or monetary unification encompasses varying
degrees of integration, and it is usually used for the union of a group of countries and
not for all countries of the world.
For an attempt to analyze the nature of this cost saving, see Ryou (2003).
Feldstein (1992) makes the case that monetary integration is not a necessary condition
for the benefits of other forms of economic integration to be enjoyed.
For details on the relevant literature, see Baldwin (2006).
One view, held by the so-called liberals, is that increased economic interdependence
among countries fosters peaceful political relations by generating economic benefits and
increasing the costs of conflict. Critics of this view, the realists, argue that economic
interdependence generates political discord, rather than fostering political cooperation,
by intensifying competition between the states and by increasing dependence on strategic goods. Many others hold the view that economic interdependence has no systematic
influence on politics, maintaining that conflicts originate mainly from the political
arena. For a comprehensive survey, see Mansfield and Pollins (2001).
Another example of a monetary union that disintegrated is the Scandinavian Monetary
Union, which was formed between the neighboring Scandinavian countries of Sweden
and Denmark in 1873. Norway joined the union in 1875. A common currency unit,
the krone, which was based on gold, circulated as legal tender in member countries.
The union was extended to cover note circulation when note-issuing banks in Norway
and Sweden agreed to accept each others notes at par in 1894. Denmark joined this
agreement in 1900. From 1905, the conditions of note circulation were amended to
allow for a commission to be charged on foreign notes. Despite this added cost, the
joint circulation of notes continued. In 1914, with the outbreak of World War I, the
redemption of bank notes was suspended, and the monetary union effectively ended
(Nielsen 1933).
The West African Economic and Monetary Union (UEMOA, from its name in French)
is the longest surviving monetary union today. France exercises direct influence over the
affairs of the union through appointees to the Board of the Bank. France also exercises
considerable indirect influence on individual members through concessive assistance
and commercial links. Broughton (1991) suggested that UEMOA does not meet the
conventional criteria for an optimal currency union. Its survival should be understood
as a result of the parity between UEMOAs CFA franc and the French franc and of the
political leadership by France.
If the two countries propose differently, then an agreement will not be reached. In this
interpretation, the off-diagonal elements in the pay-off matrix will be (0, 0). If the use of
one countrys money is imposed on the other country, then the situation will be a definite
deterioration as expressed in the pay-off matrix.
Consider the case of foreign direct investment. Investors will worry less about the possibility of hostage effect, or hold-up effect, for example.
Some nationals might even have had psychological resistance to giving up the role
of being the country that issued the dominant currency in Europe. See Jonas et al.
(2005).

83
REFERENCES
Alesina, Alberto F., and Robert J. Barro (2002). Currency Unions. Quarterly
Journal of Economics 117: 40936.
Alesina, Alberto F., Robert J. Barro, and Silvana Tenreyro (2002). Optimal
Currency Areas. NBER Working Paper no. 9072, National Bureau of Economic
Research, Cambridge, MA.
Allen, F., and D. Gale (2000). Comparing Financial Systems. Cambridge, MA:
MIT Press.
Axelrod, Robert (1981). The Evolution of Cooperation. Science 211 (4489):
13906.
Baldwin, Richard E (2006). The Euros Trade Effect. ECB Working Paper no.
594, European Central Bank, Frankfurt.
Broughton, J.M. (1991). The CFA Franc Zone: Currency Union and Monetary
Standard. IMF Working Paper no. 91/133, International Monetary Fund,
Washington, DC.
Cargill, Thomas F., and Shoichi Royama (1988). The Transition of Finance in Japan
and the United States. Stanford, CA: Hoover Institution Press.
Corden, W. Max (1972). Monetary Integration. Essays in International Finance
no. 93, Princeton University, Princeton, NJ.
Duisenberg, Wim (2005). Europe, An Engine of Peace. www.project-syndicate.
org/commentary/duisenberg2/English.
Eichengreen, Barry (1998). Exchange Rate Stability and Financial Stability.
Open Economies Review 9:S1: 569607.
Feldstein, Martin (1992). The Case against EMU. The Economist June 13:
1922.
Feldstein, Martin (1997). EMU and International Conflict. Foreign Affairs
NovemberDecember 76 (6): 6074.
Feldstein, Martin (2000). The European Central Bank and the Euro: The
First Year. Remarks prepared for presentation at the annual meeting of the
American Economic Association, January 8.
Festinger, Leon (1957). Theory of Cognitive Dissonance. Stanford, CA: Stanford
University Press.
Frankel, Jeffrey, and Andrew K. Rose (2002). An Estimate of the Effect of
Common Currencies on Trade and Income. Quarterly Journal of Economics
117: 43766.
Fukuda, Shin-ichi, and Koichi Hamada (1987). Towards Implementation of
Desirable Rules of International Coordination and Intervention. In Yoshio
Suzuki and Mitsuaki Okabe (eds), Towards a World of Economic Stability.
Tokyo: University of Tokyo Press.
Gartzke, Eric, Quan Li, and Charles Boehmer (2001). Investing in the Peace:
Economic Interdependence and International Conflict. International
Organization 55: 391438.
Glick, Reuven, and Andrew K. Rose (2002). Does a Currency Union Affect
Trade? The Time-Series Evidence. European Economic Review 46, 112551.
Glick, Reuven, and Alan M. Taylor (2005). Collateral Damage: Trade Disruption
and the Economic Impact of War. Working Paper 2005-11, FRB San Francisco,
CA.
Goto, Junichi (2002). Economic Preconditions for Monetary Integration in East
Asia. Manuscript, Kobe University, Kobe, Japan.

84
Goto, Junichi, and Koichi Hamada (1996). Regional Economic Integration and
Article XXIV of the GATT. Economic Growth Center Discussion Paper no.
754, Yale University, New Haven, CT.
Hamada, Koichi (1985). The Political Economy of International Monetary
Interdependence. Cambridge, MA: MIT Press.
Hamada, Koichi (2004). A Calculus of Participation Approach to Monetary
Integration in Asia. Paper presented at the Seoul Conference of Korea Money
and Finance Association.
Ito, Takatoshi (2003). Construction of Infrastructures for the Development
of Regional Bond Market. In Choong-Yong Ahn, Takatoshi Ito, Masahiro
Kawai, and Yung-Chul Park (eds), Financial Development and Integration in East
Asia. Seoul: Korea Institute for International Economic Policy.
Jonas, Eva, Immo Fritsche, and Jeff Greenberg (2005). Currencies as Cultural
SymbolsAn Existential Psychological Perspective on Reactions of Germans
Toward the Euro. Journal of Economic Psychology 26: 12946.
Kenen, Peter (1992). EMU After Maastricht, Washington, DC. Group of Thirty.
Kuroda, Haruhiko (2003). Overview of Financial and Monetary Cooperation in
East Asia. In Choong-Yong Ahn, Takatoshi Ito, Masahiro Kawai, and YungChul Park (eds), Financial Development and Integration in East Asia. Seoul:
Korea Institute for International Economic Policy.
Mansfield, Edward, and Brian Pollins (2001). The Study of Interdependence
and Conflict: Recent Advances, Open Questions, and Directions for Future
Research. Journal of Conflict Resolution 45: 83459.
McKinnon, Ronald (1963). Optimum Currency Areas. American Economic
Review 53: 71224.
Mundell, Robert A. (1968). International Economics. New York: Macmillan.
Myerson, Roger B. (1991). Game Theory: Analysis of Conflict. Cambridge, MA:
Harvard University Press.
Nielsen, A. (1933). Monetary Union. Encyclopedia of Social Science, Vol. 10.
New York: Macmillan.
Oneal, John R., and Bruce M. Russett (1997). The Classical Liberals Were Right:
Democracy, Interdependence, and Conflict, 19501985. International Studies
Quarterly 41: 26793.
Rose, Andrew K., and Charles Engel (2002). Currency Unions and International
Integration. Journal of Money, Credit, and Banking 34: 106789.
Ryou, Jai-Won (2003). Transaction Costs and Welfare Effects of Currency
Unions. Manuscript, Konkuk University, Seoul.
Ryou, Jai-Won, Koichi Hamada, and Seung-Cheol Jeon (2005). The Asian Bond
Market: Issues and Tasks for Financial Cooperation in Northeast Asia. In
Bank of Korea (ed.), Northeast Asian Economic Cooperation and the Korean
Economy. Seoul: Bank of Korea, pp. 2163.
Takagi, Shinji (2002). Fostering Capital Markets in a Bank-based Financial
System: A Review of Major Conceptual Issues. Asian Development Review 19:
6797.

4.
Learning by doing in market

reform: lessons from a regional
bond fund
Guonan Ma and Eli Remolona1
4.1
INTRODUCTION
Since the Asian crisis of 1997, local currency bond markets in the region
have expanded rapidly; even so, they are still seen as not achieving their
potential to intermediate between domestic savers and borrowers. Capital
flows since the crisis show that Asians have been investing largely in lowyielding foreign assets and foreigners in higher-yielding Asian assets. While
some of these flows are consistent with portfolio diversification, the broad
pattern suggests that a sizeable part of financial intermediation is being
carried out abroad at significant cost. To bring such intermediation home,
Asian policymakers perceive a need for deeper and more liquid local bond
markets.
This perception has spawned a number of regional cooperative efforts
at market reform. In this chapter, we assess one such undertakingan
unusual one in that it involved the creation of an actual bond fund, with
financial contributions from the parties concerned. The regional group
involved is the Executives Meeting of East Asia and Pacific (EMEAP)
central banks.2 The fund they have created is called the Asian Bond Fund
2 (ABF2) which was first launched in mid-2005. We argue that because
the group set up an actual fund, its reform efforts enjoyed significant
advantages from learning by doing.
In what follows, we first provide an overview of the recent development
of local currency bond markets in East Asia and describe the main impediments in those markets. We then explain the structure and features of ABF2
in the context of various regional initiatives for bond market development.
Finally, we comment on the role of the ABF2 exercise in the reform of bond
market regulation, providing examples of market impediments that have
been reduced in the process of creating the fund and describing the mechanism put in place to provide incentives for reducing impediments further.
87

88
4.2
Developing bond markets in East Asia
AN OVERVIEW OF LOCAL CURRENCY BOND

MARKETS IN EAST ASIA
In the wake of the Asian crisis of 1997, there was concern at first that
the lack of well-developed local currency markets was forcing Asians to
borrow in foreign currencies, thus making their economies more vulnerable to a speculative currency attack. Since then, however, governments in
the region have accumulated such high levels of foreign exchange reserves
that the risk of another currency crisis has ceased to be an immediate
concern. Of greater interest to policymakers in the region has been the
concern that their stockpiles of official reserves may be a sign of inefficient
domestic intermediation, since the reserve assets seem to have been earning
much less than what Asians pay when borrowing abroad.
McCauley (2003) documents that the broad pattern of gross capital
flows since the Asian crisis has indeed been one of Asians investing in lowyielding foreign assets and foreigners investing in higher-yielding assets in
the domestic markets of the region. Part of the reason for this pattern is the
fact that foreign exchange intervention has resulted in a large accumulation
of reserves by central banks, and these institutions by their nature have a
safety bias (rather than a home bias). In effect, Asian savings are being
sent abroad only to return mostly in the form of private sector foreign
investment. Thus, financial intermediation is being carried out in the more
developed financial markets of Europe and North America, and the cost of
intermediation is reflected in the large differences in returns between Asian
assets abroad and foreign assets in Asia. In principle, the importance of
local information should lead to such intermediation being done at home.
If local currency bond markets in Asia functioned as intended, Asian policymakers now seem to be asking, could they not keep such intermediation
at home and in the process save their economies some of the borrowing
costs?
The Asian crisis did have economic consequences that themselves added
impetus to the development of local currency bond markets in the region.
As economies contracted, governments in the region found themselves
faced with budget deficits. Huge levels of funding were needed for largescale bank restructuring. And this time, the governments in the region
made an effort to eschew borrowing abroad, instead borrowing locally in
local currencies. As a result, the total amount of domestic debt outstanding
in East Asia, excluding Japan, has risen nearly threefold since 1998 (Figure
4.1).3 Hence, to the extent that the sheer volume of debt helps contribute
to financial market development, the Asian crisis has contributed to the
development of local currency bond markets in the region.
Other factors, however, seem to continue to hold back these local

Learning by doing in market reform
89
1800
Total
Private*
1600
1400
1200
1000
800
600
400
200
1998
1999
2000
2001
2002
2003
2004
2005
2006
Note: Asia includes the bond markets of the eight EMEAP members: China, Hong
Kong, Indonesia, Malaysia, the Philippines, Singapore, Korea and Thailand. *Excluding
the Philippines.
Source:
Bank for International Settlements.
Figure 4.1
Domestic debt securities outstanding in Asia (in billions of US

dollars)
markets. While the strength of issuance has been beneficial to the primary
markets, the secondary markets still suffer from a lack of liquidity. A
number of market impediments, both cross-border and local, remain.
Takeuchi (2004) provides a survey of cross-border impediments in Asia.
While most of these cross-border impediments are well known in the literature on capital controls, some local impediments have been relatively less
well appreciated and thus have received insufficient attention.
Capital controls typically include a ban on investments by foreigners or
on repatriation of principal or income on these investments, restrictions
on currency conversion, and other prohibitions and regulatory hurdles for
both issuers and investors. There is evidence that such controls still bind
in Asia. Ma and McCauley (2006), for example, show that there is still not
sufficient arbitrage to equalize onshore and offshore yields in various Asian
money markets. Specific examples include tight foreign exchange conversion rules associated with Chinas qualified foreign institutional investor
scheme governing foreign portfolio inflows, Indonesian restrictions on
domestic banks purchasing local currency instruments from non-resident
issuers, Koreas ceiling on resident investment in overseas securities and
properties, restrictions on foreign investors purchasing foreign currency
from the local banking system in the Philippines, and Thailands limits on
domestic banks local-currency lending to non-residents.

90
Local market impediments may take the forms of taxes, insufficient

market development, and an inadequate clearing and settlement infrastructure. Withholding taxes and taxes on financial transactions remain a
major cost to non-resident investors in some of the local markets. Within
EMEAP, some jurisdictions exempt only non-resident investors, and
some do so only for certain instruments. Insufficient market development,
such as the lack of a broad and diversified bond investor base, issuers and
products, is not conducive to liquidity. The lack of an independent and
competent rating industry and more consistent rating standards is a barrier
to more efficient pricing of credits. In many cases, inadequate disclosure
requirements, weak accounting standards, and insufficient creditor rights
protection add to the impediment list. For deep and liquid markets in Asia,
Jiang and McCauley (2004) identify as essential such factors as market size,
the diversity of the investor base, and the availability of hedging instruments. Inadequate custody, clearing, and settlement facilities also hamper
bond market development. In most Asian bond markets, it is still rather
cumbersome and sometimes impossible to clear and settle cross-border
bond transactions.
4.3
REGIONAL COOPERATIVE INITIATIVES AND

THE ABF PROJECT
East Asia has seen several initiatives in regional cooperation to develop

domestic bond markets. One focus of the various regional initiatives has
been to open up domestic markets to foreign portfolio investment by
removing both local and cross-border impediments. Among these initiatives, only the ABF2 effort involves actually setting up funds to invest in
the local currency markets. We argue in this section that this unusual
approach leads regulators to encounter impediments in their actual operations of establishing bond funds so as to make it an effective mechanism for
regulatory reform. In principle, private investors could have also lobbied
for reforms as they set up investment funds. However, market reform is a
public good in the sense that the benefits are enjoyed by many investors.
Individually, investors would be unwilling to bear the costs of lobbying
fully for such reforms, because they would not be able to keep the benefits
to themselves.
At least three major government-sponsored regional organizations
in Asia are pursuing initiatives to promote financial development in
the region (Battellino 2004). First, under the banner of the Asia-Pacific
Economic Cooperation (APEC) forum,4 the tripartite structure of the
PECC (Pacific Economic Cooperation Council) brings together many of

91
the initiatives of government, business, and academe. One such initiative

is focusing on the development of securitization and credit guarantees.
Second, within the Association of South East Asian Nations Plus Three
(ASEAN13) framework,5 six working groups have been set up to address
a broad range of issues related to local bond markets in Asia. These issues
comprise securitized debt instruments, credit guarantee mechanisms,
foreign exchange transactions and settlements, local currency issuance by
multinational corporations, local and regional credit rating agencies, and
coordinated technical assistance. The third organization consists of the
EMEAP central banks, which have been behind the setting-up of the Asian
bond funds (ABFs).
The initiatives of these three regional organizations tend to complement each other. For example, under ASEAN13 the Asian Bond Market
Initiative has helped secure approval from four countries to allow multilateral development institutions to issue bonds in their local currencies. The
Asian Development Bank (ADB), the International Finance Corporation
(IFC), and the World Bank have already taken advantage of this by each
issuing bonds denominated in Malaysian ringgit and Chinese renminbi.6
The ADB has also issued in Thai bahts and Philippine pesos.7 Hence, these
actions are adding to the supply of paper in the local bond markets, while
the ABF2 exercise is adding to the demand for this paper.
The ABF exercises are the first initiatives in which a regional organization has contributed financial resources to setting up actual bond funds
in Asia. In June 2003, the EMEAP central banks launched the first fund,
the Asian Bond Fund 1 (ABF1), pooling USD 1 billion in international
reserves from the 11 central banks and investing in US dollar denominated bonds issued by sovereign and quasi-sovereign borrowers in eight
of the EMEAP economies.8 By design, ABF1 was set up to be restricted
to EMEAP central bank investment only and thus would not be open to
other investors. Nonetheless, it was the first regional pooling of international reserves in Asia. In June 2005, the EMEAP central banks launched
the second fund, the Asian Bond Fund 2 (ABF2), which has invested USD
2 billion of EMEAP central bank reserves in local currency denominated
sovereign and quasi-sovereign issues in the same eight EMEAP markets.9
4.4
CONCERTED LEARNING BY DOING: THE

ABF2 EXERCISE
As mentioned, the ABF2 initiative differs from the others in that it involves
the actual creation of local currency bond funds. The earlier ABF1 had
limited itself to dollar-denominated issues that are mostly traded in more

92

USD 1 billion
USD 1 billion
EMEAPs Investment in
ABF2
Pan-Asian Bond Index

Fund (PAIF)
8 single-country funds
Local currency
bond markets
China
fund
Hong
Kong
fund
China
markets
Hong
Kong
market
Figure 4.2
Indonesia
fund
Jakarta
market
Korea
fund
Seoul
market
Malaysia
fund
Kuala
Lumpur
market
Philippines
fund
Manila
market
Singapore
fund
Singapore
market
Thailand
fund
Bangkok
market
ABF2 structure
developed international bond markets. Nonetheless, that first fund was

important because it afforded the EMEAP central banks an opportunity
to work together to build trust so as to foster cooperation and to further
develop financial markets in the region (Leung 2006).
ABF2 is actually nine separate funds: a Pan-Asian Bond Index Fund
(PAIF) and eight single-market funds (Figure 4.2). The PAIF is a singleindex bond fund investing in sovereign and quasi-sovereign domestic currency denominated bonds issued in the eight EMEAP markets. The PAIF
will be quoted in US dollars on an unhedged basis (for more details about
the structure of the PAIF, see section 4.5 below). The eight single-market
funds will each invest in the respective local currency bond market. Each
of the nine funds will replicate a bond index provided by a third party, the
International Index Company (IIC), which has been a major participant
in developing the highly successful credit default swap (CDS) indices in
Europe and North America.10 All eight single markets funds have been
registered locally in their own markets. Private sector fund managers have
been designated to individually manage the PAIF and the single-market
funds. The mandate of each fund manager is then to try to replicate the
relevant index and manage the fund passively.
ABF2 has been proceeding in two phases. In Phase 1, investments in
both the PAIF and the single-market funds are confined to the international reserves of the 11 EMEAP central banks, with a total sum of USD
2 billion. The EMEAP announced the formal launch of Phase 1 of ABF2
in June 2005. In Phase 2, through public offerings, the PAIF and the eight
single-market funds will be gradually opened up to other institutional and
retail investors, both within and without the EMEAP region.

Table 4.1
93
Cumulative growth of the ABF2 funds open to non-EMEAP

investors
Funds
PAIF
Hong Kong Bond Index Fund
Malaysia Bond Index Fund
Philippine Bond Index Fund
Singapore Bond Index Fund
Thailand Bond Index Fund
Fund size (USD

million)
Growth since
inception (%)
1156
214
125
57
220
121
14
55
26
3
35
20
Note: As of April 30, 2006, fund size is measured as net asset value, while growth is based
on the number of fund units. The IPO dates vary across different funds.
Source:
Bloomberg.
The pace and timing of the opening of these nine bond funds could
vary across jurisdictions. Between the formal launch of ABF2 in June
2005 and the time of writing, six of the nine ABF funds have opened to
the public. They are the PAIF, Hong Kong Bond Index Fund, Malaysia
Bond Index Fund, Philippine Bond Index Fund, Singapore Bond Index
Fund, and Thailand Bond Index Fund. Some of the remaining three
single market funds are expected to be offered publicly in the near future.
Undertaking the project in phases has allowed the central banks to
identify market impediments in stages and deal with them on a realistic
schedule.
As more ABF2 funds became available to the public, the total size of
the ABF2 funds, including both the USD 2 billion EMEAP seed money
and new non-EMEAP investments, increased to USD 2.47 billion as of
April 2006. Since their respective listings, the growth of individual funds
has varied noticeably, ranging from a rise of 3 percent to above 50 percent
(Table 4.1). Part of the reason for the differences is the different timing
of their initial public listings. For instance, the Hong Kong Bond Index
Fund was listed in June 2005, while the listing of the Philippine Bond Index
Fund took place only in late April 2006. Still, these developments suggest
reasonable market demand for Asian local currency index bond funds,
with the size of the Philippine fund in particular more than doubling since
inception.
The advantage of creating actual funds in the process of trying to reform
markets is that an important element of learning by doing is introduced.
Informal conversations with the key individuals involved suggest that in

94
setting up ABF2 the EMEAP central banks encountered myriad market

impediments, many of them seemingly small but each one critical to the
operation of the funds. Even when these officials had been aware of the
impediments before the exercise, they found some of them to be more significant than initially thought. By undertaking the initiative as a group, the
EMEAP central banks improved their understanding of specific impediments in their local bond markets and came up with ways to overcome
them. Seeing that the authorities in neighboring jurisdictions had already
instituted certain market reforms encouraged the relevant authorities to
fast-track their own initiatives. Often the central bank officials worked
with their counterparts at the finance ministry or the securities regulator
to deal with the impediments.
Recognizing that their job is far from complete, the central banks have
also agreed on an incentive mechanism for further reducing market impediments. In particular, the scheme for allocating the portfolio to the various
local markets will take market impediments into account: the portfolio
weight in ABF2 for an individual market rises as cross-border and local
market impediments in that market are removed (see section 4.7 for a more
detailed discussion).
4.5
DESIGNING A LIQUID PAN-ASIAN BOND

INDEX FUND11
The Pan-Asian Bond Index Fund (PAIF) is designed to allow institutional

and retail investors to gain access to Asias local bond markets in a simple
and transparent manner. The main challenge was to ensure that investors
benefit from the lower transaction costs associated with passively tracking
a regional index rather than active management of a portfolio. The specific
index is the iBoxx Pan Asia Index, which is constructed to be transparent
and to comprehensively cover the eight local Asian bond markets. The
index will be quoted in US dollars on an unhedged basis. In practice, the
fund will replicate the index closely by holding a selected set of issues rather
than all the constituents of the index. Such fund management, however,
requires an infrastructure of risk management and analysis of risk factors
(Cheung 2006).
The funds regional scope meant that there was no natural domicile for
it in any of the EMEAP jurisdictions. Tax, legal, operational, and marketing considerations suggested that a Singapore-based unit trust listed on the
Hong Kong Stock Exchange would offer the best option for all investors,
and this is how it was established.
The fund is open-ended as well as exchange-listed. Hence, investors are

95
Investors
Buying and selling
Buying and selling
Stock exchange
Participating dealers
Subscription
and redemption
Liquidity
Arbitrage
PAIF trustee
Market-makers
Primary market
Secondary market
Figure 4.3
Two ways to trade the PAIF
able to trade the PAIF in two ways. First, as shown in Figure 4.3, investors can go to fund trustees through dealers to buy or redeem units at that
days closing net asset value, thus engaging in a transaction in the primary
market. Second, they can buy or sell units on the stock exchange, thus
trading in the secondary market. As has been the case with other funds
previously launched in the region, there are some restrictions on trading
PAIF units in the primary market so as to concentrate liquidity in the secondary market. Nonetheless, the primary market continues to provide an
important means for arbitrage to ensure that secondary market prices stay
in line with the funds net asset value.
In the primary market, the PAIF follows the participating dealer
model. This model limits daily subscriptions and redemptions only to
dealers who have signed an agreement with the fund manager. To help the
manager deal with cash inflows and outflows, the participating dealers may
only transact a minimum size. For cash transactions, there is a limit on the
total daily volumes, and the manager charges a dilution fee. The limit is
waived if transactions are in exchange of a basket of bonds. These transactions are known as in-kind subscriptions or redemptions. The in-kind
facility makes the PAIF similar to an exchange-traded fund (ETF), the main
difference being that in-kind dealing is more formalized with an ETF.
In the secondary market, the fund manager has appointed marketmakers to provide liquidity in the trading of units on the stock exchange.
The market-makers are expected to maintain tight bid and offer quotes on

96
the exchange and to seize arbitrage opportunities by closely monitoring the

funds net asset value and comparing it with the prices on the exchange.
To help them provide liquidity, the market-makers have priority in the
primary market for trades up to a specified limit per day, and they may
borrow fund units from institutional investors.
The opening of the PAIF to the public took place at the end of June
2005. Before listing the fund on the exchange, a unit trust was set up by
the EMEAP central banks with an in-kind transfer of the equivalent of
around USD 1 billion in local currency bonds, which was purchased during
Phase 1 of ABF2. The fund may be further enlarged through private placements by institutional investors, participating dealers, and market-makers.
It has been listed to allow all other investors to acquire the units on the
secondary market. This strategy is expected to keep costs low and avoid the
volatility usually associated with a sale-driven initial public offering. From
then on, listings on other EMEAP stock exchanges will be considered,
depending on demand from local investors and on whether local regulatory
rules permit it.
Once listed, the PAIF became the first low-cost, passively managed
investment fund invested in the eight EMEAP local bond markets. These
features should potentially make the PAIF attractive in the long term to
pension funds and retirement accounts seeking opportunities for diversification and favorable long-term returns.
4.6
MARKET IMPEDIMENTS ALREADY REDUCED
For such relatively small sums, the ABF2 initiative has apparently been
unusually effective in promoting the reform of local bond markets.12
Because of the other initiatives that are also under way, it is always difficult
to attribute regulatory changes to the ABF2 effort alone. Nonetheless,
many of the participants feel that the effort has made a significant difference. In this section, we can provide only a few illustrative examples of
reductions in impediments.
The most apparent area for reform has been in capital controls.
Malaysia, for example, has announced measures to liberalize its foreign
exchange market, so that it has now essentially restored the regime that
was in place before it imposed capital controls during the Asian crisis. The
Malaysian authorities have lifted all restrictions on non-resident hedging
activities. Companies controlled by non-residents now enjoy full access
to onshore ringgit credit facilities. Residents without domestic ringgit
borrowing can freely invest abroad. Finally, the Malaysians have permitted multilateral agencies to issue local currency bonds in the domestic

97
market and allowed these non-resident issuers to hedge exchange risks

with onshore banks.
Korea has also recently announced a series of measures to further
liberalize foreign exchange regulations. The limit on won borrowing by
foreign investors has been lifted from KRW 1 billion previously to KRW
10 billion. Moreover, Korean residents are now allowed to invest in
overseas property up to USD 1 million. In China, the investment quota
on portfolio inflows by qualified foreign institutional investors has been
raised from the initial USD 2 billion in 2003 to nearly USD 7 billion as of
May 2006. Measures have been announced to allow Chinese residents to
invest in overseas securities, subject to some quota. Chinas active participation in ABF2 also bodes well for the countrys willingness to simplify
its still extensive regulations on cross-border portfolio investment and to
lower hurdles for the still fragmented domestic bond markets regulated
by multiple authorities. In fact, the PAIF is the first foreign institutional
investor that has been given direct access to the Chinese interbank bond
market.13
Withholding taxes and other taxes are another area of reform. Thailand
has already granted non-resident investors withholding tax exemption
for income from investing in baht-denominated government bonds and,
in most cases, government-guaranteed bonds in the domestic market.
Malaysia has also announced the exemption of non-resident investors
from the withholding tax on the interest income received from investment
in ringgit-denominated debt securities onshore. In Korea, the withholding
tax on interest incomes that foreign investors earn on local currency bonds
has been reduced. So far, five of the eight EMEAP member markets have
either implemented or offered exemption from the withholding tax to nonresidents investing in local currency sovereign or quasi-sovereign issues.
In the Philippines, the documentary stamp duty on bond trading will be
removed with the introduction of the Philippine single-market fund.
One unexpected area of reform has been the legal accommodation of
national jurisdictions, so that a fund domiciled in one jurisdiction may be
sold in another. The PAIF, for example, is to be domiciled in Singapore
to take advantage of a host of factors including bilateral tax agreements
between Singapore and the other EMEAP members. However, it will be
initially listed in Hong Kong, in part to take advantage of the high degree
of liquidity and depth in that market. This combination is the first ever in
Asia, demanding a significant learning effort on the part of each regulatory
authority.14 To make the fund possible, the participating central banks and
the regulatory authorities concerned needed to cooperate in reconciling
divergent regulatory frameworks. Direct participation in local currency
bond markets by the EMEAP central banks has thus helped them to

98
further identify, understand the details of, and gauge the importance of
market impediments as well as to better appreciate the diversity of each
others regulatory frameworks. This appreciation should in turn set the
stage for further streamlining of market regulation in the region.
4.7
INCENTIVES TO FURTHER REDUCE

IMPEDIMENTS
The central banks involved in ABF2 have devised a mechanism to provide

incentives to further reduce impediments in their own local bond markets.
These incentives are built into the determination of the market weights in
the portfolio of the PAIF and the single-market funds. As the assigned
weight on a particular market increases, a larger portion of the investments
into these index bond funds will be allocated to that market.
Determining the Market Weights in ABF
The portfolio allocations of the PAIF and the initial EMEAP investment
in the eight single-market funds will be determined in large part by market
weights that take account of various factors. There are specifically four
such factors: the size of the local market, the turnover ratio in that market,
the sovereign credit rating,15 and a market openness factor. Starting from
an equal allocation for each local market, the allocation will be adjusted to
take account of these four factors. The allocation for a given market will be
adjusted upwards if the adjustment factors score better than the averages
for the eight markets. In the adjustment, market size, turnover ratio, and
credit rating will each carry a 20 percent weight. The greatest part of the
adjustment will be determined by the market openness factor, which will
be assigned a 40 percent weight.
The Market Openness Factor
IIC, the company generally responsible for the bond market indices to be
replicated by the nine ABF2 funds, has constructed a qualitative factor
that gauges the relative openness of the eight markets (IIC, 2005a). This
factor is measured by a so-called impediments index (Table 4.2): the
fewer the impediments present, the better the openness factor would score.
In creating the index, IIC consulted with a number of international and
domestic market participants through its Asian Index Committee and
Asian Oversight Committee, as a means to help ensure the credibility
and market acceptance of the indices (EMEAP 2005).

99
Table 4.2
Five sub-factors of the impediments index
Sub-factors
Description
Regulatory environment
Degree of investor protection and freedom of

action, such as restrictions on foreign investments,
currency and capital controls, etc.
Legal restrictions or laws discriminating against
specific investor classes, corporate governance
Tax treatment and other monetary duties
State of development of the local capital market and
its influence on asset management
Degree of sophistication and competition in back
office functionalities such as settlement, custody, etc.
Legal environment
Fiscal environment
Market infrastructure
Clearing and settlement
infrastructure
Note: The higher the score of each sub-factor, the fewer the market impediments present.
These scores are derived from polls among members of the iBoxx Asian Index Committee.
The impediments index is a weighted average of the five sub-factors.
Source:
International Index Company (2005a, 2005b).
Any assessment of market openness might be expected to take account

of such considerations as the absence of capital controls, the level of withholding taxes, the availability of hedging instruments, the facilities for
real-time gross settlement, and the ability to clear local bonds internationally. The higher the market openness factor assigned for a given market,
the more the portfolio allocation is adjusted in favor of that market. The
country weights, and thus the market openness factor, will be reviewed
every September. As impediments are removed, the changes can be
expected to be reflected in a rebalancing of the regional portfolio.
In constructing the market openness factor, the iBoxx Asian Index
Committee will treat the regulatory environment as the most important consideration. This sub-factor, which includes capital controls and
provisions for investor protection, will alone account for 25 percent
of the openness score. Notably, the two considerations that have been
most neglected in the academic literaturemarket infrastructure and the
payment systemwill together receive a weight of 37.5 percent. The legal
and fiscal environments will account for the remaining 37.5 percent.
An Illustration
Figure 4.4 shows the effects of such weight adjustments. It compares
the weights based on raw market capitalization data and the adjusted
weights in the PAIF portfolio at the time of launch. The allocation after

100

A. Market weights based on
raw capitalization data
B. Market weights based on

the iBoxx PAIF index
Thailand
Singapore
Thailand
Philippines
Malaysia
China
China
Singapore
Hong
Kong
Philippines
Indonesia
Hong Kong
Korea
Note:
Source:
Indonesia
Malaysia
Korea
Weights add up to 100% in total.

International Index Company.
Figure 4.4
Market weights for the Asian Bond Fund
adjustments differs considerably from the weights on the basis of raw

market capitalization data. The allocations in the PAIF to China and
Korea fall noticeably below their capitalization-based weights, while the
remaining six markets gain. In particular, the Hong Kong and Singapore
allocations rise more than fourfold.
There are at least two ways in which the ABF2 indices will help the
development of Asian local currency bond markets. First, they provide a
dynamic mechanism to encourage the eight EMEAP members to continue
their efforts at market liberalization. As discussed above, market openness
is the single most important adjustment factor in the allocation weights,
and these will be reviewed annually. As new market liberalization measures are introduced, the market openness factor would score better, thus
increasing the weights on the market in question. Second, the improved
transparency, replicability, and credibility of these market indices will
provide the kinds of benchmarks that have proven useful elsewhere for the
development of markets in corporate bonds.
4.8
CONCLUSION
The ABF2 initiative is a regional cooperative effort aimed at fostering local

currency bond markets in Asia. It differs from other such efforts in that it

101
involved actually setting up bond funds. Hence, it contains an important

element of learning by doing, which has enabled EMEAP to identify in
detail significant market impediments that had not been well appreciated
before. The process has already helped ease various market impediments,
both cross-border and local. The exercise also provides incentives to
further reduce market impediments.
An important test for the exercise will be whether it sets the stage for
the development of local currency markets in corporate bonds. Already,
ABF2 is bringing new instruments to the local markets. As a listed openended index fund, the PAIF is a relatively low-cost, low-denomination,
and transparent fund, which would be potentially appealing to a broad
spectrum of institutional and retail investors. Thus, the PAIF may help
broaden both the investment menu and the investor base. The ABF
may help explore the potential role of ETFs. As of April 2006, six of the
eight single-market funds have been exchange-listed.16 China is currently
working its own single-market fund to be an ETF in the second phase of
ABF2. The introduction of these funds along with a set of transparent and
replicable benchmark indices for Asian local currency bond markets may
facilitate the development of other fixed income and derivative products,
including corporate bonds and credit default swaps.
NOTES
1.
2.
3.
4.
5.
6.
We thank Claudio Borio, Norman Chan, Guy Debelle, Mr Gudmundsson, Robert

McCauley, Frank Packer, Sakkapopp Panyanukul, Ramona Santiago, Atsushi
Takeuchi, Philip Wooldridge, and Sunny Yung for helpful discussions. We have
also benefited from the comments and suggestions of participants at the HWWA/
HWWI Conference on East Asian Monetary and Financial Integration in Hamburg in
December 2005. The views expressed here are those of the authors and do not necessarily
reflect those of the Bank for International Settlements.
The 11 EMEAP central banks and monetary authorities are the Reserve Bank of
Australia, Peoples Bank of China, Hong Kong Monetary Authority, Bank Indonesia,
Bank of Japan, Bank of Korea, Bank Negara Malaysia, Reserve Bank of New
Zealand, Bangko Sentral ng Pilipinas, Monetary Authority of Singapore, and Bank of
Thailand.
For a review of bond market development since the 1997 crisis, see Committee on the
Global Financial System (1999), McCauley and Remolona (2000), Jiang and McCauley
(2004), Battellino (2004), and Sheng (2005).
APEC has 21 member economies: Australia, Brunei Darussalam, Canada, Chile, China,
Hong Kong SAR, Indonesia, Japan, Korea, Malaysia, Mexico, New Zealand, Papua
New Guinea, Peru, the Philippines, Russia, Singapore, Taiwan (China), Thailand, the
United States, and Vietnam.
The members of ASEAN are Brunei Darussalam, Cambodia, Indonesia, Laos,
Malaysia, Myanmar, the Philippines, Singapore, Thailand, and Vietnam. The 13 are
China, Japan, and Korea.
There have been four issues of supranational local currency bonds in Malaysia since
late 2004: MYR 400 million by the ADB, MYR 500 million by the IFC, and MYR

102
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.

760 million by the World Bank. The ADB issue in Thailand in 2005 amounted to THB
4 billion. The ADB came back to Malaysia for a second ringgit offering of MYR 500
million in 2006. In China, IFC and ADB issued so-called panda bonds worth RMB 1.13
billion and RMB 1 billion in 2005.
ADB issued PHP 2.5 billion and THB 4 billion bonds, respectively, in the Philippines
and Thailand in 2005.
The three EMEAP countries in which the ABFs will not invest are Australia, Japan, and
New Zealand.
The Bank for International Settlements (BIS) acts as manager for ABF1 and as administrator for ABF2.
CDS indices are now the most actively traded instruments in credit markets. The main
CDS index for Europe is DJ iTraxx and for North America DJ CDX. Both are the
result of a merger between two competing families, Trac-x and iBoxx. See Amato and
Gyntelberg (2005).
This section draws heavily from a box written by Pierre Cardon in Ma and Remolona
(2005).
In this respect, the small sums involved help in that they avoid the problem of a passive
investor taking away from the market too much of the available volume of tradable
instruments.
Previously, qualified foreign institutional investors were allowed to directly invest in
bonds and stocks traded on the smaller Shanghai Stock Exchange and Shenzhen Stock
Exchange.
In the future, the PAIF may be listed in another market in the region, and some of the
eight single-market funds could be managed in jurisdictions other than those where they
are registered and listed.
Local currency long-term debt ratings of the three international rating agencies (Fitch,
Moodys, and Standard & Poors) are applied.
The PAIF, and the single-market funds of Hong Kong, Singapore, Malaysia, Thailand
and the Philippines.
REFERENCES
Amato, Jeffrey D., and Jacob Gyntelberg (2005). CDS Index Tranches and the
Pricing of Credit Risk Correlations. BIS Quarterly Review March: 7387.
Battellino, Ric (2004). Recent Developments in Asian Bond Markets. Speech
given at the 17th Australian Finance and Banking Conference, December.
Cheung, Hon (2006). The ABF2 Funds: Pushing the Frontiers in Asian Bonds.
Asiamoney May: 3641.
Committee on the Global Financial System (1999). How Should We Design
Deep and Liquid Markets? The Case of Government Securities, October. Basel:
CGFS.
Executives Meeting of East Asia and Pacific Central Banks (2005). The Asian
Bond Fund 2 Has Moved Into Implementation Phase. www.emeap.org/
press//12May05.htm.
International Index Company (2005a). iBoxx ABF Index Family Guide, Version
2.0, April. www.indexco.com/news/Attach/DM73/iBoxx_ABF_IndexGuide.
pdf.
International Index Company (2005b). iBoxx ABF Bond Indices Launched.
www.indexco.com/news/Attach/DM74/iBoxx_ABF2_BondIndices.pdf.
Jiang, Guorong, and Robert N. McCauley (2004). Asian Local Currency Bond
Markets. BIS Quarterly Review June: 6779.

103
Leung, Julia (2006). Developing Bond Markets in Asia: Experience With ABF2.
BIS Papers No. 26, Develop Corporate Bond Markets in Asia, March: 749.
Ma, Guonan, and Robert N. McCauley (2006). Are Chinas Capital Controls Still
Binding. Paper presented at the Seoul National University/Korea International
Economic Policy Institute Conference China and Emerging Asia: Reorganising
the Global Economy, Seoul, May.
Ma, Guonan, and Eli M. Remolona (2005). Opening Markets Through a Regional
Bond Fund: Lessons from ABF2. BIS Quarterly Review June: 8192.
McCauley, Robert N (2003). Capital Flows in East Asia Since the 1997 Crisis.
BIS Quarterly Review June: 4155.
McCauley, Robert N., and Eli M. Remolona (2000). Size and Liquidity of
Government Bond Markets. BIS Quarterly Review November: 528.
Sheng, Andrew (2005). Corporate Debt in Asia and Asian Financial Market
Development. Speech given at the conference on Advancing East Asian
Integration, March 34.
Takeuchi, Atsushi (2004). Study of Impediments to Cross-border Investment and
Issuance in Asian Countries. Interim Report for the ASEAN13 Asian Bond
Market Initiative, November.

5.
Currency denomination in Asian

bond markets
Eiji Ogawa and Junko Shimizu1
5.1
INTRODUCTION
One of the main features of financial crises, especially those that occurred
in the 1990s around the world, is that they tend to spill over from one
country to neighboring countries. This was the case during the Asian currency crisis of 199798, when contagion throughout the regional economies
was fuelled by strong trade linkages among the East Asian countries. The
Asian currency crisis brought about a strong awareness of the necessity for
regional cooperation, and, accordingly, several policy proposals have been
developed for the strengthening of regional financial cooperation.
Kuroda and Kawai (2003) proposed a more effective surveillance
process and considered the option of creating a common pool of foreign
exchange reserves in order to allow flexible financial support during
times of crisis and contagion, which would also reduce the problem of
moral hazard. The monetary authorities in East Asian countries in 2000
established, and recently have augmented, the Chiang Mai Initiative, a
swap agreement meant to prevent future currency and financial crises by
boosting short-term liquidity.2 One approach to strengthening regional
financial markets and avoiding future financial crises in East Asia that
we feel deserves more investigation is the call for a more efficient regional
bond market.
Ever since the Asian currency crisis, East Asian monetary authorities
have recognized the underlying problems caused by a double overdependence on the banking sector in their financial systems on the one hand and
on the US dollar in their currency systems on the other hand. Even though
there is an abundance of both savings and profitable investment opportunities in East Asian emerging market countries, the inefficiency with which
savings and investments are matched within East Asia through regional
financial markets has proven to be a problem. East Asian debtors (but also
creditors) have been facing foreign exchange risks because they are only
able to denominate liabilities (or assets, respectively) in foreign currency
104

105
when they borrow (or lend) internationally. The establishment and development of regional bond markets in East Asia has been proposed in order
to directly circulate regional savings into regional investments (Ito 2004,
Ito and Park 2004). In addition, the possibilities of introducing a common
currency basket unit and even a common currency in East Asia have been
discussed.3
The monetary authorities of East Asian countries have already started
promoting the development of regional bond markets through various initiatives such as the EMEAPs Asian Bond Fund4 and ASEAN13s Asian
Bond Market Initiative.5 Recent discussions of the Asian Bond Market
Initiative have focused on the choice of denomination currency for bond
issues and the establishment of a credit guarantee and rating agency. One
choice is to denominate regional bond issues in US dollars. This is what
the EMEAP central banks did when they first launched the Asian Bond
Fund 1 (ABF1) in June 2003 as a basket of US dollar denominated bonds
issued by Asian sovereign and quasi-sovereign issuers in EMEAP economies (excluding Australia, Japan, and New Zealand). Ito (2003) proposed
another option, namely, an Asian bond designed as a fund of the local
currency denominated bonds issued by governments of East Asian countries. The EMEAP has subsequently worked to extend the ABF concept to
bonds denominated in local currencies, which were launched as the ABF2
in June 2005.6
In previous research, Ogawa and Shimizu (2002, 2004) assessed the effectiveness of basket currency denominated bonds in East Asian countries
from the points of view of bond issuers and of foreign investors. Conducting
empirical analyses, we compared between Asian bonds denominated in
terms of a G3 currency basket (a currency basket composed of the US
dollar, the euro, and the Japanese yen) with those denominated in terms
of a single currency (the US dollar, the euro, the Japanese yen, or one of
seven East Asian currencies) in terms of both relative risk and liquidity.
The results showed that issuing currency basket denominated bonds would
reduce foreign exchange risk for bond issuers in the East Asian countries
under investigation, with the exception of the dollar-pegging countries of
Malaysia, China, and Hong Kong.
Shimizu and Ogawa (2005) considered another type of currency basket,
the Asian Monetary Unit (AMU), which is composed of the currencies of
the ASEAN14 countries (ASEAN13 and Hong Kong). As Kawai et al.
(2004) noted, an AMU composed of East Asian currencies would be equivalent to the G3 currency basket if all East Asian countries were to use the
G3 currency basket as a reference or target in conducting their exchange
rate policies, though this is not currently the case. Ogawa and Shimizu
(2005) proposed an AMU as well as AMU Deviation Indicators to assist

106
with the coordination of exchange rate policies by East Asian monetary

authorities. Shimizu and Ogawa (2005) investigated the risk properties of
AMU denominated Asian bonds by comparing them with those of local
currency denominated bonds issued in East Asian countries.
Building on our previous empirical research into the choice of denomination currency for Asian bonds, in this chapter we focus on the risk
properties of AMU denominated bonds as well as the expected liquidity of
AMU bond markets. Defining the AMU as a currency unit comprising a
basket of East Asian currencies, we simulate a currency basket composed
of the currencies of the ASEAN5 countries (Indonesia, Malaysia, the
Philippines, Singapore, and Thailand) plus those of Japan, China, Korea,
and Hong Kong. Our results indicate that the portfolio effects of holding
AMU denominated bonds can reduce foreign exchange risk for both US
and Japanese investors. These results, however, depend on the currency
systems of the East Asian countries.
This chapter is organized as follows: the next section analyzes foreign
exchange risk aspects of Asian bonds for bond issuers, and the third section
analyzes foreign exchange risk aspects of Asian bonds for foreign investors. The reduction of foreign exchange risk through the portfolio effects
of holding AMU denominated bonds is pointed out in the fourth section.
In the fifth section, we focus on liquidity aspects of Asian bond markets.
Finally, we summarize and conclude this chapter in the last section.
5.2
FOREIGN EXCHANGE RISKS FOR BOND

ISSUERS
Methodology and Data

We investigate how much foreign exchange risk bond issuers would face
when issuing bonds denominated in terms of various currencies, focusing
on the volatility of foreign borrowing costs.7 We compare the volatilities
of bonds denominated in terms of the currencies of nine East Asian countries (Singapore, Thailand, Malaysia, the Philippines, Indonesia, Taiwan,
Korea, Hong Kong, and China) against those of bonds denominated in
terms of three major currencies (the US dollar, the euro, and the Japanese
yen) and those of several types of bonds denominated in terms of currency
baskets composed of the three major currencies.
For the currency baskets, we consider several different currency shares
based on exchange rate movements and the trade weights of the nine
East Asian countries with the United States, Japan, and the euro area
countries.

107
We also consider two types of baskets. The first type of basket is a

common basket for the East Asian region as a whole. This kind of basket
is based on the proposal for a so-called AMU which was developed by
Kawai et al. (2004) and Ogawa and Shimizu (2005). Like the European
Currency Unit (ECU) under the European Monetary System during the
period from 1979 to 1998, the monetary authorities of East Asian countries
could consider creating a common regional unit of account that could be
used for trade invoicing, bond issues, and as a reference for their exchange
rate policies.
For this common basket, we calculate two different kinds of trade
weights, named trade-weights I and II, based on the calculation methods of
Ogawa and Kawasaki (2003). These trade weights reflect the shares of the
total amount of trade between ASEAN514 and the United States, Japan,
and the euro area countries. They differ in their inclusion or exclusion of
the trade between the nine East Asian countries and the rest of the world.
Trade-weight I is based on the share of total trade between the nine East
Asian countries and the United States, Japan, and the euro area countries
only. Trade-weight II also takes into account trade with the rest of the
world. In this case, trade with the rest of the world is added to the share of
the United States based on the assumption that trade with the rest of the
world is invoiced in US dollars. We calculate both trade weights by using
annual data and apply the average share of annual results during the period
from 1990 to 2002. The resulting currency basket shares based on tradeweights I and II are calculated as the US dollar: the Japanese yen: the euro
5 35.7 percent: 35.1 percent: 29.2 percent and 69.3 percent: 16.7 percent:
14.0 percent, respectively.
The second type of basket is country-specific, tailored specifically for
each East Asian country. In order to reflect the characteristics of each
nations trade relationship with the countries of the three major currencies, we calculate the basket shares based on trade intensity, using the
method put forward by Petri (1993).8 The index of trade intensity, which
can measure the bilateral trade linkages among countries (or regions), is
defined as follows:
Ij,k 5 (Tj,k/Tj)/(Tk/Tw),
(5.1)
where Ij,k is the index of trade intensity between country j and country k,
Tj,k is the volume of trade between country j and country k, Tj is the total
volume of trade of country j, Tk is the total volume of trade of country k,
and Tw is the total volume of trade in the world.
The index of trade intensity measures the closeness of bilateral trade
linkages between countries j and k by comparing bilateral trade volumes

108
with the respective countries trade volumes with the rest of the world. An
index close to unity can be interpreted as a neutral bilateral trade relationship. Indices less than or greater than one mean that the nations trade
relationship is biased in comparison with the nations trade with the rest
of the world. Values greater than one mean that the trade relationship
between countries j and k is biased toward stronger interdependence than
their trade with the rest of the world, while values less than one indicate that
the nations trade relationship is biased toward weaker interdependence
than their trade with the rest of the world.
Our index of trade intensity is based on 2000 trade data.9 The index
of trade intensity between every East Asian country and Japan is larger
than one. The index of trade intensity between Indonesia and Japan is the
greatest, while Hong Kong registers the lowest value of trade intensity with
Japan among East Asian nations. The indexes of trade intensity between
some East Asian countries and the United States are greater than one but
always smaller than those between the East Asian countries and Japan.
The indexes of trade intensity between the East Asian countries and the
euro area countries are all less than one.
Thus, we design three types of currency baskets with different weights for
each of the East Asian currencies. With the aim of indicating empirically
how the currency basket denominated bonds could contribute to reducing foreign exchange risks, we use historical data on exchange rates and
interest rates to simulate foreign borrowing costs by applying these basket
shares, which are calculated ex post. Applying these different types of currency basket shares, we can investigate not only whether the hypothetical
currency basket denominated bonds would more effectively contribute to
decreasing the volatilities of foreign borrowing costs but also which type
of currency basket would be more effective in decreasing the volatilities for
bond issuers in each of the East Asian countries.
Three-month money market interest rates and daily close exchange rates
are used to calculate the series of foreign borrowing costs for three-month
periods (90 days). Then we calculate the means and the standard deviations of these series for whole sample period. The standard deviation is
regarded as the volatility of the foreign borrowing costs. The three-month
foreign borrowing costs evaluated in terms of the issuers home currency
are calculated under the uncovered interest rate parity condition.
For example, the costs of three-month borrowing from the United States
for issuers in country A are calculated as follows:
e
{ 1US$ 1 (1 3 r3 month,us) 3 90/360 } 3 (A/US$) t190 days

1US$ 3 (A/US$) t
2 1 f 3 100 (%)
(5.2)

109
where (A/US$) t is the exchange rate between the US dollar and currency
A at period t.
Our empirical analysis covers the period from January 1, 1999 to March
31, 2006. We calculate three-month foreign borrowing costs for each day
during the period. As a result, we have 1888 observations for each type of
borrowing pattern during the period under analysis.10
First, we investigate three-month foreign borrowing costs for the
nine East Asian countries, namely, Singapore, Thailand, Malaysia, the
Philippines, Indonesia, Taiwan, Korea, Hong Kong, and China. For each
country, over the sample period we calculate eight different types of foreign
borrowing costs associated with issuing bonds denominated in terms of
four single currencies (the home currency, the US dollar, the euro, the
Japanese yen) and in terms of four types of currency baskets. The currency
basket denominated bonds are a kind of portfolio of bonds denominated
in terms of the US dollar, the euro, and the Japanese yen.
Results
Table 5.1 shows the means and standard deviations of 3 month foreign borrowing costs that debtors in each of the nine East Asian countries would
face when issuing bonds denominated in terms of the home currency, the US
dollar, the euro, the Japanese yen, and the eight different types of currency
baskets. We regard the foreign borrowing cost in terms of the US dollar to
be a benchmark and statistically test the differences between the means and
standard deviations of bonds denominated in terms of the US dollar and
bonds denominated in terms of currency baskets.11,12
We obtained two important results. First, the standard deviations of
foreign borrowing cost incurred by issuing home currency denominated
bonds are far lower than those incurred by issuing foreign currency denominated bonds for the nine East Asian countries. This implies that the foreign
exchange risk has a heavy weight in the total cost of borrowing by issuing
foreign currency denominated bonds. Second, the issuing of currency
basket denominated bonds could contribute to decreasing foreign borrowing costs as well as risks. The volatility of foreign borrowing costs faced
when issuing currency basket denominated bonds is lower than when US
dollar denominated bonds are issued for Singapore, Thailand, Indonesia,
Taiwan, and Korea, since the foreign exchange risks of the three major currencies partially offset each other. The costs of issuing the currency basket
denominated bonds are lower than those of issuing the US dollar denominated bond for all of the sample countries. The costs of issuing the currency basket denominated bonds are lower than those of issuing the home
currency denominated bonds; this is especially true for dollar-pegging

110
Table 5.1
Borrowing
from
Singapore
Max
Min
Mean
Std dev.
Thailand
Max
Min
Mean
Std dev.
Malaysia
Max
Min
Mean
Std dev.
Philippines
Max
Min
Mean
Std dev.
Indonesia
Max
Min
Mean
Std dev.
Taiwan
Max
Min
Mean
Std dev.
South
Korea
Max
Min
Mean
Std dev.
Hong Kong
Max
Three-month borrowing costs in East Asian Countries (%), Jan

1999Mar 2006
Single currency type
Home
currency
USD
0.766
0.109
0.392
0.193
6.627
4.593
0.631
2.267
13.848 18.174
9.445 10.129
0.944 0.053
4.322
3.942
6.797
3.727
0.541
1.902
5.481
3.491
0.587
1.516
11.214
5.557
0.272
2.300
1.938
0.250
0.631
0.369
14.210 16.051 31.819

8.312 12.676 11.779
1.048
1.340
0.379
3.799
4.836
5.409
19.198
8.958
0.955
3.528
16.602
8.514
1.003
3.370
25.092
7.726
0.657
4.183
1.425
0.713
0.811
0.133
1.665 14.741 17.765

2.886 10.365 10.811
0.704
1.090
0.083
0.543
5.366
4.864
9.046
5.012
0.658
2.782
4.544
2.767
0.682
1.308
11.698
6.817
0.346
3.260
4.181
1.075
1.945
0.566
17.204 26.763 26.522 15.081

7.320 12.567 13.115 10.467
1.868
2.221
1.254
1.813
3.695
5.917
6.324
4.375
8.500
2.295
3.280
1.404
1.450
0.225
0.727
0.429
2.068
0.870
1.325
0.348
1.871
Euro
Basket type
Yen
Trade- Trade- Tradeweight Ia weight intensityc

IIb
15.129 19.170
8.822 10.274
1.841
1.548
3.754
4.792
31.218 31.210 51.154 37.036 34.007 45.950

24.957 27.877 26.575 25.802 24.611 26.089
1.386
1.659
0.670
1.273
1.331
0.879
8.842
9.070
9.179
8.467
8.539
8.788
8.116 16.915 15.738
5.951 10.122 10.780
0.766
1.113
0.084
2.796
5.302
4.290
15.710 28.557
11.052 12.209
0.137
0.492
4.216
6.429
1.679
14.750
8.559
5.968
0.688
2.948
8.013
4.791
0.729
2.474
10.622
8.100
0.330
3.247
22.852
9.857
0.575
4.681
16.805
8.196
0.054
4.212
16.176
8.356
0.097
3.959
15.793
7.896
0.253
3.870
17.897
9.043
4.541
11.668

Table 5.1
Borrowing
from
Min
Mean
Std dev.
China
Max
Min
Mean
Std dev.
111
(continued)
Single currency type
Home
currency
USD
Euro
Basket type
Yen
Trade- Trade- Tradeweight Ia weight intensityc

IIb
0.002
0.753
0.582
1.006 10.351 10.821

0.768
1.152
0.145
0.514
5.334
4.829
4.901
0.721
2.741
2.150
0.745
1.259
6.825
0.439
3.355
0.697
0.428
0.470
0.067
1.609
14.737 17.763
1.565 10.377 10.813
0.665
1.053
0.047
0.603
5.399
4.914
9.041
5.755
0.621
2.836
4.539
3.436
0.644
1.375
10.553
6.075
0.359
3.034
Notes: All data calculated by authors. Sample period is 1/1/19993/31/2006. All exchange
rate and interest rate data are from Datastream.
a
The basket share of trade-weight I is USD : yen : euro 5 35.7 : 35.1 : 29.2.
b
The basket share of trade-weight II is USD : yen : euro 5 69.3 : 16.7 : 14.0.
c
The basket share of trade-intensity is USD : yen : euro 5 31.5 : 57.5 : 11.0 in Singapore, USD
: yen : euro 5 25.5 : 65.9 : 8.6 in Thailand, USD : yen : euro 5 29.0 : 62.9 : 8.2 in Malaysia,
USD : yen : euro 5 35.3 : 56.7 : 8.0.
countries. Accordingly, issuing currency basket denominated bonds could

make both foreign borrowing costs and their volatilities much lower than
those of issuing the US dollar denominated bonds in many cases.
Obviously, the foreign exchange risk is the smallest for bond issuers in
all of the nine East Asian countries when bonds are denominated in the
home currency. The currency basket denominated bonds have the second
lowest foreign exchange risks for bond issuers in all countries except for the
Philippines and the three dollar-pegging countries, Malaysia, Hong Kong,
and China. Thus, currency basket denominated bonds would reduce the
foreign borrowing costs for bond issuers in most East Asian countries.
The foreign borrowing costs of the currency basket denominated bonds
with country-specific shares based on trade intensity are the lowest for the
bond issuers in all nine countries. The volatilities of the currency basket
denominated bonds with the regionally used common shares based on
trade-weight II are the lowest for bond issuers in all East Asian countries
except Malaysia and Indonesia.13 The currency basket share based on
trade-weight II would most effectively contribute to decreasing the volatility for bond issuers in Singapore, Thailand, and Taiwan. The currency
basket share based on trade-weight I would most effectively contribute
to decreasing the volatility for bond issuers in Indonesia. The currency

112
basket share based on trade intensity would most effectively contribute

to decreasing the volatility for Korea. These results suggest that optimal
basket shares should be based on the trade relations between each of the
East Asian countries and its trading partner countries, although it might
be difficult to fix common basket shares across the East Asian countries in
order to establish an Asian bond market.
5.3
FOREIGN EXCHANGE RISKS FOR FOREIGN

INVESTORS14
We also calculate the returns of local currency denominated bonds for

foreign investors and evaluate their risk properties using yield data for
benchmark local bonds issued in the nine East Asian countries, namely,
the ASEAN5 nations, Japan, China, Korea, and Hong Kong. A partial
equilibrium approach is applied to analyzing an investor model under the
assumption that interest rates and exchange rates are taken as given (and,
therefore, returns and risks are taken as given). This means that investor
behavior affects bond returns and risks.
We treat bonds as an asset class in isolation from all other assets in order
to consider AMU denominated bonds, US dollar denominated bonds, and
local currency denominated bonds as objects of investment for foreign
investors. In addition to the standard deviation of bond returns, we also
use the Sharpe ratio as an indicator of risk, since we focus exclusively on
bond portfolios from the standpoint of sub-optimal principalagency
problems. Also, it is useful to measure risk-adjusted performance among
the various kinds of bonds. The Sharpe ratio is basically a measure of portfolio returns, and AMU denominated bonds themselves are considered to
be a type of portfolio fund.15
In order to make the source of returns clear, we divide bond returns into
interest returns and foreign exchange returns. We compare the returns of
the AMU denominated bonds with the returns of each of the local currency
denominated bonds in terms of interest rate and foreign exchange risks.
Risk Properties of Single Local Currency Denominated Bonds in East Asia
First, we calculate returns for international investors who evaluate their
returns in terms of the major currencies. We suppose that there are two types
of international investors: US investors, who evaluate their returns in terms
of US dollars, and Japanese investors, who evaluate their returns in terms of
Japanese yen. We show the average return, the standard deviation of returns,
and the ratio of returns to risk (the Sharpe ratio) for each of the bonds.

113
For an investor who evaluates returns in terms of US dollars, returns

on investing in bonds are calculated as follows: the investor exchanges an
initial fund of US dollars for a local currency at the relevant exchange rate,
purchases local currency denominated bonds at a price in terms of the local
currency, holds them for one month, sells them at their price in terms of
the local currency one month later, and exchanges the revenue in terms of
the local currency into US dollars.16, 17 The returns are then divided into
interest rate (bonds yield) returns and foreign exchange returns. A similar
calculation is conducted for investors who evaluate their returns in terms
of Japanese yen.
Our formula for calculating the value of a local currency denominated
bond in terms of US dollars for a one-month investment is represented as
follows:
BondValuet (USdollarEquivalent) 5
5 100 1 a
100 3 Et 3 (1 1 Yt)
Et11
100 3 Et 3 Yt
100 3 Et
2 100b
b1a
Et11
Et11
(5.3)
(5 principal 1 interest return 1 foreign exchange return)

where Et is the closing rate of the local currency against the US dollar at
month t, and Yt is the closing rate of bond yields on a monthly basis at
month t.
The interest returns for each of the bonds are calculated from the yield
data for local bonds, and the foreign exchange returns are calculated from
actual ex post returns that are uncovered by forward transactions at the
beginning of the month and realized when the bond values are converted
back into US dollars at the end of the month. Values for one-month investments in local currency denominated bonds in terms of Japanese yen are
calculated in the same way.
Table 5.2 shows bond values and returns for local currency denominated bonds for one-month investments evaluated in terms of US dollars
over the period beginning January 1, 1999 and ending March 31, 2006.
The data for the sample period clearly show that the standard deviations
of the value of the local currency denominated bonds issued by dollarpegging countries (Malaysia, China, and Hong Kong) are far lower than
those of bonds issued by the other countries. Their fluctuations are mainly
attributed to fluctuations in interest returns for bonds issued in terms
of the dollar-pegged currencies. The standard deviation of Hong Kong
dollar denominated bonds is the highest among the three dollar-pegging
countries.

114

106.58
94.21
100.39
0.39
1.94
0.20
0.695
0.240
0.438
0.097
4.52
6.023
6.280
0.044
1.941
0.02
103.78
96.90
100.35
0.35
1.24
0.28
0.394
0.154
0.293
0.057
5.11
3.468
3.384
0.057
1.230
0.05
5.533
6.417
0.241
2.234
0.11
0.808
0.316
0.507
0.135
3.77
2.24
0.33
106.12
94.18
100.75
0.75
Singapore Thailand South

Korea
5.000
9.677
0.313
1.866
0.17
1.278
0.422
0.767
0.170
4.51
1.89
0.24
105.96
91.16
100.45
0.45
Philippines
4.374
4.253
0.005
2.026
0.00
1.271
0.578
0.801
0.197
4.07
2.12
0.38
105.29
96.37
100.80
0.80
0.888
0.399
0.031
0.166
0.18
0.459
0.145
0.239
0.046
5.25
0.17
1.58
101.16
99.84
100.27
0.27
0.717
0.282
0.002
0.102
0.02
0.588
0.201
0.396
0.115
3.45
0.14
2.80
101.00
99.97
100.39
0.39
Indonesia Malaysia Hong

Kong
One-month bond value of Asian local bonds in terms of US dollar, Jan 1999Mar 2006
Bond value
Max
Min
Average (m)
Return (%), (m
100/100)
Std dev. (s)
Return/s
Interest return
Max
Min
Average (m)
Std dev. (s)
m/s
Foreign exchange return
Max
Min
Average (m)
Std dev. (s)
m/s
Table 5.2
1.471
0.014
0.035
0.176
0.20
0.554
0.000
0.336
0.235
1.43
0.22
2.54
101.98
100.45
100.55
0.55
China
7.108
6.404
0.023
2.577
0.01
0.167
0.043
0.117
0.027
4.42
2.58
0.05
107.27
93.70
100.14
0.14
Japan
115

Notes: All values are calculated by authors. Sample period is 1/1/19993/31/2006. All exchange rate and interest rate data are from Datastream.
Bond value for 1 month starts from 100 at the beginning of the month. Then it is invested into every Asian local bond for 1 month and
converted in terms of US dollars by using every foreign exchange rate against the US dollar at the end of month. All series without Indonesia and
China are from 2/1999 to 12/2003, and the number of the sample is 59. Indonesia and China start from 8/2002 and 6/2001, and the number of the
samples is 17 and 31, respectively.
Interest returns are calculated by every bond yield. Foreign exchange returns are the actual ex post foreign exchange related returns that are
uncovered by forward transaction at the beginning of the period and realized when the bond values are converted in terms of US dollars at the end
of the period.
116
The standard deviations of bond value for the non-dollar-pegging

country bonds are relatively high in comparison with those for the bonds of
dollar-pegging countries, and this volatility is mainly attributed to fluctuations in foreign exchange returns, which are far larger than those of interest returns. Among the non-dollar-pegging countries, Singaporean bonds
have the lowest standard deviation of bond value (1.24 percent), Japanese
bonds have the highest (2.58 percent), and Korean bonds have the second
highest value (2.24 percent).18 Indonesian bonds have the highest average
bond value (100.80), while Japanese bonds have the lowest (100.14). The
Sharpe ratios of the dollar-pegged currency denominated bonds are much
larger than those of the non-dollar-pegged currency denominated bonds.
Among the non-dollar-pegged currency denominated bonds, the Japanese
bonds have the lowest Sharpe ratio (0.05), and the Indonesian and Korean
bonds have the highest (0.38 and 0.33, respectively).
Table 5.3 shows bond values and returns for single local currency
denominated bonds for investments evaluated in terms of Japanese yen
over the sample period. In contrast to investments evaluated in terms of US
dollars, the standard deviations of bond values evaluated in terms of yen
are not so different between the dollar-pegging countries and the others,
but their average returns are much higher than those evaluated in terms
of US dollars. Excluding Japanese bonds, the Singaporean bonds have the
lowest standard deviation (2.13 percent), while Philippine bonds have the
highest (3.07 percent) among the local bonds. The fluctuations are mainly
attributed to fluctuations in foreign exchange returns. Korean bonds have
the highest average bond value (100.77), while Malaysian bonds have the
lowest (100.31). As for the Sharpe ratio, Korean bonds and Indonesian
bonds have the highest value (0.30), and Malaysian bonds have the lowest
(0.12), though the difference between them is small.
Risk Properties of AMU Denominated Bonds
When we simulate returns for AMU denominated bonds, which are composed of the nine East Asian countries government bonds, in order to
investigate their risk properties, we need to choose the allocation of weights
given to each countrys government bond.
An AMU denominated bond composed of an equal share of each of the
East Asian countries government bonds was used as a benchmark. This is
an abstraction from the case of real-world international bond funds, which
investors often use as an index for international portfolio investments, in
which fund composition weights are practically based on market capitalization or on the outstanding amount of bonds.19 The total value of currently
outstanding Japanese government bonds is much larger than that of the

117

106.11
91.49
100.42
0.42
2.63
0.16
0.709
0.236
0.438
0.098
4.48
5.631
8.994
0.018
2.616
0.01
105.92
94.41
100.37
0.37
2.13
0.18
0.394
0.154
0.293
0.057
5.16
5.567
5.916
0.082
2.131
0.04
Singapore Thailand
5.448
8.808
0.259
2.547
0.10
0.845
0.314
0.507
0.134
3.78
2.55
0.30
106.29
91.77
100.77
0.77
South
Korea
7.171
8.733
0.275
2.997
0.09
1.318
0.408
0.769
0.178
4.33
3.07
0.16
108.18
92.11
100.49
0.49
Philippines
4.676
6.319
0.044
2.404
0.02
1.258
0.580
0.801
0.197
4.07
2.49
0.30
105.60
94.29
100.76
0.76
6.814
6.638
0.073
2.579
0.03
0.468
0.143
0.240
0.047
5.13
2.59
0.12
107.07
93.59
100.31
0.31
Indonesia Malaysia
6.850
6.680
0.041
2.567
0.02
0.589
0.202
0.396
0.115
3.44
2.58
0.17
107.31
93.86
100.44
0.44
Hong
Kong
6.849
6.637
0.077
2.587
0.03
0.581
0.000
0.336
0.235
1.43
2.27
0.24
107.01
95.81
100.55
0.55
China
One-month bond value of Asian local bonds in terms of Japanese yen, Jan 1999Mar 2006
Bond value
Max
Min
Average (m)
Return (%),
(m 100/100)
Std dev. (s)
Return/s
Return on interest rate
Max
Min
Average (m)
Std dev. (s)
m/s
Foreign exchange
return
Max
Min
Average (m)
Std dev. (s)
m/s
Table 5.3
0.000
0.000
0.000
0.000
0.000
0.167
0.044
0.117
0.026
4.47
0.03
4.47
100.17
100.04
100.12
0.12
Japan
118

(continued)
Notes: All values are calculated by authors. Sample period is 1/1/19993/31/2006. All exchange rate and interest rate data are from Datastream.
Bond value for 1 month starts from 100 at the beginning of the month. Then it is invested into every Asian local bond for 1 month and
converted in terms of US dollars by using every foreign exchange rate against the US dollar at the end of month. All series without Indonesia and
China are from 2/1999 to 12/2003, and the number of the sample is 59. Indonesia and China start from 8/2002 and 6/2001, and the number of the
samples is 17 and 31, respectively.
Interest returns are calculated by every bond yield. Foreign exchange returns are the actual ex post foreign exchange related returns that are
uncovered by forward transaction at the beginning of the period and realized when the bond values are converted in terms of US dollars at the end
of the period.
Table 5.3
119
other East Asian countries, meaning that Japanese bonds would occupy an
extremely large part of the AMU denominated bonds if their composition
were decided on the basis of currently outstanding government bonds in
the market.
In addition to the equally weighted AMU, we calculate two more kinds
of AMUs, each with different country weights. For these, we use two kinds
of quarterly data on external debts classified by country and obtained from
the Quarterly Review of the Bank for International Settlements (BIS). The
first is International Debt Securities by Nationality of Issuer (Table 12A in
the Quarterly Review), which we call BIS1, and the other is International
Bonds and Notes by Country of Residence (Table 14B in the Quarterly
Review), which we call BIS2.
We next use these three types of AMU denominated bonds to simulate
their returns for US and Japanese investors in the same way that was done
for single local currency denominated bonds. Our formula for calculating
the value of AMU denominated bonds in terms of US dollars for a onemonth investment is represented as follows:20
9
100 3 Et,i 3 (1 1 Yt,i)
BondValuet (USdollarEquivalent) 5 a wi # a
b (5.4)
E
i51
t11,i
9
9
100 3 Et,i 3 Yt,i
100 3 Et,i
5 100 1 a wi # a
2 100b
b 1 a wi # a
E
E
i51
t11,i
i51
t11,i
(5 principal 1 interest return 1 foreign exchange return)

where Et,i is the closing rate of the exchange rate of the currency of country i
against the US dollar at month t and Yt,i is the closing bond yield of country
i on a monthly basis at month t.
Values of the AMU denominated bonds in terms of Japanese yen for
one-month investments are calculated in the same way as values in terms
of US dollars.
First, we compare the AMU denominated bonds with single local currency denominated bonds in terms of risk and Sharpe ratio. Table 5.4
shows bond values, returns, and risks evaluated in terms of US dollars
during the sample period January 1, 1999 to March 31, 2006. The standard
deviation of the equally weighted AMU denominated bonds is lower than
those of the individual government bonds in the six non-dollar-pegging
East Asian countries.
There are two reasons why the equally weighted AMU bonds risk is
lower. First, the equal weighting of the government bonds of all nine countries causes a large reduction in foreign exchange risk through a portfolio
effect. Second, the inclusion of the Japanese, Malaysian, and Singaporean

120
Table 5.4
One-month AMU denominated Asian bond value in terms of

US dollar, Jan 1999Mar 2006
AMU denominated AMU denominated AMU denominated
Asian bond
Asian bond
Asian bond
(equally weighted)
(BIS1)
(BIS2)
Bond value
Max
Min
Average (m)
Return (%),
(m 100/100)
Std dev. (s)
Return/s
Interest return
Max
Min
Average (m)
Std dev. (s)
m/s
Foreign
exchange return
Max
Min
Average (m)
Std dev. (s)
m/s
102.79
97.21
100.41
0.41
105.58
94.89
100.29
0.29
104.27
95.54
100.35
0.35
0.99
0.42
1.85
0.16
1.51
0.23
0.539
0.342
0.431
0.048
9.00
0.301
0.176
0.250
0.028
8.81
0.412
0.220
0.312
0.046
6.79
2.333
3.194
0.018
0.987
0.02
5.277
5.331
0.037
1.850
0.02
3.858
4.745
0.037
1.507
0.02
Notes: All values are calculated by authors. Sample period is 1/1/19993/31/2006. All
exchange rate and interest rate data are from Datastream.
There are three types of currency basket ratio. The equally weighted AMU denominated
Asian bond is composed with all countries bonds at the same ratio. The AMU
denominated Asian bond BIS1 is composed with all countries bonds by using the basket
weights based on the amounts of International Debt Securities by Nationality of Issuer
(BIS Table 12A). The AMU denominated Asian bond BIS2 is composed with all countries
bonds by using the basket weights based on the amounts of International Bonds and Notes
by Country of Residence (BIS Table 14B).
government bonds reduces interest risk because their interest rates are quite
stable. The AMU bonds Sharpe ratio (0.42) substantially exceeds that of
any of the government bonds of non-dollar-pegging countries. The results
suggest that for investors who evaluate their returns in terms of US dollars,
both foreign exchange risk and interest risk are reduced by investing in AMU
denominated bonds rather than in individual local currency denominated
bonds.
There exist differences in the standard deviations of bond returns among

121
the three types of AMU denominated bonds, with the standard deviation
of the BIS1 type of AMU denominated bonds having the highest value
(1.85 percent). This is because the weight accorded to Japanese government
bonds is 50 percent to 70 percent, which is much higher than the weight of
the others. With the exception of Singapore (1.24 percent), the standard
deviation of the BIS2 type of AMU denominated bonds (1.51 percent) is
lower than the returns on investment in the single currency denominated
bonds of the non-dollar-pegging countries. The share of Japanese government bonds in the BIS2 type of AMU denominated bonds is 30 percent
to 50 percent, which is still higher than the Japanese share in the equally
weighted AMU denominated bonds. The results suggest that a smaller
share of the Japanese government bonds is better for investors who evaluate their returns in terms of US dollars, because the returns of Japanese
government bonds are low with high fluctuations.21
Table 5.5 shows that the standard deviation of AMU denominated
bonds for investors who evaluate their returns in terms of Japanese yen is
lower than those of the single local currency denominated bonds, except
for Japanese bonds. Also, with the exception of Japanese and Korean
bonds, the Sharpe ratios of AMU denominated bonds are higher than the
single local currency denominated bonds, indicating that investors who
evaluate their returns in terms of Japanese yen can improve their return per
unit risk by investing in AMU denominated bonds. Among the three types
of AMU denominated bonds, the BIS1 type of AMU denominated bonds
has the lowest standard deviation (0.83 percent) and the highest Sharpe
ratio (0.34). This is because the BIS1 type of AMU denominated bonds
has the highest share of Japanese government bonds. The results suggest
that, for investors who evaluate their returns in terms of the Japanese yen,
a higher share of Japanese government bonds has a risk-reducing effect.
These investors could obtain less risky and more profitable outcomes by
investing in AMU denominated bonds, because they consist of both East
Asian government bonds, whose returns and risks are high, and Japanese
government bonds, whose returns and risks are quite low.
5.4
THE FOREIGN EXCHANGE RISK REDUCTION

EFFECT OF THE AMU DENOMINATED BONDS
In the previous section, we conducted a simulation analysis to show that

investors would be able to reduce overall risk by investing in AMU denominated bonds. In this section, we focus specifically on foreign exchange risk
and give a theoretical explanation of how investing in AMU denominated
bonds can contribute to reducing foreign exchange risk.

122
Table 5.5
One-month AMU denominated Asian bond value in terms of

Japanese yen, Jan 1999Mar 2006
AMU denominated AMU denominated AMU denominated
Asian bond
Asian bond
Asian bond
(equally weighted)
(BIS1)
(BIS2)
Bond value
Max
Min
Average (m)
Return (%),
(m 100/100)
Std dev. (s)
Return/s
Interest return
Max
Min
Average (m)
Std dev. (s)
m/s
Foreign
exchange return
Max
Min
Average (m)
Std dev. (s)
m/s
105.56
95.25
100.43
0.43
102.23
98.24
100.28
0.28
103.43
97.10
100.35
0.35
2.04
0.21
0.83
0.34
1.26
0.28
0.575
0.328
0.426
0.057
7.44
0.303
0.177
0.250
0.028
8.79
0.419
0.221
0.312
0.046
6.73
5.109
5.221
0.005
2.021
0.00
1.974
2.031
0.034
0.829
0.04
3.096
3.228
0.043
1.256
0.03
Notes: All values are calculated by authors. Sample period is 1/1/19993/31/2006. All
exchange rate and interest rate data are from Datastream.
There are three types of currency basket ratio. The equally weighted AMU denominated
Asian bond is composed with all countries bonds at the same ratio. The AMU
denominated Asian bond BIS1 is composed with all countries bonds by using the basket
weights based on the amounts of International Debt Securities by Nationality of Issuer
(BIS Table 12A). The AMU denominated Asian bond BIS2 is composed with all countries
bonds by using the basket weights based on the amounts of International Bonds and Notes
by Country of Residence (BIS Table 14B).
International portfolio diversification is widely practiced by investors

who seek to reduce their investment risks. Recently, international investors
have been turning to foreign markets to obtain greater scope for diversification than can be offered in domestic markets. However, when investing
in foreign currency denominated bonds it is important to consider not only
the portfolio effects on interest returns and risks but also foreign exchange
returns and risks.

123
To expand on this, we first use basic portfolio theory to explain the

reduction of investment risk by the portfolio effect. The return (Rp) on a
portfolio investment in bonds of country i (with return Ri) with a portfolio
| at time t is represented by
share w
n
|#R
Rp,t 5 a w
i,t
(5.5)
i51
| 5 (w ,w ,...,w ) .
where w
1 2
n
This portfolio has the following expected returns and variance:
mp 5 n a wimi
(5.6)
i51
s2p 5 a a sij
i,j
5 (w1,...,wn)
s11,s12,...,s1n
w1
( ( (
(
sn1,sn2,...,snn
wn
|TWw
|
5w
where mi is the expected value of Ri,s2i is the variance of Ri,sij is the covariance with Ri and Rj,rij is the correlation coefficient with Ri and Rj, and W is
the variancecovariance matrix of sij.
Then we separate the variance (V 2p) of the portfolio return into a sum of
variances and a sum of covariances as follows:
n
V 2p 5 a w2i s2i 1 2 a a wiwj sij

i51
(5.7)
i,j
The portfolio effect means that the variance of portfolio returns should
be smaller when investors increase bond diversity, because individual fluctuations in bond returns partly cancel out with each other. The variance of
returns on the equally weighted AMU denominated bonds is represented
as follows:
1 1 n
2
s2p 5 a a s2i b 1 2 a a sij
n n i51
n
i,j
(5.8)
1
n (n 2 1)
(average of sij)
5 (average of s2i ) 1
n
n2
In the above equation, the first term is the average foreign exchange risk
for each of the local government bonds. The AMU denominated bonds
are composed of the non-dollar-pegging as well as the dollar-pegging

124
countries government bonds. The former have a high foreign exchange

risk against the US dollar, while the latter have a near-zero exchange risk.
Accordingly, the foreign exchange risk of investing in AMU denominated
bonds is substantially lower than that of investing in individual local government bonds in non-dollar-pegging countries. The second term is almost
equal to the average covariance of exchange rate returns among the local
government bonds. The covariance (sij) is calculated as
sij 5 rij # si # sj
(5.9)
If the correlation coefficient rij is negative and, in turn, sij also is negative, portfolio investments reduce foreign exchange risk more. The AMU
denominated bonds consist of the dollar-pegged currency denominated
bonds and the non-dollar-pegged currency denominated bonds. Even the
currencies of the non-dollar-pegging countries in East Asia are highly correlated to movements in the US dollar. Correlation coefficients between
East Asian currencies, especially the dollar-pegged currencies and the
Japanese yen, tend to be negative in terms of foreign exchange returns.
Table 5.6 shows correlation coefficient matrices for one-month investments in government bonds in terms of US dollars. It shows that most of
the correlation coefficients between the dollar-pegging countries and the
non-dollar-pegging countries are negative. Accordingly, for investors who
evaluate their returns in terms of US dollars, the AMU denominated bonds
should be effective in reducing foreign exchange risk.
For investors who evaluate their returns in terms of Japanese yen, the
average foreign exchange risk in the first term of equation (5.8) could be
reduced by including Japanese government bonds in the AMU denominated bonds. The higher the share of Japanese government bonds, the lower
the foreign exchange risk of the AMU denominated bonds. Accordingly,
AMU denominated bonds are likely to have the portfolio effect of reducing
foreign exchange risk regardless of whether the investor evaluates returns
in terms of US dollars or Japanese yen.
5.5
LIQUIDITY IN ASIAN BOND MARKETS22
We compare the liquidities of bonds denominated in terms of the three

major currencies and the East Asian currencies. Because of constraints on
bidask spread data for bond markets across the sample countries, we use
the bidask spreads in foreign exchange markets as a measure of liquidity.23
We regard the foreign exchange bidask spreads as a measure of liquidity
in the Asian bond markets because bond issuers must conduct spot and

125

1.0000
0.7176
0.0986
0.1090
0.3942
0.4969
0.2604
0.1348
0.6403
Singapore
0.7176
1.0000
0.2196
0.1076
0.4022
0.5395
0.1306
0.1742
0.6928
Thailand
0.0986
0.2196
1.0000
0.3084
0.1445
0.3756
0.0268
0.1109
0.0444
Malaysia
0.1090
0.1076
0.3084
1.0000
0.0363
0.1661
0.0488
0.1536
0.0187
Philippines
0.3942
0.4022
0.1445
0.0363
1.0000
0.1477
0.0815
0.1226
0.2754
Indonesia
0.4969
0.5395
0.3756
0.1661
0.1477
1.0000
0.2567
0.0565
0.3680
Korea
0.2604
0.1306
0.0268
0.0488
0.0815
0.2567
1.0000
0.3148
0.3302
Hong Kong
0.1348
0.1742
0.1109
0.1536
0.1226
0.0565
0.3148
1.0000
0.0509
China
0.6403
0.6928
0.0444
0.0187
0.2754
0.3680
0.3302
0.0509
1.0000
Japan
Correlation matrix for one-month Asian government bond returns in terms of US dollar, Jan 1999June 2005
Source:
Authors calculations.
Note: We make the sample period up to June 2005 because Chinese bond returns in terms of the US dollar changed dramatically as a result of
their change of currency system.
Singapore
Thailand
Malaysia
Philippines
Indonesia
Korea
Hong Kong
China
Japan
Table 5.6
126
forward transactions on the foreign exchange market in order to convert

their foreign borrowings into their home currencies and because their
choice of bond denomination currency is affected by the bidask spreads
in the foreign exchange market.24
Bidask spreads are caused by three factors: (1) order processing costs,
(2) inventory holding costs, and (3) information costs for market making
(Hartmann 1998). These costs are lowered by a higher volume of trade.
Order processing costs are subject to economies of scale because the costs
of purchasing the electronic market information needed for processing
orders are fixed. The average costs of holding inventory diminish according
to the law of large numbers, as statistically independent orders increase.
Economies of scale are also present because of substantial fixed information costs for market making. Thus, the average bidask spread is reduced
by a large volume of trading in liquid markets.
Basically, bidask spreads for trades between the major currencies and
the US dollar and for those between frequently quoted major cross currencies, such as the euro and the Japanese yen, the UK pound and the
Japanese yen, the Swiss franc and the Japanese yen, and the UK pound and
the euro, are very narrow. On the other hand, direct exchanges between
so-called exotic currencies such as the other East Asian currencies may
cause some inconvenience and incur higher transaction costs, especially in
forward outright trading.
It is usually the case that the East Asian currencies are quoted against
the US dollar in foreign exchange markets and not against the euro or the
Japanese yen. Therefore, we calculate the cross rates of the East Asian currencies vis--vis the Japanese yen by using exchange rates of the US dollar
vis--vis the Japanese yen and the US dollar vis--vis East Asian currencies. We use the same procedure to calculate cross rates of East Asian currencies vis--vis the euro, using the exchange rates of the US dollar vis--vis
the euro and of the US dollar vis--vis East Asian currencies. In the case of
quoting the forward outright rate, we use the same procedure to calculate
cross swap rates. Accordingly, the bidask spreads for the forward outright
rates of the East Asian currencies vis--vis the euro or the Japanese yen are
wider in comparison with the other major currencies. This seems to be a
major reason why borrowers in the East Asian countries did not use swap
transactions to cover their foreign borrowings in terms of foreign currencies before the Asian currency crisis in 1997.
We compare the bidask spreads for the forward swap rates of the
seven East Asian currencies vis--vis the three major currencies, as well as
those of the euro and the Japanese yen vis--vis the US dollar. In order to
compare the bidask spreads, we express them in terms of percentage of
transaction-based cost. Transaction-based cost refers to the bidask spread

127
one must pay to deal one forward outright transaction. We calculate the
bidask spreads for eight East Asian currencies for forward transactions
with terms of one month, three months, and six months vis--vis the US
dollar, the euro and the Japanese yen. In order to compare these with the
bidask spreads for major traded currencies, we calculate spreads for the
euro and the Japanese yen vis--vis the US dollar as well as those for the
Japanese yen vis--vis the euro. We use the spot and forward rates from
the Bloomberg Currency Composite pages and the brokers page of Prebon
Yamane Asia Region on Bloomberg dated September 13, 2002, February
6, 2003, and May 18, 2006.
Table 5.7 shows the transaction-based bidask spreads. The bidask
spreads for all of the East Asian currencies vis--vis the US dollar are
lowest for all periods, while the bidask spreads for all of the East Asian
currencies vis--vis the euro are highest for the one-month and threemonth forward swap contracts. In addition, the bidask spreads for all of
the East Asian currencies vis--vis the US dollar are far higher than those
for the euro and the Japanese yen vis--vis the US dollar. The bidask
spreads for the euro and the Japanese yen vis--vis the US dollar are 3
to 4 basis points (bp)25 for one transaction, while those of the East Asian
currencies vis--vis the US dollar range from 5 bp for the Taiwan dollar to
almost 50 bp for the Philippine peso and the Indonesian rupiah vis--vis
the US dollar. The bidask spreads for all of the East Asian currencies vis-vis the euro and the Japanese yen are almost 3 to 4 bp higher than those
vis--vis the US dollar, and they are higher than those of the Japanese yen
vis--vis the euro, too. As for the differences resulting from terms, there is
not much difference between the bidask spreads for contracts with terms
of one month, three months, or six months for the Hong Kong dollar, the
Singapore dollar, the Thai baht, and the Korean won and the spreads for
contracts with those terms for the euro and the Japanese yen. On the other
hand, the bidask spreads for the rest of the East Asian currencies become
higher as the terms become longer.
Among the eight East Asian currencies, the Hong Kong dollar and the
Taiwan dollar have the lowest bidask spreads. The currencies with the
next lowest bidask spreads include the Singapore dollar, the Thai baht,
and the Korean won. The currencies with the highest bidask spreads
include the Malaysian ringgit, the Philippine peso, and the Indonesian
rupiah. It is interesting that the Malaysian ringgit had the highest onemonth forward contract bidask spread among the nine East Asian countries as of September 13, 2002. It seems that there is not much demand for
forward swaps in the Malaysian ringgit because the monetary authorities
of Malaysia have been adopting the dollar-peg system by pegging the spot
rate of the Malaysian ringgit to the US dollar.

128

6m
1m
3m
6m
(sample day: May 18,

2006)
Spreads (%)
1m
3m
(average)
Spreads (%)
6m
0.2982 0.3103 0.3354 0.1675 0.2969 0.5384 0.1988 0.2822 0.3672 0.2215 0.2965 0.4136
0.3168 0.3287 0.3499 0.1795 0.3207 0.5710 0.2017 0.2863 0.3736 0.2327 0.3119 0.4315
0.3016 0.3129 0.3404 0.1809 0.2963 0.5073 0.2019 0.2851 0.3701 0.2281 0.2981 0.4059
0.1415 0.1620 0.1384 0.0500 0.0710 0.0509 0.1108 0.1374 0.1388 0.1008 0.1234 0.1094
0.2763 0.2895 0.3158 0.1421 0.2737 0.5368 0.1927 0.2753 0.3579 0.2037 0.2795 0.4035
0.1601 0.1803 0.1526 0.0604 0.0872 0.0586 0.1137 0.1413 0.1443 0.1114 0.1362 0.1185
0.1450 0.1652 0.1455 0.0662 0.0848 0.0708 0.1138 0.1399 0.1409 0.1084 0.1300 0.1191
0.1129 0.1239 0.1240 0.0611 0.0646 0.0621 0.0359 0.0448 0.0528 0.0700 0.0778 0.0796
0.1194 0.1405 0.1171 0.0234 0.0409 0.0234 0.1049 0.1311 0.1311 0.0826 0.1042 0.0905
0.1313 0.1417 0.1373 0.0714 0.0806 0.0693 0.0384 0.0476 0.0565 0.0804 0.0900 0.0877
0.1168 0.1281 0.1327 0.0774 0.0787 0.0825 0.0394 0.0487 0.0576 0.0778 0.0852 0.0909
0.0907 0.1020 0.1020 0.0344 0.0344 0.0344 0.0298 0.0380 0.0444 0.0516 0.0581 0.0603
3m
1m
1m
6m
(sample day: Feb 6,

2003)
(sample day: Sept 13,

2002)
3m
Spreads (%)
Spreads (%)
Forward swap bidask spreads (% in transaction basis) in seven East Asian currencies against three major
currencies
Singaporian agst US
dollar
dollar
agst euro
agst JP
yen
average
Thailand
agst US
baht
dollar
agst euro
agst JP
yen
average
Malaysian agst US
ringgit
dollar
agst euro
agst JP
yen
average
Currency
Table 5.7
129

agst US
dollar
agst euro
agst JP
yen
average
Indonesian agst US
rupiah
dollar
agst euro
agst JP
yen
average
New Taiwan agst US
dollar
dollar
agst euro
agst JP
yen
average
South
agst US
Korean won dollar
agst euro
agst JP
yen
average
Philippine
peso
0.1135 0.1129 0.1452 0.1029 0.1061 0.1030 0.1110 0.1637 0.1630 0.1091 0.1275 0.1371
0.1322 0.1315 0.1604 0.1135 0.1229 0.1117 0.1145 0.1688 0.1719 0.1200 0.1410 0.1480
0.1169 0.1158 0.1505 0.1188 0.1189 0.1210 0.1129 0.1638 0.1587 0.1162 0.1328 0.1434
0.0804 0.1092 0.1672 0.0727 0.1049 0.1310 0.1000 0.1634 0.1634 0.0844 0.1258 0.1539
0.0914 0.0914 0.1246 0.0764 0.0764 0.0764 0.1056 0.1584 0.1584 0.0911 0.1087 0.1198
0.0988 0.1270 0.1804 0.0830 0.1207 0.1379 0.1030 0.1672 0.1703 0.0949 0.1383 0.1629
0.0843 0.1134 0.1759 0.0890 0.1190 0.1516 0.1029 0.1660 0.1630 0.0921 0.1328 0.1635
0.2437 0.4041 0.5037 0.1836 0.3067 0.4618 0.7627 0.8128 0.8105 0.3967 0.5079 0.5920
0.0581 0.0872 0.1453 0.0460 0.0748 0.1035 0.0942 0.1570 0.1570 0.0661 0.1063 0.1353
0.2639 0.4302 0.5365 0.1952 0.3287 0.4855 0.8000 0.8607 0.8679 0.4197 0.5399 0.6300
0.2443 0.3918 0.4729 0.1975 0.3094 0.4486 0.6989 0.7320 0.7180 0.3802 0.4778 0.5465
0.2130 0.2114 0.3962 0.2104 0.3924 0.5607 0.1950 0.1949 0.1955 0.2061 0.2662 0.3841
0.2230 0.3903 0.5018 0.1580 0.2821 0.4514 0.7892 0.8455 0.8455 0.3901 0.5060 0.5996
0.2321 0.2314 0.4172 0.2226 0.4170 0.5894 0.1993 0.2008 0.2043 0.2180 0.2831 0.4036
0.2157 0.2115 0.3885 0.2233 0.3896 0.5368 0.1955 0.1935 0.1921 0.2115 0.2649 0.3725
0.1914 0.1914 0.3827 0.1853 0.3706 0.5559 0.1902 0.1902 0.1902 0.1889 0.2507 0.3763
130

(continued)
6m
1m
3m
6m

2006)
Spreads (%)
1m
3m
(average)
Spreads (%)
6m
0.0668 0.0661 0.0664 0.0799 0.0902 0.0826 0.0173 0.0184 0.0225 0.0547 0.0582 0.0572
0.0261 0.0262 0.0308 0.0429 0.0442 0.0479 0.0096 0.0108 0.0135 0.0262 0.0271 0.0307
0.0505 0.0503 0.0542 0.0949 0.0726 0.0955 0.0228 0.0261 0.0277 0.0561 0.0496 0.0592
0.0408 0.0402 0.0362 0.0371 0.0463 0.0352 0.0086 0.0094 0.0117 0.0288 0.0320 0.0277
0.0690 0.0683 0.0680 0.1053 0.0888 0.1032 0.0254 0.0288 0.0313 0.0666 0.0620 0.0675
0.0543 0.0543 0.0625 0.1110 0.0864 0.1151 0.0264 0.0300 0.0325 0.0639 0.0569 0.0701
0.0282 0.0282 0.0321 0.0683 0.0425 0.0683 0.0168 0.0193 0.0193 0.0378 0.0300 0.0399
3m
1m
1m
6m
(sample day: Feb 6,

2003)
(sample day: Sept 13,

2002)
3m
Spreads (%)
Spreads (%)
Sources: Calculated by authors. All spot rates and forward rates are collected from Bloomberg currency composite pages and Prebon Yamane
Asia Region pages on sample days. Forward swap spreads are calculated by bid and ask spreads on both spot and forward rates.
Hong Kong agst US

dollar
dollar
agst euro
agst JP
yen
average
Euro
agst US
dollar
Japanese
agst US
yen
dollar
Japanese
agst euro
yen
Currency
Table 5.7
131
As explained earlier, an investment in the currency basket denominated bonds means a portfolio investment in the bonds denominated in
terms of the US dollar, the euro, and the Japanese yen. It is possible to
use weighted averages of the bidask spreads for the three currencies as a
proxy for the liquidity of the currency basket denominated bonds. Except
for the Indonesian rupiah and the Malaysian ringgit, the differences in the
bidask spreads between the US dollar and the currency basket are around
0.03bp to 0.08 bp for one-month swap transactions, 0.02 bp to 0.03 bp for
three-month swap transactions, and 0 bp to 0.01 bp for six-month swap
transactions.
Thus, we use data on the bidask spreads of swap transactions to compare
the liquidities of the three major currencies and the currency basket. Our
findings show that the US dollar has the highest degree of liquidity for all
of the nine East Asian currencies. However, the differences between the US
dollar and the currency basket are not very large. This is especially true for
three- and six-month swap transactions, though the differences are larger
for one-month swap transactions.
Next, we choose two East Asian currencies, the Singapore dollar and
the Thai baht, both of which are relatively actively traded in East Asian
markets. We calculate forward bidask spreads vis--vis the US dollar,
the euro, and the Japanese yen from 2000 to 2006 for the two currencies.
Table 5.8 shows the change in transaction-based forward bidask spreads
from 2000 to 2006.
Comparing the same calculations for the bidask spreads for the euro
and the Japanese yen vis--vis the US dollar and the Japanese yen vis-vis the euro over this time period, it is clear that the bidask spreads for
the two East Asian currencies become lower. For example, the six-month
contract bidask spread for the Singapore dollar vis--vis the Japanese yen
was 16 bp in June 30, 2000, and it was 6 bp in May 18, 2006. Similarly, the
six-month contract bidask spread for the Thai baht vis--vis the Japanese
yen was 31 bp on June 30, 2000, and it was 14 bp on May 18, 2006. It means
that the conditions of foreign exchange markets in East Asian countries are
improving, and that the forward outright contracts for major East Asian
currencies are gradually becoming less expensive to trade.
5.6
CONCLUSION
This chapter has explained the advantages and disadvantages of choosing a

common currency basket such as the AMU over an international currency
as a denomination currency for bond issuers and foreign investors in terms
of both foreign exchange risks and liquidity. Performance of the currency

132

US dollar
euro
JP yen
average
US dollar
euro
JP yen
average
US dollar
US dollar
euro
Singapore
dollar
0.0608
0.1139
0.0992
0.0913
0.1532
0.2064
0.1912
0.1836
0.0530
0.0386
0.0914
0.0926
0.1658
0.1453
0.1346
0.2042
0.2787
0.2539
0.2456
0.0735
0.0518
0.1244
0.0984
0.1817
0.1569
0.1457
0.2553
0.3416
0.3062
0.3010
0.0840
0.0565
0.1384
0.0329
0.0901
0.0684
0.0638
0.2869
0.3443
0.3195
0.3169
0.0572
0.0354
0.0925
0.0604
0.1461
0.0998
0.1021
0.3311
0.4176
0.3596
0.3694
0.0861
0.0395
0.1247
3m
0.0659
0.1371
0.1224
0.1084
0.4414
0.5138
0.4706
0.4753
0.0720
0.0564
0.1268
6m
1m
1m
6m
(sample day: July 1,

2001)
(sample day: June 30,

2000)
3m
Spreads (%)
Spreads (%)
0.0240
0.0847
0.0591
0.0559
0.1445
0.2051
0.1793
0.1763
0.0608
0.0350
0.0957
1m
0.0282
0.0837
0.0701
0.0607
0.1686
0.2237
0.2092
0.2005
0.0557
0.0417
0.0970
3m
0.0395
0.0997
0.0899
0.0764
0.1927
0.2520
0.2402
0.2283
0.0608
0.0501
0.1099
6m
(sample day: July 2,

2002)
Spreads (%)
0.0344
0.0714
0.0774
0.0611
0.0234
0.0604
0.0662
0.0500
0.0371
0.0429
0.0799
1m
0.0344
0.0806
0.0787
0.0646
0.0409
0.0872
0.0848
0.0710
0.0463
0.0442
0.0902
3m
0.0344
0.0693
0.0825
0.0621
0.0234
0.0586
0.0708
0.0509
0.0352
0.0479
0.0826
6m
(sample day: Feb 6,

2003)
Spreads (%)
0.0298
0.0384
0.0394
0.0359
0.1049
0.1137
0.1138
0.1108
0.0086
0.0096
0.0173
1m
0.0380
0.0476
0.0487
0.0448
0.1311
0.1413
0.1399
0.1374
0.0094
0.0108
0.0184
3m
0.0444
0.0565
0.0576
0.0528
0.1311
0.1443
0.1409
0.1388
0.0117
0.0135
0.0225
6m

2006)
Spreads (%)
2 bid price
) 3 100 3 1/181
6 month bid2ask spread cost (daily based, %) 5 (ask price
spot rate
2 bid price
) 3 100 3 1/91
spot rate
2 bid price
) 3 100 3 1/30
spot rate
Sources: Calculated by authors. All spot rates and forward rates are collected from Bloomberg Currency Composite pages. Forward swap bid
ask spreads calculated by bid and ask quotations on both spot and forward rates.
Euro
JP yen
JP yen
Thai baht
Against
The change of forward swap bidask spreads (% in transaction basis) in two East Asian currencies against
three major currencies
Currency
Table 5.8
133
basket denominated bonds was compared with that of bonds denominated

in terms of the US dollar, the euro, and the Japanese yen.
Currency basket denominated bonds would have the second lowest foreign
exchange risks of foreign borrowing costs in Thailand, the Philippines,
Indonesia, and Korea. The foreign exchange risk would be smaller than that
of issuing the US dollar denominated bonds. Moreover, currency basket
denominated bonds would, in general, reduce the foreign borrowing costs
for bond issuers in all of the East Asian countries. Especially for bond issuers
in the dollar-pegging countries, the foreign borrowing costs of issuing the
currency basket denominated bonds would be lower than those of issuing
home currency denominated bonds. Thus, issuing the currency basket
denominated bonds would contribute to decreasing foreign borrowing costs
and foreign exchange risks for bond issuers in East Asian countries.
For the currency basket shares, we investigated two types of basket
shares based on trade weights. One was a common basket share for all
bond issuers across East Asian countries, while the other was a countryspecific basket share for bond issuers in each of the East Asian countries.
The volatilities of the currency basket denominated bonds with the
common basket share are the lowest for bond issuers in most of the East
Asian countries except for Indonesia and Korea. On the other hand, the
means of foreign borrowing costs are the lowest when currency basket
denominated bonds are issued with the basket shares based on countryspecific trade intensities. Thus, issuing the currency basket denominated
bonds would contribute to decreasing foreign borrowing costs and their
foreign exchange risks in many cases. However, US dollar denominated
bonds are still the best choice for bond issuers in the dollar-pegging countries. These results suggest that it might be difficult to fix common currency
basket shares across the nine East Asian countries.
For both investors who evaluate their returns in terms of the US dollar
and investors who evaluate their returns in terms of the Japanese yen, the risk
of investing in AMU denominated bonds is likely to be lower than that of
investing in local currency denominated bonds in all but the dollar-pegging
East Asian countries. Especially for investors who evaluate their returns
in terms of US dollars, investing in equally weighted AMU denominated
bonds is effective both in reducing risk and in improving the Sharpe ratio in
comparison with investing in each of the East Asian countries government
bonds, with the exception of the bonds of the dollar-pegging countries. The
reduction of total risk is a result of the reduction of foreign exchange risk
from the portfolio effect. It is important to note, however, that these results
depend on the currency system in place in the East Asian countries.
On the other hand, the composition weight of the Japanese yen denominated bonds in the AMU denominated bonds is critical for investors who

134
evaluate their returns in terms of Japanese yen. Because of the higher

composition weight of Japanese government bonds in the BIS1 type of
AMU denominated bonds, these bonds should have the lowest risk and
the highest Sharpe ratio for investors who evaluate their returns in terms
of Japanese yen, in contrast to investors who evaluate their returns in terms
of US dollars. In addition, for both investors who evaluate their returns in
terms of US dollars and those who evaluate them in terms of Japanese yen,
the effect on the interest return from including high yield bonds in AMU
denominated bonds is larger for longer investment periods.
When investors evaluate their returns on investments in single currency
denominated government bonds, their risks derive from foreign exchange
risk rather than interest risk. Therefore, the introduction of AMU denominated bonds could prevent contagion selling of local currency denominated
bonds caused by a sudden depreciation of one local currency.
As far as liquidity is concerned, data on the bidask spreads of swap
transactions show that the US dollar has the highest degree of liquidity for
bond issuers in all of the nine East Asian currencies. However, the differences between the US dollar and the currency basket are not very large for
three- and six-month swap transactions, though the differences are larger
for one-month swap transactions.
Accordingly, it is true that bond issuers in East Asia face a trade-off
between foreign exchange risk and higher liquidity when they issue the currency basket denominated bonds. Although the currency basket denominated bonds would contribute to decreasing foreign exchange risk, bond
issuers might prefer the US dollar denominated bonds to the currency
basket denominated bonds as long as they care about liquidity more than
foreign exchange risks. When we used forward swap bidask spreads as
a measure of liquidity, we found that the forward outright contracts for
major East Asian currencies have recently become less expensive to trade.
Our analyses show that US dollar denomination of bond issues provides
the highest degree of liquidity as a result of the US dollars dominance in
the world financial markets and that the US dollar denominated bonds
are still the best choice, especially for bond issuers in the dollar-pegging
countries in East Asia. However, we also find that the currency basket
denominated bonds would contribute to reducing foreign borrowing costs
and their foreign exchange risks in many cases. In order to activate regional
bond markets and establish a more solid base for intra-regional capital
flows in East Asia, markets for regional bonds denominated in terms of a
common currency basket should be developed.
Our results were obtained after making some simplifying assumptions
in our analysis. However, these assumptions will be reconsidered in future
research. First, we assumed a partial equilibrium, and therefore we did

135
not consider the possibility of bond returns and risks being affected by
investors behavior. Examining the effects of the widespread adoption of
AMU bonds in East Asian countries on bond returns and risks is therefore a topic for future discussion. Second, we focused only on the risk
reduction effects of AMU denominated bonds. However, other important
factors, such as transaction costs, market liquidity, and credit rating, all
influence investors decisions, so they should also be investigated in future
analyses of AMU denominated bonds.26 Finally, our results indicate that
the risk-reduction effect of AMU denominated bonds largely depends on
the country share, meaning that, while taking into account returns from
AMU denominated bonds, further discussion on optimal share levels for
risk reduction will be called for in the near future.
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
The authors are grateful to Beate Reszat, Koichi Hamada, and Ulrich Volz for their
valuable comments and suggestions.
The Chiang Mai Initiative, initiated in May 2000, created bilateral swap agreements
worth almost USD 40 billion. In 2005, the value of the agreements grew to over USD
70 billion.
Research on the common currency basket system has been conducted by Ito et al. (1998);
Bnassy-Qur (1999); Williamson (2000); Ogawa and Ito (2002); and Ogawa et al. (2004).
EMEAP (Executives Meeting of East-Asia and Pacific Central Banks) is a cooperative
organization of central banks and monetary authorities in the East Asia and Pacific
region. The 11 EMEAP central banks and monetary authorities are the Reserve
Bank of Australia, Peoples Bank of China, Hong Kong Monetary Authority, Bank
Indonesia, Bank of Japan, Bank of Korea, Bank Negara Malaysia, Reserve Bank of
New Zealand, Bangko Sentral ng Pilipinas, Monetary Authority of Singapore, and
Bank of Thailand.
The ASEAN member countries are Brunei, Cambodia, Indonesia, Laos, Malaysia,
Myanmar, the Philippines, Singapore, Thailand, and Vietnam. ASEAN13 refers to
the ASEAN member nations plus China, Japan, and Korea. The Asian Bond Market
Initiative established six working groups, which include new securitized debt instruments, credit guarantee mechanisms, foreign exchange transactions and settlement issues,
issuance of bonds denominated in local currency, local and regional rating agencies,
and technical assistance coordination. One of the working groups focuses on currency
denomination in Asian bond markets. The Joint Ministerial Statement of the ASEAN13
Finance Ministers Meeting on May 4, 2005 in Istanbul stated that the possible issuance
of Asian currency-basket bonds could be explored under the auspices of the roadmap.
See the previous chapter by Ma and Remolona for an extended discussion of the
ABF2.
Usually ex ante borrowing costs in the real world should be calculated from the prevailing benchmark interest rate, risk premium, and foreign exchange rate fluctuations.
However, since our aim is to compare the borrowing costs in different currencies or
currency baskets for one country, we assume the risk premium to be common for one
country and concentrate on the foreign exchange risk only here.
Kawai and Takagi (2000) calculated the trade intensity within East Asia and Western
Europe and found that intensity is extremely high for many trading pairs in East Asia,
frequently exceeding the corresponding figures for European pairs.

136
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.

All trade data except those for Taiwan are from the Direction of Trade Statistics (IMF).
The Taiwan data are from the National Statistics of Taiwan.
All data are obtained from Datastream. Exchange rates are daily closing rates, and
interest rates are the middle rate of three months money market rate. The details of the
data are as follows: US TB 3 months, Japan CD 3 months (new issue), euro Interbank 3
months, Singapore Interbank 3 months (MAS), Thailand Deposit 3 months, Malaysia
Interbank 3 months, Indonesia Deposit 3 months, Philippine Treasury Bill 91 days,
Taiwan Money Market 90 days, Korea Commercial Paper 91 days, Hong Kong
Interbank 3 months, China Time Deposit Rate 3 months.
The standard deviations might be undervalued because we calculate them by using
overlapping data. However, we just compare these values relatively rather than evaluate
their absolute value. We conducted test differences among the standard deviations of
foreign borrowing costs by issuing different type of bonds. Accordingly, we regard our
tests as having no problem even though we used the overlapping data.
We use Students t and Tukey-Kramer HSD statistics for the test that the means are
equal, and the OBrien test, the Brown-Forsythe test and the Levene test for testing if
the variances are equal.
We calculate and compare the volatilities of the daily borrowing costs instead of threemonth borrowing costs with the same data and the same sample periods.
See Shimizu and Ogawa (2005).
Actually, the fund of funds, which invests in a portfolio of several funds (or investment
trusts), has recently become more common and increasingly popular for private investors in Japan.
We suppose the case in which investors do not use forward swap transactions for covering foreign exchange risk.
We suppose each local bond to be a zero coupon bond. In addition, the yield data give
rates on an annual basis, so we convert them to a monthly basis for calculation.
Because of the limited sample number, we do not compare the standard deviations of
Chinese and Indonesian bonds.
For example, the Asian Dollar Bond Index (ADBI) of HSBC is calculated on the basis
of shares of the total market capital in each of East Asian countries.
We suppose the AMU denominated Asian bond is a portfolio investment in the nine
East Asian countries government bonds in this chapter.
In this chapter we focus on the risks and returns of bonds. However, a higher share
for Japanese bonds should be desirable if we take into account other factors, such as
transaction costs, grade and liquidity in the market.
See Ogawa and Shimizu (2004).
A commentator indicates that the bidask spreads in foreign exchange forward swap
markets account for the hedging costs.
We consider bidask spreads and foreign borrowing costs separately, since bidask
spreads should not be included in the measure of foreign borrowing costs, as they are
one of the transaction costs.
1 basis point (bp) 51/100 percent.
For example, concerning the influence of transaction costs, when a transaction cost is
required for holding overseas assets, the efficient frontier shifts to the lower part equal
to the transaction cost amount and the investors utility is seemed to decrease.
REFERENCES
Bnassy-Qur, Agnes (1999). Optimal Pegs for East Asian Currencies. Journal
of the Japanese and International Economies 13: 4460.
Hartmann, Philipp (1998). Currency Competition and Foreign Exchange

137
Markets: The Dollar, the Yen and the Euro. Cambridge: Cambridge University
Press.
Ito, Takatoshi (2003). Construction of Infrastructures for the Development of
Regional Bonds Market. In Choong-Yong Ahn, Takatoshi Ito, Masahiro
Kawai, and Yung-Chul Park (eds), Financial Development and Integration
in East Asia. Seoul: Korea Institute for International Economic Policy, pp.
20621.
Ito, Takatoshi (2004). Promoting Asian Basket Currency Bonds. In Takatoshi
Ito and Yung Chul Park (eds), Developing Asian Bond Markets. Canberra: Asian
Pacific Press at the Australian National University, pp. 6789.
Ito, Takatoshi, Eiji Ogawa, and N. Yuri Sasaki (1998). How Did the Dollar Peg
Fail in Asia? Journal of the Japanese and International Economies 12: 256304.
Ito, Takatoshi, and Yung Chul Park (2004). Overview: Challenges and Strategies.
In Takatoshi Ito and Yung Chul Park (eds), Developing Asian Bond Markets.
Canberra: Asian Pacific Press at the Australian National University, pp. 115.
Kawai, Masahiro, Eiji Ogawa, and Takatoshi Ito (2004). Developing New
Regional Financial Architecture: A Proposal. Mimeograph.
Kawai, Masahiro, and Shinji Takagi (2000). Proposed Strategy for a Regional
Exchange Rate Arrangements in Emerging East Asia. Policy Research Working
Paper no. 2502, World Bank, Washington, DC.
Kuroda, Haruhiko, and Masahiro Kawai (2003). Strengthening Regional
Financial Cooperation in East Asia. PRI Discussion Paper Series no. 03A-10,
Policy Research Institute, Japanese Ministry of Finance, Tokyo.
Ogawa, Eiji, and Takatoshi Ito (2002). On the Desirability of a Regional Basket
Currency Arrangement. Journal of the Japanese and International Economies
16: 31734.
Ogawa, Eiji, Takatoshi Ito, and N. Yuri Sasaki (2004). Costs, Benefits, and
Constraints of the Currency Basket Regime for East Asia. In Asian Development
Bank (ed.), Monetary and Financial Integration in East Asia. The Way Ahead,
Volume 2. New York: Palgrave, pp. 20939.
Ogawa, Eiji, and Kentaro Kawasaki (2003). What Should be Weights on the
Three Major Currencies for a Common Currency Basket in East Asia? Paper
presented at the Regimes and Surveillance in East Asia Conference in Kuala
Lumpur, March.
Ogawa, Eiji, and Junko Shimizu (2002). Roles of Regional Currency in Bonds
Markets in East Asia. Faculty of Commerce Working Paper no. 81, Hitotsubashi
University, Tokyo.
Ogawa, Eiji, and Junko Shimizu (2004). Bonds Issuers Trade-off for Common
Currency Basket Denominated Bonds in East Asia. Journal of Asian Economics
15: 71938.
Ogawa, Eiji, and Junko Shimizu (2005). A Deviation Measurement for
Coordinated Exchange Rate Policies in East Asia. RIETI Discussion Paper
no. 05-E-017, Research Institute Economy, Trade and Industry, Tokyo.
Petri, Peter A. (1993). The East Asian Trading Bloc: An Analytical History. In
Jeffrey Frankel and Miles Kohler (eds), Regionalism and Rivalry: Japan and the
United State in Pacific Asia. Chicago: University of Chicago Press, pp. 2152.
Shimizu, Junko, and Eiji Ogawa (2005). Risk Properties of AMU Denominated
Asian Bonds. Journal of Asian Economics 16: 590611.
Williamson, John (2000). Exchange Rate Regimes of Emerging Markets: Reviving the
Intermediate Option. Washington, DC: Institute for International Economics.


6.
East Asias role in the revived

Bretton Woods system
Michael P. Dooley, David Folkerts-Landau,
and Peter Garber
6.1
INTRODUCTION AND SUMMARY OF RESULTS
In this chapter, we set out in greater detail how we think about the emergence of China and Asia as major players in world capital and foreign
exchange markets. Our approach, which has come to be known as the
Bretton Woods II view, provides a coherent explanation for the current
structure of interest rates, exchange rates, and current account balances
and also provides a contingent playbook for future interest and exchange
rate movements.
We argued in 2003 and 2004 that the de facto global system was analogous to that during the Bretton Woods period because of the existence
of a macroeconomically important periphery that fixed an undervalued
exchange rate to the dollar to promote an export-driven development
policy. The US willingly played the role of center country again, running
balance of payments deficits and serving as the global financial intermediary. Further, contrary to most forecasts at the time, we claimed that the
system would last for a long time because of the incentives, especially those
of China, to absorb a vast pool of underemployed labor into the industrial
system.1
Conventional analyses are based on the assertion that the Bretton
Woods II system cannot hold together for much longer. This may or may
not turn out to be correct, but it does not offer any guidance for further
analysis if the system does survive for an extended time period, as we
believe it will.
For simplicity, our framework divides the world into three regions,
emerging Asia, the US, and Euroland.2 Euroland includes all countries
outside the US with open capital markets and market-determined exchange
rates. We use the euro to stand for the currencies of these countries since
it is the dominant currency among them and the renminbi to stand for the
currencies of emerging Asia.
141

142
Exchange rate policies in East Asia
The analysis leans on four assumptions. We believe these assumptions

are realistic, and they dramatically simplify the dynamics of a three region
analysis.
First, emerging Asian financial markets are assumed to be poorly integrated with those of the other two regions because of capital controls and
because Asian assets are imperfect substitutes for US and Euroland assets.
This allows Asia to manage the dollar-renminbi exchange rate so that the
renminbi appreciates in real terms slowly over an adjustment period of
many years.
Second, the US and Euroland financial markets, in contrast, are assumed
to be very well integrated and their respective assets close substitutes, an
assumption consistent with a great deal of empirical work. The US and
Euroland do not manage the eurodollar exchange rate.
Third, the dominant changes in the economic environment that are
driving the main features of the world economy are the rapid growth of
savings rates and the level of savings in emerging Asia and their exportation
to the rest of the world.
Fourth, the US and Euroland differ in their capacities to utilize Asian
savings, with the US having a much greater absorptive capacity.
Our analysis departs significantly from the conventional approach that
still dominates the official and academic discourse, which includes the
following assumptions contrary to ours.
First, conventional analysis considers Asian financial markets sufficiently
integrated with international markets so that Asian governments will not
be able to manage real exchange rates at reasonable costs. Moreover, they
will not want to distort real exchange rates to encourage export-led growth
for much longer.
Second, conventional analysis assumes that US and Euroland financial
markets are not well integrated. For this reason, diversification of Asian
reserves is thought to have an important effect on the dollareuro rate.
This assumption seems to us inconsistent with substantial evidence that
intervention and reserve management by US and Euroland authorities
have not had a large or lasting effect on industrial country exchange
rates.
Third, conventional analysis usually identifies a fall in the US household
savings rate or a rise in the government fiscal deficit rate as the driving force
behind the US current account deficit and the global imbalances.
Fourth, conventional analysis predicts a rise in US interest rates, which
in reality has been contradicted by falling interest rates. To circumvent
this contradiction, it is conventionally asserted that interest rates and asset
prices are driven by incorrect expectations, a misunderstanding of the
dangerous nature of the system, or bubbles.

143
The interesting results of our analysis following a sudden, long-term rise

in exports of Asian savings provide a contingent playbook for future real
interest and real exchange rate movements.
First, at the start of the system, once it becomes generally understood,
there is a substantial immediate appreciation of the euro against the dollar.
Second, real interest rates in the US and Euroland will remain low relative
to historical experience but converge slowly toward world rates as Asian
markets become integrated with international markets. Third, the dollar and
the euro will gradually depreciate relative to the renminbi but, after the initial
euro appreciation against the dollar, remain constant relative to each other.
Fourth, more rapid expected growth in Europe would depreciate the
euro relative to the dollar and renminbi and raise interest rates in the US
and Europe. Fifth, more rapid expected growth in the US, or more investment demand following exogenous shocks such as Hurricane Katrina in
2005, would tend to depreciate the dollar relative to the euro and renminbi.
Because the dollarrenminbi rate is managed, the dollar would not fall
immediately but would begin to depreciate more rapidly. The euro would
appreciate immediately against the dollar and then match the dollars more
rapid rate of depreciation against the renminbi. Sixth, shifts in the currency
composition of Asian reserves from dollars to euros would have little or
no lasting effect on the dollareuro exchange rate. Seventh, effective protection in the US and Euroland or a fall in the savings rate in Asia would
generate a stronger dollar in the long run. The immediate effect would
be less rapid dollar depreciation against the renminbi. The euro could go
either way against the dollar.
Eighth, high oil prices and high consumption by oil exporters would generate a slower rate of dollar depreciation against the renminbi and higher
interest rates in the US and Euroland. The dollareuro rate could go either
way. Ninth, a decision by Asian governments to manage their exchange
rates relative to a dollar/euro basket would reduce the volatility of the
dollareuro exchange rate but not its current or long-term level. Finally,
in real terms, the dollar will eventually have to depreciate relative to the
renminbi. But most of the adjustment in the US trade account will come as
US absorption responds to increases in real interest rates. Slow adjustment
in the composition of US output toward traded goods over an extended
time period will not require unprecedented dollar depreciation.
6.2
THE CONVENTIONAL APPROACH
Current account deficits and surpluses (i.e. exports and imports of savings)
have clearly moved far outside of historical norms and have therefore

144
attracted a great deal of attention. Let S stand for national savings and
I for investment. A current account deficit must be associated with S < I
(United States), a current account surplus with S > I (Asia), and a current
account balance with S 5I (approximately, Euroland).3 It follows that if
we could explain why net savings are imported by the US, exported by
Asia, and not traded much by Euroland, we would understand the current
state of the international monetary system. Moreover, if we could explain
how trade in savings will evolve over time we could make meaningful asset
price forecasts.
The conventional approach starts with the assumption that the fundamental cause of the US current account deficit has been an exogenous
fall in national savings in the US. Two explanations for the change in US
behavior are offered. First, the jump in the US fiscal deficit reduces government savings. Second, bubbles in asset values, for example housing, reduce
household savings. (Both are sometimes linked to an even more fundamental moral decline.) The US is able to live beyond its means (an emotional
way to say S < I) by pulling in goods from abroad.
The US current account deficit is identically equal to the increase in book
value of US net foreign debt, so the deficit pushes US debt into foreign
portfolios. This profligacy, it is argued, is bound to lead to crisis and
ever-rising ultimate disaster the longer it lasts.
A more complete explanation for the US current account deficit would
also describe the transmission of the fall in US savings to other regions.
In short, why did Asia supply savings and goods to the US? That is, why
is S > I in Asia, and why did Asia buy US securities? Moreover, why did
Euroland do neither?
The standard analysis of the effects of a decline in US savings is straightforward. First, the market reaction to the decline in US savings is a fall in
the value of the dollar relative to other currencies. The depreciation of the
dollar tends to reduce the US current account deficit below what it would
otherwise be, and the cheaper dollar in the medium term will generate a net
improvement in trade to help pay the interest on increased US debt.
This basic story has been recently augmented by the idea that foreigners really do not want to lend to the US and will do so only if the dollar
is expected to rise in value relative to other currencies. This is sometimes
called a home bias in investment preferences, a bias that is quite clear
in portfolios in the US and the rest of the world. Home bias implies that
the dollar has to go down even more now so that investors can reasonably expect it to rise in the future. The expected appreciation of the dollar
increases the expected yield on dollar assets relative to assets denominated
in other currencies and overcomes the reluctance to lend abroad.
The dollar has declined against the euro and other floating currencies,

145
as the conventional approach predicts, but not against Asian currencies,

because governments in Asia intervene in markets to support the dollar
and invest the proceeds in US assets. Most experts believe that this artificial
tampering with markets cannot last long and will end with a costly speculative attack on the fixed exchange rate arrangements.
The US will then suddenly have to live within its means and repay debt.
This sudden adjustment will be very costly in terms of both unemployment
and large declines in the real value of the dollar as resources are shifted from
domestic to traded goods industries. This is certainly a good description of
what has happened to other fixed rate systems in recent years. If the US
were just like Argentina, the formal story would be finished. Economists
warn their neighbors, are perplexed by an indifferent financial industry,
and wait for the unhappy end. If the model is right, market expectations
must be wrong. If it does not end this year, we are even more certain it will
end next year, and with a bigger bang. These warnings have been forthcoming for almost six years, even as the global system soldiers on.
6.3
BRETTON WOODS II
But there are problems with this story, and we have emphasized two. First,
if a fall in the US savings rate is driving the system, we should expect US
real interest rates to rise. In fact, US rates have fallen and remain so low
that they constitute as large a mystery as the very large current account
deficit. Second, if the current situation is artificial and cannot last long,
this expectation should be apparent in longer term interest rates. But again,
the yield curve reflects the expectation that US rates will rise very slowly
over the next 10 years. For, example implied forward rates on one-year
Treasury inflation-protected securities (TIPS) as of September 17, 2005
rise gradually from 1.07 percent to 2.35 percent over 10 years. Strongly
contradicting the expectations underlying these price data, the adherents of
the conventional model assume that investors do not understand the world
and will be surprised when the end comes.
In our framework, the fundamental shock to the system is a change in
the supply of savings from developing Asia and a suspension of the usual
home bias in allocating these savings across world markets. It may not
seem all that important to decide whether it was the fall in US savings
or the increase in Asian savings that has driven the pattern of current
accounts we now see. But it is in fact crucial for understanding the system
and the direction it will take.
The first obvious departure from the conventional analysis is the observation that Asian real exchange rates are not market-determined prices

146
but are heavily and successfully managed by Asian governments. As noted

above, the conventional analysis assumes this troublesome fact will soon
go away. We argue that this policy behavior will eventually go away but is
a central feature of Asian development policies and will not dissipate for
a long time.4 It follows that if the rest of the world is to adjust now to a
savings shock emanating from Asia, the primary adjustment mechanism
will not be changes in Asian real exchange rates.
To manage real exchange rates, Asian governments must intervene in
foreign exchange markets. That part of the intervention that is sterilized is
in fact intervention in credit markets. Asian finance ministries or central
banks sell domestic securities, reducing the supply of loanable funds to
domestic borrowers, and buy foreign securities, thereby increasing the
supply of loanable funds in the US and Euroland. The resulting shift in
interest differential is possible because of effective capital controls. That is,
Asian governments can manage exchange rates and interest rates because
their domestic assets are made imperfect substitutes for foreign assets in
private portfolios by policy, if not by private preference.
Because Asian exchange rates are managed, adjustment must proceed
through current account balances and real interest rates. Recall that to
understand current accounts we have to understand savings and investment. The question then is: how are savings and investment changed in the
US, Euroland, and elsewhere as Asian savings are offered to the rest of the
world? In particular, can we understand why real interest rates might fall
in both the US and Euroland while current account balances adjust by very
different amounts? In our view, this is a very easy case to understand.
We can illustrate our approach first with a set of figures focusing on
interest rates and current accounts for Asia, the US, and the Euroland and
then with another set focusing on net foreign debt positions and exchange
rates. Figure 6.1 shows real interest rates for the US, Euroland, and Asia
on the vertical axes. The horizontal axes are the current accounts for these
US
r
S9
r0
Euroland
S
r
Figure 6.1
S9
r1
CA
Deficit
I
Surplus
Asia
r
r1
r0
r0
r1
S9
I
0
Surplus
CA
Deficit
0
Surplus
CA
Deficit
Current account and interest rates

147
three regions. The upward sloping curves labeled S are national savings.
The curves labeled S9 are national savings augmented by imports or
exports of savings. The downward sloping curves labeled I are investment.
For convenience, we start with balanced current accounts at a common
interest rate, but any starting point for the separate economies will do as
long as real rates are the same in the US and Euroland.
A policy to divert Asian savings to the US and Euroland reduces the
supply of savings available in Asia and shifts the Asian domestic supply
curve to the left. A current account surplus is generated and interest rates
in Asia rise.
In the US and Euroland, savings supply curves shift to the right as Asian
savings push in. The real interest rate in the US and Euroland falls as we
move down the investment demand curves. The demand curves are downward sloping because investment increases relative to domestic savings
as interest rates fall. Moreover, consumption rises with a fall in interest
rates so domestic savings fall as well. The rise in consumption and investment is matched by an inflow of foreign savings, and the current account
deficit increases. The increase in Asias current account surplus is matched
by the sum of the increases in the current account deficits of the US and
Euroland.
In the US, the increase in savings demanded is large because investment
and savings are quite sensitive to the rate of interest.5 Euroland sees the
same qualitative changes, but investment and the current account deficit
increase only slightly because there are few profitable investment opportunities and consumption is not very responsive. The fundamental factor
driving the different responses of the US and Euroland current account
deficits is the difference in opportunities for efficiently utilizing foreign
savings as the interest rate falls in both regions.
An important aspect of the adjustment process is the equalization of real
rates of return on capital invested in the US and Euroland through private
arbitrage. When we turn to exchange rate determination below, we use
the assumption that real interest rates are equalized by flows of savings.
It is clear, however, that expected rates of return on capital in the US and
Euroland could be equalized by expected real exchange rate changes in
addition to real interest rates. This apparent indeterminacy between real
interest rates and expected changes in real exchange rates during the adjustment period is resolved at the end of the period. When the new equilibrium is
established there is no reason to predict that the real exchange rate between
the euro and the dollar would continue to change over time. Since the
capital stocks must have the same expected rate of return looking forward
at the end of the adjustment period, it follows that real interest rates must
be the same at that time. Arbitrage across time will ensure that any capital

148
put in place in the US and Euroland during the adjustment period that will
remain in place in a new steady state must have the same rate of return.
The optimal policy over time for Asian governments is to allow gradual
real exchange rate appreciation. This reduces over time their intervention
in credit markets and their exports of savings. By the end of the adjustment
period real interest rates will have equalized across the three regions.
We now turn to the foreign exchange markets. There are three keys to
understanding the three cross exchange rates. First, for some years Asian
governments can and will manage the real dollar value of their currencies. They can do so because capital controls make Asian domestic assets
imperfect substitutes for US and Euroland assets in private portfolios.
Their ability to manage their real exchange rate will erode over time as
capital controls become less effective and their domestic asset markets are
integrated with international capital markets. Their desire to maintain
the system will also erode as their surplus labor is absorbed. But they will
manage rates as long as they can because undervaluation is an important
part of their development strategy.
Second, in the long run, say 10 years more or less, the real value of the
three currencies will have to adjust to changes in the international investment positions of the three regions generated during the adjustment period.
Asias net asset position will improve, while the US and Euroland positions
will deteriorate by relatively large and small amounts, respectively.
The relationship between the long run exchange rate and the net foreign
debt position of each region is not controversial. As net foreign debt
increases, larger trade balance surpluses are needed to service net debt
(balance the current account). So a fall in net foreign assets is associated
with a depreciation of the real exchange rate. The implication is that the
dollar and the euro must depreciate against the renminbi, but the dollar must
depreciate by more. Therefore, the dollar must depreciate against the euro.6
Third, exchange rates today would normally reflect these long run expectations to some degree. But intervention by Asian governments is sufficient
to manage strictly the dollarrenminbi exchange rate. Intervention will not
keep the renminbi undervalued forever, but it can extend the adjustment
period. As we have argued elsewhere, the optimal path (from Chinas perspective) for Asian real exchange rates is a gradual appreciation toward
their new long run values.
In contrast, the euro cross-rates both today and along the adjustment
path are determined by private investors. The relevant context for these
portfolio choices is that dollar and euro assets are close substitutes. The
key implication is that once the system comes to be understood the euro
and the dollar must depreciate at the same rate over time relative to the
renminbi. Recall that real interest rates on capital invested in the US and

149
RMB/
USD A
0
RMB/
EUR E
C
Figure 6.2
Exchange rates
Euroland are equalized by net savings flows. It follows that investors must
expect the eurodollar exchange rate to remain unchanged. Put another
way, both currencies must depreciate, and be expected to depreciate, at the
same rate against the renminbi.
The result of a shift in Asian savings exports is then an immediate euro
appreciation against the dollar and the renminbi followed by a constant
dollareuro rate. This means that there will be immediate, maximal political pressure for relief in a Euroland unable to absorb the shock easily, and
continuous, though declining, pressure thereafter.
These results are illustrated in Figure 6.2. Starting from an initial value A
for the renminbidollar rate in the top panel and C for the renminbieuro
rate in the bottom panel, we can follow the effects of an increase in Asian
savings together with intervention. The increase in savings combined with
intervention raises interest rates in Asia and lowers interest rates in the US
and Euroland. Asia generates a current account surplus matched by deficits
in the US and Euroland. This continues until Asian savings and intervention return to normal levels. In Figure 6.2, this interval is from 0 to T.
The eventual fall in the dollar against the renminbi from A to B is
required in order to generate the trade surplus needed to service the higher
level of US debt at time T and after.

150
Without intervention, we would expect an immediate depreciation of the

dollar; but this can and will be delayed by intervention.7 Along the adjustment path AB, the dollar is supported by a flow of intervention. Private
investors know the dollar will depreciate but nevertheless are willing to
hold the stock of dollars, reduced by Asian purchases of US assets.8 US
debt to foreigners is growing more rapidly than it would have if the fall in
interest rates had been partially offset by a market-determined depreciation of the dollar.
The renminbieuro rate starts at C and must eventually move to D, a
much smaller depreciation. Like the US, Euroland will accumulate debt (or
reduce net assets below their previous path) during the adjustment period.
But in this case Asian governments are not intervening to manage the
exchange rate either at point C or along the adjustment path. The question
is then: where will the market set euro exchange rates?
We can make our analysis more realistic and much more transparent
by assuming that US and Euroland assets are close substitutes in private
portfolios. This is a very important departure from the usual portfolio
balance model because it implies that the currency composition of Asian
intervention is of secondary importance. If euro and dollar assets are
close substitutes in private portfolios, Asian governments could intervene
in either dollars or euros to stabilize the dollar value of their currencies.
Moreover, diversification of Asian reserves would have little or no lasting
effect on the dollareuro exchange rates, contrary to a key conclusion
of the conventional view. This view is consistent with a very large body
of empirical evidence that sterilized intervention has no lasting effect on
exchange rates among industrial countries.9
The practical importance of this assumption is that the two adjustment
paths in Figure 6.2 must have about the same slope. If they did not, more
rapid dollar depreciation against the renminbi, relative to euro depreciation against the renminbi, would imply expected depreciation of the dollar
against the euro. Since interest rates in the US and Euroland are the same,
arbitrage would be profitable. Private investors would immediately bid for
euros against dollars and would do so until the euro jumps to E. From this
initial appreciation the euro would then depreciate against the renminbi
at the same rate as the dollar. Note that along this adjustment path the
euro would remain overvalued relative to the dollar and the renminbi
throughout the adjustment interval, although the degree of overvaluation
would shrink over time.
We can now iterate through the current account analysis. If the euro
has appreciated against the renminbi and the dollar, Eurolands current
account deficit, already increased by the fall in interest rates, would tend
to widen. The dollar would remain unchanged against the renminbi and

151
would have depreciated against the euro, so the already increased US

current account deficit would be reduced. The Asian surplus would be
increased by the euros appreciation. These second round effects on the
current account positions of the three regions would not alter our basic
story assuming the reactions of absorption to interest rates are very
different in the US and Euroland.
6.4
WHERE ARE WE NOW COMPARED TO THESE

PICTURES?
The discussion above suggests that the dollar should have depreciated
against the euro when market participants realized that US imports of
savings from Asia would generate a substantial increase in US net investment income payments. But there has been no obvious correspondence
between the US current account deficit or the increase in net US international debt and the value of the dollar. The current account deteriorated
sharply in 1996, and net debt began to grow at a historically unprecedented
rate, but the dollar was strong until 2002.
One explanation is found in Figure 6.3, which shows US net investment
income payments. As of mid-2004 the US still earned more on its stock of
gross financial assets than it paid on its larger stock of gross international
Million USD
25 000
20 000
15 000
10 000
5 000
2005 Q1
2004 Q1
2003 Q1
2002 Q1
2001 Q1
2000 Q1
1999 Q1
1998 Q1
1997 Q1
1996 Q1
1995 Q1
1994 Q1
1993 Q1
1992 Q1
1991 Q1
1990 Q1
1989 Q1
1988 Q1
1987 Q1
1986 Q1
1985 Q1
1984 Q1
1983 Q1
Note: Net investment income 5 Income receipts on US owned assets abroad Income
payments on foreign-owned assets in the United States.
Source:
Deutsche Bank.
Figure 6.3
US net investment income

152
debt. Indeed, in the eight most recent quarters (Q2/2003Q1/2005), US net

earnings totaled USD 86 billion, compared to USD 47 billion in the previous eight quarters.
The lesson is clear: it is not enough to forecast changes in the US net debt
position. We also need to know the stocks of gross international assets and
liabilities and the rates of return on these stocks. Our framework suggests
that over time the growth in net US debt will generate net investment payments in part because we expect net debt to grow and US interest rates to
rise over the adjustment period. But we also expect very substantial capital
gains on US-owned foreign direct investment, and this tends to limit the
growth in the market value of US net debt and reduces net investment
income payments. Forming correct expectations about the long run value
of net investment income payments requires a difficult evaluation of the
evolution of net and gross investment positions and rates of return.10
Moreover, it would not be surprising if market participants were uncertain about the longevity of US current account deficits. The conventional
story still insists that a speculative attack on Asian currencies that will force
a revaluation is long past due. If it had occurred, the path of US debt would
have already started to reverse. Moreover, the first analysis that we know
of that predicted large and sustained US current account deficits as part of
a rational global system was not published until mid-2003.11
Perhaps the best way to link the analysis with the experience of the past
10 years is to assume that, as the US deficit has moved to a historically high
level and stayed there for an extended period, market participants have
placed increased weight on the probability that the very gradual appreciation of Asian currencies requires the substantial near term rise of the euro
against the dollar described above. It would not be surprising if this learning process has some way to go. If so, there will be periodic upward pressure on the euro. As the learning curve flattens, the euro should stabilize
against the dollar.
6.5
THE BW II PLAYBOOK: HOW WILL BUMP

SCENARIOS MOVE INTEREST AND
EXCHANGE RATES?
Of course, many changes in conditions could shift the dollareuro

exchange rates along the adjustment path set out in the previous section.
The framework developed above is useful for evaluating changes in the
economic environment during the adjustment process, and the peculiar
nature of the global system produces some remarkable and unanticipated
results.

153
RMB/
USD A
F
G
B
t1
RMB/
EUR E
H
D
I
J
t1
Figure 6.4
Exchange rates
Bump 1: A Stronger Euroland Outlook

Suppose, for example, that at time t1 an improved outlook for profits in
Euroland were to generate a positive shift in the demand for investment in
Euroland. Figure 6.1 suggests that Asian savings would be shifted from the
US to Euroland for the balance of the adjustment period and that interest
rates in both regions would rise.
The effects on exchange rates are illustrated in Figure 6.4. With more
Asian savings going to Euroland and less to the US, at the end of the
adjustment period (at T) the euro would be weaker and the dollar stronger
than would have otherwise been the case. If Asian intervention at t1 were
to keep the dollar at F in Figure 6.4, the euro would depreciate sharply
at t1 for two reasons. First, it would then need to reach level J at T. And
second, it would then depreciate more slowly to match the dollars reduced
rate of depreciation.
Bump 2: A Weaker Euroland Outlook
A weaker outlook for Euroland investment would have symmetric effects.
In this case there would be a deterioration in the final expected debt position

154
of the US and an improvement in the final debt position of Euroland. This

would require a more rapid rate of dollar depreciation against the renminbi
and another lurch up for the euro. Interest rates in both regions would fall.
Bump 3: A Stronger US Outlook
Changes in US growth and investment would have similar effects. As US
growth increases, so would the expected stock of US debt. The greater
long run depreciation would not affect the current dollarrenminbi rate
but would require a more rapid appreciation of the renminbi against the
dollar for the balance of the adjustment period. The euro would appreciate against the renminbi and the dollar for two reasons. First, its long run
level would jump up. Second, it would have to appreciate immediately in
order to match the dollars higher expected depreciation rate against the
renminbi.
This is illustrated in Figure 6.5. The expected RMB/USD at T would
shift down from B to G, and the expected RMB/EUR would move up from
D to K. The euro would immediately jump from H to I as again the change
in the euro would be amplified by arbitrage between dollar and euro assets.
Interest rates in both regions would rise.
RMB/
USD A
F
B
G
t1
RMB/
EUR E
I
H
K
D
Figure 6.5
t1
Exchange rates

155
Bump 4: Exogenous Shocks to the US Economy

A shock to the US economy such as Hurricane Katrina in 2005 would be
a one-off and would increase US demand for foreign savings, which would
lead to an increased US indebtedness at T. Therefore, Bump 3 analysis
applies. The euro would appreciate against the dollar.
Bump 5: Protectionism Surges; Oil Exporters Start Consuming Asias
Surplus Savings
Bump 5a
For example, effective protection against Asian exports in both the US and
Euroland would reduce net savings transfers to the US and Euroland from
Asia by forcing a reduction in Asias net trade surplus.
Bump 5b
Similarly, a decline in net Asian savings exported to the US and Euroland
would occur if a larger share of US, Euroland, and Asian income were
transferred to oil exporters via terms of trade shifts. As the oil exporters
would start to consume a high fraction of this transfer, fewer excess savings
would be available to fund US and Euroland debt accumulation.
Each of these developments can be analyzed as illustrated in Figure
6.6. It turns out that both would have the same impacts on interest and
exchange rates. In these events expected US debt at T would be reduced,
which would increase the terminal exchange rate from B to G. Euroland
debt would also fall, which would raise the terminal RMB/EUR from D
to K. We assume that the new path for the RMB/USD would not jump up
at t1 but that the rate of dollar depreciation would be reduced so that the
new path for the RMB/USD would be FG. The RMB/EUR rate must reach
K at T, and the path from t1 must have the same slope as FG; that is, the
RMB/EUR must have the same expected rate of depreciation as the RMB/
USD. The conclusion is that the euro could either depreciate or appreciate
immediately against the dollar, depending on the relative change in debt
stocks in response to the new environment. There is no necessary direction
of effect for this key exchange rate. Interest rates would rise both in the US
and Euroland because of the reduction in available savings.
A useful rule of thumb is that events that change expected US and Euroland
debt stocks and real exchange rates in opposite directions generate large and
immediate changes in the dollareuro rate. The market rate changes in the
same direction as the change in the expected future rates. Events that move
both expected debt stocks in the same direction have ambiguous effects on
the exchange rate at the point where expectations change.

156

RMB/
USD A
F
G
B
t1
RMB/
EUR E
C
H
K
D
Figure 6.6
t1
Exchange rates
Bump 6: A Serious Basket Peg

The results discussed above would be altered if Asian authorities ever really
did adjust the dollarrenminbi rate in reaction to changes in the dollar
euro rate via the adoption of a basket peg. In the absence of new shocks to
the equilibrium path, a basket peg would have no effect because the dollar
euro exchange rate is constant during the adjustment period. However,
in the face of the other shocks discussed above, a basket peg would tend
to reduce the volatility of the dollareuro exchange rate and would either
increase or decrease the average real value of the renminbi, depending on
the nature of shocks. There is no implication that a move from a straight
dollar peg to a basket peg would weaken the dollar.
6.7
CONCLUDING REMARKS
It is generally accepted that adjustment must occur to ultimately remove

the imbalances from the international monetary system. The dispute has
been between a view that the system will end abruptly and soon and a
view that it will last for years more with a smooth adjustment in interest

157
rates and exchange rates. Here, we presume that the basic system that has
dominated at least the last five years will continue for years more, while
converging to a sustainable equilibrium. With this system as a background,
we examine the behavior of interest rates and exchange rates following
a variety of shocks to the international monetary system. Our analysis
suggests that real interest rates in the US and Europe will remain low relative to historical cyclical experience for an extended period but converge
slowly toward normal levels. During this adjustment interval, the US will
continue to absorb a disproportionate share of world savings. Moreover,
after a substantial initial appreciation, the floating currencies will remain
constant relative to the dollar in the undisturbed background system. An
improvement in the investment climate in Europe during the adjustment
period would generate an immediate depreciation of the euro relative to
the dollar. In real terms, the dollar and the floating currencies will eventually have to depreciate relative to the managed currencies. But most of the
adjustment in the US trade account will come as US absorption responds
to increases in real interest rates.
NOTES
1.
2.
3.
4.
5.
6.
7.
For a detailed presentation of this system, see Dooley et al. (2003a, 2003b, and 2004b).
This current note is a shorter version of Dooley et al. (2005).
Because there is no necessity for geographic contiguity, we have referred to these regions
in other essays from the functional viewpoint as the trade account region, the center
country, and the capital account region.
Of course, Euroland, as we have defined it, includes some countries with large surpluses,
such as Canada, or deficits, such as the UK. We consider the impact of the new oil
exporter surpluses as a separate issue.
Our critics sometimes claim that we predict that Asian governments will fix real
exchange rates and finance a US current account deficit forever. This, of course, would
make no sense. The tactic of the development strategy is a controlled rise in dollar wages
to world levels at the end of a long adjustment period. We have argued that Bretton
Woods II might be replaced by a Bretton Woods III as a new set of periphery countries
replace Asian graduates.
This means that there are lots of viable projects or confident consumers ready to go with
a small improvement in financing costs relative to Euroland.
In our view, the amount of the eventual dollar depreciation is often overestimated. Recall
that the primary factor driving the increase in the US trade and current account deficits
is the relatively strong response of US investment and consumption to a decline in interest rates. Over the adjustment period interest rates will rise, thereby causing an equally
strong reverse effect; this will help reduce the US deficit. The exchange rate adjustment
therefore must be consistent with a slow shift in US output toward traded goods.
We could replace time with net debt on the horizontal axis and have a diagram similar
to that presented in Blanchard et al. (2005). The case we present here is similar to their
discussion of intervention following a shift in preferences away from US goods. The
interested reader is encouraged to work through their analysis of an imperfect substitutes model. Their analysis assumes that interest rates are unchanged, and changes in
absorption are assumed to be related to fiscal policies.

158
8.
9.
10.
11.

The portfolio balance equilibrium is based on the idea that residents of all countries
prefer home assets but can be moved away from their preferred portfolio by differences
in expected yieldsthat is, by interest differentials adjusted for expected changes in
exchange rates.
We have also explored the effects of diversification under the assumption of imperfect
substitution between dollar and euro assets. Our conclusion was that it is not in the
interests of Asian governments to diversify. See Dooley et al. (2004a). The argument
presented here suggests that Asian governments can diversify if they choose to do so but
that this would have little effect on dollar exchange rates.
See Gourinchas and Rey (2005) for a detailed breakdown of the denomination of US
gross foreign assets and liabilities.
See Dooley et al. (2003a).
REFERENCES
Blanchard, Olivier, Francesco Giavazzi, and Filipa Sa (2005). The U.S. Current
Account Deficit and the Dollar. NBER Working Paper no. 11137, National
Bureau of Economic Research, Cambridge, MA.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2003a). Dollars and
Deficits. Where Do We Go from Here? Deutsche Bank, June 17.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2003b). An Essay
on the Revived Bretton Woods System. NBER Working Paper no. 9971,
National Bureau of Economic Research, Cambridge, MA.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2004a). The Revived
Bretton Woods System: The Effects of Periphery Intervention and Reserve
Management on Interest Rates and Exchange Rates in Center Countries.
NBER Working Paper no. 10332, National Bureau of Economic Research,
Cambridge, MA.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2004b). Direct
Investment, Rising Real Wages, and the Absorption of Excess Labor in the
Periphery. NBER Working Paper no. 10626, National Bureau of Economic
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2005). Interest
Rates, Exchange Rates, and International Adjustment. NBER Working Paper
no. 11771, National Bureau of Economic Research, Cambridge, MA.
Gourinchas, Pierre Olivier, and Helene Rey (2005). From World Banker to World
Venture Capitalist: US External Adjustment and the Exorbitant Privilege.
Cambridge, MA.

7.
Current account surpluses and

conflicted virtue in East Asia: China
and Japan under the dollar standard
Ronald McKinnon and Gunther Schnabl1
7.1
INTRODUCTION
The 199798 Asian crisis triggered a still ongoing debate about what the
proper exchange rate policy should be in East Asia. Before 1997, all the
East Asian countrieswith the important exception of Japaninformally
pegged to the dollar at both high and low frequencies of observation. But
opinions diverge as to whether this soft dollar pegging aggravated the
crisis.
The International Monetary Funds (IMF) position is that these soft
dollar pegs accentuated moral hazard in poorly regulated domestic banks
(Fischer 2001). In the absence of immediate foreign exchange risk, they
over-borrowed by accepting foreign currency deposits at lower interest
rates in order to make higher-yield loans in their domestic currencies.
However, floating the exchange rate may not be the correct policy response
for curbing moral hazard in banks. McKinnon and Pill (1999) suggest
that the differential between domestic and foreign interest rates might
actually widen under a volatile float, thus aggravating the temptation to
over-borrow.
While the academic discussion about the pros and cons of more
exchange rate flexibility continues, the focus of the political discussion
about the proper exchange rate regimes in East Asia has shifted toward
the pros and cons of allowing appreciations of the East Asian currencies
against the dollar. Some authors have arguedin particular with respect to
Chinathat countries in the economic catch-up process that are running
high saving surpluses should allow their exchange rates to appreciate to
curtail excessive surpluses in their current accounts (Cline 2005, Frankel
2006).
As we show, the East Asian countries are conflicted in making the decision to allow an appreciation of their currencies, as appreciation need not
reduce the trade surpluses but could cause serious deflation. Therefore,
159

160
to maintain macroeconomic stability, the East Asian countries are well

advised to maintain their dollar pegs, as they have mainly done since
1998.
7.2
THE POST-CRISIS RETURN TO

HIGH-FREQUENCY PEGGING
The evidence in favor of the East Asian return to exchange rate stabilization against the dollar in the new millennium has two facets. The first has
to do with the weight that the smaller East Asian economies assign to the
dollar compared to the weights assigned to other major currencies such as
the yen or euro in their currency baskets, that is, the weights directing
their foreign exchange interventions. The second has to do with the volatility of East Asian dollar exchange rates now as compared to that before the
199798 crisis. Let us discuss each in turn.
The Composition of Currency Baskets
Before the 199798 crisis, the East Asian currencies had de jure pegs to
baskets of major currencies, but typically the weights assigned to various
currencies in the official basket were not announced. Instead, they had de
facto or soft pegs to the dollar, as Figure 7.1 shows. Although many East
Asian countries such as Korea, Thailand, and the Philippines officially
shifted toward more flexible exchange rates after the crisis, most East Asian
countries seem to have returned to having high dollar weights in their
exchange rate strategies.
Using an econometric technique proposed by Frankel and Wei (1994),
McKinnon and Schnabl (2004) use an outside currencythe Swiss
francas a numraire for measuring the exchange rate volatility of the
East Asian countries (except Japan) in order to detect the composition of
their currency baskets.2
For example, if changes in the Korean won to Swiss franc exchange rate
are largely explained by the changes in the US dollar to Swiss franc rate,
the US dollar will have a weight close to one (100 percent) in the Korean
currency basket. In other words, the Korean won can be assumed to be
virtually pegged to the US dollar. Alternatively, if movements in the won
against the Swiss franc are largely explained by movements in the yen
against the Swiss franc, then the yen will have a weight close to one in
Koreas currency basket; and similarly with the euro.3
Figure 7.2 plots 130-day rolling regressions for each East Asian countrys exchange rate against the Swiss franc on the dollars exchange rate

161

n
Ja
02
n
Ja
04
n
Ja
06
n
Ja
Ja
Figure 7.1
94
96
n
Ja
98
n
Ja
00
96
n
Ja
98
Ja
00
Taiwan dollar
n
Ja
Ja
n
Ja
Malaysian ringgit
Ja
02
02
n
Ja
n
Ja
Ja
04
04
04
300
Ja
50
100
150
200
250
300
n
Ja
150
200
250
Ja
90
90
90
n
Ja
n
Ja
n
Ja
92
92
92
n
Ja
94
Ja
96
n
Ja
98
n
Ja
00
n
Ja
96
n
Ja
98
n
Ja
00
Ja
n
Ja
98
n
Ja
Thai baht
96
00
02
n
Ja
Ja
06
06
Ja
04
04
04
Ja
n
Ja
02
02
Ja
Ja
Philippine peso
Ja
Indonesian rupiah
94
94
Ja
Ja
East Asian exchange rate pegs against the dollar, 1990:012006:02 (monthly)
IMF: IFS, central bank of China.
Note: Index 1990.01 5 100. Note different scale for Indonesia.
Source:
n
Ja
94
02
50
92
Ja
100
00
50
Ja
n
Ja
50
90
98
100
100
02
04
06
96
00
98
n
n
n
n
n
n
Ja
Ja
Ja
Ja
Ja
Ja
Singapore dollar
Ja
150
94
96
150
Ja
Hong Kong dollar
Ja
200
Ja
92
94
200
92
n
Ja
Ja
250
90
92
300
Ja
Ja
Ja
250
90
00
Korean won
n
Ja
90
800
700
600
500
400
300
200
100
0
300
n
Ja
50
n
Ja
50
98
06
100
n
Ja
n
Ja
100
96
04
150
n
Ja
150
Ja
02
200
94
n
Ja
250
00
200
Ja
n
Ja
250
92
98
300
n
Ja
n
Ja
300
90
96
Chinese yuan
n
Ja
50
94
50
n
Ja
100
100
92
150
150
n
Ja
200
200
90
250
250
n
Ja
300
300
162

90
1.
90
1.
93
1.
Bloomberg.
90
1.
.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
90
1.
.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
90
1.
.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
96
02
05
99
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Malaysian ringgit
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Hong Kong dollar
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Taiwan dollar
93
1.
.0
01
93
1.
.0
01
93
1.
.0
01
.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
90
1.
90
1.
.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
01
90
1.
.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
93
99
02
05
96
1.
1.
1.
1.
1.
.0
.0
.0
.0
.0
01
01
01
01
01
Thai baht
93
96
99
02
05
1.
1.
1.
1.
1.
.0
.0
.0
.0
.0
01
01
01
01
01
Philippine peso
93
96
99
02
05
1.
1.
1.
1.
1.
.0
.0
.0
.0
.0
01
01
01
01
01
Indonesian rupiah
Figure 7.2 Dollars weight in East Asian currency baskets, 130-trading-day rolling regressions, 1990:012005:12 (daily)
Source:
6
.9
99
02
05
1.
1.
1.
.0
.0
.0
01
01
01
Korean won
1
.0
.9
99
02
05
1.
1.
1.
.0
.0
.0
01
01
01
Chinese yuan
1
.0
01
01
93
96
99
02
05
1.
1.
1.
1.
1.
.0
.0
.0
.0
.0
01
01
01
01
01
Singapore dollar
.0
93
1.
.0
01
01
Note: 1 corresponds to 100 percent.
01
.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
.0
01
90
1.
.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
01
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Current account surpluses and conflicted virtue in East Asia
163
against the Swiss franc using daily exchange rate observations over the
period January 1990 to December 2005. The regression coefficients for
all East Asian countries dollar exchange rates are close to one for the
observed period before the crisis. While the pegged exchange rate regimes
fell apart during the 199798 crisis, following the crisis the coefficients have
again remained high for all East Asian countries. The weights of the dollar
in the currency baskets have remained close to one for China, Hong Kong,
Malaysia, the Philippines, Taiwan, and (although the coefficient is rather
erratic) Indonesia. For Korea, Singapore, and Thailand the weight seems
to decline to some extent but remains well above 50 percent.
Thus, in the new millennium, the dollar retains its predominant weight
in all East Asian currency basketswith the currencies of China, Hong
Kong, and Malaysia remaining tightly fixed to the dollar despite the recent
shift of China and Malaysia toward slightly more flexible exchange rates
(McKinnon 2005). This is not to deny that the yen and, more recently, the
euro (Schnabl, Chapter 10, this volume) are accorded some small weights
in these currency baskets, which have become larger (although much less
than the dollars weight) post-crisis, in particular in the cases of Korea,
Singapore, and Thailand.4
Exchange Rate Volatility against the Dollar
However, knowing the currency basket weights used in the East Asian
exchange rate strategies is not the whole story on exchange rate volatility. In complementary tests, we measure volatility as the percentage daily
change of the national currency against the dollar (log first differences)
from January 1990 through December 2005. We have two standards for
assessing this volatility.
First, we compare the volatilities of the dollar exchange rates of the
smaller East Asian economies with those of the Japanese yen,5 the euro
(German mark), and the Swiss franc.6 Figure 7.3 shows that in the noncrisis periods the daily volatilities of the dollar exchange rates of these more
developed industrial economies are an order of magnitude higher than
those of the smaller East Asian countries. Not only are the daily exchange
rate volatilities of these industrial countries very high, but they do not
change significantly over time. In contrast, the volatilities of the East Asian
currencies are generally much lowerbut with greater variability over
time, with the crisis periods showing up very starkly in the graph.
Second, we compare pre-crisis and post-crisis exchange volatilities. Table
7.1 reports the standard deviations of daily exchange rate fluctuations
against the dollar. In the pre-crisis period (before June 1, 1997), the standard
deviations of the day-to-day exchange rate volatilities against the dollar of

164

90
1.
90
1.
Thai baht
02
05
93
96
99
1.
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
.0
01
99
02
05
1.
1.
1.
.0
.0
.0
01
01
01
Chinese yuan
96
1.
.0
01
93
96
99
02
05
1.
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Malaysian ringgit
.0
01
.9
1
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
93
1.
.0
01
93
1.
.0
01
02
05
99
1.
1.
1.
.0
.0
.0
01
01
01
Japanese yen
96
1.
.0
01
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Philippine peso
99
02
05
96
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Hong Kong dollar
93
1.
.0
01
90
1.
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
90
1.
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Singapore dollar
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Indonesian rupiah
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Euro (Deutsche mark)
93
1.
.0
01
93
1.
.0
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
93
1.
.0
01
99
02
05
1.
1.
1.
.0
.0
.0
01
01
01
Korean won
96
1.
.0
01
02
05
99
1.
1.
1.
.0
.0
.0
01
01
01
Swiss franc
96
1.
.0
01
05
96
99
02
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
New Taiwan dollar
93
1.
.0
01
93
1.
.0
01
Exchange rate volatility against the US dollar of selected crisis and non-crisis currencies, 1990:012005:12
(daily)
Bloomberg.
Volatility is daily percentage changes against the dollar.
Figure 7.3
Source:
Note:
01
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
90
1.
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
Table 7.1
165
Standard deviations of daily exchange rate fluctuations against

the dollar
Chinese yuan
Hong Kong dollar
Indonesian rupiah
Korean won
Malaysian ringgit
Philippine peso
Singapore dollar
New Taiwan dollar
Thai baht
Japanese yen
Swiss franc
Pre-crisis
Crisis
Post-crisis
2004/2005
0.03
0.02
0.17
0.22
0.25
0.37
0.20
0.19
0.21
0.67
0.60
0.69
0.01
0.03
4.43
2.35
1.53
1.31
0.75
0.50
1.55
1.00
0.58
0.66
0.05
0.03
1.03
0.46
0.03
0.47
0.29
0.24
0.36
0.64
0.65
0.68
0.09
0.03
0.51
0.42
0.05
0.23
0.29
0.28
0.28
0.59
0.61
0.69
Note: Percent changes. Pre-crisis 5 February 1, 1994May 30, 1997, crisis 5 June 1,
1997December 31, 1998, post-crisis 5 January 1, 1999May 17, 2004, 2004/2005 5
January 1, 2004December 8, 2005.
Source:
Bloomberg.
all East Asian currencies are much smaller than the standard deviations of
the so-called free floaters (Japan, Euroland, and Switzerland). The standard
deviations of the hard pegs (China and Hong Kong) are close to zero during
and after the crisis. For Indonesia, Korea, Malaysia, the Philippines, and
Thailand, the standard deviations in Table 7.1 increase massively during the
crisis period (June 1, 1997 to December 31, 1998), while those of the noncrisis economies Singapore and Taiwan increase somewhat less.
From the end of the crisis to December 2005, the standard deviations of all
affected countries decline again (Table 7.1). Except in the case of Malaysia,
which fixed firmly to the dollar at the end of 1998, the dollar exchange rate
volatilities of the crisis economies for the whole post-crisis period (19992005)
are slightly larger than before the crisis. In 200405 (the right hand column
in Table 7.1), with the exception of Malaysia and the Philippines, exchange
rate volatilities in East Asia remain somewhat above the pre-crisis levels,
but they are significantly lower than those of the freely floating benchmark
economies. China has allowed for slightly more exchange rate volatility since
July 2005, but has kept its currency tightly fixed to the dollar.
With the benefit of hindsight, the post-crisis return to high-frequency
dollar pegging is hardly surprising. But how is this exchange rate stabilization at high frequencies linked to the exchange rate fluctuations at lower

166
Table 7.2
Standard deviations of monthly exchange rate fluctuations

against the dollar
Chinese yuan
Hong Kong dollar
Indonesian rupiah
Korean won
Malaysian ringgit
Philippine peso
Singapore dollar
New Taiwan dollar
Thai baht
Japanese yen
Swiss franc
Pre-crisis
Crisis
Post-crisis
0.25
0.08
0.26
1.01
1.06
1.19
0.76
1.01
0.43
3.66
2.20
2.62
0.03
0.07
26.54
11.53
6.69
5.25
2.88
2.63
8.88
3.64
2.33
2.60
0.18
0.09
4.57
1.90
0.14
1.57
1.09
1.13
1.60
2.32
2.43
2.40
Note: Percent changes. Pre-crisis 5 February 1994May 1997, crisis 5 June

1997December 1998, post-crisis 5 January 1999February 2006.
Source:
IMF: IFS.
frequencies? Learning from their past vulnerability to yen/dollar depreciation (McKinnon and Schnabl 2003), many East Asian countries seem to
be allowing more dollar exchange rate variability at lower frequencies in
the post-crisis period, as shown by Hernndez and Montiel (2002). Indeed,
Figure 7.1 shows more dollar exchange rate drift after than before the
crisis on a month-to-month basis. For Indonesia, Korea, the Philippines,
Singapore, Taiwan, and Thailand monthly exchange fluctuations are
greater than beforealthough those for China, Hong Kong, and Malaysia
remain close to zero, even after China and Malaysias shift to (slightly)
more exchange rate flexibility in July 2005.
Table 7.2 compares the standard deviations of monthly percentage
exchange rate fluctuations against the dollar in the pre-crisis and post-crisis
periods. We observe that for all East Asian countries except China, Hong
Kong and Malaysia the monthly exchange rate variability against the
dollar is still somewhat higher post-crisis than in the pre-crisis period. In
part, this could be due to the fact that all appreciated together against the
dollar after their over-shooting depreciations in early 1998, but attempts
to reduce exchange rate fluctuations against the Japanese yen and more
recently the euro (Schnabl, Chapter 10, this volume) may also play a role.
There is an alternative way to measure exchange rate smoothing at low
frequencies in the post-crisis era. Suppose we use fluctuations in the more


0
10
15
20
25
30
167
35
Chinese yuan
Hong Kong dollar
Indonesian rupiah
Korean won
Malaysian ringgit
Philippine peso
Singapore dollar
New Taiwan dollar
Thailand baht
Japanese yen
Euro
Swiss franc
Note:
Source:
Positive values correspond to depreciation.

Bloomberg.
Figure 7.4
Exchange rate changes against the US dollar, 1999:01

2001:12 (percent)
freely floating eurodollar exchange rate as the benchmark for measuring

general market pressure against the dollar. We assume that the European
Central Bank, which only conducts inflation policy, leaves the eurodollar
rate to market forces. We partition the data into the following two subperiods: 19992001 (Figure 7.4), when the dollar appreciated generally
against the euro, and 2002 to December 2005 (Figure 7.5), when the dollar
mostly depreciated against the euro (though tending to appreciate during
2005; Figure 7.6).
Figure 7.4 plots the cumulative depreciation of the euro and all East
Asian currencies over the period of dollar appreciation. All East Asian
currencies, except the Philippine peso but including the Japanese yen,
depreciated less than the euro against the dollar. From the beginning of
2002, when the dollar started depreciating, the picture reverses. Figure 7.5
shows that all East Asian currencies except the Korean won appreciated
considerably less against the dollar than did the euro. In order to resist this
exchange market pressure for local currency appreciation, each East Asian
central bank intervened by buying dollarsas reflected in the rise of official
foreign reserves shown in Figure 7.6. Official foreign reserves in East Asian
countries have increased surprisingly quickly in all East Asian countries.7

168
30

25
20
15
10
5
Chinese yuan
Hong Kong dollar
Indonesian rupiah
Korean won
Malaysian ringgit
Philippine peso
Singapore dollar
New Taiwan dollar
Thailand baht
Japanese yen
Euro
Swiss franc
Note:
Source:
Negative values correspond to appreciation.

Bloomberg.
Figure 7.5
Exchange rate changes against the US dollar, 2002:01

2005:12 (percent)
This accumulation of foreign reserves flattened in 2005 as US interest

rates considerably increased (Figure 7.6). Nevertheless, the long-term
trend of reserve accumulation in East Asia persists. In contrast, the official foreign exchange reserves of the benchmark free floaters, Germany
(Euroland, after 1999) and the United States itself, hardly changed over
that time period. Far beyond simply rebuilding their pre-crisis levels of
exchange reserves, East Asian governments have evidently been intervening on a massive scale to prevent their exchange rates from appreciating.
7.3
THE RATIONALE FOR DOLLAR PEGGING

ORIGINAL SIN AND CONFLICTED VIRTUE
The rationale for dollar pegging, as observed in section 7.2, is not primarily based on strong trade ties between East Asia and the United States.
As outlined by McKinnon and Schnabl (2003), the US accounts for only
about one fifth to one quarter of the overall exports of the smaller East
Asian economiesand for much less of their imports. More important is
the widespread use of the dollar as the invoice currency for most of East
Asian trade, even though Japanese trade in the region is as large as that of
the United States (McKinnon and Schnabl 2004).

169

Ja
10
20
30
40
50
90
90
90
n
Ja
Ja
Ja
94
94
94
98
98
98
Thailand
Ja
02
02
02
Ja
Ja
Ja
Malaysia
Ja
China
Ja
n
Ja
06
06
06
n
Ja
Ja
Ja
800
700
600
500
400
300
200
100
0
Ja
18
16
14
12
10
8
6
4
2
0
n
Ja
140
120
100
80
60
40
20
0
90
90
90
n
Ja
Ja
Ja
Ja
98
n
Ja
Ja
98
n
Ja
94
98
Japan
Ja
n
Ja
Philippines
94
02
02
02
Hong Kong
94
06
06
06
n
Ja
Ja
Ja
Ja
120
100
80
60
40
20
0
Ja
120
100
80
60
40
20
0
Ja
40
35
30
25
20
15
10
5
0
90
90
90
n
Ja
Ja
n
Ja
94
94
94
98
98
n
Ja
98
n
Ja
Germany
Ja
Singapore
Ja
02
02
02
n
Ja
Indonesia
n
Ja
n
Ja
n
Ja
06
06
06
Ja
n
Ja
60
50
40
30
20
10
0
Ja
280
240
200
160
120
80
40
0
n
Ja
200
160
120
80
40
0
90
90
90
n
Ja
Ja
Ja
94
94
94
98
98
98
US
n
Ja
n
Ja
02
02
02
n
Ja
Ja
Taiwan
n
Ja
Korea
n
Ja
n
Ja
06
06
06
n
Ja
n
Ja
Official foreign exchange reserves of crisis and non-crisis countries in billions of dollars, 1990:012006:02
(monthly)
IMF: IFS.
Billion dollars. Note different scales on the y-axis.
Figure 7.6
Source:
Note:
n
Ja
80
70
60
50
40
30
20
10
0
Ja
800
700
600
500
400
300
200
100
0
170
This currency asymmetry extends to the financial markets. The dollar

is the dominant inter-bank currency for clearing international payments
and denominating short-term capital flows, and it is the principal official
intervention and reserve currency for governments (McKinnon 2004).
And using a common international money, sometimes called a key currency, in international markets for goods and finance greatly facilitates
commerce, particularly within the increasingly integrated East Asian
economy.
If, in addition, the key currency is itself stable in terms of its purchasing
power over tradable goods and servicesas the US dollar was in the 1950s
and 1960s, and as it has become once more from the mid-1990s into the new
millenniumthen governments in the peripheral countries are induced to
peg their currencies to the key currency in order to better anchor their own
price levels. Ideally, the center country should have no exchange rate objectives of its own, allowing it to follow an independent monetary policy more
purely focused on stabilizing its own price level. In East Asia, the fact that
most of the increasingly integrated economies of the region more or less
peg to the dollar greatly widens the effective dollar areathus enhancing
the anchoring effect of any one of them pegging to the dollar.
In summary, the benign face of the East Asian dollar standard arises out
of the dollars microeconomic role in facilitating international commerce
on the one hand and out of its macroeconomic role as nominal anchor
for national price levels on the other. Both increase incentives for Asian
governments to peg to the dollar (McKinnon 2005). However, this dollar
dominance also has a more malign aspect that, paradoxically, also induces
governments in peripheral countries to peg to the dollar. Domestic financial markets, particularly bond markets, may remain underdeveloped in
part because financial instruments denominated in domestic currencies are
not used in international transactions. Dollar dominance creates unhedgeable foreign exchange risks in countries whose private sector is a significant
net debtor or a significant net creditor internationally.
Before 199798, Indonesia, Korea, Malaysia, the Philippines, and
Malaysia all ran substantial current account deficits (see Table 7.3) that
had accumulated into large, mainly dollar-denominated, net debts at the
time of the great crisis. Post-crisis, these five crisis economies were forced
to run current-account surpluses, and four of them, Korea, Malaysia,
the Philippines, and Malaysia, are rapidly transforming themselves into
creditorsparticularly in their net liquid dollar asset positions. In parallel,
Table 7.3 also shows the chronic international creditors, Japan, Singapore,
Taiwan, and China (a special case discussed more fully below), running
current account surpluses from 1990 to 2005. So all the East Asian countries are now, or are close to being, net international creditors in dollars.

171

1991
Source:
Note:
3.02
7.24
3.14
1.33
0.23
4.46
5.55
5.09
1.94
1993
2.72
16.17
2.66
1.58
0.95
6.06
4.60
5.60
1.28
1994
2.10
17.56
2.07
3.18
1.68
9.71
2.67
8.07
0.23
1995
1.40
15.03
3.91
3.37
4.16
4.43
4.77
8.07
0.88
1996
IMF: IFS, WEO.
2.25
15.58
2.43
2.27
1.62
5.92
5.28
2.00
4.09
1997
1999
2000
3.02 2.57 2.52

22.22 17.85 12.84
1.29 2.78 2.86
4.29 4.13 4.84
11.69 5.51 2.39
13.19 15.92 9.40
2.37 9.48 8.24
12.73 10.13 7.59
3.30 2.11 1.90
1.50 6.28 4.14
1998
2002
2003
2004
2005
2.11 2.83 3.17 3.68 3.59

16.78 17.77 29.03 25.95 28.75
6.36 9.09 9.98 5.74 4.73
4.20 3.91 3.53 1.23 1.07
1.67 0.99 1.96 4.06 2.09
8.28 7.55 12.87 12.57 15.56
1.86 5.71 1.76 2.42 3.01
5.36 5.53 5.56 4.10 2.16
1.46 2.72 3.12 3.99 7.13
5.88 7.58 10.40 9.48 10.69
2001
50.1 84.8 121.6 113.7 124.9 140.9 214.0 300.1 416.0389.5 475.2 519.7 668.1 804.9
117.09 117.63 132.70 93.56 43.83 129.16 244.22 231.28 212.26 176.99 233.44 297.88 357.16 430.48
0.79 1.27 1.72 1.54 1.60 1.70 2.45 3.24 4.24 3.85 4.54 4.74 5.69 6.45
2.97
11.87
4.14
2.00
1.30
3.67
1.89
5.66
1.36
1992
East Asia does not include Hong Kong from 1990 to 1997.
Percent of GDP
Japan
1.45 1.96
Singapore
8.45 11.32
Taiwan
6.96 7.11
Indonesia
2.61 3.32
Korea
0.80 2.85
Malaysia
1.97 8.51
Philippines 6.08 2.28
Thailand
8.53 7.71
China
3.13 3.32
Hong
Kong
United
1.36 0.05
States
Billions of US dollars
Total East 54.28 73.35
Asia
Total US 78.9
2.8
1990
Table 7.3 East Asian and US current accounts in comparison, 19902005
172
For countries on the dollars periphery, we consider the problems facing

debtors and creditors in turn.
Debtor Countries with Original Sin
In their original formulation, Eichengreen and Hausmann (1999) call
original sin a situation in which the domestic currency cannot be used
to borrow abroad (international original sin) or to borrow long term, even
domestically (domestic original sin). Because of incomplete and fragile
financial markets, domestic investors face a maturity mismatch (long-term
projects are financed by short-term loans) or a currency mismatch (projects
that generate domestic currency are financed with dollars). According to
Eichengreen and Hausmann, these mismatches are not primarily because
banks and enterprises are not prudent enough to hedge their foreign
exchange risk exposures. Instead, less developed countries whose liabilities are denominated in foreign currency are unable to hedge because a
typical foreign investor wont accept claims denominated in their domestic
currencies.
Original sin renders monetary authorities less willing to let the exchange
rate move both at low and at high frequencies. At low frequencies,
Eichengreen and Hausmann argue that dollar liabilities increase macroeconomic instability by increasing the credit risk for national balance
sheets. When debt is denominated in foreign currency, exchange rate
fluctuations strongly affect the servicing cost of this debt in terms of the
domestic currency. Sharp depreciations can force indebted enterprises and
financial institutions into default, with considerable risk for the viability of
the whole domestic financial system.
In addition, McKinnon and Schnabl (2004) argue that original sin
affects exchange rate stabilization in emerging markets at high frequencies.
As domestic capital markets are shallow and incomplete, an active forward
market in foreign exchange against the dollaror any other currency
is absent in most developing countries and emerging markets. Forward
transactions by risk-averse traders wanting to hedge their open positions
in foreign currency are difficult, even when the private sector may not be
a net debtor.
Potential market makers such as banks cannot easily cover transactions
involving selling the domestic currency forward for, say, dollars, because
a convenient array of interest-bearing domestic bonds liquid at different
terms to maturity is unavailable for foreigners to hold.8 But the problem
becomes more acute if the economy is a net debtor. Because economies
with uncovered net short-term dollar debts typically have high risk premia
in their domestic currency interest ratesas was the case with Indonesia,

173
Korea, Malaysia, the Philippines, and Thailand before the 199798 crisis
(McKinnon and Pill 1999)individual owners of dollar liabilities see the
cost of forward cover (i.e. the premium on buying dollars forward with the
domestic currency) to be too high. Thus, they typically dont hedge.
This induces the government to provide an informal hedge by keeping
the exchange rate stable to offset the non-existent private market in forward
exchange. If short-term exchange rate fluctuations are low, private banks
and enterprises can then repay their short-term foreign currency debts,
which are largely denominated in dollars, with minimal exchange rate risk.
Thus, if a countrys financial markets are condemned by original sin, its
monetary authorities are induced to undertake high-frequency exchange
rate pegging in order to mitigate payments riskand are induced to hold
official exchange reserves that offset at least part of the private sectors net
international indebtedness.
Creditor Countries with Conflicted Virtue
The original sin argument is adequate for explaining the pervasive fear
of floating in the less developed East Asian debtor countries before the
199798 crisis, as well as in developing countries more generally (Calvo
and Reinhart 2002). But post-crisis, when all East Asian countries ran
current-account surpluses and joined the creditor club (Table 7.3), why
has exchange rate stabilization against the dollar persisted? Even highly
industrialized Japan, which has had a sustained current account surplus
since the early 1980s and has built up an immense international investment
position in US dollars, increased its efforts to stabilize the yen against the
dollar in the new millennium (Hillebrand and Schnabl 2006).
What is the motivation for exchange rate stabilization in the East Asian
creditor countries? In Euroland, most private claims on foreigners are
denominated in euros. In contrast, East Asian countries hold dollars as the
financial counterpart to their cumulative current account surpluses and net
inflows of foreign direct investment. Because of underdeveloped financial
markets or residual capital controls, private or public investors in most
East Asian economies find it more attractive to invest in dollar assets than
in claims on foreigners denominated in their home currencies. Even Japan,
which has a more highly developed capital market, has built up the lions
share of its international portfolio assets in US dollars.
To put it the other way around, foreigners (among whom Americans
are dominant) are disinclined to build up debts denominated in minor
Asian currencies (including the Chinese yuan) as the counterpart of the
huge and still rising US current account deficit (Table 7.3). (Some US firms
are more willing to issue bonds in euros in the now highly developed euro

174
bond markets. But Euroland is not as big a net creditor relative to its GDP
as are the Asian countries collectively.)
There is two-way causation between private portfolio preferences for
dollars and government exchange rate policies. Once an Asian government
sees its private sector accumulating dollar assets (dollar debts in the case of
debtor economies), it becomes more anxious to stabilize its exchange rate
against the dollar. And when neighboring countries that are close trading
partners are also stabilizing their dollar exchange rates, this enhances the
attractiveness of holding dollar assets within any one country. So dollar
dominance helps finance the large US current account deficit. The upshot is
that the main foreign exchange risk associated with this enormous transfer
of capital is shifted to the largely Asian creditor countries.9
Any international creditor country that cannot lend in its own currency
accumulates a currency mismatch that we call the syndrome of conflicted
virtue, the mirror image of the concept of original sin in debtor economies.
Countries that are virtuous by having a high saving rate (most unlike the
United States!) tend to run surpluses in the current account of their international balance of payments, that is, lend to foreigners. With the passage of
time, two things happen. First, as the stock of dollar claims accumulates,10
domestic holders of dollar assets worry more about a self-sustaining run
on the domestic currency forcing an appreciation. Second, foreign governments start complaining that the countrys ongoing flow of trade surpluses
is unfair and the result of having an undervalued currency.
Of course, both interact. The greater the foreign mercantilist pressure
for appreciation of the domestic currency, the greater the concern of the
domestic private holders of dollar assets. This induces them to convert
dollars into domestic currency. As runs on the domestic currency and out
of dollars begin, the government is conflicted because an appreciation
would dampen exports and, if repeated, could set in train serious deflation
ending with a zero interest liquidity trap. But foreign governments may
threaten trade sanctions if the creditor country in question does not allow
its currency to appreciate. Hence, the syndrome of conflicted virtue.
Notice that conflicted virtue would not arise in international creditor countries whose money is internationally accepted. Britain was the
worlds dominant creditor country in the 19th century, with huge net
capital outflows. But sterling was used to denominate most British claims
on foreigners, sometimes with gold clauses. Similarly, for two and half
decades after World War II, the US had large trade surpluses and was the
worlds biggest creditor, with its claims on foreigners denominated mainly
in dollars.
However, the East Asian economies are historically unusual in
being significant international creditor economies whose currencies are

175
relatively little used. In Japan, current account surpluses have persisted

since the early 1980s. Taiwan and Singapore have exhibited current
account surpluses since the late 1980s. Chinas export surpluses, although
not as large relative to its GDP, have persisted since 1994. Since 1998,
all East Asian countries have run current account surpluses (reflecting
their virtuously high saving rates) resulting in high net capital exports
(Table 7.3).
With the longest modern history of current account surpluses, the
build-up of international assets (which are largely dollar-denominated)
continues in Japan (Figure 7.7). Japans net international investment position, as reported by the Japanese Ministry of Finance, has increased since
the early 1980sreaching a new record high of 1.8 trillion dollars at the
end of 2004. Although Chinas build-up of liquid dollar claims has a much
shorter history than Japans, it was and continues to be accelerated by
large inflows of foreign direct investment (FDI)a relatively illiquid longterm liability. McKinnon and Schnabl (2006a) estimate Chinas liquid
dollar-denominated assets at roughly USD 1000 billion by the end of 2005.
Including all the smaller economies as well, the cumulative joint current
account surplus of all East Asian countries since 1990 (Table 7.3) amounts
to nearly USD 3 trillion.
Potential balance sheet losses from the fluctuation of the dollar against
the domestic currency increase as these dollar claims accumulate. Should
the domestic currency appreciate, unhedged individual or institutional
holders of such large dollar assets would be increasingly at risk. Although
all East Asian holders of dollar assets like their higher yields compared to
those on yen assets, they would suffer capital losses if their domestic currency appreciated. Thus, as the economys overall dollar assets accumulate, dollar asset holders become more fearful of a flight from dollars into
the domestic currency, followed by appreciation.
Like in most other East Asian countries, in China the natural currency habitat of domestic nationals is their home currency. Household
consumption expenditures, wages, and claims on financial intermediaries
such as banks (deposits) and insurance companies (annuities) are mainly
in yuan. Chinese firms and households will hold dollar assets only if there
is a substantial business convenience in doing so or if the interest rate on
dollar assets is higher. The primary downside risk is for the yuan to appreciate against the dollar and thus reduce the yuan value of dollar assets.
Depending on how sensitive domestic holders of dollar assets are to this
risk, periodic flights from dollars into yuan could occur just on rumors of
appreciation (McKinnon and Schnabl 2006a).
Should the domestic currency actually appreciate when the world price
level measured in dollars is itself quite stable, each East Asian government

176

2000
1800
total
public
private
Billions of dollars
1600
1400
1200
1000
800
600
400
200
0
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004
Source:
Japanese Ministry of Finance.
Figure 7.7
Net international investment position of Japan, 19802004

(billions of dollars)
worries about the sudden loss of competitiveness of its exporters followed

by a domestic deflationary spiralas experienced by Japan from the mid1980s through the 1990s as a result of the erratically appreciating yen
(McKinnon and Ohno 1997).
Implications for Interest Rates: The Negative Risk Premium
Governments in creditor economies with conflicted virtue may cut domestic short-term interest rates to forestall or slow the conversion of privately
held dollar assets into domestic currency assets. Insofar as people believe
that low short-term rates would persist, domestic long-term interest rates
would also be bid down. Can a new portfolio equilibrium be found in
which, at any given exchange rate, private agents would be willing to
finance the ongoing current account surplus by building up liquid dollar
claims on foreigners rather than the government having to accumulate
official exchange reserves?
Japan has the longest experience with current account surpluses, and
the associated build up of dollar claims, from the early 1980s into the new
millennium. The upper panel of Figure 7.8 shows that interest rates on
long-term (10-year) Japanese government bonds (JGBs) are on average
about 3 to 4 percentage points less than those on long-term US treasuries;
and the lower panel shows more volatile short-term money market rates,

177
16
Japan
US
14
Percent per annum
12
10
8
6
4
2
0
Jan 80
Jan 84
Jan 88
Jan 92
Jan 96
Jan 00
Long-term: 10-year US treasuries and JGBs
Jan 04
20
Japan (call money rate)
US (federal funds rate)
18
16
Percent per annum
14
12
10
8
6
4
2
0
Jan 80
Figure 7.8
Jan 85
Jan 90
Jan 95
Short-term: money market rates
Jan 00
Jan 05
Interest rates in the US and Japan, 1980:012006:02 (percent

per annum)
with Japanese rates, on average, also being substantially lower than US

interest rates.
To account for the sustained interest differential between yen and dollar
assets, we postulate the augmented interest parity relationship:

178
i 5 i* 1 se 1 f,
(7.1)
where i is the (endogenously determined) Japanese long-term nominal

interest rate, i* is the (exogenously given) US long-term nominal interest
rate, s is the yen price of one dollar, se is the expected depreciation of
the yen, and f is the risk premium on yen assets. From the 1970s to the
mid-1990s the interest differential, i 2 i*, had been driven primarily by the
negative se term, when the erratically appreciating yen peaked out in April
1995 (McKinnon and Ohno 1997). Since the mid-1990s, se < 0, and the
interest differential has been driven primarily by the f term, which is also
negative (Goyal and McKinnon 2003).
The coefficient f is the excess yield on dollar assets that a Japanese
investor demands for bearing foreign exchange risk. For a private Japanese
financial institution holding net dollar assets, fluctuations in the yendollar
exchange rate result in fluctuations in the yen value of the net dollar assets,
and hence in the yen value of the net worth of the financial institution.
From this perspective, the dollar asset is a risky one because its liabilities
are denominated in yen. So f captures the excess yield, over and above
expectations of ongoing yen appreciation, on the dollar asset necessary to
induce the domestic financial institution to hold it.
It follows that f is negative for a creditor country such as Japan with
private-sector assets denominated in foreign currency. Conversely, f is
positive for a debtor country whose private sector debts are denominated
in foreign currency. The size of f depends on the countrys net foreign currency position and on the expected variance in its exchange rate against the
dominant foreign money. With a credibly fixed exchange rate, f approaches
zero. However, if fluctuations in the yendollar exchange rate are normal
and if interest rates on yen assets are sufficiently below those on dollar
assets, the Japanese private sectorbanks, insurance companies, trust
funds, and so oncan still be persuaded to fund Japans ongoing current
account surpluses by building up their stocks of higher-yield dollar assets.
But there are limits on how negative this risk premium on yen assets,
and on how wide the associated interest differential, can become. When
American interest rates fell to abnormally low levels, with short rates at
just 1 to 1.5 percent, in 2003 and 2004 and Japanese interest rates were
bounded from below by zero11 (lower panel of Figure 7.8), the spread
simply was not big enough. Then the Japanese private sector refused to
keep acquiring enough dollar assets, which they saw to be riskier, to cover
Japans ongoing current account surplus. Indeed, private agents in Japan
started unloading previously accumulated dollar assets in order to acquire
near zero-yield yen assets! In 2003, the Bank of Japan intervened massively
in the foreign exchange markets, so that official reserves rose by 201.33

179
billion dollars while Japans current-account surplus was just 136.4 billion
dollars (Table 7.4). In effect, the Japanese government over-funded
Japans current account surplus in 2003 and almost in 2004although not
in earlier years.12
The problem of governments over-funding their current account surplusesconflicted virtue in an extreme formhas been common in East
Asia in recent years. Table 7.4 shows the same phenomenon in Taiwan
(2002, 2003, and 2004), Korea (2002, 2003, and 2004), Malaysia (in 2004
and almost in 2003), Thailand (in 2004), and China (2001 to 2005).13
For East Asia as a whole, official reserve accumulation in 2003 and 2004
was 434 and 472 billion dollars, respectively, while the current account
surpluses were just 255 and 362 billion dollars, respectively. But apart
from the extreme case of over-funding in 2003 and 2004, East Asian
governments have been trapped into unduly heavy reserve accumulation for several years because of the reluctance of their private sectors to
accumulate nearly enough dollar assets, as well as their inability to lend in
their domestic currencies to foreigners in order to fully cover their current
account surpluses.
Recently, even more than in Japan, the over-funding problem has
become obvious in China. In the face of increasing current account surpluses and inflows of foreign direct investment, Chinese private firms and
individuals have become less willing to accumulate dollar assets because
of the threat that the yuan might be appreciated. In contrast to Japans,
the Chinese exchange rate had remained tightly pegged to the dollar up to
July 2005, and it has only appreciated slightly since then. But the threat is
there, this is, se < 0. Chinese wealth holders see dollar assets to be riskier
than yuan assets. In addition, even if there is no unidirectional expectation
that the yuan will appreciate, f becomes more negative the more volatile
the exchange rate and the greater the private holdings of dollar assetsand
Chinese interest rates are driven down further.
If the renminbi were to remain credibly pegged to the dollar, as is almost
the case in Hong Kong, then both se and f would equal zero and shortterm money market interest rates (as in Hong Kong) would converge
toward the US federal funds rate. In contrast, suppose the exchange rate
were to remain fixed but investors face the risk that the renminbi might
appreciate. For portfolio equilibrium, the interest rate on renminbi assets
must be less that on dollar assets by se 1 f, which reflects the size of
any expected discrete appreciation, the probability that it will occur, how
distant the event is, and what the subsequent turmoil in exchange market
fluctuations would be. Given an increasing uncertainty about exchange
rate appreciation, the Chinese interest rate would fall below the US interest
rateas has been the case since early 2005 (Figure 7.9).

180

China
Thailand
Philippines
Malaysia
Korea
Indonesia
Taiwan
Singapore
Japan
Table 7.4
CA
RC
CA
RC
CA
RC
CA
RC
CA
RC
CA
RC
CA
RC
CA
RC
CA
RC
44.1
8.5
3.1
7.4
10.9
0.8
3.0
2.0
2.0
0.5
0.9
1.9
2.7
0.5
7.3
3.8
12.0
11.6
68.2 112.6 131.6 130.3 111.0 65.8 96.8 118.8 114.6 119.7
7.7
0.1 26.8 26.4 57.3 34.9
0.5 4.7 74.5 69.5
4.9
5.9
4.2 11.4 14.7 13.9 14.9 18.3 14.8 11.9
6.4
5.7
8.4
9.8 10.5
8.1 5.6
3.5
1.9
3.4
12.5
8.6
7.0
6.5
5.5 10.9
7.1
3.4
8.0
8.9
10.0 0.1
1.3
8.9 2.1 2.3 4.5
6.8 15.9
0.5
4.3 2.8 2.1 2.8 6.4 7.7 4.9
4.1
5.8
8.0
1.8
1.0
0.8
0.8
1.5
4.5 1.7
6.3
3.8
2.0
8.4 4.1
0.8 4.0 8.7 23.2 8.4 40.4 24.5 12.3
1.2
3.3
3.1
5.3
6.9
1.3 13.5 32.3 21.7 22.2
4.2 2.2 3.0 4.5 8.6 4.5 5.9
9.5 12.6
8.5
1.1
6.4 10.0 1.9 1.9
3.2 6.1
4.7
4.9 1.1
1.0 1.0 3.0 3.0 2.0 4.0 4.4
1.6
7.2
6.3
2.3
1.1
0.3
1.3
0.4
3.7 2.8
2.0
4.0 0.2
7.6 6.3 6.4 8.1 13.6 14.7 3.0 14.2 12.4
9.3
4.0
2.7
4.1
4.8
6.6
1.7 11.5
2.7
5.4 1.9
13.3
6.4 11.6
6.9
1.6
7.2 37.0 31.5 21.1 20.5
14.1 23.2
1.8 30.4 22.0 31.5 34.9
5.1
9.7 10.9
87.8 112.5 136.2 172.1 163.9

40.5 63.7 201.3 171.5
4.6
14.4 15.7 27.0 27.9 33.6
4.8
6.5 13.6 16.5
3.9
17.9 25.6 29.3 18.5 16.4
15.5 39.5 45.0 35.1 11.6
6.9
7.8
8.4
3.1
3.0
1.2
3.7
4.0 0.0 2.0
8.0
5.4 12.0 27.6 16.6
6.6 18.3 33.7 43.7 11.8
7.3
7.2 13.4 14.9 20.4
1.0
3.7 10.2 21.9
4.3
1.3
4.4 11.4
2.1
2.9
0.4 0.3
0.3 0.5
2.8
6.2
7.0
7.9
6.6 3.8
0.4
5.7
2.9
7.5
2.0
17.4 35.4 45.9 68.7 158.6
46.6 74.2 116.8 206.7 208.9
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
East Asian current accounts (CA) and changes in official foreign reserves (RC), 19902005 (billions of
dollars)
181

CA
RC
CA 54.3
RC 16.4
4.2
6.4
7.8
6.3
6.2
73.4 117.1 117.6 132.7 93.6
35.0
3.5 64.3 92.2 107.1
2.5 10.3
7.0
9.8 12.4 16.5 15.7 19.0
8.4 29.0 3.2
6.6 11.3
3.6
0.7
6.5
5.2
0.7
43.8 129.2 244.2 231.3 212.3 177.0 233.4 297.9 357.2 430.5
95.1 18.6 55.6 148.5 116.7 108.5 215.8 434.3 472.5 237.0
Source:
IMF: IFS, central bank of China.
Note: East Asia does not include Hong Kong from 1990 to 1997. Shaded areas indicate that reserve accumulation is larger than the current
account balances.
Hong
Kong
East Asia
182
13
US (federal funds rate)
China (interbank rate)
12
11
Percent per annum
10
9
8
7
6
5
4
3
2
1
0
Jan 90
Source:
Jan 92
Jan 94
Jan 96
Jan 98
Jan 00
Jan 02
Jan 04
IMF: IFS, Bloomberg.
Figure 7.9
Short-term interest rates in the US and China (19902005)
All in all, in 2005 and 2006 the appreciation pressure on the East Asian
currencies was somewhat released as sharply rising interest rates in the
US provided private investors with a greater incentive to invest in the US.
In Japan, foreign exchange intervention abated, and the Bank of Japan
started to shorten the supply of liquidity to the Japanese economy. In 2006,
it was believed that the Bank of Japan might even terminate its zero interest
rate policy, and the Japanese money market interest rate may follow the
US interest rate path.
Figure 7.10 shows how short-term interest rates in nine East Asian
countries (not including Japan) have also fallen sharply in the new millennium relative to those of the 1990s. (Unlike that of Japan, their long-term
bond markets are too underdeveloped to get meaningful interest-rate
quotations.) By the standards of less developed countries, their central
banks have allowed short rates to fall to unusually low levels in an attempt
to stem upward pressure on their currencies and excess accumulation of
official reserves.
Ultra-low American interest rates in 200204 made it difficult to establish a negative risk premium as in Japan, but domestic interest rates in
Hong Kong and Singapore approached zero, and those in Taiwan and
Thailand were near the low American level. China, Korea, and Malaysia
were just a percentage point or two above American rates. The Philippines

183

90
90
90
n
Ja
n
Ja
n
Ja
93
93
93
China
United States
Ja
Ja
99
n
Ja
99
02
n
Ja
05
Ja
05
n
Ja
14
12
10
8
6
4
2
0
n
Ja
12
10
8
6
4
2
0
Ja
16
14
12
10
8
6
4
2
0
90
90
90
Ja
Ja
Ja
93
93
93
Ja
Ja
Ja
99
n
Ja
Ja
99
02
n
Ja
05
n
Ja
02
n
Ja
05
Taiwan
United States
Malaysia
Ja
Taiwan
96
96
Malaysia
United States
99
02
05
n
n
n
Ja
Ja
Ja
Hong Kong
96
Hong Kong
United States
Ja
10
15
20
25
Ja
30
25
20
15
10
5
0
n
Ja
90
80
70
60
50
40
30
20
10
0
90
90
90
Ja
Ja
Ja
93
93
93
n
Ja
Ja
Ja
Ja
99
99
Ja
Ja
99
Ja
02
Ja
05
n
Ja
05
02
Ja
05
Thailand
United States
02
Philippines
United States
Ja
Indonesia
United States
Philippines
Ja
Thailand
96
96
Indonesia
96
Monthly money market interest rates, 1990:012006:01 (percent per annum)
n
Ja
02
Singapore
United States
n
Ja
Korea
n
Ja
Singapore
96
96
Korea
United States
96
99
02
05
n
n
n
n
Ja
Ja
Ja
Ja
China (Bank Rate)
IMF: IFS. (Note different scales).
Figure 7.10
Source:
n
Ja
10
9
8
7
6
5
4
3
2
1
0
n
Ja
10
15
20
25
n
Ja
12
10
8
6
4
2
0
184
and Indonesia were outliers, with an apparently significant positive risk

premium. The sharp US interest rate increase since 2004 has triggered
respective adjustments. Interest rates in Hong Kong, Singapore, and
Thailand have followed the path of the Federal Funds Rate closely while
remaining below the US level. In China, Korea, Malaysia, and Taiwan
short-term interest rates are now below the US level, reflecting those countries growing roles as creditors. The Philippines and Indonesia remain
outliers, as their interest rates remain significantly above the US level.
The recent increase in US interest rates hastemporarily?relieved the
threat in China and East Asia, that, like in Japan, interest rates might be
compressed toward zero, that is, forcing the East Asian countries into the
liquidity trap. In the long term, however, this threat will persist as long as
the East Asian countries accumulate dollar-denominated assets, the liquidity trap potentially rearing up whenever US interest rates start to decline
again.
The imperfect solution for ameliorating this risk is for each East Asian
government to keep its dollar exchange rate as stable as it can. This then
reduces the risk faced by unhedged dollar creditors within the economy in
question. Hence monetary authorities resorting to the soft or informal
dollar pegging we observe in non-crisis periodsand which was documented above. To be sure, some governments would like to give full assurance that the domestic exchange rate is never going to change. But not
even China dares to commit itself to an absolutely fixed dollar exchange
rate when its neighbors, who are close trading partners, have not done so.
The spillover effects from other countries changing their exchange rates
are just too great for any country to risk becoming completely inflexible
in responding to either an appreciation or a depreciation by a neighboring
country.14 So, short of the adoption of a full-fledged system of regional
dollar parities (a difficult exercise in collective action although a potentially
great public good for East Asia), soft pegging is the result.
7.4
THE SUSTAINABILITY OF THE DOLLAR

STANDARD
The large collective trade surplus of the East Asian countries with the
United States, matched by the huge East Asian net capital exports to the
US, has become a stylized fact. But noting this fact alone does not tell us
anything about causality or sustainability. Do the East Asian governments
acquire dollar bonds in a conscious effort to keep their exchange rates
undervalued in order to promote exports and development? Or is the huge
current account deficit of the United States, reflecting low American saving

185
and an unlimited line of credit with the rest of the world, forcing more and
more countriesbut particularly those in East Asiato run with current
account surpluses that lead to conflicted virtue?
The DFG Approach
Dooley, Folkerts-Landau, and Garber (DFG; 2003, 2004, Chapter 6,
this volume) argue that the causality runs from East Asian nations trade
surplusesand their governments mercantilist policiesto the US trade
deficit. Their argument has deeper historical roots. Under the world dollar
standard since 1945, DFG see the world divided into a center (the United
States) and its periphery. For the 1950s and 1960s, when exchange rates
were more or less fixed under the Bretton Woods agreement, they argue
that the important periphery was Western Europe and Japan.
In order to recover more quickly from the ravages of the war, these
countries cooperated (implicitly) to keep their currencies undervalued in
order to promote both high export growth in manufactures and investment
in higher tech industrial export activities. (In the 1950s and 1960s, less
developed countries were then mainly producers and exporters of primary
products or embarking on import-substituting industrialization. As such
they were not a mercantile threat to the center country.) The cost to the
European countries was the rapid build-up of low-yield, if highly liquid,
dollar assets.
But this cost, if any, may have been small or nonexistent. Despres et al.
(1966) put forward what they called the minority view that the United
States, with its more highly developed long-term capital market, was simply
providing financial intermediation services to the Europeans by lending
long in illiquid forms, including foreign direct investment, while borrowing
back short in more liquid forms as Europeans built up dollar bank accounts
and official foreign exchange reserves.15 In the 1950s and 1960s, the US had
a current account surplus. This intermediation argument still holds in a
more limited way today for countries with less developed domestic financial markets, such as China. The United States is an important source of
foreign direct investment into China, financed in large measure by Chinas
huge build-up of more liquid dollar claims on the United States. The difference now, of course, is that the United States, with its large current-account
deficit, is no longer a net lender to developing countries
In the new millennium, DFG suggest that European countries have
matured from being peripheral into having their own fully developed
financial marketsparticularly with the advent of the euro. In our terminology, European countries are now redeemed from original sin (for those
that are debtors) and from conflicted virtue (for those that are creditors).

186
Therefore, they are quite willing to allow their currencies (principally the
euro) to float relatively freely against the dollar without any inordinate
build-up of dollar exchange reserves. DFG label these relatively free floaters, including Canada and Australia, capital-account countries.
However, DFG identify a new periphery of the United States composed
of what they call trade-account countries: mainly our high-growth East
Asian economies. DFG claim that these trade-account countries intervene
heavily in the foreign exchanges to keep their currencies undervalued in
order to stimulate export expansion into the American market, with a
consequent stimulus to investment in these export activities. (As befits
the center country in the world dollar standard, the United States has no
independent exchange rate objective.)
The East Asian countries are willing to bear the opportunity cost, in the
form of a huge build-up of low-yield dollar reserves, of their trade surpluses
for this development objective. At the same time, the American governments own borrowing constraints on financing wars, cutting taxes, and
so on, has been greatly softened because of the large (incidental) foreign
capital inflow embodied in the large US current account deficits. Because
both sides see themselves as benefiting from this arrangement despite
mercantilist conflicts in particular industries, DFG see the East Asian
countries large trade surpluses and undervalued exchange rates to be
sustainable for a sustained period of time.
The MCS Approach
In contrast to DFG, McKinnon and Schnabl (MCS) hypothesize that the
relatively high-saving East Asian countries are collectively being forced into
running current account surpluses with the relatively low-saving United
States. Because of the asymmetrical nature of the world dollar standard,
the US alone has a virtually unlimited dollar line of credit with the rest of
the world (McKinnon 2004). With low American household saving and the
large deficits of the US federal government, the United States is drawing on
this dollar line of credit with a vengeance: Americas current account deficit
was about 6.5 percent of GDP in 2005 and was set to approach 7 percent
in 2006. But this American savingsinvestment imbalance is not a result
of mercantilist governments in East Asia undervaluing their exchange
rates. The East Asian countries are covering a considerable part (about
half), but by no means all, of the US current account deficit. An increasing
share of the US current account deficit is covered by oil-exporting countries
such as Russia and the Middle East.
Why, then, are East Asian governments intervening so heavily (Figure
7.6) to keep their currencies from appreciating if it is not to generate a trade

187
surplus? MCS argue that they are trapped by conflicted virtue. A sharp
appreciation by any one East Asian country would: first, impose capital
losses on domestic holders of dollar assets and make them even more
reluctant to finance future current account surpluses; second, cause an
immediate loss of mercantile competitiveness to East Asian neighbors with
whom they are all closely integrated in trade and with whom they compete
in third markets; and third, risk a macroeconomic slowdown followed by
deflation, the more so if the appreciation is repeated.
However, undergirding the willingness of East Asian countries to maintain their dollar pegs is the presumption that the purchasing power of the
dollar will remain fairly stable through time. Although American fiscal
policy seems to be out of control, these countries presume that Federal
Reserve Chairman Ben Bernanke will continue using Americas monetary policy to stabilize the US price level, as in the 1990s, into the new
millennium. So MCS see exchange rate policy as an important adjunct
of the domestic monetary policy of a peripheral country. On the dollars
periphery, an appreciating currency is a recipe for ongoing deflation with
a potential liquidity trap for interest ratesas Japan experienced with its
forced appreciations from the mid-1980s to the mid-1990s (Goyal and
McKinnon 2003, McKinnon and Schnabl 2006a).
But monetary cum exchange rate policy cannot predictably influence a
countrys net trade balance, which comes down to saving and investment
propensities at home relative to those abroad (McKinnon and Schnabl
2006b). Thus, we part company with DFGs claim that the currencies of
the East Asian countries are now undervalued in the new millennium,
and we also disagree with their claim that the currencies of the Western
European countries and Japan were undervalued in the 1950s and 1960s.
Putting aside Balassa-Samuelson effects associated with rapidly growing
economies, price inflation in Europe and Japan in the 1950s and 1960s was
about the same as that in the United States. By this criterion, the Bretton
Woods exchange rates were more or less right as long as the American
price level remained stablewhich it was until the very end of the 1960s.
Similarly, price inflation in most East Asian countries today is about the
same as it is in the United States: their soft dollar pegs are not obviously
misaligned. But with conflicted virtue, the amount their currencies would
appreciate if floated has no well-defined upper bound.
When DFG link undervalued currencies to trade surpluses, they (like
most economists) probably have something like the textbook elasticities
model of the balance of trade in their minds. This intuitively appealing
model predicts that an appreciation will slow a countrys exports while
stimulating its imports, resulting in the reduction of its trade surplus.
However, the elasticities model only applies to an economy that is insular:

188
its foreign trade is a fringe activity, it is closed to international investment

flows, and setting the exchange rate is separable from monetary policy.
On the other hand, for a fully open economy, McKinnon and Ohno (1997,
Chs 6 and 7) show that an appreciation could cause a sufficient slump in
domestic investment for imports to fall more than exports, so that the net
trade balance would improve. They conclude that the effect of an exchange
rate change on the net trade balance is ambiguous.16
However, MCS agree with DFG that the present dollar-based regime
with large East Asian trade surpluses and huge trade US deficits is sustainable, or at least is not on the verge of an imminent breakdown. After all,
something like this regime has been around since early 1980s when Japan
was the dominant external creditor of the United States. But DFG believe
it to be a product of export mercantilism in peripheral countries, whereas
MCS see them as being trapped by improvident US saving behavior.
7.5
CONCLUSION
The collective macroeconomic consequences of each East Asian government opting individually to peg to the dollar, if only softly, enlarges the
effective zone of stable dollar prices far beyond each countrys direct trade
with the United States (McKinnon 2005). Thus each national central bank
can lean more heavily on its own stable dollar exchange rate to anchor its
domestic price levelwhich in turn helps its neighbors, with whom it is
closely connected in trade, to stabilize their own. A virtuous circle for a
change!
From the 1980s up to the Asian crisis of 199798, most East Asian countries could pin down their wholesale price levels by anchoring their currencies to the dollar (McKinnon and Schnabl 2004). Only the wholesale price
indices of Indonesia and the Philippineswhich allowed their currencies to
depreciate continually, but in a controlled fashionrose significantly. This
common dollar anchor was more robust because all East Asian countries
except Japan participated in it. International commodity arbitrage within
the whole of East Asia, and not just with the United States, helped pin
down the price level of each participating country.
In the 1990s and earlier, the need for a nominal anchor in East Asia
reflected the concern of the smaller countries with the ever-present threat
of inflation. (Japan, of course, had been mired in a deflationary spiral for
more than a decade and only began to show signs of recovery in 2005.)
In the new millennium, however, the increasing need is for a common
monetary anchor against the threat of deflation. As the East Asian economies build up their liquid dollar balances in the process of transforming

189
themselves from being dollar debtors into dollar creditors (providing the
counterpart to huge American trade deficits), the threat of forced appreciation followed by deflation, that is, the problem of conflicted virtue, has
become more acute.
Here, again, collective dollar exchange rate stability in East Asia is a
public good. As long as all the now more highly integrated East Asian
economies stand firm against foreign pressure to appreciate, it will be easier
for any one country to avoid appreciating. However, if a major country
loses control and allows its currency to appreciate, this could set in train a
new East Asian currency crisis with a round of contagious appreciations as
private holders of dollar assets in other East Asian countries become more
nervous and dump them.
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
We thank Wolfgang Bretschneider and Adrian Hhl for excellent research assistance.
An earlier version of this chapter was published in International Finance 7 (2004):
169201.
Any other outside currency, such as the pound sterling, could serve equally well as
numraire.
German mark before January 1999.
For a more detailed econometric analysis of the evolution of these basket weights before
and after the 199798 crisis, see Schnabl (Chapter 10, this volume).
Hillebrand and Schnabl (2006) show that the exchange rate volatility of the yen against
the dollar has declined since 1999.
The euro and the Swiss franc are widely regarded as floating freely against the dollar.
Singapore hides parts of its overseas assets in its Government Investment
Corporation.
In contrast, forward exchange transactions between any two industrial countries can
thrive because each country has a well-developed domestic bond market denominated
in its domestic currency. Long-term forward markets, with a well-defined forward
premium equal to the interest differential between the two national bond markets at
each term to maturity, can thrive at much lower cost.
During 2005 and 2006 the importance of the oil-exporting countries as counterparts of
the US current account deficit increased significantly.
For empirical estimates of the stocks of these liquid dollar claims see Goyal and
McKinnon (2003), for the case of Japan, and McKinnon and Schnabl (2006a) for the
case of China.
The astute reader will note that the story of how foreign exchange risk associated with
conflicted virtue creates a negative risk premium on yen assets is also an explanation of
why Japan has fallen into a zero interest liquidity trapthe macroeconomic implications of which are spelled out in Goyal and McKinnon (2003) and McKinnon (2005).
Hillebrand and Schnabl (2006) analyze the impact of Japanese foreign exchange intervention on the yen-dollar exchange rate.
In China, conflicted virtue has been rendered more acute by heavy inflows of foreign
direct investment (McKinnon and Schnabl 2006a).
Think of what happened to Argentina in the late 1990s when Brazil and Chile allowed
large depreciations of their currencies.
For a more up-to-date assessment of Despres et al. (1966) see Bisignano (2004).

190
16.

See also our critique of William Clines (2005) advocacy (based on the elasticities model)
of massive dollar devaluation to reduce Americas trade deficit (McKinnon and Schnabl
2006b).
REFERENCES
Bisignano, Joseph (2004). Machlup Was Right: Despres, Kindleberger and Salant
Briefly Revisited. Manuscript, Bank for International Settlements, Basel.
Calvo, Guillermo, and Carmen Reinhart (2002). Fear of Floating. Quarterly
Journal of Economics 117: 379408.
Cline, William (2005). The United States as a Debtor Nation. Washington, DC:
Institute for International Economics.
Despres, Emile, Charles Kindleberger, and Walter Salant (1966). The Dollar and
World Liquidity: A Minority View. The Economist, February 5.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2003). An Essay on
the Revived Bretton Woods System. NBER Working Paper no. 9971, National
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2004). The Revived
Cambridge, MA.
Eichengreen, Barry, and Ricardo Hausmann (1999). Exchange Rates and
Financial Fragility. NBER Working Paper no. 7418, National Bureau of
Economic Research, Cambridge, MA.
Fischer, Stanley (2001). Exchange Rate Regimes: Is the Bipolar View Correct?
Journal of Economic Perspectives 15: 324.
Frankel, Jeffrey (2006). On the Yuan. The Choice between Adjustment under
a Fixed Exchange Rate and a Flexible Exchange Rate. CESifo Studies 52:
24675.
Frankel, Jeffrey, and Shang-Jin Wei (1994). Yen Bloc or Dollar Bloc? Exchange
Rate Policies in East Asian Economies. In Takatoshi Ito and Anne Krueger
(eds), Macroeconomic Linkages: Savings, Exchange Rates, and Capital Flows.
Chicago: University of Chicago Press, pp. 295329.
Goyal, Rishi, and Ronald McKinnon (2003). Japans Negative Risk Premium
in Interest Rates: The Liquidity Trap and Fall in Bank Lending. The World
Economy 26: 33963.
Hernndez, Leonardo, and Peter Montiel (2002). Post-crisis Exchange Rate
Policy in Five Asian Countries: Filling in the Hollow Middle? Journal of the
Japanese and International Economies 17: 33669.
Hillebrand, Eric, and Gunther Schnabl (2006). A Structural Break in the Effects
of Japanese Foreign Exchange Intervention on Yen/Dollar Exchange Rate
Volatility. ECB Working Paper no. 650, European Central Bank, Frankfurt.
McKinnon, Ronald (2004). The World Dollar Standard and Globalization: New
Rules for the Game? In Leo Michaelis and Mark Lovewell (eds), Exchange
Rates, Economic Integration, and the International Economy. Toronto: APF
Press, pp. 328.
McKinnon, Ronald (2005). Exchange Rates under the East Asian Dollar Standard:
Living with Conflicted Virtue. Cambridge, MA: MIT Press.

191
McKinnon, Ronald, and Kenichi Ohno (1997). Dollar and Yen: Resolving Economic
Conflict between the United States and Japan. Cambridge, MA: MIT Press.
McKinnon, Ronald, and Huw Pill (1999). Exchange Rate Regimes for Emerging
Markets: Moral Hazard and International Overborrowing. Oxford Review of
Economic Policy 15: 42364.
McKinnon, Ronald, and Gunther Schnabl (2003). Synchronized Business Cycles
in East Asia and Fluctuations in the YenDollar Exchange Rate. The World
Economy 26: 106789.
McKinnon, Ronald, and Gunther Schnabl (2004). The East Asian Dollar
Standard, Fear of Floating, and Original Sin. Review of Development Economics
8: 33160.
McKinnon, Ronald, and Gunther Schnabl (2006a). Chinas Exchange Rate and
International Adjustment in Wages, Prices, and Interest Rates: Japan Dj Vu?
CESifo Studies 52: 276303.
McKinnon, Ronald, and Gunther Schnabl (2006b). Devaluing the Dollar: A
Critical Analysis of William Clines Case for a New Plaza Agreement. Journal
of Policy Modeling 28: 68394.


8.
Three cases for monetary

integration in East Asia
Ulrich Volz1
8.1
INTRODUCTION
Since the Asian financial crisis, monetary and financial integration has
become a much-discussed topic in East Asia.2 The past years have seen
a proliferation of proposals for fostering East Asian integration, ranging
from currency baskets and regional exchange rate systems to monetary
union. But unlike in Europe, which saw long-standing discussions on
the costs and benefits of monetary unification, a proper debate is not yet
underway in East Asia. Instead, the classical European arguments for
and against monetary integration are implicitly being adopted, despite the
very different histories of economic development of East Asian countries
as compared to those of European countries. The implications of monetary
integration for East Asia have barely been explored.
The discussion centers very much on whether East Asia qualifies as an
optimum currency area (OCA), even though OCA theory has been rightfully criticized for its static point of view (Frankel and Rose 1998, Schelkle
2001a, 2001b). It is basically concerned with an ex ante analysis of the
costs and benefits of monetary integration and does not take into account
changes in economic activity that are induced through a policy of integration. Moreover, OCA theory is limited to an analysis of the allocative
effects of monetary integration but fails to address potential accumulative
effects. The development context of monetary integration has largely been
ignored. But monetary integration must not be analyzed solely in terms of
whether it is advantageous given the present conditions. It is also important to ask whether and in what ways monetary integration can contribute
to overcoming structures that present obstacles to economic development
(Roy and Betz 2000). In this context, this chapter investigates three cases
for pursuing monetary integration in East Asia in the long run.
First, the chapter discusses the potential trade-creating effects of monetary integration. This argument is grounded on the improvement of allocative efficiency within a region and was intensively discussed for Europe in
195

196
Toward monetary integration in East Asia
the 1990s. For East Asiaa region for which intra-regional trade accounts
for almost 50 percent of overall tradeexchange rate spillover effects from
one country to another are also of great importance. This is even more
the case as East Asian countries also compete against each other in third
markets. Because of neighborhood effects, nations decisions to fix or float
should not be made independently of each other. Using a simple gravity
model, the chapter shows that the similarity of currency regimes has a
significant and large positive effect on trade in the region.
Second, the chapter draws attention to the potential role of regional
monetary integration in overcoming the problems of original sin and
conflicted virtue, an issue that has rarely been analyzed so far. While
original sin refers to a countrys inability to borrow in its own currency,
whether internationally or domestically, conflicted virtue describes a countrys inability to lend in its own currency. Both original sin and conflicted
virtue force weak currency countries to accumulate a currency mismatch.
Estimations of the causes of original sin and conflicted virtue, as well as
the existing literature on the determinants of key currency status, suggest
that the relative size of an economy is an important explanation for these
phenomena. Forming a larger entity through economic integration might
therefore be a way for economically small countries to remedy the two
problems.
Third, the chapter highlights regional monetary integration as a way of
(re)gaining some degree of monetary independence in the region. In the
European discourse, opponents of monetary union argued that giving up
ones own currency would mean the loss of a forceful policy instrument.
By analyzing the policy decisions of national central banks, the chapter
shows that member countries of the European Monetary System (except
Germany) had already abandoned monetary and exchange rate autonomy
long before entering the European Monetary Union (EMU) and that
Germany has, in fact, been the only country that has actually lost monetary
autonomy through EMU. Virtually all other euro countries have regained
a voice in monetary policy decisions through EMU membership. Likewise,
the chapter argues that the current (informal) East Asian dollar standard
(i.e. the common practice of (soft) pegging to the US dollar) has created a
situation in which East Asian countries (except Japan) have largely abandoned monetary policy autonomy. Through the creation of a common
currency, which could float freely against the dollar, East Asian countries
could potentially (re)gain some degrees of shared monetary independence.
The rest of the chapter develops these arguments. None of these three
cases, however, is sufficiently strong to support monetary union in East Asia
on its own. The final decision to move toward monetary unificationor
the refusal to do sowill eventually be based on political considerations.

8.2
197
MONETARY INTEGRATION AND TRADE

The greatest thing you ever can do now
Is trade a smile with someone whos blue now
Its very easy just...
(Led Zeppelin in Friends, from the album Led Zeppelin III, 1970)
Over the past few decades, the share of intra-regional trade in total trade
has reached high levels in East Asia (Table 8.1), and the region is set to
increase these trading links even more. With the exception of Cambodia,
trade with neighboring countries is of great importance for all East Asian
nations, with the share of regional exports in total exports ranging from
around 40 percent (Laos, China, Japan, Korea, Vietnam) to over 50
percent (Indonesia, Malaysia, Myanmar, the Philippines, Singapore,
Thailand, Brunei). In recent years, East Asian countries have been busy
signing trade agreements with each other to further encourage trade. In the
largest of these, the ASEAN countries have agreed with China to create the
worlds largest free trade area by 2012, with more than 1.7 billion people.
Moreover, further ASEAN negotiations are under way with Japan and
Korea.3
In addition to trade linkages, intra-regional interdependence in foreign
direct investment (FDI) has also increased dramatically (Kawai and
Urata 2004). Extensive regional production networks have developed in
East Asia. In particular, multinational corporations have formed regional
supply chains and production networks, taking advantage of intraregional divisions of labor and promoting the specialization of production
by breaking the production process down into different sub-processes
within the same industry.
For a region as economically intertwined as East Asia, exchange rate
spillover effects from one country to another are of great importance.
There are numerous problems faced by countries that are close trading
partners but follow different exchange rate regimes.4 First, exchange rate
disagreements could lead to reduced exports from the country that loses
competitiveness to its partners. This could evoke increased protectionism
and even a scaling back or elimination of trade agreements. The country
that loses competitiveness as a result of a real exchange rate appreciation
vis--vis its trading partners may employ anti-dumping or other administrative measures (if tariffs are precluded through trade agreements) to
protect domestic firms. This, in turn, could trigger a trade war as well as a
round of beggar-thy-neighbor devaluations. Second, regional trade agreements may spark fierce competition for the location of investment, and
swings in the bilateral exchange rates may have important consequences
for the location of new investments and might even shift the location of

198

Source:
Note:
8.64
55.89
21.09
16.95
12.28
22.73
16.96
0.62
18.77
0.00
18.17
12.96
16.10
20.19
17.24
2.59
25.87
18.08
13.99
12.63
15.78
14.94
28.55
12.57
15.80
17.18
14.46
14.87
23.59
16.49
EU
55.60
40.04
44.78
36.56
51.34
58.39
54.77
53.24
50.41
39.64
47.12
73.94
12.55
41.31
73.89
12.26
24.31
57.67
53.66
33.78
37.63
36.56
45.38
57.23
46.84
43.41
45.30
38.05
43.28
ASEAN14 ASEAN13
8.55
1.35
10.10
5.17
20.12
6.44
13.98
13.60
12.38
38.11
3.50
12.39
5.32
22.31
Japan
0.24
3.50
0.01
2.80
4.11
1.93
2.37
3.89
13.96
0.24
4.69
2.18
6.75
7.82
Korea
8.37
19.32
26.76
2.11
12.65
7.09
14.61
18.39
12.49
10.59
11.52
4.55
1.34
China &
HK
1.94
6.26
7.14
0.00
5.97
1.16
7.93
9.82
5.12
1.59
4.94
0.05
0.29
17.00
HK
IMF/DTS.
ASEAN13 includes ASEAN countries plus China, Japan, and Korea. ASEAN14 also includes Hong Kong.
Brunei
Cambodia
China
Hong Kong
Indonesia
Japan
Korea
Laos
Malaysia
Myanmar
Philippines
Singapore
Thailand
Vietnam
Unweighted average
USA
Table 8.1 Export as a percentage of total exports in 2004
44.01
6.44
13.06
19.61
2.10
6.69
5.93
6.69
8.57
7.38
9.01
10.39
4.50
1.05
China
17.32
7.47
7.23
6.15
18.17
12.90
9.47
32.86
25.09
45.27
17.23
24.30
22.01
13.07
18.47
ASEAN
199
existing investments. Third, and finally, a change of the exchange regime in

one of the partner countries may cause an exchange rate crisis in the other.
Exchange rate depreciation in one country may reduce the credibility of
the partners commitment to a fixed parity and can generate speculative
attacks on its currency. A country may thus be forced to abandon its
preferred exchange rate policy because of the exchange rate disagreement.
This, in turn, could have repercussions for the country that sparked the
crisis.5
Hence, the deeper trade relations become and the more comprehensive trade and investment agreements are, the more important the
question of macroeconomic policy coordination becomes. Increasing
intra-regional trade turns formerly second-order effects into issues that
have to be addressed with first-order preference. In highly integrated areas,
policy coordination to achieve real exchange rate consistency is therefore
essential.
Exchange rate cooperation becomes even more important if countries
also compete against one another in third markets, as is the case in East
Asia. Hence, McKinnon (2005) describes mutual exchange rate stability in
East Asia as the quintessential public good. Because of neighborhood
effects, he reasons, a nations decision to fix or float should not be made
independently of other nations decisions.
While these arguments favor exchange rate cooperation over flexible exchange rates, there is no implicit logic for currency integration.
Eichengreen (1998) points out that the need for monetary integration
depends on the degree of trade integration that is to be achieved. A customs
union can be sustained despite the existence of separate national currencies that fluctuate against one another.6 But deeper integration, including
the free movement of goods, services, capital, and people, implies an even
greater degree of openness of domestic markets and more intense crossborder competition, making exchange rate changes more disruptive. The
European Union (EU) has always given great importance to this issue.
Indeed, the EUs reasoning for monetary union can be described as one
market, one money (Emerson et al. 1992). If East Asian leaders want to
foster deeper integration akin to Europes single market, they will need to
contemplate monetary integration like in Europe.
More recently, the potential trade-creating effect of a common currency
union has gained wide attention in the literature, reinforcing the trade
argument for monetary integration. Rose (2000) turned the EUs logic
upside down and established the case for one money, one market. While
there has been a lot of controversy about the actual magnitude of the tradecreating effect of monetary union, a broad agreement has emerged that a
currency union has a significant positive effect on trade.7

200
While it is impossible to accurately predict the trade creating-effect of a

potential monetary union in East Asia, we can estimate the effect that the
similarity of currency regimes has on bilateral trade in the region. To do
this, we apply the following gravity model to the trade data from 13 East
Asian countries over the period 19802003:8
Tijt 5 b0 1 b1 Fijt 1 b2 Dijt 1 b3 I1ijt 1 eijt,
where Tijt is an indicator for the level of bilateral trade integration between
countries i and j at time t, Fijt is an indicator for the intensity of bilateral
FDI, and Dijt is an indicator for the similarity of currency regimes of the
two countries in question. I1ijt is a vector comprising the standard variables
of the gravity model, and eijt is a well-behaved error term.
The gravity model has become standard in the international trade literature, and hence one can draw on a rich body that employs it. We include
the standard variables of the gravity modelthat is, the product of the
natural logarithms of the populations of the two countries under investigation, a variable describing their combined GDP computed as the product
of the natural logarithms of each countrys GDP, the natural logarithm of
the distance between the two countries, and a dummy variable indicating
whether the countries share a common land border.9
As in Frankel and Rose (1998), bilateral trade integration is measured
by computing
Tijt 5 (xijt 1 mijt) / (Xit 1 Mit 1 Xjt 1 Mjt),
where xijt are the exports from country i to j during period t, mijt are the
imports to country i from j, and Xit, Mit, Xjt, Mjt are the total exports/
imports of country i/j respectively. A higher value of Tijt thus represents
a greater trade intensity between countries i and j. As is common in the
literature, we take the natural logarithm of the ratio. The data are from the
IMFs Direction of Trade Statistics. Data for Taiwan are from the Taiwan
Statistical Data Book 2004.
Similarly, FDI intensity is measured as
Fijt 5 (fdiijt 1 fdijit) / (GDPi 1 GDPj),
where fdiijt denotes FDI to country i from country j and fdijit denotes FDI
to country j from i. The bilateral FDI flows are divided by the joint GDP
of the two countries. As with the trade measure, Fijt takes a higher value
the more integrated the countries are. The bilateral FDI data are taken
from the UNCTAD Foreign Direct Investment Statistics and CEIC Asia

201
Database. The GDP data are from the IMFs International Financial
Statistics and the Taiwan Statistical Data Book 2004.
To measure the similarity of the currency regimes of the two countries in
question, hypothetical currency baskets are estimated in order to compute
the following indicator (see Volz 2005):
Dij 5 1 |wit wjt|,
where wit is the weight of the US dollar in a hypothetical currency basket of
country i. To estimate the weights of the currencies included in the hypothetical currency baskets against which East Asian countries manage their
currencies, we follow Frankel and Wei (1994) and regress each East Asian
currency e on a constant c, the US dollar, the euro, and the Japanese yen:
ln (e/CHF) t 5 c 1 b1 ln (USD/CHF)t 1 b2 ln (EUR/CHF)t
1 b3 ln (JPY/CHF)t 1 et.
The Swiss franc, which can be assumed to be uncorrelated with the three
basket currencies as well as with the East Asian currencies, is used as
numraire in order to minimize multicolinearity problems. The b coefficients are the weights of the basket. stands for the first-difference operator
and t for time. All variables are in natural logarithms. b1 is used to compute
the above indicator (Dij); that is, b1 5 wit. If both countries give the same
weight to the dollar in their (hypothetical) currency basket, Dij is one; if one
of the countries chooses a hard fix to the dollar (b1 5 1) while the other one
chooses a zero dollar weight in its currency basket (b1 5 0), Dij is zero.10
The results presented in Table 8.2 show a strong trade-creating effect of
similar currency regimes in East Asia. The coefficients for the similarity of
the currency regime are large and significant. The trade-creating effect gets
smaller if China is excluded from the regression, which can be explained by
a Japan effect: Japan is a driving force in intra-regional trade but is the
only country in the region that does not adhere to the informal East Asian
dollar standard. If Japan is omitted from the regression, the currency effect
becomes even larger and highly significant. That is, exchange rate stability
is beneficial for trade within East Asia, and, given the previous findings in
the literature that monetary union has an even larger effect on trade than
a hard peg, it is also sensible to expect a significant trade-creating effect of
monetary unification in East Asia.
The coefficients of the other variables are in line with theoretical expectations. Strong FDI links, a shared land border, and larger GDP all
increase bilateral trade whereas distance has the inverse effect. Population

202
Table 8.2
Determinants of trade in East Asia
FDI
Similarity of
currency regime
Distance
Common
border
GDP
Population
Constant
Number of
observations
R-squared
Full sample
Without
China
Without
Japan
Without China
and Japan
0.1012***
(0.0369)
1.1036***
0.0921***
(0.0367)
0.8280*
0.0610
(0.0432)
2.0410***
0.0356
(0.0405)
1.5680**
(0.4633)
20.5866***
(0.1701)
0.7391***
(0.4633)
20.3055*
(0.1813)
1.1925***
(0.6743)
20.6808***
(0.2023)
0.7495***
(0.6854)
20.2130
(0.2097)
1.3917***
(0.2808)
0.0210***
(0.0017)
20.0062**
(0.0028)
212.2067***
(1.4469)
128
(0.3283)
0.0199***
(0.0018)
20.0048
(0.0037)
213.9071***
(1.5486)
104
(0.2853)
0.0239***
(0.0025)
20.0073**
(0.0032)
214.0290***
(1.6989)
100
(0.3255)
0.0229***
(0.0024)
20.0054
(0.0039)
217.3330***
(1.8779)
79
0.5921
0.5951
0.5957
0.6135
Note: *** denotes statistical significance at the 1 percent level, ** at the 5 percent level,
and * at the 10 percent level. Standard errors are in parentheses.
apparently is not a major factor for trade, indicating that the overall size
of the economy rather than the size of the population matters for tradea
notion that is intuitive given that many of the less populous East Asian
economies are among the regions most successful traders.11
8.3
MONETARY INTEGRATION, INVESTMENT

CONDITIONS, AND THE DEVELOPMENT OF
REGIONAL CAPITAL MARKETS
The best things in life are free
But you can keep em for the birds and bees.
(The Beatles in Money (Thats What I Want),
from the album With The Beatles, 1963)12
In the classic as well as neo-classic traditions, money is regarded as a veil

over the real economy.13 According to this school of thought, money is

203
merely a means of payment on spot markets and a store of value. The

neutrality of money implies that the central bank cannot affect the real
economy, and money, therefore, also has no impact on the development
of a countrys economy. But if money is regarded as a means of deferred
payment in an uncertain world with enterprises that finance investment
through capital markets, the denomination of debt becomes crucial for
production, employment, and growth (Nitsch 2006).
Acknowledging the central role of money in economic development, an
analysis of monetary integration between emerging market and developing
nations, such as those in East Asia, has to address different questions than
does an analysis of monetary integration between developed countries, such
as those in the euro area. The former has to take into account the specific
problems developing countries face.14 Two of the features that characterize
any less developed country are a weak currency status (Schelkle 2000) and
a lack of financial maturity (Bordo and Flandreau 2003). As will be argued
below, developing countries lack of an international currency puts a
number of constraints on their development prospects and increases financial fragility. This section of the chapter therefore investigates whether
and to what extent monetary integration could be used as a strategy to
overcome these impediments in order to create favorable macroeconomic
conditions, that is, to promote accumulation and to reduce the risk associated with asset and liability dollarization.
The international monetary system is characterized by currency competition and a hierarchy of currencies. Cohen (1998, p. 114) likens currency
competition to a vast, three-dimensional pyramid: narrow at the top,
where a few popular currencies dominate; increasingly broad below,
reflecting varying degrees of competitive inferiority. Different currencies exhibit different qualities in their function as storage of wealth. In
an open economy, the central bank has to defend the national currency
against competition with other currencies in the portfolios of economic
agents. Because of devaluation expectations, an agent holding a weaker
currency will demand a higher premium, that is, a higher real interest
rate. The international hierarchy of currencies thus sets off a cumulative
process by which investments in hard currency countries can be financed
most cheaply of all, while market forces cause higher interest rates to
be demanded in weaker currency countries, making credits dearer and
thus retarding growth and development (Nitsch 1999). High real interest
rates are a problem for some East Asian countries, but not all. While
least developed countries such as Cambodia (11.6 percent), Laos (17.5
percent), and Indonesia (6.6 percent) suffer from relatively high real interest rates, others such as China (0.7 percent), Malaysia (0.0 percent), the
Philippines (3.7 percent), and Thailand (0.9 percent) have managed to

204
establish relatively low levels of real interest despite their weak currency
status.15
Low productive power domestically corresponds to low monetary power
internationally. This is demonstrated by the fact that developing countries
currencies are not capable of entering into international contracts (Riese
2004). Weak currency countries are confronted with what Eichengreen and
Hausmann (1999) have termed original sin.16 They describe original sin
as a situation in which the domestic currency cannot be used to borrow
long term, even domestically. As a result, financial fragility is unavoidable because all domestic investment will have either a currency mismatch
(projects that generate domestic currency will be financed with an international currency) or a maturity mismatch (long-term projects will be
financed with short-term loans). Both currency and maturity mismatches
increase the danger of financial crises. Indeed, original sin is widely
regarded as one of the causes of the Asian crisis.
Moreover, in a phenomenon McKinnon (2005) has named conflicted
virtue, weak currency countries are unable to lend in their own currencies, forcing creditor countries with weak currencies to cumulate
currency mismatches. As most East Asian countries have turned into
creditor countries after the Asian crisis, the excess build-up of foreign
exchange assets has created a conflicted virtue problem for them. While
Japan, Singapore, and Taiwan have had current account surpluses for
more than two decades and China has had more modest current account
surpluses since 1995, even the five former crisis economies (Indonesia,
Korea, Malaysia, the Philippines, and Thailand), which had large current
account deficits before 1997, have now accumulated large stocks of liquid
dollar assets in both private and public portfolios. With mounting dollar
claims, non-US holders of dollar assets have to worry more about domestic currency runs, which would cause a domestic currency appreciation
and hence a decline of their net wealth. Countries are therefore inclined
to avoid large-scale appreciation of their currencies, which might invoke
protests from deficit countries about unfair competition through an
undervalued currency.
A weak currency status therefore brings with it multiple problems,
from retarding financial and real development and increasing the risk of
financial crisis to triggering potential trade conflict. A hardening of the
currency, or a pyramid-climbing, is therefore a crucial precondition
for creating favorable investment conditions and enabling sustainable
economic development and for overcoming the problems of original sin
and conflicted virtue. To analyze if and how monetary integration could
contribute to a hardening of currencies, it is important to understand what
makes a currency an international currency.

205
As Hyman Minsky observed, anyone can create money by issuing

IOUs,17 but the difficult part is to get them to be generally accepted (see
Kregel 2006). The easiest way to have ones IOUs accepted is to generate
liabilities that can only be extinguished through possession of these IOUs.
A government can domestically enforce the acceptance of its currency
through the fiscal system. It can create a tax liability on its citizens that
can only be redeemed by rendering the governments IOU in the form of
money issued by the government. But this does not work internationally, as
a government can only tax its own citizens who are subject to government
regulations but cannot force non-residents to hold claims. The only way to
make the currency internationally accepted is by building an expectation
that these liabilities will act as perfect substitutes for other governments
liabilities.
A number of conditions can be identified that contribute to building
such expectations. First, the confidence in a currencys future value is
dependent on the political stability of the country of origin (Cohen 2000).
This is the quintessential precondition for establishing a track record of
relatively low inflation and low inflation variability. Second, countries
need to develop sound and credible fiscal institutions.18 In conjunction
with non-inflationary income policies, an austere fiscal framework lays
the groundwork for a non-inflationary monetary environment with low
nominal as well as real interest rates.
Third, countries need to establish credible monetary regimes.
Unpredictable monetary policy makes agents unsure about the future real
value of their assets issued in domestic currency and may lead them to
denominate them in international currency (Jeanne 2005). Establishing a
strong (de facto) independent central bank with strong inflation aversion
and a clear monetary policy objective is an important way to keep down
inflationary expectations and to reduce this uncertainty.
Fourth, not running into international debt but instead striving for a
surplus in the trade and current account favors national autonomy and
employment and helps to create expectations of an appreciation of the
national currency. From a long-term development perspective, it is not the
short-term stabilization of the exchange rate that is of central importance
but rather the durable enhancement of a currencys quality (Fritz 2002).
The quality of a nations currency is undermined when a currency regime
is chosen that achieves price and exchange rate stabilization at the cost of
an increase in the countrys foreign debt. Instead, countries need to develop
the ability to generate foreign reserves by generating export surpluses. Such
a strategy requires a tendency toward an undervaluation of the currency
(Riese 2004). Keeping domestic money scarce represents the necessary
condition for securing the undervaluation of the currency becauseas a

206
result of its deflationary effects internallyit allows a country to achieve

a price advantage over foreign countries, providing the precondition for
the stimulation of exports. Successful examples of this development strategy are Western Germany in the 1950s and Japan in the 1970s. The East
Asian tiger economies have very successfully followed this strategy more
recently.
But developing sound fiscal and monetary institutions and generating
export surpluses might not be enough. The literature on the determinants
of key currency status points to another factor, namely, the size of the
economy. Matsuyama et al. (1993) explain the international use of currencies and, succinctly, the determinants of key currency status as a function
of relative country size and the degree of international economic integration. Because of network externalities and transaction costs, the global
portfolio is concentrated in very few currencies. In some ways money is
comparable to language, whose usefulness is also dependent on the number
of people with whom one can communicate; similarly, a currencys utility
rises with the number of other market participants using the same currency (Dowd and Greenaway 1993). A currencys attractiveness increases
with its transactional liquidity, which in turn is dependent on the existence
of well-developed and broad financial markets that offer a wide range of
short- and long-term investment opportunities as well as fully operating
secondary markets (Cohen 2000). Eichengreen et al. (2005) point out that
larger countries offer significant diversification possibilities, while smaller
countries add fewer diversification benefits relative to the additional costs
they imply.
As a result, the global portfolio is concentrated in a small number of
currencies (those at the top of the international currency pyramid) for
reasons partly beyond the control of even the countries that follow sound
domestic policies. Developing key currency status is hence a very difficult
and maybe even impossible endeavor for small economies.19 There is also
empirical support for this view. Eichengreen et al. (2005) show that larger
economies have less of a problem with original sin than smaller ones. Using
three different measures of size (log of total GDP, log of total domestic
credit, and log of total trade), their estimates suggest that all measures of
size are robustly correlated to original sin.
In the face of these constraints, regional monetary integration could be
employed by developing nations as a strategy for overcoming their weak
currency status and their inability to enter into international contracts with
domestic currency. While a hardening of the national currency could in
principle also be achieved by each country alone through austere monetary
and fiscal policies and through the generation of export surpluses, it might
not suffice to develop an international currency. Monetary integration,

207
however, could address both issues at the same time. One the one hand, it
would place an external constraint on countries participating in the monetary integration process, facilitating the domestic policy adjustment necessary for a hardening of the currency. In the European context, Giavazzi
and Pagano (1988) have termed this the advantage of tying ones hands.
Monetary integration could be used as a disciplinary device for inflationprone countries by forcing policy-makers to pursue more restrictive fiscal
and monetary policies than they would otherwise. In return, countries
would enjoy potential credibility gains. This strategy worked reasonably
well in Europe in the run-up to monetary union.20
At the same time, monetary integration would address the problems of
original sin and conflicted virtue by creating a larger economic entity with
vast investment opportunities which would be hard for international asset
managers to ignore and which, if backed by austere monetary and fiscal
policies, would increase the chances of entering the club of international
currencies. The underlying logic is that the whole would be equal to more
than the sum of its parts. To be sure, size alone is not enough: Russia,
Argentina, and Brazil are examples of large emerging economies that,
despite their size, face problems of original sin. But, as explained before,
good domestic policies are necessary as well, and these three countries have
not been prime examples for prudent economic policy-making.
Bordo et al. (2005), in an analysis of how five former British colonies (the
US, Canada, Australia, New Zealand, and South Africa) overcame original sin, point to another interesting factor: the role of shocks such as wars
and massive economic disruptions. For instance, the onset of World War
I essentially closed the London capital market and led Canada, Australia,
New Zealand, and South Africa to suspend the gold convertibility of
their domestic currencies and raise funds domestically; that is, the disruption of the war basically forced these countries to create domestic bond
markets.21 An interesting parallel can be drawn with East Asia, where the
financial crisis had been such a major shock. As one reaction to the crisis,
the region has begun trying to develop regional bond markets. The Asian
Bond Funds I and II are first attempts to bundle bond issues by East Asian
countries in order to make it more attractive for international investors to
include them in their portfolios.22
This might be a step in the right direction for East Asia, and pursuing
a similar cooperative path in monetary policy could help overcome the
problems of original sin and conflicted virtue that are associated with a
weak currency status. Individually, most East Asian nations will have
little prospect of escaping this trap. For instance, it is hard to imagine how
a small developing economy like Vietnam will be ever able to develop its
national currency, the dong, into an internationally accepted currency.

208
Yet, united, the region is far too important to be ignored by currency

traders. As a possible first move in this direction, in December 2005 the
Asian Development Bank announced the launch of a currency unit comprised of a basket of regional currencies akin to the European Currency
Unit. This virtual basket currency may give observers a taste of the kind
of standing a regional East Asian currency could achieve in international
financial markets.23
8.4
MONETARY INTEGRATION AND THE

RECOVERY OF MONETARY POLICY
INFLUENCE
When you got nothing, you got nothing to lose.
(Bob Dylan in Like a Rolling Stone,
from the album Highway 61 Revisited, 1965)
In this last section, the potential for regional monetary integration to be

a way for East Asian nations to (re)gain some degrees of monetary influence in the region is highlighted. Monetary cooperation and integration
are not necessarily equivalent to a loss of autonomy, as a country can only
lose what it has. In the discourse preceding the creation of EMU, opponents of monetary unification argued that giving up ones own currency
would mean losing a forceful policy instrument. However, the discussion
in Europe had been somewhat unreal. As Charles Goodhard (1995, p.
458) pointed out, through their membership in the European Monetary
System (EMS) EMS countries (except Germany) had already abandoned
discretionary monetary policy long before monetary union was finalized,
meaning that there would be virtually no economic cost in doing so formally and completely by moving to a full monetary union.24
For a formal test of this hypothesis, the following simple Taylor rule is
estimated:
rct 5 b1 1 b2 IPct-2 1 b3 ct-2 1 b4 rBt-2 1 e,
where rct is the central bank interest rate of country c at time t, IPct-2 is country
cs lagged industrial production, and ct-2 is the lagged inflation rate. rBt-2 is
the lagged Bundesbank discount rate.25 The results for the period ranging
from August 1971 to December 1998, that is, from the breakdown of the
Bretton Woods system to the launch of the euro, are presented in Table
8.3. The estimates show that the monetary policy decisions of all European
countries analyzed here, with the exceptions of Greece and Norway (both
of which were not members of the EMS26), were driven to a large extent by

Table 8.3
Determinants of monetary policy decisions, August 1971

December 1998
Bundesbank
discount rate
Austria
209
CPI
0.6916*** 0.0597
(0.0463)
(0.0385)
0.9371*** 0.1575***
Belgiuma
(0.1248)
(0.0497)
Denmark
0.7214*** 0.1933***
(0.0483)
(0.0379)
Finland
0.3226*** 0.1439***
(0.0464)
(0.0226)
France
0.7435*** 0.4341***
(0.0785)
(0.0420)
Greece
0.2366
0.1978***
(0.1561)
(0.0565)
Italy
0.4584*** 0.3403***
(0.1462)
(0.0582)
Luxembourga 0.9789*** 0.0449
(0.1181)
(0.0737)
Netherlands
0.9881*** 20.0457
(0.0594)
(0.0295)
Norway
0.1592
0.3018***
(0.1365)
(0.0949)
Portugal
0.7889*** 0.5391***
(0.2442)
(0.0626)
Spain
0.4329*
0.2147**
(0.2354)
(0.1045)
Sweden
0.4431*** 0.3301***
(0.0968)
(0.0437)
Switzerland
0.5760*** 0.1695***
(0.0674)
(0.0347)
UK
0.4792*** 0.2338***
(0.1348)
(0.0526)
US
20.03
0.5844***
(0.1124)
(0.0692)
Industrial
production
Observations
R-squared
20.0117*
(0.0064)
20.0285*
(0.0150)
20.0391***
(0.0069)
20.0407***
(0.0066)
0.0372***
(0.0116)
0.2839***
(0.0261)
0.0548***
(0.0177)
20.0842***
(0.0151)
20.0199**
(0.0080)
0.0619***
(0.0160)
0.1824***
(0.0206)
0.0614**
(0.0295)
0.0155*
(0.0084)
n.a.
327
0.7776
327
0.431
298
0.787
327
0.6986
327
0.678
327
0.5864
327
0.3219
327
0.5
327
0.7033
327
0.1089
327
0.4552
327
0.0725
327
0.4673
327
0.6292
20.0039
(0.0260)
0.0016
(0.0121)
327
0.318
327
0.5202
Notes: Data are from IFS and Global Financial Data. Newey West standard errors in
parentheses. *** denotes statistical significance at the 1 percent level, ** at the 5 percent level,
and * at the 10 percent level.
a
Luxembourg and Belgium have had a monetary union, the BelgiumLuxembourg
Economic Union, since 1922.

210
the interest rate policy of the Bundesbank.27 While domestic inflation and
industrial production also seem to have played a role in the conduct of many
of these countries monetary policies, the Bundesbank discount rate had a
far greater impact for all countries. US interest rate policy, in contrast, and
as one would expect, was not influenced by the Bundesbank.
One can push Goodhards argument further and argue that the European
countries (except Germany) that entered monetary union not only did not
lose monetary autonomy but actually (re)gained some degrees of monetary
policy influence through EMU membership. Indeed, it is arguable that
Germany was the only country that actually lost monetary autonomy.
Virtually all other EMU member countries have obtained a voice in monetary policy decisions. Instead of following the Bundesbanks policy stance, all
EMU member countries are now represented through their national central
bank governors in the European Central Banks (ECB) Governing Council,
the ECBs monetary policy-making body. Under current rules, the central
bank governors of each euro area member, along with the six members of the
ECBs Executive Board, have the right to vote at each council meeting, providing smaller countries such as Austria and Luxembourg with the same de
jure influence on monetary policy as the larger members France, Germany,
Italy, and Spain.28 Among the central banks of the old EU member countries
that did not join EMU, the monetary authorities of Denmark and Sweden
have basically been forced to shadow ECB policy, without exerting any
influence themselves on the policy they have to follow.29
A different but also Dylanesque situation applies for most East Asian
countries when it comes to the conduct of their monetary policy. The
regions exchange rate policy can be described by the East Asian dollar
standard (see McKinnon 2005), a situation in which all East Asian countries with the exception of Japan operate more or less tight pegs to the US
dollar. The pegs were temporarily lifted during and after the Asian crisis
(except in the cases of China, Hong Kong, and Singapore), but over the last
few years there has been a resurrection of the dollar standard (McKinnon
2001, Schnabl, Chapter 10, this volume), even though most of the countries
officially declare their exchange rate regimes as (managed) floats.
Table 8.4 gives an overview of East Asian countries official and de
facto exchange rate regimes. The estimated weights of the US dollar in
the hypothetical currency baskets of East Asian countries presented in
column three show that China, Hong Kong, Malaysia, and Vietnam
maintain fixed exchange rates vis--vis the dollar. Cambodia, Laos,
Myanmar, and the Philippines also manage tight pegs to the dollar, with
the weights of the dollar in their hypothetical currency baskets all about
90 percent. Indonesia, Korea, Singapore (and hence also Brunei, which
maintains a currency board vis--vis the Singapore dollar), and Thailand

Table 8.4
211
De jure and de facto exchange rate regimes in East Asia

IMF classification
Brunei dollar
Currency board arrangement (vis-vis the Singapore dollar)

Cambodian
Managed floating with no preriel
announced path for the exchange
rate
Chinese
Conventional pegged arrangement
renminbi
(vis--vis the US dollar)
Hong Kong
Currency board arrangement (vis-dollar
vis the US dollar)
Indonesian
Managed floating with no prerupiah
announced path for the exchange
rate
Japanese yen
Independently floating
Korean won
Independently floating
Lao kip
Managed floating with no preannounced path for the exchange
rate
Malaysian
Conventional pegged arrangement
ringgit
(vis--vis the US dollar)
Myanmar kyat Managed floating with no preannounced path for the exchange
rate
Philippine peso Independently floating
Singapore
Managed floating with no predollar
announced path for the exchange rate
Thai baht
Managed floating with no preannounced path for the exchange
rate
Vietnamese
Managed floating with no predong
announced path for the exchange rate
Estimated weight of the

USD in hypothetical
basket (percent)
65.28
92.56
99.42
99.37
81.47
42.74
74.03
89.15
98.12
93.94
91.04
65.28
69.23
100.40
Notes: The estimates in column three were calculated using Frankel and Wei (1994)
as presented in section 8.2 of this chapter with daily exchange rates from 7/21/2005
4/24/2006.* All dollar weight estimates are significant at the 1 percent level. The yen
was regressed only on the dollar and the euro. Data are from Datastream (Reuters and
Tenfore). Classifications are from IMF (2005).
* On July 21, 2005 the Chinese central bank officially announced that it had abandoned
the 11-year-old peg to the dollar and instead linked the yuan to an undisclosed basket of
currencies. Changing the sample period does not alter the results.

212
allow considerably more flexibility toward the dollar than the previously
mentioned countries but nevertheless display a strong dollar orientation in
their exchange rate regimes, with dollar weights ranging from 65 percent to
80 percent. The only exception to this pattern is Japan, with a dollar weight
of only 43 percent.
Although most East Asian countries maintain some form of capital controls that, in principle, should allow for monetary policy autonomy in the
face of an exchange rate target, historical evidence suggests quite strongly
that capital controls are porous and are easily circumvented (Edwards
1999). Furthermore, there is a growing amount of empirical literature that
puts into doubt even the traditional argument that countries with flexible
exchange rates are able to isolate their domestic interest rates from changes
in international interest rates.
Frankel et al. (2004), for instance, find that while floating regimes afford
greater monetary independence than fixed regimes, they offer only temporary monetary independence.30 That is, while the speed of adjustment of
domestic rates toward the long run, one-for-one relation with international
interest rates is generally lower under floating than under fixed regimes,
even floating regimes cannot exert autonomous monetary policy. Their
findings suggest that besides the US, Germany (now the euro zone) and
Japan appear to be the only countries that can independently choose their
own interest rates in the long run. Fratzscher (2002) also observes that even
under flexible exchange rate arrangements it becomes ever more difficult to
pursue independent monetary policy.31
Moreover, the underdevelopment of domestic financial markets hampers
the conduct of monetary policy, creating a situation wherein the vast
majority of East Asian countries have not been able to effectively pursue
independent monetary policy.32 The Peoples Bank of China (PBC), for
instance, did not change interest rates for more than a decade until October
2004, and even this change was very modest and more symbolic than of
practical import.33 The most extreme case of a loss of monetary autonomy
in East Asia is Hong Kong. To maintain its currency board vis--vis the US
dollar, the Hong Kong Monetary Authority has to move in tandem with
the Federal Reserve, even though local inflation development has been
very different from US inflation over recent years.
If this argument is correct and the status quo level of East Asian monetary independence is limited, then the costs of monetary integration in
East Asia, at least for the economically small and developing countries,
are much lower than commonly assumed.34 A common agency approach
to monetary unification, as pursued in Europe, could pave the way for
greater monetary policy flexibility in East Asia. Through the creation of
a common currency, which could float freely against the dollar and the

213
euro, East Asian countries could potentially gain some degrees of shared
monetary independence. While they would still face an external constraint
on domestic economic policies, the great difference between multilateral
monetary union in East Asia and a continued (informal) dollar pegging
under the East Asian dollar standard (or even full dollarization) is that the
former would give members of the currency union a say in the common
policy. Through a pooling of sovereignty, each member of the currency
union would have a share in the central banks policy-making. In contrast,
continued dollar pegging (or dollarization), while basically requiring the
same sacrifices in domestic policy autonomy as monetary union, would
mean that all monetary policy influence is abandoned (permanently).35
One potentially delicate political problem would be the institutional
design of a common central bank and the apportionment of power in an
East Asian monetary union. A representational structure akin to that of
the EMU, with equal rights between all members, would be unrealistic if
either China or Japan were involved, making such an arrangement less
attractive for the smaller Southeast Asian countries because of the potential dominance of the bigger member countries. Conversely, neither Japan
nor China (nor Korea) would be likely to accept the disproportional representation of the economically smaller Southeast Asian member countries
the way Germany did in Europe.36 Even though they are also highly heterogeneous, a solution would probably be easier to reach among ASEAN
member nations.37
Because the exit costs associated with a common monetary union are
very high, a quick rush into monetary unification is not advisable if the
potential partner countries have not had the chance to develop mutual trust
as well as an understanding of each others policy preferences.38 Therefore,
a gradual approach to monetary unification would be preferable, as it
would allow East Asian countries to get to know their potential partners
more closely. A gradual move toward monetary union in East Asia could
first involve a coordinated regional adoption of currency baskets,39 flanked
by a strengthening of financing facilities under the Chiang Mai Initiative
and a further enhancement of regional surveillance mechanisms. A further
option would be the introduction of a parallel basket currency for all
participating countries, which could be used as an invoicing currency for
trade and bond issues (see Eichengreen 2006). With time, the composition of the baskets could be harmonized among East Asian countries,
and exchange rate bands could be introduced, developing a more formal
regional exchange rate system. If countries were still committed to regional
monetary unification after having experienced what close monetary and
exchange rate cooperation really means, they could eventually form an
East Asian monetary union.

214
8.5
CONCLUDING REMARKS
The three cases for monetary integration in East Asia presented in this
chapter are in principle also valid for developing and emerging countries in
other regions contemplating monetary integration, such as Latin America or
the Middle East. However, this essay should not be mistaken for a straightforward and unambiguous recommendation for each and every region,
including East Asia, to enter into monetary union. Important preconditions
need to be in place before embarking on such a big endeavor. Above all, the
countries involved need to have a far-reaching consensus on policy preferences as well as a firm agreement about the conduct of macroeconomic and,
in particular, monetary policy (Volz 2006b). A very strong commitment is
required from all parties willing to engage in regional monetary integration.
The willingness and ability to subordinate internal economic objectives to
the objective of successful monetary integration is essential.
Moreover, the arguments put forward in this chapter have different
value for different East Asian countries. The economic size argument
certainly does not apply to Japan and China, and applies only to a limited
extent to Korea. Also, Japan already has a strong international currency
and enjoys monetary policy autonomy. Anyhow, China, Japan, and Korea
are very particular cases, and their potential involvement has to be seen in
the context of competition for regional leadership. The case for monetary
integration is, in general, much stronger for the ASEAN countries, all of
which are relatively small in economic terms and dependent on regional as
well as global developments.
Finally, one needs to recognize that the rationale for (and indeed
against) regional monetary integration goes beyond purely economic considerations. In Europe, for instance, economic integration was meant to a
large extent to be a means to overcoming centuries of war. From the beginnings of European integration, with the formation of the European Coal
and Steel Community in 1951 and the European Economic Community in
1957, economic integration was regarded as much more than a scheme for
promoting economic prosperity within Western Europeit was regarded
as a way of building a lasting peace. This has certainly been the greatest
and most important achievement of the European Union and cannot be
measured in economic terms.
NOTES
1.
Earlier versions of this chapter were presented at a workshop on Currency Conflicts

and Currency Cooperation in the Global Economy held at the University of British

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
215
Columbia, Vancouver, February 910, 2006 and at the 9th Annual Harvard East
Asia Society Conference in Cambridge, MA, February 1718, 2006. I would like to
thank Joshua Aizenman, Yin-Wong Cheung, Jerry Cohen, Manfred Nitsch, Federico
Ravenna, and participants at the UBC and Harvard conferences and at seminars at
UC Santa Cruz and Yale for valuable comments and suggestions. All errors are my
own. Financial support by the Fox International Fellowship at Yale and the Max Kade
Foundation is gratefully acknowledged.
In this chapter, East Asia refers to the 10 member countries of the Association of the
Southeast Asian Nations (ASEAN) as well as China, Hong Kong, Japan, and Korea.
The members of ASEAN are Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar,
the Philippines, Singapore, Thailand, and Vietnam.
On the patterns and structure of trade in East Asia see Kawai and Urata (2004).
See Fernndez-Arias et al. (2004). These problems are nicely illustrated for Brazil and
Argentina in the 1990s by Eichengreen (1998).
Glick and Rose (1998) have shown that countries that are closely linked through
international trade, both as trading partners and competitors, are especially prone to
contagious effects.
The North American Free Trade Agreement (NAFTA) is a case in point.
See Rose (2002, 2004), Frankel and Rose (2002), Nitsch (2002, 2004), Berger and Nitsch
(2005), and Baldwin (2006).
The countries are Cambodia, China, Hong Kong, Indonesia, Japan, Korea, Laos,
Malaysia, the Philippines, Singapore, Taiwan, Thailand, and Vietnam. The sample is
divided into three sub-sample periods, 198089, 199096, and 19992003, to capture the
change over those years, with the crisis years 199798 excluded.
GDP and population data are from WDI. Taiwan data are from the Taiwan Statistical
Data Book 2004. Distance was calculated using the distance calculator of the US
Department of Agriculture/Agricultural Research Service Phoenix, Arizona (www.wcrl.
ars.usda.gov/cec/java/capitals.htm).
As an alternative measure for the similarity of currency regimes, we compute bilateral
exchange rate volatility, that is, the standard deviation of the first differences of the
natural logarithms of the nominal exchange rate between the two countries in question.
The estimates for the gravity model using this measure, which are not reported here, are
almost identical to those in Table 8.2, confirming the robustness of the results. For both
indicators, we use monthly exchange rates from the IMF/IFS and the Central Bank of
China for Taiwan.
Singapores share of intra-regional trade, for instance, is particularly high because of the
fact that it engages considerably in intra-regional entrept trade.
Money was actually written by Berry Gordy and Janie Bradford and was a 1959 hit
single by Barrett Strong for the Tamla label (soon to be renamed Motown).
This view dates back to the 18th century, when David Hume (1752) initiated the famous
oil-in-the-machine illustration of the neutrality of money.
Fritz and Metzger (2006) hence distinguish southsouth cooperation from north
north cooperation, where north refers to a countrys ability to accumulate debt in
its own currency and south refers to its inability to do so.
All data are for 2004 and are taken from the World Banks World Development
Indicators. For an explanation how the threat of currency appreciation has pushed
the interest rates of China and several other East Asian countries below the US interest
rate level see McKinnon and Schnabl, Chapter 7, this volume.
The term original sin has been criticized by Goldstein and Turner (2004) and Nitsch
(2006) because it suggests that it cannot be overcome.
An IOU is a promise of money, goods, services, or other items of value, which may
be either written or verbal. The name derives from the phonetic pronunciations of the
respective letters, which sound like the phrase I owe you.
See, for instance, Corsetti and Makowiak (2005).
Eichengreen et al. (2005, pp. 2501) argue that Switzerland and the UK can be regarded
as special cases that achieved key currency status due to their unique historical roles.

216
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.

After entering monetary union, however, the governments of several EMU member
countries have manifestly shrugged off fiscal discipline and effectuated a weakening of
the Stability and Growth Pact.
The case of the USA is quite different from that of the other four former colonies. See
Bordo et al. (2005).
On the Asian Bond Fund see Ma and Remolona, Chapter 4, this volume.
On the Asian Currency Unit see Kawai, Chapter 12, this volume.
Goodhard (1995) does mention the costs of losing seignorage, but he argues that the
value of seignorage to a stable country with low inflation is small and that the arrangements made under the Maastricht Treaty for returning seignorage to the constituent
national central banks suggests that the net loss or gain to most European countries is of
secondary importance. One should also highlight the loss of the national central banks
ability to act as a lender of last resort, but in the EMU this function is still fulfilled by
the European System of Central Banks with the European Central Bank at the center.
Data are monthly and are taken from the International Financial Statistics and Global
Financial Data. As is usual for backward-looking Taylor rules, the estimates are
obtained using ordinary least squares with heteroskedasticity- and autocorrelation-consistent standard errors. See, for instance, Orphanides (2001) and Carare and Tchaidze
(2005). Changing the lags does not significantly alter the results.
The Greek drachma joined the EMS only in March 1998.
This result is confirmed by Frankel et al. (2004) and Chinn et al. (1993), who find that
interest rates in European countries had become completely insensitive to US interest
rates but fully sensitive to German interest rates.
The current voting rules are soon to be replaced by a rotating system to ensure efficient
decision-making in the face of an increase of the Councils current size of 18 members
to 30 or more in the process of euro area expansion. The new voting scheme will give
more weight to the larger euro area economies. But, irrespective of the distribution of
voting rights, the ECB statutes stipulate that national central bank governors sit on the
Governing Council in a personal and independent capacity, not as representatives of
their own countries.
The situation is different for the UK, where the Bank of England has managed to continue its independent monetary policy tailored to the national economys needs. This
can be attributed to the size of the UKs economy as well as its financial sector, which
give it a greater pull compared with smaller economies like Denmark and Sweden.
This is not inconsistent with Kim and Lees (2004) finding that the sensitivity of local
interest rates to US rates has declined for Korea and Thailand since they moved toward
less rigid exchange rate regimes after the Asian crisis.
See also Calvo and Reinhart (2001 and 2002).
This does not apply to Japan, Korea, and Singapore.
One must note, however, that China has managed to sterilize the large increase of
foreign reserve holdings very well and has thus been able to control inflation. The PBC
has also made use of other monetary policy instruments, such as reserve requirements
for domestic banks and credit ceilings. In October 2004, when announcing the raising of
benchmark rates on one-year yuan loans to 5.58 percent from 5.31 percent and the rate
on one-year deposits to 2.25 percent from 1.98 percent, the PBC said in a statement that
[t]his interest rate rise [. . .] is to make bigger use of economic measures in resource allocation and macro-adjustment (Xu 2004), indicating the countrys intention to increasingly deploy its macroeconomic policy instruments. In April 2006, the PBC raised the
interest rate on one-year yuan loans by another 0.27 percentage points to 5.85 percent.
This does not imply, however, that there are no costs involved with abandoning the
national currency. First, monetary union brings about the loss of the exchange rate as an
instrument for coping with idiosyncratic shocks to the national economy. Second, there
would still be the political cost of giving up formal monetary independence, that is, the
loss of monetary policy autonomy illusion. Furthermore, giving up de jure independence
could involve diplomatic costs in the form of political dependency on foreign nations.

35.
36.
37.
38.
39.
217
Third, the cost of exiting a monetary union can be assumed to be much higher than that
of exiting a conventional exchange rate peg.
On the advantages of multilateral monetary union over dollarization see Alexander and
von Furstenberg (2000).
On the power structure within the EMU see Berger and De Haan (2002).
Volz (2006a) discusses the factors that push ASEAN countries toward greater
integration.
As Friedrich Schiller versified in The Song of the Bell, Whoeer would form eternal
bonds should weigh if heart to heart responds.
Schnabl (Chapter 10, this volume) analyzes the use of currency baskets as a way to
diversify exchange risk in East Asia and finds hints of a move by several East Asian
countries toward a basket strategy. For proposals for currency baskets in East Asia see
Williamson (1999 and Chapter 11, this volume) and Ogawa and Ito (2002).
REFERENCES
Alexander, Volbert, and George M. von Furstenberg (2000). Monetary
IntegrationA Superior Alternative to Full Dollarization in the Long Run.
The North American Journal of Economics and Finance 11: 20525.
Baldwin, Richard (2006). The Euros Trade Effects. Proceedings of the June 2005
Workshop on What Effects is EMU Having on the Euro Area and its Member
Countries?, ECB Working Paper no. 594, European Central Bank, Frankfurt
am Main.
Berger, Helge, and Jacob De Haan (2002). Are Small Countries Too Powerful
Within the ECB? Atlantic Economic Journal 30: 26380.
Berger, Helge, and Volker Nitsch (2005). Zooming Out: The Trade Effect of
the Euro in Historical Perspective. CESifo Working Paper No. 1435, CESifo,
Munich.
Bordo, Michael D., and Marc Flandreau (2003). Core, Periphery, Exchange Rate
Regimes, and Globalization. In Michael D. Bordo, Alan M. Taylor, and Jeffrey
G. Williamson (eds), Globalization in Historical Perspective. Chicago: University
of Chicago Press, pp. 41772.
Bordo, Michael D., Christopher M. Meissner, and Angela Redish (2005). How
Original Sin was Overcome: The Evolution of External Debt Denominated in
Domestic Currencies in the United States and the British Dominions, 18002000.
In Barry Eichengreen and Ricardo Hausmann (eds), Other Peoples Money: Debt
Denomination and Financial Instability in Emerging Market Economies. Chicago:
University of Chicago Press, pp. 12253.
Calvo, Guillermo A., and Carmen M. Reinhart (2001). Fixing for Your Life. In
Susan Collins and Dani Rodrik (eds), Brookings Trade Forum 2000. Washington,
DC: Brookings Institution, pp. 139.
Calvo, Guillermo A., and Carmen M. Reinhart (2002). Fear of Floating.
Quarterly Journal of Economics 117: 379408.
Carare, Alina, and Robert Tchaidze (2005). The Use and Abuse of Taylor Rules:
How Precisely Can We Estimate Them? IMF Working Paper no. 05/146,
International Monetary Fund, Washington, DC.
Chinn, Menzie D., Jeffrey A. Frankel, and Steve Philips (1993). Financial
and Currency Integration in the European Monetary System: The Statistical
Record. In Francisco Torres and Francesco Giavazzi (eds), Adjustment and

218
Growth in the European Monetary Union. Cambridge: Cambridge University

Press, pp. 270306.
Cohen, Benjamin J. (1998). The Geography of Money. Ithaca, NY: Cornell
University Press.
Cohen, Benjamin J. (2000). Life at the Top: International Currencies in the
Twenty-First Century. Essays in International Economics no. 221, Princeton
University, Princeton, NJ.
Corsetti, Giancarlo, and Bartosz Makowiak (2005). A Fiscal Perspective
on Currency Crises and Original Sin. In Barry Eichengreen and Ricardo
Hausmann (eds), Other Peoples Money: Debt Denomination and Financial
Instability in Emerging Market Economies. Chicago: University of Chicago
Press, pp. 6894.
Dowd, Kevin, and David Greenaway (1993). Currency Competition, Network
Externalities and Switching Costs: Towards an Alternative View of Optimum
Currency Areas. The Economic Journal 103: 11809.
Edwards, Sebastian (1999). How Effective are Capital Controls? Journal of
Economic Perspectives 13: 6584.
Eichengreen, Barry (1998). Does Mercosur Need a Single Currency? NBER
Working Paper no. 6821, National Bureau of Economic Research, Cambridge,
MA.
Eichengreen, Barry (2006). The Parallel Currency Approach to Asian Monetary
Integration. AEA Papers and Proceedings 96: 4326.
Financial Fragility. Proceedings, Federal Reserve Bank of Kansas City,
August, pp. 32968.
Eichengreen, Barry, Ricardo Hausmann, and Ugo Panizza (2005). The Mystery
of Original Sin. In Barry Eichengreen and Ricardo Hausmann (eds), Other
Peoples Money: Debt Denomination and Financial Instability in Emerging
Market Economies. Chicago: University of Chicago Press, pp. 23365.
Emerson, Michael, Daniel Gros, Alexander Italianer, Jean Pisani-Ferry, and Horst
Reichenbach (1992). One Market, One Money. An Evaluation of the Potential
Benefits and Costs of Forming an Economic and Monetary Union. Oxford: Oxford
University Press.
Fernndez-Arias, Eduardo, Ugo Panizza, and Ernesto Stein (2004). Trade
Agreements, Exchange Rate Disagreements. InVolbert Alexander, Jacques
Mlitz and George M. von Furstenberg (eds), Monetary Union and Hard
Pegs. Effects on Trade, Financial Development, and Stability. Oxford: Oxford
University Press, pp. 13550.
Frankel, Jeffrey A., and Andrew K. Rose (1998). The Endogeneity of the Optimum
Currency Area Criteria. The Economic Journal 108: 100925.
Frankel, Jeffrey A., and Andrew K. Rose (2002). An Estimate of the Effect of
Common Currencies on Trade and Income. Quarterly Journal of Economics
117: 43766.
Frankel, Jeffrey A., Sergio L. Schmukler, and Luis Servn (2004). Global
Transmission of Interest Rates: Monetary Independence and Currency Regime.
Journal of International Money and Finance 23: 70133.
Rate Policies of the East Asian Economies. In Takatoshi Ito and Anne Krueger

219
Fratzscher, Marcel (2002). The Euro Bloc, the Dollar Bloc, and the Yen Bloc:
How Much Monetary Policy Independence can Exchange Rate Flexibility Buy
in an Interdependent World? ECB Working Paper no. 154, European Central
Bank, Frankfurt am Main.
Fritz, Barbara (2002). Entwicklung durch wechselkurs-basierte Stabilisierung? Der
Fall Brasilien. Marburg: Metropolis Verlag.
Fritz, Barbara, and Martina Metzger (eds) (2006). New Issues in Regional Monetary
Coordination. Understanding NorthSouth and SouthSouth Arrangements.
Houndmills: Palgrave Macmillan.
Giavazzi, Francesco, and Marco Pagano (1988). The Advantage of Tying Ones
Hands: EMS Discipline and Central Bank Credibility. European Economic
Review 32: 105582.
Glick, Reuven, and Andrew K. Rose (1998). How Do Currency Crises Spread?
FRBSF Economic Letter, 9825, San Francisco: Federal Reserve Bank of San
Francisco.
Goldstein, Morris, and Philip Turner (2004). Controlling Currency Mismatches in
Emerging Markets. Washington, DC: Institute for International Economics.
Goodhard, Charles (1995). The Political Economy of Monetary Union. In Peter
Kenen (eds), Understanding Independence. The Macroeconomics of the Open
Economy. Princeton. NJ: Princeton University Press, pp. 448505.
Hume, David (1752). Of Money. Reprinted in David Hume (1754). Essays:
Moral, Political and Literary. London and Edinburgh.
IMF (2005). Annual Report on Exchange Arrangements and Exchange Restrictions.
Washington, DC: International Monetary Fund.
Jeanne, Olivier (2005). Why Do Emerging Economies Borrow in Foreign
Currency? In Barry Eichengreen and Ricardo Hausmann (eds), Other Peoples
Money: Debt Denomination and Financial Instability in Emerging Market
Economies. Chicago: University of Chicago Press, pp. 190217.
Kawai, Masahiro, and Shujiro Urata (2004). Trade and Foreign Direct Investment
in East Asia. In Gordon de Brouwer and Masahiro Kawai (eds), Exchange Rate
Regimes in East Asia. London: Routledge, pp. 15102.
Kim, Chang-Jin, and Jong-Wha Lee (2004). Exchange Rate Regimes and
Monetary Independence in East Asia. In Gordon de Brouwer and Masahiro
Kawai (eds), Exchange Rate Regimes in East Asia. London: Routledge, pp.
30219.
Kregel, Jan (2006). Chances and Limits of SouthSouth Monetary Coordination.
In Barbara Fritz and Martina Metzger (eds), New Issues in Regional Monetary
Coordination: Understanding NorthSouth and SouthSouth Arrangements.
Houndmills: Palgrave, pp. 4253.
Matsuyama, Kiminori, Nobuhiro Kiyotaki, and Akihiko Matsui (1993). Toward
a Theory of International Currency. Review of Economic Studies 60: 283307.
McKinnon, Ronald I (2001). After the Crisis, the East Asian Dollar Standard
Resurrected: An Interpretation of High-Frequency Exchange Rate Pegging.
In Joseph E. Stiglitz and Shahid Yusuf (eds), Rethinking the East Asian Miracle.
Oxford: Oxford University Press, pp. 197246.
McKinnon, Ronald I (2005). Exchange Rates Under the East Asian Dollar Standard.
Living With Conflicted Virtue. Cambridge, MA: MIT Press.
Nitsch, Manfred (1999). Entwicklungstheorie unter Unsicherheit. Das
Investitionsrisiko als Motor und Strquelle von Entwicklung. In Rheinold E.
Thiel (ed.), Neue Anstze zur Entwicklungstheorie. Bonn: DSE, pp. 31220.

220
Nitsch, Manfred (2006). Comment on Perspectives for a Monetary Between

Argentina and Brazil. In Barbara Fritz and Martina Metzger (eds), New Issues
in Regional Monetary Coordination: Understanding NorthSouth and South
South Arrangements. Houndmills: Palgrave Macmillan, pp. 11625.
Nitsch, Volker (2002). Honey, I Shrunk the Currency Union Effect on Trade.
The World Economy 25: 45774.
Nitsch, Volker (2004). Comparing Apples and Oranges: The Effect of Multilateral
Currency Unions on Trade. In Volbert Alexander, Jacques Mlitz, and George
M. von Furstenberg (eds), Monetary Union and Hard Pegs. Effects on Trade,
Financial Development, and Stability. Oxford: Oxford University Press, pp.
89100.
16: 31734.
Orphanides, Athanasios (2001). Monetary Policy Rules Based on Real-Time
Data. American Economic Review 91: 96485.
Riese, Hajo (2004). Development Strategy and Economic Theory: Comments
on a Neglected Topic. In Jens Hlscher and Horst Toman (eds), Money,
Development and Economic Transformation. Selected Essays by Hajo Riese.
Houndmills: Palgrave Macmillan, pp. 84113.
Rose, Andrew K. (2000). One Money, One Market: The Effect of Common
Currencies on Trade. Economic Policy 15: 945.
Rose, Andrew K. (2002). Honey, the Currency Union Effect on Trade Hasnt
Blown Up. The World Economy 25: 4759.
Rose, Andrew K. (2004). The Effect of Common Currencies on International
Trade: A Meta-Analysis. In Volbert Alexander, Jacques Mlitz, and George M.
von Furstenberg (eds), Monetary Union and Hard Pegs. Effects on Trade, Financial
Development, and Stability. Oxford: Oxford University Press, pp. 10112.
Roy, Tobias, and Karl Betz (2000). Whrungskooperation im Mercosur.
Diskussionsbeitrge des Fachbereich Wirtschaftswissenschaft der Freien
Universitt Berlin, no. 2000/15 Volkswirtschaftliche Reihe, Free University,
Berlin.
Schelkle, Waltraud (2000). Regional Integration Among Less Developed
Economies: Discordant Variations on an Evergreen. In Martina Metzger and
Birgit Reichenstein (eds), Challenges for International Organizations in the 21st
Century. Houndmills: Macmillan, pp. 6588.
Schelkle, Waltraud (2001a). Monetre Integration. Bestandsaufnahme und
Weiterentwicklung der neueren Theorie. Heidelberg: Physica-Verlag.
Schelkle, Waltraud (2001b). The Optimum Currency Area Approach to European
Monetary Integration: Framework of Debate or Dead End? South Bank
European Papers no. 2/2001, South Bank University, London.
Volz, Ulrich (2005). The Effects of Real and Financial Integration on Output
Correlations in East Asia. Paper presented at the HWWA/HWWI Conference
on East Asian Monetary and Financial Integration in Hamburg, December
1516.
Volz, Ulrich (2006a). Economic Cooperation in ASEAN and the Rise of China.
YCIAS Working Paper FOX 01, Yale University, New Haven.
Volz, Ulrich (2006b). On the Feasibility of a Regional Exchange Rate System for
East Asia: Lessons of the 1992/93 EMS Crisis. Journal of Asian Economics 17:
110727.

221
Williamson, John (1999). The Case for a Common Basket Peg for East Asian
Currencies. In Stefan Collignon, Jean Pisani-Ferry, and Yung Chul Park (eds),
Exchange Rate Policies in Emerging East Asian Countries. London: Routledge,
pp. 32743.
Xu, Binlan (2004). Bank Rates Raised for First Time in a Decade. China Daily.
October 29.

9.
Monetary and exchange rate policy

coordination in ASEAN11
William H. Branson and Conor N. Healy
9.1
INTRODUCTION
This chapter develops an argument in favor of monetary and exchange rate

coordination in East Asia as part of a package of monetary integration that
would be aimed at supporting growth and poverty reduction. This could
be achieved directly through coordinated exchange rate stabilization and
indirectly through the implications coordination would have for reserve
pooling and investment in a new Asian Development Fund and through
the development of an Asian bond market. By monetary and exchange
rate coordination, we are here referring to flexible joint management
of exchange rate movements against a common basket, with the aim of
maintaining real effective exchange rates (REERs) near their equilibrium
values as underlying real economic conditions evolve. The chapter focuses
on establishing the conditions for such coordination.
In the next section, we examine the policy of monetary coordination
within ASEAN. We address the costs incurred when monetary stabilization is not accompanied by the coordination of sustainable underlying
policies. We emphasize a distinction between monetary stabilization with
and without such policies. The section uses an analytical narrative of the
key macro developments in ASEAN since the early 1990s as its vehicle.
Macro and exchange-rate policy coordination could have at least cushioned the effects of the 199799 crisis and prevented at least partially the
growth slowdown that followed. We use this as a starting point for identifying the direct gains from stabilization with sustainable macro policies,
inquiring how the gains from cooperative stabilization would have ruled
out the unintended competitive devaluations and cascading of speculation from market to market that was seen in the crisis. Explicit movement
toward coordination could also help the development of regional bond
markets, support surveillance and reserve-sharing under the Chiang Mai
Initiative, and, thus, free up reserves for additional investment through an
Asian Development Fund.
222

Monetary and exchange rate policy coordination in ASEAN11
223
In section 9.3 the trade structure of ASEAN and China is laid out in
terms of both geographic sources of imports and markets for exports, and
of the commodity structure of trade. The structure of trade by commodity
and sources and markets is also summarized for major commodity groups.
The similarities of the geographic trade structures across the region are
consistent with adoption of a common currency basket for stabilization.
The similarities of commodity structure in trade across the region are consistent with an argument for monetary integration across the region along
the lines of Mundell (1961) on optimum currency areas. The even distributions of trade across sources and markets for the major commodities
show that stabilization against a common basket would not create strong
differential sectoral strains across the region.
Section 9.4 draws on the geographic data of section 9.3 to construct
currency baskets and real effective exchange rates for the countries in the
region. Since their trade patterns are quite similar and their policies are
already implicitly coordinated, their REERs tend to move together. This
means that ASEAN and China are already moving toward integration in
practical effect, providing support for the implementation of a common
(externally based) currency basket. Section 9.4 also draws attention to
the effective coordination of monetary policy by studying correlations
among monthly movements in nominal exchange rates and changes in
reserve money, representing monetary policy. The correlations are positive
and quite strong. This is consistent with common reactions to common
shocks or with attempts to maintain exchange rates within a stable zone
within the region. In this case, monetary policy coordination is already
implicit. Making the coordination explicit, or even formal, could yield the
benefits of ruling out competitive devaluations and forestalling cascading
speculation. Section 9.5 ends with some tentative conclusions about the
desirability of monetary cooperation in East Asia.
9.2
MACRO POLICY COORDINATION
This section looks at the policy of coordinated stabilization of real exchange

rates, relating it to the Asian crisis as well as to the pre-crisis and postcrisis periods. It is important to make clear that when we consider exchange
rate stabilization we are not referring to the unilateral stabilizations that
occurred beforeand contributed tothe crisis. We argue that while there
are substantial benefits to be gained from stabilization, policies of stabilization will be more successful (and less prone to risk) if accompanied by
regional cooperation. They are also likely to be successful if accompanied by
underlying policies that are more sustainable than those that existed before.

224
Thus, before discussing the explicit benefits of cooperative stabilization, we

highlight the costs of maintaining exchange rate stability without the necessary underlying policies and without regional cooperation. We discuss losses
from failed stabilization in terms of the events leading to the crisis, the loss
of output during the crisis, and the depressed investment and slower growth
that have followed the crisis. These are losses that could have been prevented
or at least minimized by sustainable underlying macro policies and coordinated exchange-rate policies. We then go on to discuss potential gains from
cooperative stabilization, namely, the end of competitive depreciations and
the removal of individual country targets for speculative pressure. We also
use this discussion as the starting point for our broader discussion of the
benefits of reserve sharing and regional bond issues.
Exchange Rate Stabilization Without Sustainable Underlying Policies
A necessary condition for successful stabilization is a sustainable fiscal
position that frees monetary policy to target inflation with exchange rate
stability. A focus on exchange rate stabilization thus requires macro policy
sustainability, a lesson learned (and being re-learned) in Europe. The data
for the ASEAN countries since the early 1990s provide an illustration
of the problem of exchange rate stabilization without underlying macro
sustainability and the resulting losses from failure of stabilization.
Figures 9.1 and 9.2 show monthly USD exchange rates for the core
ASEAN countries, indexed to the beginning of 1990. These are defined as
units of home currency per dollar, so an upward movement is a depreciation of the home currency. As the extreme movement of the Indonesian
rupiah in 1997 distorts the picture in Figure 9.1, Figure 9.2 eliminates
Indonesia. Both graphs show the stable and roughly parallel movement of
nominal dollar exchange rates before and after the crisis, interrupted by
the crisis explosion in 199799. Before the crisis, USD exchange rates were
stable in nominal terms. But the economies were experiencing investment
booms not accompanied by fiscal adjustment, as shown in the tables in the
next section. They were overheating, with inflation causing real appreciation of their currencies, although they were stable in nominal terms against
the USD. They had large and growing current account deficits and growing
external debts denominated in foreign exchange. Thus, the growth in the
current account deficits was accompanied by appreciating real effective
exchange rates, an unsustainable scenario that can be interpreted as a
bubble in the foreign exchange market.1
Exchange rate stability was restored in the period after the crisis. However,
this period has seen a major depression in investment, with the formation
of current account surpluses. This is consistent with real appreciation but


800
Indonesia
Thailand
Philippines
700
225
Singapore
Malaysia
600
Indonesia
500
400
300
Philippines
200
Thailand
Malaysia
Singapore
100
04
03
D
ec
02
ec
D
01
D
ec
00
ec
D
99
D
ec
98
D
ec
97
ec
D
ec
96
95
ec
D
ec
94
93
ec
D
92
ec
D
90
D
ec
91
Source:
D
ec
ec
D
ec
89
Global Financial Data.
Figure 9.1
Monthly exchange rates for core ASEAN, December 31,

1989December 31, 2004
300
Thailand
Singapore
Malaysia
Philippines
Philippines
250
200
Thailand
150
Malaysia
100
Singapore
ec
89
D
ec
90
D
ec
91
D
ec
92
D
ec
93
D
ec
94
D
ec
95
D
ec
96
D
ec
97
D
ec
98
D
ec
99
D
ec
00
D
ec
01
D
ec
02
D
ec
03
D
ec
04
50
Source:
Global Financial Data.
Figure 9.2
Monthly exchange rates for core ASEAN, excluding

Indonesia, December 31, 1989December 31, 2004

226
240
Malaysia
Philippines
Indonesia
220
Singapore
Thailand
200
180
160
140
120
100
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source:
World Bank, World Development Indicators.
Figure 9.3
GDP (constant, 2000 LCU)
slower real growth. The effects of the lack of sustainable policies and the
breakdown of exchange rate stability in the crisis are shown in Figure 9.3.
The paths of real GDP are interrupted by the crisis, with serious recessions
in all the ASEAN countries. More importantly, none of the countries has
recovered back to its original growth path, and in all cases the underlying
growth rate has been reduced. The unsustainable earlier policies and the
crisis have depressed investment, lowered and slowed the real GDP growth
path, and substantially reduced the potential for poverty reduction.
Investment, saving, the current account, and REERs since 1990
The current account, investment-saving balance, and external debt data
for the core ASEAN countries since 1990 are summarized in Table 9.1
and Figures 9.49.7. The table shows in the first three lines the evolution
of the current account balance (CAB), investment (I), and saving (S), all
as percent of GDP. From the GDP accounts, these are connected by the
equation CAB 5 I S. The current deficit is the excess of investment over
saving. The current deficit must be financed by borrowing abroad, expanding the external debt. The last line of each table shows the path of the debt/
GDP ratio for each country. The paths of the effective real exchange rates
of the core ASEAN countries are shown in Figure 9.8. These are the total
trade-weighted indexes from Section 9.4 below. We now analyze in turn
the data for the pre-crisis period 19901996, the crisis period 199799, and
the post-crisis period 19992004.

227

Ext. balance,
goods &
services (%
GDP)
Gross capital
formation (%
GDP)
Gross domestic
savings (%
GDP)
Ext. balance,
goods &
services (%
GDP)
Gross capital
formation (%
GDP)
Gross domestic
savings (%
GDP)
External debt,
total (DOD,
current US$)
GDP (current
US$)
Ext. debt, %
GDP
Malaysia
Indonesia
34.1
62.1
61.1
34.5
128 168
114 427
37.8
79 548
69 872
32.4
33.2
32.3
23.7
31.6
30.7
2.1
1.7
1991
1.6
1990
36.7
35.4
1.4
63.3
139 116
88 002
33.4
30.5
2.9
1992
39.1
39.2
20.1
56.4
158 007
89 172
32.5
29.5
3.0
1993
39.6
41.2
21.6
61.0
176 892
107 824
32.2
31.1
1.1
1994
39.7
43.6
23.9
61.5
202 132
124 398
30.6
31.9
21.3
1995
Table 9.1 External balance, investment, savings and debt
42.9
41.5
1.4
56.7
227 370
128 937
30.1
30.7
20.6
1996
48.7
26.7
43.0
43.9
22.0
158.5
95 446
151 236
26.5
16.8
9.8
1998
0.9
63.1
215 749
136 161
31.5
31.8
20.3
1997
47.4
22.4
25.1
108.0
140 001
151 201
19.5
11.4
8.1
1999
47.3
27.3
20.0
96.1
150 196
144 407
25.6
16.1
9.5
2000
42.3
23.9
18.4
93.7
143 034
134 045
24.9
17.4
7.4
2001
42.1
23.8
18.3
76.2
172 971
131 755
22.2
15.7
6.6
2002
42.3
21.4
21.0
64.5
208 312
134 389
21.5
16.0
5.5
2003
228

Ext. balance,
goods &
services (%
GDP)
Ext. balance,
goods &
services (%
GDP)
Gross capital
formation (%
GDP)
Gross domestic
savings (%
GDP)
External debt,
total (DOD,
current US$)
GDP (current
US$)
Ext. debt, %
GDP
Singapore
External debt,
total (DOD,
current US$)
GDP (current
US$)
Ext. debt, %
GDP
Philippines
Table 9.1
34.8
34.8
32 494
45 417
71.5
30 580
44 331
69.0
10.6
17.2
18.4
6.9
20.2
24.2
23.0
49 134
44 024
25.8
17 080
1991
15 328
1990
(continued)
9.7
62.7
52 977
33 220
16.4
21.3
24.9
33.8
59 151
20 018
1992
7.8
66.5
54 368
36 143
15.5
24.0
28.4
39.1
66 894
26 149
1993
14.9
62.8
64 085
40 257
17.8
24.1
26.3
40.7
74 481
30 336
1994
16.1
53.1
74 120
39 391
14.6
22.5
27.8
38.7
88 832
34 343
1995
14.4
53.1
82 847
44 031
15.2
24.0
28.8
39.3
100 852
39 673
1996
12.2
61.6
82 343
50 746
14.4
24.8
210.3
47.1
100 169
47 228
1997
21.0
82.1
65 172
53 529
13.7
20.3
26.6
58.8
72 175
42 409
1998
17.8
76.2
76 157
58 063
18.6
18.4
0.2
52.9
79 148
41 903
1999
16.1
80.2
75 913
60 850
23.1
21.2
1.9
46.4
90 320
41 941
2000
19.4
81.2
72 043
58 499
17.5
20.6
23.1
50.7
88 001
44 612
2001
22.8
77.1
77 954
60 090
18.8
19.3
20.5
51.3
95 164
48 833
2002
33.3
77.8
80 574
62 663
16.2
18.7
22.5
47.3
103 737
49 074
2003
229

4 369
43 191
10.1
3 772
36 901
10.2
42.8
36.3
37 703
98 234
38.4
41.4
33.8
28 095
85 345
32.9
26.5
45.1
43.3
27.5
34.5
36.4
37.5
111 453
42.1
125 009
52 638
35.8
36.0
41 784
40.0
24.2
9.5
58 355
5 524
45.2
37.4
40.0
24.0
9.2
49 863
4 582
45.5
35.8
45.3
144 527
65 533
35.4
40.3
24.8
10.8
70 610
59.6
167 896
100 039
35.4
42.1
26.7
10.0
83 933
8 368
50.2
48.0
7 594
34.1
33.1
62.1
181 689
112 838
35.5
41.8
26.3
10.6
92 221
9 802
50.3
35.9
72.7
150 892
109 699
35.1
33.7
1.4
12.4
95 395
11 803
51.4
39.2
93.8
111 860
104 917
36.3
20.4
15.9
14.8
81 911
12 093
53.3
32.3
79.1
122 338
96 770
33.1
20.5
12.6
16.8
81 381
13 701
50.3
32.4
64.9
122 725
79 710
31.4
22.8
8.6
17.1
91 476
15 623
48.5
32.4
Sources: All World Bank, World Development Indicators, except Singapore external debt from Economist EIU data.
Ext. balance,
goods &
services (%
GDP)
Gross capital
formation (%
GDP)
Gross domestic
savings (%
GDP)
External debt,
total (DOD,
current US$)
GDP (current
US$)
Ext. debt, %
GDP
Gross capital
formation (%
GDP)
Gross domestic
savings (%
GDP)
External debt,
(current US$m)
GDP (current
US$)
Ext. debt, %
GDP
Thailand
58.1
115 536
67 181
30.6
24.1
6.5
21.6
84 871
18 361
44.6
25.2
46.9
126 770
59 459
31.1
23.9
7.2
23.4
88 275
20 657
43.9
21.2
36.2
142 953
51 793
32.0
25.2
6.7
24.3
91 342
22 218
46.7
13.4
230
Pre-crisis, 199096
With the exception of Singapore and Indonesia in the early 1990s, the
data show substantial and growing current deficits in the period before the
crisis. These were generated by investment booms, with investment exceeding domestic saving, where investment and saving both include private and
public. The excess of investment over saving could have been remedied
by an increase in public saving, that is, fiscal tightening. In the absence of
fiscal adjustment, the current deficits persisted and grew, maintaining or
increasing the debt/GDP ratios.
Thailand provides a good illustration of the stabilization problem before
the crisis. From 1990 to 1996 investment was 4142 percent of GDP,
while saving was around 35 percent. Thus, the current deficit stayed near
5 percent of GDP, increasing to over 6 percent over 199596, before the
crisis. The ratio of external debt to GDP increased from 32 percent in 1990
to 62 percent in 1996 and to 73 percent in 1997. The market could see that
this path was unsustainable, and speculative pressure against the Thai baht
began the crisis in July 1997. The depreciation of the baht led to contagion
across the region and to the cascade of devaluations shown in Figures 9.1
and 9.2 above. Thus, the crisis was the combined effect of unsustainable
underlying macroeconomic policies and the lack of cooperative macro
policy management.
The pattern is basically the same for all the core ASEAN countries
except Singapore. The investment ratios in Figure 9.4 are high and rising
over the period 199097. The picture for saving, illustrated in Figure 9.5, is
less clear. Thailand, Indonesia, and, especially, the Philippines have saving
ratios that are flat or falling. The saving ratios of Malaysia and Singapore
increase all the way until 1998, but Malaysias saving is lower than its rising
investment through 1995 while Singapores saving exceeds investment over
the entire period.
The external consequences of the investment boom underfinanced by
domestic saving are shown in Figures 9.6 and 9.7. Figure 9.6 shows the
decreasing current account balances for all the core ASEAN countries
except Singapore. The deficits of Thailand and the Philippines are the
largest. Those of Malaysia and Indonesia are closer to zero, with Malaysias
deficit diminishing after 1995. These current account balance patterns are
reflected in the debt/GDP ratios of Figure 9.7. With rapid GDP growth, the
debt ratios of all but Thailand are stable until 199697. Thailands debt ratio
rises throughout the period 199098, with an increase in its growth rate in
1994. This rapidly rising debt ratio signaled the potential unsustainability
which led to the onset of the crisis in 1997.
The depreciation of the Thai baht in July 1997 put competitive pressure on the other ASEAN countries. This pressure was perceived by the

231
45
40
35
30
25
20
Indonesia
Malaysia
Philippines
15
Singapore
Thailand
10
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source:
Figure 9.4
Gross capital formation (% GDP)
55
50
45
40
35
30
Indonesia
Malaysia
Philippines
25
20
Singapore
Thailand
15
10
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source:
Figure 9.5
Gross domestic savings (% GDP)
markets, leading to cascading speculation and devaluations. The source

of the competitive pressure was the similarity of ASEAN nations trade
patterns, which we discuss below. The contagion of this pressure is one
argument for cooperative exchange rate management.
Movements in the REERs of the core ASEAN countries since 1990

232

35
Indonesia
Malaysia
Philippines
30
25
Singapore
Thailand
20
15
10
5
0
5
10
15
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source:
Figure 9.6
External balance on goods and services (% GDP)
160%
Indonesia
Malaysia
Philippines
140%
Thailand
Singapore
120%
100%
80%
60%
40%
20%
0%
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source:
Economist, EIU data.
Figure 9.7
External debt (% GDP)
are shown in Figure 9.8.2 The REER paths generally show a downward
concavity from 1990 to 1996. There is an initial period of depreciation
from 1990 to 1992, a more or less flat period from 1992 to 1994, and then
an appreciation in 199596. Again, the main exception is Singapore, with

233
Core ASEAN real effective exchange rate, total trade basis

(markets outside ASEAN + China/HK)
160
Indonesia
Malaysia
Philippines
150
140
Thailand
Singapore
130
120
110
100
90
80
70
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Sources: Global Financial Data; IMF Direction of Trade Statistics; World Bank, World
Development Indicators.
Figure 9.8
Real effective exchange rate
a real appreciation from 1989 to 1996, consistent with its rapid growth
shown in Figure 9.1 earlier. Combined with the growing current account
deficits, the REER paths are consistent with the unsustainable position
that developed into a bubble in the ASEAN foreign exchange markets, as
shown analytically in Branson (2005).
The initial real depreciations over 199093 were broadly consistent with
stable adjustment. Further depreciation would have contributed to the correction of the current deficits. But the flattening out of the REER paths and
the turn to real appreciation were a signal that the economies were moving
away from equilibrium. The real appreciations would contribute to further
growth in the current account deficits, rather than correcting them. As
the markets saw this growing contradiction, speculation on depreciation
developed, with the initial pressure on Thailand, the country with the
rapidly growing debt ratio. The unsustainability of the underlying macro
policies combined with the lack of coordination of exchange rate policy set
the stage for the crisis.
Crisis, 199799
The crisis is clearly visible in Tables 9.19.2 and Figures 9.49.8. The collapse
of the REERs of all the core ASEAN countries except Singapore is evident in

234
Table 9.2
Regional aggregates for percentage of poor at USD 1 and USD

2 per day in East Asia 5*
USD 1 per day

USD 2 per day
1990
1996
1999
2000
2001
2002
15.6
53.3
6.9
38.5
9.0
46.2
8.0
44.7
7.0
42.8
6.2
41.1
* East Asia 5 comprises Indonesia, Malaysia, the Philippines, Korea, and Thailand.
Source:
World Bank, East Asia Update, September 2000.
Figure 9.8. The real depreciations range from 20 percent for the Philippines
to 40 percent for Indonesia. With external debt denominated in foreign currency and domestic assets in home currency, this collapse led to financial
institution failures and a breakdown in credit. This phenomenon, in which
currency crisis spreads to the financial sector, is well known in the economic
literature as a twin crisis. As the currencies collapsed, so did the local
financial institutions, whose large dollar exposures were no longer matched
by equivalent local currency assets. The financial collapse was precipitous. In
Thailand, for example, 56 out of 91 finance companies were eventually liquidated. Similarly dramatic levels of collapse were seen elsewhere, especially
in Korea and Indonesia (see, for example, Radelet and Sachs 1998). All of
this combined with a major increase in uncertainty, leading to the investment
collapses shown in Figure 9.4. Investment fell by more than 50 percent in
Indonesia, Malaysia, and Thailand and 25 percent in the Philippines. It fell
even more steeply from 1997 to 2003 in Singapore. This investment collapse
generated the recessions in GDP shown earlier in Figure 9.3.
The effects of the crisis on poverty are well documented (e.g., World
Bank 2000). The crisis generated recessions, financial failures, and rising
import prices, all leading to sharp increases in poverty across the region.
Moreover, its effects on poverty have often been understated, as, in the
absence of the crisis, poverty rates would have declined further below precrisis levels. Table 9.2 shows the substantial progress in reducing poverty
made across the region between 1990 and 1996, with the percentage of
people living on less than USD 1 and USD 2 per day falling markedly over
this period. The crisis reversed this trend between 1996 and 1999, leaving
the percentage living on less than USD 2 a day in 2002 still well above the
1996 level.
The effects on the external balances can be seen in Figure 9.6. With
saving ratios fairly stable and investment collapsing, the current balances
all moved sharply into surplus. This is essentially a macro result, resulting from the real currency depreciations experienced by all of the core

235
ASEAN nations during the crisis. Figure 9.7 shows the crisis results for
external debt ratios. With debt mainly denominated in foreign exchange,
the depreciations directly increased the debt ratios. The recessions in
real GDP added to the increases, most markedly in Indonesia. Thus, the
investment collapse in the crisis led to the severe recessions and the shift
to current account surpluses, while the depreciations increased the debt
ratios. The combination of real recession, financial failures, and rising
import prices contributed to significant increases in poverty, reversing the
trend of a decade.
Post-crisis, 19992004
Since the crisis, the ASEAN economies have stabilized, with lower and
slower paths of real growth (Figure 9.3), much lower investment ratios
(Figure 9.4), current account surpluses in all but the Philippines (Figure
9.6), and currencies depreciated in real terms (Figure 9.8). The lower
investment paths are a serious source of concern, as capital formation was
a driving force behind the Asian miracle.3 The domestic saving ratios
remain high, as shown in Figure 9.5. Thus, excess saving in the region
now generates current account surpluses and growth in reserves, while the
central banks hold currencies stable.
The relationship between the paths of current account balances in Figure
9.6 and REERs in Figure 9.8, the main source of instability before the
crisis, is mixed. The only country that stands out is Singapore, with a rising
surplus and a depreciating currency, an unstable combination. Indonesia
and Malaysia have substantial but decreasing surpluses with currencies
appreciating in real terms, a stable combination. Thailand has a falling
surplus with a slowly depreciating currency, and the Philippines has a small
deficit with real depreciation.
In summary, the region has settled on a lower and slower, but stable
growth path. Macro coordination centered on exchange rate management
might have averted the crisis by leading to sustainable macro policies and
reduced vulnerability to speculation.
Gains from Cooperative Stabilization
The direct gains from stabilizing real exchange rates come from the
stability provided by the underlying macro policies and the resulting
minimization of exchange risk to investors.4 Cooperative exchange rate
stabilization against the USD can yield two immediate potential benefits.5
First, it rules out competitive depreciations. Figures 9.1 and 9.2 illustrate
the problem. Once the pressure from speculation forced the devaluation
in Thailand in July 1997, the other ASEAN countries had to follow. The

236
market understood this, so the pressure moved to Indonesia and Malaysia.

These depreciations were not competitive in the sense that they were led by
policymakers seeking a competitive edge; they were forced by the market.
But they had the same result. Once one goes, the others have to follow.
This led to overshooting, as is evident in the graphs. A coordinated policy
could have attained an orderly group devaluation, if needed, without the
disorganization that at least partially contributed to financial failures and
the depth of the recessions. The competitive nature of the depreciations is
based on the similarity in the trade structures of the ASEAN economies in
terms of both commodity composition and markets (which are discussed
in the next section).
The second benefit is the removal of convenient individual country
targets from potential speculative pressure. In Asia in 1997, as in Europe
in 1992, speculators targeted the weakest country in the group and then
moved sequentially from country to country. Coordinated stabilization
with reserve sharing could remove this targeting convenience. If the coordination were based on an explicit agreement, the market would understand
that it is facing a cooperative based on shared reserves. In this case, the
group would come to a considered decision on how to deal with speculative
pressures.6 Cooperative stabilization would be facilitated by the development of a common currency basket for exchange rate management. The
feasibility of an agreement on such a basket is supported by the similarity of
the compositions of trade of the core ASEAN countries and China in terms
of markets and commodities, as shown in the next section. Thus, a common
currency basket weighted according to export destination or import source
could be developed as the basis for exchange rate coordination, as well as
for regional bond issues. These baskets are discussed in section 9.4.
Explicit exchange rate coordination would also support the reserve
sharing of the Chiang Mai Initiative (CMI) and help the development of
an Asian Bond Market (ABM), two initiatives that have received substantial attention to date.7 The CMI was developed by the ASEAN finance
ministers together with their counterparts from China, Japan, and Korea
(known as the ASEAN13 grouping) in May 2000 in Chiang Mai, Thailand.
It comprises a system of bilateral short-term financing facilities in the form
of swap arrangements in the event of a financial crisis within the group.
To date, 16 bilateral swap agreements between ASEAN13 countries have
been successfully concluded, for a combined total size of USD 75 billion.
The political cooperation involved in the CMI and the rhetorical ambition
of recent ASEAN13 statements have been taken by many to suggest a
new era in region-wide monetary coordination. In practical terms, though,
actual progress has been more limited. The CMI swap agreements have not
been activated, and the ABM is still mostly denominated in Singapore or

237
Macro policy and

exchange rate coordination
Stability, growth and

poverty reduction
Sustainable macro
policies and
surveillance
Exchange rate
stability
Common basket
Leveraged ADF
ABM
Figure 9.9
Reserve sharing (Chiang Mai)
Derivatives and futures

Issuance in common
basket
The relationship of policy coordination, reserve sharing, and

the ABM
US dollars. Regional macro policy and exchange rate coordination would

strengthen surveillance under the CMI and can thus be regarded as a necessary condition for further development of such reserve pooling.
Furthermore, exchange rate stabilization against a common basket of
currencies could reduce risk in bond markets and facilitate development
of derivatives and futures.8 Just as was the case with European monetary
coordination and the use of the European Currency Unit (ECU), the
common basket could act as a focal point for monetary coordination.
Moreover, it could also act as a spur for the ABM, particularly through
the creation of a broadly accepted local currency for bond issues. These are
necessary conditions for the development of an active ABM denominated
in the currencies of the region.
The relationship of exchange rate and macro policy coordination with
reserve sharing and the ABM is summarized in Figure 9.9. Macro policy
coordination can contribute directly to stabilization and growth through
crisis prevention and the reduction of uncertainty. It could provide the
basis for surveillance, activating reserve sharing. Concerted reservesharing would reduce the demand for reserves at the individual country
level. This could permit the release of reserves for investment as paid-in
capital into a newly created Asian Development Fund (ADF), leveraging
it in international markets to become a real influence in regional development. Exchange rate stabilization and the additional flow of investment
resources could stimulate development of the ABM, increasing the efficiency of investment in growth and poverty reduction. Thus, macro policy

238
coordination could be viewed as a necessary condition for further development both of reserve pooling via the CMI and of the ABM.
9.3
TRADE STRUCTURE: ASEAN AND CHINA
This section presents and analyzes the trade structures of the ASEAN
countries and China (ASEAN11) with two objectives in mind. The first
objective is to develop a basis for assigning weights for ASEAN currency
baskets in terms of ASEANs trade flow. The weights would be based on
the currency composition of external trade determined by the geographic
distribution of trade. For individual countries, these weights would be
used for calculating effective exchange rates, both nominal and real. For
ASEAN as a group, the weights would define a common basket for coordinated management of exchange rate policy. This common basket could
also be used as the currency of issue in the Asian Bond Market.
The second objective is to evaluate the case for some form of monetary
integration of ASEAN11 based on the commodity composition of trade.
Here we reflect Mundells (1961) original view that countries with similar
structures of trade by commodity are good candidates for integration into
a currency area.
The last step is to present a summary of the structure of trade by
market and commodity. This summarizes the similarities and differences
in trade structure across the region. It can also yield an idea of the effects
of stabilizing exchange rates against a geographic basket.
Geographic Structure of Trade
The geographic structure of trade in 2003 for the core ASEAN economies
and China is summarized in Tables 9.3 for exports and 9.4 for imports.
Trade structure does not change very quickly over time, so the particular choice of year is not crucial. Each table shows the total trade of the
ASEAN11 countries in billions of USD and its percentage distribution
across import and export markets. We focus on the large ASEAN countries, because inclusion of the small, newer members does not change the
numbers at all significantly.
Table 9.3 shows the structure of exports by market for ASEAN11. The
first column gives total exports, and the rest give the percentage distribution
across major markets. Taiwan and Australia are included among the major
markets because of their regional importance; their shares of ASEAN11
exports are similar to Koreas. In moving to weights for REERs, we concentrate on the six markets that are external to ASEAN11, treating trade

239

Source:
60.995
104.966
36.225
80.521
144.121
438.25
490.358
426.828
22.3
10.7
15.9
14.2
6.7
13.6
14.6
Japan
(%)
IMF DOTS 2004 Yearbook.
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China 1 HK
Core ASEAN
Total
($bn)
7.1
2.9
3.6
2.0
4.2
4.6
5.0
Korea
(%)
12.1
19.6
20.1
17.0
14.3
21.1
27.4
US (%)
13.1
12.1
16.3
14.7
13.4
16.5
20.8
EU (%)
Table 9.3 Export distribution by market, ASEAN11, 2003 data
3.7
3.6
6.9
3.2
4.8
2.1
2.9
Taiwan
(%)
2.9
2.5
0.9
2.7
3.2
1.4
1.8
Australia
(%)
17.6
24.8
18.2
20.6
27.9
7.1
9.1
ASEAN
(%)
6.2
6.5
5.9
7.1
7.0
China
(%)
15.0
17.3
12.3
18.5
18.4
33.7
18.3
Other
(%)
240

Source:
13.0
17.3
20.4
24.1
12.5
18.0
19.1
Japan
(%)
IMF DOTS 2004 Yearbook.
32.544
82.726
37.5
75.809
127.996
412.836
533.002
356.575
Total
($bn)
4.7
5.5
6.4
3.9
3.7
10.4
10.2
Korea
(%)
8.3
15.5
19.8
9.5
14.1
8.2
8.8
US (%)
10.9
11.8
8.0
10.0
12.5
12.9
13.6
EU (%)
Import distribution by market, ASEAN11, 2003 data
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China 1 HK
Core ASEAN
Table 9.4
2.7
5.0
5.0
4.3
5.1
12.0
12.3
Taiwan
(%)
5.1
1.5
1.3
2.1
1.7
1.8
1.7
Australia
(%)
23.8
24.4
17.1
16.6
28.6
11.5
14.0
ASEAN
(%)
9.1
8.8
4.8
8.0
8.7
China
(%)
22.5
10.1
17.3
21.6
13.3
25.3
20.4
Other
(%)
241
in the region as internal. Here we begin with the broader picture. We note
for future reference and research that the total value of Singapores trade
is exaggerated by its extensive engagement in entrept trade. It is not clear
what, if any, effect this has on its trade distribution.
One thing to notice in Table 9.3 is that the distribution of Chinas
exports across markets is very similar to ASEANs. The shares of the main
six markets in the last row of the table fit right into the ASEAN pattern.
This also holds for import sources in Table 9.4. This similarity between
the geographic distribution of trade of ASEAN and that of China is a first
indicator that the two may be good candidates for monetary coordination.
We also note that the total value of ASEAN exports (USD 427 billion) is
about the same as Chinas (USD 438 billion).
Turning to some of the notable details of the table, we note that the
share of ASEAN internal exports ranges from 17 to 28 percent. The share
of Chinas exports to ASEAN is smaller, as is ASEANs to China, around
67 percent. Thus, as of 2003, ASEAN was a denser trade area than
ASEANChina. The total share of the three largest markets (Japan, US,
Europe) for all seven countries ranges between 42 percent (for Malaysia)
and 52 percent (for the Philippines). Again, China fits into the ASEAN
distribution here. If we look at China separately, we see that Chinas share
of ASEAN exports falls between that of the largest three markets (Japan,
US, Europe), and that of the smaller three (Korea, Taiwan, Australia).
The export market data would support the adoption of a common
currency basket by both ASEAN and China, composed of the currencies
either of the three largest export markets (Japan, US, Europe) or of all six
(the top three plus Taiwan, Korea, and Australia).
Table 9.4 shows the structure of imports by source for ASEAN11. As
in Table 9.3, the geographic structure of Chinas imports is quite similar to
ASEANs. Chinas shares of imports from Korea and Taiwan are larger
than ASEANs, while the shares of the US and Australia are smaller. But
the overall impression is that Chinas import pattern for the major markets
fits into the ASEAN pattern, potentially supporting an argument for coordination. The total value of ASEANs imports (USD 356 billion) is about
the same as Chinas (USD 413 billion). Thus, in terms of trade volumes,
aggregate ASEAN and China are about the same size.
The core ASEAN member nations shares of imports from other
ASEAN countries range from 17 to 29 percent, and the range is the same
in the case of intra-ASEAN exports. ASEANs share of Chinese imports is
smaller, as is Chinas share of ASEAN imports. Thus, on the import side,
intra-ASEAN trade is also denser than that between ASEAN and China.
The total share of the ASEAN11 nations imports from their three largest
trade partners (Japan, the EU, and the US) ranges from 39 percent (for

242
China) to 48 percent (for the Philippines). The total import shares for the
largest sources are smaller than their shares of exports, especially in the
case of China, for whom trade with Japan, the EU, and the US accounts
for 51 percent of its exports and 39 percent of its imports. The USs shares
of imports in Table 9.4 are all smaller than its export shares in Table 9.3.
The imbalance between the export share and the import share for the three
largest trading partners is a reflection of the US trade deficit. As in the case
of exports, Chinas share of ASEAN imports falls between that of the three
largest sources (Japan, US, Europe) and the three smaller ones (Korea,
Taiwan, Australia).
The import source data of Table 9.4 would also support the adoption of
a common currency basket by both ASEAN and China.
Commodity or Sector Structure of Trade
The commodity structure of trade at the one-digit SIC level of the ASEAN
countries and China in 2002 is summarized in Table 9.5 for exports and
Table 9.6 for imports. The tables separate the founding core ASEAN
countries from the later entrants plus Brunei. The latter are much smaller
individually as well as in aggregate and have very different trade structures.
China and China including Hong Kong are also presented separately,
mainly because of the entrept nature of Hong Kongs trade.9 The inclusion of Hong Kong with China does not affect the data or conclusions
significantly at this level of aggregation.
Table 9.5 for exports shows clearly the results of the Asian miracle.
SIC 7 is the dominant export sector for all of core ASEAN but Indonesia.10
The largest share of Indonesias exports is in SIC 3, which includes oil. But
even if SIC 2 and 3 for Indonesia are excluded, the 16.8 percent in SIC 7
becomes 25 percent, still much less than the other core ASEAN countries.
Chinas share of exports in SIC 7 is about the same as the shares of the
Philippines and Thailand. If SIC 7 and 8 are aggregated, the commodity
structure of Chinas exports fits into the ASEAN pattern, as it did with
market distribution earlier. China and the recent ASEAN entrants also
have large shares in SIC 8, which includes apparel. This is a major difference between those countries and the core ASEAN countries.
Table 9.6 for imports shows a pattern that is dominated by manufactures
(SIC 68), as is normal for industrializing countries. The core ASEAN
countries (except Indonesia) and China show particularly high shares
in SIC 7. This is also the case for Laos and Brunei. Indonesias import
structure more closely resembles that of Vietnam. However, no particular pattern stands out in Table 9.6. Again, the structure of Chinas trade
resembles that of the ASEAN countries.

243

Source:
7.00
60.30
8.60
1.20
19.20
16.80
14.50
0.60
42.30
7.00
39.70
2.90
1.00
1.00
38.176
3.90
0.20
0.90
1.10
10.60
15.60
39.90
10.20
5.20
0.10
75.043
12.30
0.20
3.90
2.10
4.20
8.90
64.50
3.70
8.70
0.20
127.894
1.50
0.80
0.60
6.90
Sing.
Other ASEAN
7.60
39.70
7.10
5.30
1.10
0.00
16.847
17.40
0.10
2.40
19.20
32.60
27.90
1.10
2.60
0.10
0.10
2.639
21.50
0.50
12.60
1.10
1.60
93.60
0.20
2.30
0.00
0.00
1.909
0.80
0.10
1.40
0.00
Vietnam Myanmar Cambodia
World Trade Analyzer, Statistics Canada. International Trade Division.
4.60
5.00
4.60
5.20
101.797
2.10
0.30
2.30
8.50
61.264
6.30
0.50
8.00
24.40
Indonesia Malaysia Philippines Thailand
Core ASEAN countries
49.50
40.40
0.10
0.70
0.00
0.00
0.33
4.70
0.00
4.60
0.00
Laos
0.20
6.10
4.40
1.10
0.00
0.00
4.074
0.00
0.00
0.00
88.20
Brunei
Commodity exports to the world, % of total exports for each sector by country, 2002
Total, $ bn
0-Food
1-Beverages
2-Crude materials
3-Fuels, lubricants &
related materials
4-Animal &
vegetable oils, fats
& waxes
5-Chemicals &
related products,
n.e.s.
6-Manufactured
goods classified
chiefly by material
7-Machinery &
transport equipment
8-Miscellaneous
manufactured articles
9-Commodities &
trans. not classified
elsewhere
SIC code
Table 9.5
0.20
31.60
38.60
16.40
4.60
0.00
358.565
4.40
0.30
1.40
2.60
China
0.50
32.80
40.20
15.40
4.60
0.00
564.977
3.20
0.30
1.20
1.70
China1HK
China
244

0.40
6.30
9.60
63.30
6.10
3.90
0.80
8.60
12.80
0.20
14.60
16.70
29.80
3.60
1.80
1-Beverages
2-Crude materials
3-Fuels, lubricants &

related materials
4-Animal & vegetable
oils, fats & waxes
5-Chemicals & related
products, n.e.s.
6-Manufactured goods
classified chiefly by
material
7-Machinery &
transport equipment
8-Miscellaneous
9-Commodities &
trans. not classified
elsewhere
Source:
4.10
11.10
6.50
5.20
48.00
11.90
8.30
0.20
7.60
2.50
2.00
7.80
30.682
4.00
6.60
49.90
16.40
10.80
0.10
4.70
3.50
0.50
3.50
53.332
5.00
8.90
57.90
7.70
5.50
0.30
10.70
0.60
0.70
2.60
114.596
Sing.
1.50
6.00
33.40
26.70
13.60
0.50
8.70
2.20
2.30
5.10
15.443
2.245
2.60
5.50
30.50
24.60
9.30
2.50
16.10
0.80
2.70
5.40
1.80
8.90
17.80
44.80
5.00
0.40
8.80
2.10
8.90
1.50
1.546
Cambodia
Other ASEAN
Vietnam Myanmar
1.90
0.50
4.00
28.143
0-Food
84.884
Indonesia Malaysia Philippines Thailand
8.20
5.50
40.90
14.70
3.60
0.00
6.70
0.30
16.70
3.30
0.186
Laos
3.70
10.70
46.90
20.50
4.40
0.30
1.40
0.50
2.50
9.20
1.884
Brunei
Commodity imports from the world, % of total imports for each sector by country, 2002
Total, $ bn
SIC code
Table 9.6
1.60
7.60
46.70
17.80
12.30
0.60
5.10
6.60
0.10
1.70
328.045
China
1.90
11.10
46.80
18.10
10.20
0.40
4.10
4.70
0.30
2.30
510.141
China1HK
China
245
Disaggregated SIC categories 68 at the 2-digit level are shown for

exports and imports in Table 9.7. Several patterns stand out. On the
export side, the importance of SIC 7577 for the core ASEAN countries
and China is clear. Here again, Chinas trade fits in with ASEANs. The
importance of SIC 8485 for the other ASEAN countries and to a smaller
degree for China is also clear. These countries are the apparel exporters
of the region. On the import side, SIC 77 stands out in Table 9.7 for the
core ASEAN countries and to a lesser extent for China, reflecting intraindustry trade in electrical machinery. SIC 65 stands out for the other
ASEAN countries. Their apparel industry imports textiles, to some extent
from China (see Chinas 5 percent in imports in SIC 65 in Table 9.7), and
exports the finished product. These disaggregated data reinforce the similarity of Chinas trade to ASEANs, an important condition for monetary
coordination.
Trade by Sector and Market
We now analyze the pattern of trade in SIC 68 of the core ASEAN countries and China by sector, export destination, and import source in order
to understand the potential differential effects on sectors of coordinating
exchange rate policies through a common basket. The same movement of
the exchange rate would have differing effects on the aggregate economy
of each country depending on the sectors with the largest differences in
trade shares across markets and sources. Similar distributions of trade by
sectors and markets or sources will minimize differential impacts within the
economies; dissimilar distributions will increase these differences, making
coordination more costly.
Tables 9.89.10 show the distributions of manufactured exports by
markets for the core ASEAN countries and China in SIC 68. The entries
in each line show first the fraction of total exports in the SIC category and
then the share of total exports in that category separately to each market.
The total shares for SIC 7 in Table 9.9 are the largest for the core ASEAN
countries except Indonesia, and for China. More than half of total exports
are SIC 7 for Malaysia and Singapore. With some exceptions, exports in
Tables 9.89.10 are evenly distributed across markets. Notable exceptions are the small share of the US in Malaysias exports in SIC 6 and the
large share in SIC 7 as well as the small share of the US in the Philippines
exports of SIC 7. These sectors could be differentially affected by movements of the US dollar within a common basket.
Tables 9.119.13 show the same import source distributions. The import
distributions are even more similar than those for exports, with the exception of the low US shares of SIC 6 exports to Malaysia and the Philippines

246

material
65X-Textile yarn,
fabrics, made-up art.,
related products
7-Machinery and
transport equipment
75X-Office mach.
& automatic data
processing equip.
76X-Telecommunic.
& sound recording
apparatus
77X-Elec. machinery,
apparatus & appliances
n.e.s.
8-Miscellaneous
84X-Articles of apparel
and clothing accessories
851-Footwear
Imports
Exports
SIC code
11.40
25.90
8.60
2.10
0.10
5.90
4.10
14.50
7.10
2.00
60.30
16.80
19.10
1.10
5.10
3.80
7.00
19.20
Indonesia Malaysia
0.10
3.90
7.00
16.40
3.00
17.10
39.70
0.70
2.90
Philipp.
1.20
5.40
15.60
11.70
6.70
12.30
39.90
2.10
10.20
Thai.
0.10
1.40
8.90
29.80
6.40
21.00
64.50
0.60
3.70
Sing.
16.90
14.60
39.70
3.60
0.70
0.50
7.10
1.80
5.30
0.80
26.10
27.90
0.30
0.10
0.00
1.10
0.40
2.60
Vietnam Myanmar
7.50
85.60
93.60
0.10
0.00
0.00
0.20
1.30
2.30
Camb.
Other ASEAN
Table 9.7 Breakdown of key sectors for commodity exports and imports, %, 2002
1.30
38.90
40.40
0.00
0.00
0.00
0.10
0.10
0.70
Laos
0.00
5.90
6.10
0.10
0.10
0.00
4.40
0.60
1.10
Brunei
2.40
9.80
27.90
15.10
9.20
12.20
44.00
6.10
15.40
China
4.40
8.10
29.40
15.50
9.60
11.70
43.80
6.10
15.00
China1HK
China
247

Source:
63.30
37.40
3.20
1.40
5.00
29.80
9.60
16.70
22.90
48.00
3.60
11.90
17.60
49.90
2.30
16.40
24.50
57.90
0.90
7.70
5.90
33.40
9.30
26.70
material
65X-Textile yarn,
fabrics, made-up art.,
related products
7-Machinery and
transport equipment
77X-Elec. machinery,
apparatus & appliances
n.e.s.
3.40
30.50
12.10
24.60
1.10
17.80
36.70
44.80
5.10
40.90
8.80
14.70
4.60
46.90
8.30
20.50
18.20
45.50
5.00
16.50
18.40
45.90
5.40
17.00
248

15.2
9.7
6.8
14.5
3.1
22.9
18.8
Korea
13.3
2.6
2.3
11.3
0.6
13.2
11.1
US
18.3
6.0
1.5
11.7
1.9
13.8
11.3
Europe
19.2
7.9
1.5
13.1
3.4
17.2
12.2
Taiwan
19.0
10.0
5.9
13.1
4.1
19.1
16.4
Australia
25.0
8.3
3.7
8.8
5.4
20.4
20.2
Other
11.6
42.9
53.2
39.4
59.5
32.4
33.7
Japan
6.2
48.1
26.7
43.7
71.9
30.6
37.4
Korea
19.2
80.6
33.5
39.4
75.7
39.4
35.5
US
16.8
60.3
39.7
39.9
64.5
38.6
40.2
Total
18.3
62.9
54.9
45.8
71.4
40.0
38.3
Europe
7.4
64.7
33.1
55.3
75.2
50.3
55.9
Taiwan
10.4
38.3
58.3
44.2
47.2
32.2
32.1
Australia
22.6
56.0
31.3
36.4
59.1
40.7
44.9
Other
SIC 7, Machinery and transport equipment export shares to markets outside core ASEAN11 countries
(percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Source:
16.3
9.1
3.4
8.7
1.9
12.2
11.1
Japan
Table 9.9
Source:
19.2
7.0
2.9
10.2
3.7
16.4
15.4
Total
SIC 6, Manufactured good export shares to markets outside core ASEAN11 countries (percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Table 9.8
249

1.9
2.4
0.9
5.8
9.1
16.5
17.6
Korea
43.6
12.0
18.1
31.9
11.0
41.2
49.0
US
29.0
14.0
6.1
26.5
7.9
35.7
42.2
Europe
2.1
3.6
0.6
3.4
5.2
12.9
16.8
Taiwan
7.0
12.6
8.3
11.3
12.3
38.1
42.6
Australia
19.5
15.0
11.2
18.7
9.0
18.5
16.4
Japan
36.3
13.7
17.0
21.5
9.2
24.8
24.3
Korea
7.0
3.0
3.0
7.6
3.8
6.0
7.6
US
16.7
9.6
11.9
16.4
7.7
17.8
18.1
Total
12.0
9.0
11.4
16.6
9.9
12.7
16.3
Europe
36.6
13.1
18.8
28.0
6.9
24.6
22.8
Taiwan
15.5
26.7
13.3
26.1
5.4
17.8
18.4
Australia
SIC 6, Manufactured good import shares by source, outside core ASEAN11 countries (percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Source:
5.6
7.5
3.7
14.8
19.1
35.6
38.3
Japan
Table 9.11
Source:
14.5
8.6
7.0
15.6
8.9
31.6
32.8
Total
14.2
8.8
13.1
14.1
7.9
18.0
18.8
Other
7.2
6.6
2.2
4.6
7.5
24.5
20.8
Other
SIC 8, Miscellaneous manufactured article export shares to markets outside core ASEAN11 countries
(percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Table.9.10
250

56.6
64.6
66.2
60.9
64.2
54.2
55.4
24.0
75.1
59.9
57.9
71.6
43.5
46.4
Korea
35.7
79.4
66.1
51.2
66.4
54.3
52.2
US
29.8
63.3
48.0
49.9
57.9
46.7
46.8
Japan
44.5
65.1
49.5
46.3
55.7
60.6
52.8
Europe
29.5
72.0
27.2
46.3
80.2
47.5
51.6
Taiwan
13.6
13.0
6.9
5.8
13.0
5.2
6.6
Australia
15.9
56.8
36.0
45.6
53.1
40.4
41.9
Other
Source:
3.6
6.1
5.2
6.6
8.9
7.6
11.1
3.7
7.3
4.4
6.5
8.2
8.7
10.2
Japan
5.4
2.0
2.6
2.9
3.1
4.7
4.6
Korea
2.7
6.4
4.3
8.7
11.2
9.0
10.6
US
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Total
3.5
6.6
6.2
10.3
12.3
7.7
12.4
Europe
4.1
4.3
3.4
7.6
5.0
8.7
7.9
Taiwan
1.3
3.6
3.9
2.7
5.4
2.6
4.8
Australia
3.8
6.1
6.4
5.3
8.5
7.3
13.1
Other
Table 9.13 SIC 8, Miscellaneous manufactured article import shares by source, outside core ASEAN11 countries (percent)
Source:
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Total
Table 9.12 SIC 7, Machinery and transport equipment import shares by source, outside core ASEAN11 countries (percent)
251
(Table 9.11). To summarize, our analysis of trade patterns by geography

and sector does not show serious imbalances in the distributions of trade by
sectors. Thus, coordinated exchange-rate stabilization against a common
basket based on these sources and markets should not create substantial
differential sectoral pressures within ASEAN or China.
9.4
CURRENCY BASKETS AND REAL EFFECTIVE

EXCHANGE RATES
This section discusses the construction of currency baskets and REER

indexes for the core ASEAN countries and China. These are based on trade
of ASEAN11 outside this region, since the focus of the study is first on
intra-ASEAN coordination and then on coordination between ASEAN
and China. Similar baskets and REERs can be calculated for the ASEAN
members largest trading partners, regardless of region, or for ASEAN11
or ASEAN13 against major trading partners outside the region. We begin
by presenting the alternative weights for exports, imports, and total trade,
based on the data from the previous section. Next, the alternative REER
measures are shown and discussed.
The REERs combine the movements of nominal exchange rates and
domestic price levels, both of which are at least partially under the control
of the domestic monetary authority, with those of trading partners price
levels, which are independent of domestic policy. Therefore, the similarity
in REER movements across countries implies some form of policy management of movements of domestic inflation and nominal exchange rates.
This is presumably aimed at maintaining competitiveness in the region
without competitive devaluations. To analyze this policy management, we
show the decomposition of movements of the import-weighted REERs at
the end of the section.
Weights for Currency Baskets
The currency basket weights based on exports, imports, and total trade of
ASEAN plus China are shown in Tables 9.149.16. Tables 9.14 and 9.15
are taken directly from Tables 9.3 and 9.4, scaled to 100. Table 9.16 is based
on the total shares from Tables 9.3 and 9.4. The trade shares are quite
similar across countries in each table as well as across tables. Exceptions
are Indonesia, with a high weight for Japan and a low weight for the US on
the export side, and Thailand, with a high weight for Japan on the import
side. The weights for China are well within the ASEAN distribution in all
three tables.

252

19.8
38.1
31.6
31.6
30.6
35.7
37.8
US
21.4
23.6
25.5
27.3
28.7
27.8
28.6
Europe
10.5
9.7
10.5
7.2
7.4
16.5
15.5
Korea
18.6
27.4
32.5
17.6
28.5
13.0
13.4
US
29.1
30.5
33.5
44.8
25.2
28.4
29.1
Japan
24.5
20.9
13.2
18.6
25.2
20.3
20.7
Europe
Import shares for weighting real effective exchange rates (percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Source:
11.6
5.6
5.7
3.7
9.0
7.7
6.9
Korea
Table 9.15
Source:
36.5
20.8
25.0
26.4
14.4
22.9
20.1
Japan
Export shares for weighting real effective exchange rates (percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Table 9.14
6.0
8.8
8.2
7.9
10.2
18.9
18.7
Taiwan
6.0
7.0
10.8
6.0
10.3
3.5
4.1
Taiwan
11.3
2.7
2.2
3.9
3.5
2.8
2.5
Australia
4.8
4.8
1.4
5.0
7.0
2.4
2.5
Australia
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
253

Source:
11.3
7.5
8.1
5.4
8.2
7.7
11.2
Korea
19.5
33.1
32.0
24.8
29.6
35.7
25.7
US
22.2
22.3
19.4
23.1
27.0
27.8
24.7
Europe
34.4
25.3
29.2
35.3
19.6
22.9
24.5
Japan
Trade shares for weighting real effective exchange rates (percent)
Indonesia
Malaysia
Philippines
Thailand
Singapore
China
China1Hong Kong
Table 9.16
6.0
7.8
9.5
6.9
10.2
3.5
11.3
Taiwan
6.6
3.8
1.8
4.5
5.3
2.4
2.5
Australia
100.0
100.0
100.0
100.0
100.0
100.0
100.0
254
The similarities between the ASEAN countries weights suggest that a

common currency basket could fit ASEAN. Clearly, common weights would
have to be the subject of eventual discussion and negotiation within ASEAN.
The similarity of ASEANs weights to those of China suggests that eventual
coordination of ASEAN exchange-rate policy using a basket similar to that
of China could be feasible. A common basket would fit both fairly well.
REER Indexes
Figures 9.109.12 show the movements of the REER indexes using the
alternative weights from Tables 9.149.16. The export-weighted and
total trade indexes use GDP price indexes, which include exports but not
imports. The import-weighted indexes use CPIs, which include imports
but not exports. The ASEAN indexes for total trade from Figure 9.12
are the ones used earlier in Figure 9.8. All three graphs show the stable
pattern of REERs before and after the crisis of 199798. The REERs of
the core ASEAN countries except Singapore show sharp depreciations
during the crisis period. Singapore had a milder depreciation of about
1520 percent from 1996 to 1999 on the bases of export and total trade but
only 5 percent on an import basis. Chinas REER moved differently, with
180
Malaysia
Philippines
Thailand
170
160
Singapore
Indonesia
China + HK
150
140
130
120
110
100
90
80
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Figure 9.10
Real effective exchange rate, export basis (markets outside

ASEAN1China/HK)

255
180
Indonesia
Malaysia
Philippines
170
160
Thailand
Singapore
China + HK
150
140
130
120
110
100
90
80
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Figure 9.11
Real effective exchange rate, import basis (markets outside

ASEAN1China/HK)
180
Malaysia
Philippines
Thailand
160
Singapore
Indonesia
China + HK
140
120
100
80
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Figure 9.12
Real effective exchange rate, total trade basis (markets

outside ASEAN1China/HK)

256
an early appreciation in 199597, followed by a small depreciation. Thus,

Singapore and China escaped the contagion of the crisis.
The graphs show the highly similar movements of the country indexes
across sets of weights. This is expected from Tables 9.149.16, where
country weights are similar across baskets. They also show similar
movements across the ASEAN countries, with China moving somewhat
independently. Before and after the crisis, though, the REER indexes,
including Chinas, show no trend or tendency to diverge. They tend to
stay within a fairly well-defined range. After 2000, they seem to converge.
This suggests that policies are at work in the background that tend to
stabilize REERs in the region and even make them tend to converge. Such
policies would be consistent with an implicit coordination of nominal
exchange rate movements aimed at maintaining REER stability with
a view to maintaining competitive positions. This policy stance can be
seen when the REER movements are decomposed into their nominal
components.
The broad synchronicity of the REER movements across the ASEAN11
countries tells us that it would be possible to construct a common basket
and opt for explicit exchange rate stabilization in East Asia with relative
ease. The relative export and total trade weightings behind these graphs
support this case. The trade structure of these countries is broadly similar,
with each country having roughly the same proportions of trade with
their major trading partners. It would, therefore, be possible to construct
a basket currency based on external or total trade without fundamentally changing each countrys current policies. The regional governments
could use this common basket currency as a benchmark against which to
coordinate their exchange rate policies. In a similar way, the countries of
the European Monetary System used the basket ECU to coordinate their
exchange rate policies (though with the distinction that the ECU was an
internally based basket currency). Such a basket currency could be used
to manage regional stabilization and to address such specific regional
concerns as export competitiveness and the response to externally driven
shocks. It would also be a natural unit of denomination for ABM issues.
Nominal Components of REERs
The REER is the nominal exchange rate adjusted for relative price levels.
Specifically, here it is constructed as the product of a nominal exchange
rate index times the ratio of a trading partners price index to the domestic price index. In symbols, the REER index e is defined as e 5 EP*/P,
where E is the nominal effective exchange rate (NEER) index and P* is the
trading partners price index, both defined by the same set of weights, and

257
P is the domestic price index. The trading partners price index P* is not
substantially affected by domestic policies, but the NEER and P are. Thus,
any policy to stabilize a REER must work on the co-movement of E and
P to be consistent with that of P*. For example, with a stable P*, domestic
inflation in P must be matched by depreciation of E. This is illustrated in
Figures 9.139.15, which show the components of the REERs weighted by
import shares. These are chosen because they use CPIs as the price indexes.
Inflation perceptions and policies tend to focus more on CPIs than on GDP
deflators.
Figure 9.13 depicts the movement of the weighted CPIs for trading
partners (the P* indexes). They move closely together, since they differ
only in their weights across countries (which, as we have seen, are quite
similar). The graph shows a general inflationary trend of about 2 percent
per year for trading partners. Figure 9.14 shows highly divergent movements of domestic price indexes (the Ps). These range from nearly flat for
Singapore to an average of about 11.5 percent a year for Indonesia, with
a jump during the crisis. Figure 9.15 shows the movement of the nominal
exchange rate indexes E. These are similar to the movements of the price
140
135
130
125
120
115
110
Indonesia
Malaysia
Philippines
105
Thailand
Singapore
China + HK
100
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Figure 9.13
Weighted CPI indices for trading partners, core ASEAN and

China 1 Hong Kong

258
500
Indonesia
Malaysia
Philippines
450
Thailand
Singapore
China + HK
400
350
300
250
200
150
100
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Figure 9.14
CPI indices for core ASEAN and China 1 Hong Kong
indexes in the previous graph. Thus, policy pressures on domestic price

levels and nominal exchange rates tend to produce parallel movements that
result in the stability of REERs shown earlier.
Coordination of macro policies in ASEAN would work on both
nominal variables P and E. Coordination of sustainable macro policies
domestically can work to stabilize price levels P across countries around
an acceptable rate of inflation. This is the objective of surveillance.
Coordination of exchange-rate policies can work to stabilize E. The combination would maintain stable and competitive REERs with an acceptable inflation trend in the region. In the absence of sustainable domestic
macro policies that stabilize P around the acceptable trend, price levels
will diverge, as in Figure 9.14, and require changes in nominal exchange
rates to maintain stability in REERs. A combination of surveillance of
sustainable domestic macro policies and coordinated exchange rate policies would be a macropolicy coordination package promoting growth and
poverty reduction.

259
520
Indonesia
Malaysia
Philippines
470
Thailand
Singapore
China + HK
420
370
320
270
220
170
120
70
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Figure 9.15
Nominal effective exchange rate, import basis (markets

outside ASEAN1HK/China)
Implicit Exchange Rate and Monetary Policy Coordination

There is already a surprising degree of implicit coordination of exchange
rate policies among the ASEAN countries and, even further, among the
ASEAN13. Table 9.17 shows the correlations of monthly changes of USD
exchange rates for the ASEAN13 countries for the post-crisis period. All
the correlations are positive. The correlations for China and Malaysia are
small, since these countries essentially fixed their currencies against the USD
during this period. The others show generally strong positive correlations,
indicating common co-movements against the dollar.
Similar results are shown in Table 9.18 for monthly changes in reserve
money. These are all positive and quite large for core ASEAN and
ASEAN11, indicating monetary policy movements consistent with effective exchange rate coordination. These results show that a movement to
explicit coordination would not require a large change in actual policy
positions. The ASEAN countries and China already conduct monetary
policy in such a way as to maintain stability in exchange rates and REERs,

260

Source:
1
0.15
0.39
0.45
0.54
0.07
0.12
0.05
0.10
0.54
0.18
0.27
0.11
0.25
1
0.09
0.23
0.32
0.15
0.19
0.00
0.02
0.34
0.11
0.29
0.07
0.20
1
0.57
0.33
0.31
0.16
0.01
0.15
0.33
0.27
0.24
0.36
0.19
1
0.65
0.05
0.41
0.08
0.00
0.65
0.07
0.53
0.62
0.47
1
0.03
0.34
0.06
0.22
0.98
0.15
0.64
0.33
0.47
1
0.00
0.17
0.05
0.02
0.17
0.06
0.37
0.13
1
0.21
0.03
0.33
0.07
0.35
0.28
0.21
1
0.65
0.04
0.03
0.06
0.10
0.05
1
0.19
0.08
0.19
0.01
0.08
1
0.15
0.58
0.33
0.46
1
0.06
0.11
0.01
1
0.36
0.57
1
0.60
Indonesia Malaysia Philippines Thailand Singapore Vietnam Myanmar Cambodia Laos Brunei China Japan Taiwan Korea
Correlation matrix of monthly changes in dollar exchange rates, January 2000June 2005
Global Financial Data; Datastream.
Indonesia
Malaysia
Philippines
Thailand
Singapore
Vietnam
Myanmar
Cambodia
Laos
Brunei
China
Japan
Taiwan
Korea
Table 9.17
261

1
0.07
0.35
0.33
0.37
0.22
0
0.08
0.19
0.37
0.5
0.07
0.35
1
0.1
0.34
0.73
0.55
0.15
0.2
0.22
0.5
0.34
0.46
0.08
1
0.21
0.29
0.15
0.06
0.03
0.11
0.27
0.4
0.07
0.15
1
0.51
0.07
0.1
0.11
0.18
0.47
0.65
0.15
0.42
1
0.45
0.02
0.22
0.2
0.64
0.61
0.37
0.29
1
0.13
0.05
0.1
0.45
0.26
0.39
0.07
1
0.06
0.13
0.14
0.27
0.1
0.01
1
0.04
0.13
0.13
0.04
0.07
1
0.03
0.11
0.17
0.12
Indonesia Malaysia Philippines Thailand Singapore Vietnam Myanmar Cambodia Laos
1
0.51
0.33
0.28
China
1
0.04
0.41
1
0.18
IMF IFS; Taiwan CBC.
Japan Taiwan Korea
Correlation matrix of monthly changes in reserves, denominated in LCU (starting Jan 1992)
China starting July 99, Vietnam missing JulyDec 99; Laos missing Mar 05; not including Brunei (pegged to Singapore dollar).
Sources:
Note:
Indonesia
Malaysia
Philippines
Thailand
Singapore
Vietnam
Myanmar
Cambodia
Laos
China
Japan
Taiwan
Korea
Table 9.18
262
as discussed above. The policies are implicitly coordinated, as seen in the

correlations here. A movement to explicit coordination would present a
common posture to the markets and diffuse speculative pressure when it
arises.
9.5
CONCLUSIONS
This chapter has examined the prospects for macro policy and exchange rate
coordination in Asia (in particular in the ASEAN and in the ASEAN plus
China area) and looked at the potential implications of such coordination
for reserve sharing and financial developments such as the ADF and the
ABM. We argue that better coordination of the underlying macro policies
could have prevented the Asian crisis or minimized its costs, in particular
through beneficial consequences for pre-crisis policy, the control of exposure to speculative contagion, and swifter post-crisis investment recovery.
We argue that the existing geographic and commodity structure of
the regions trade means that REERs already move together, ensuring
a minimum autonomy loss from policy coordination. We looked at the
geographic structure of trade, using this to develop weights for baskets
of ASEAN currencies in terms of major trade partners outside ASEAN
and China. Such weights are broadly similar for the main ASEAN countries. The movements of real effective exchange rates are also broadly
synchronous. This highlights an existing implicit coordination among
the ASEAN11 countries and suggests that the use of a common basket
for exchange rate coordination would not mark a radical departure from
existing policies or from domestic autonomy. We additionally looked at
the commodity composition of trade for the ASEAN11 countries. Once
again, this composition is broadly similar for the larger ASEAN11 countries (SIC 7 being the dominant export sector, while the manufactures of
SIC 68 dominate imports). Such a similar commodity structure implies
greater benefits from coordination, as argued by Mundells optimal currency area theory. Overall, the trade data by both sector and geography
show no serious imbalances across major sources and markets. A coordinated exchange rate stabilization against a common basket based on these
sources and markets shouldnt create substantial differential pressures
within ASEAN or China.
We construct real and nominal effective exchange rate measures for
the countries under examination. These rates demonstrate that moves
toward integration are already in practical effect in the region and underline the ease of any potential transition toward greater coordination. In
particular, they highlight how an externally based currency basket might

263
easily be constructed, offering a focal point for greater regional integration. Potentially playing a role similar to that of the ECU within the
European Monetary System, such a basket currency could permit greater
coordination of regional policies, addressing particular regional concerns
regarding competitiveness and the response to externally driven shocks
or cascading speculation. It would yield direct benefits for growth and
poverty reduction, as well as setting the foundation for further cooperation in the areas of reserve sharing and ABM development. The regional
currency unit would also provide a natural basis for the denomination
of ABM issues.
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
See, for instance, Branson (2005).

These are the REERs based on total trade weights for the six major markets outside
ASEAN11, indexed to 1989 5 100. The methodology is explained in detail in section
9.4. For clarity, we focus here on the total trade-weighted REERs.
See Young (1995) for a fundamental discussion of investment and growth in Asia.
For a discussion of the rationale for nominal exchange rate stabilization see also
Gunther Schnabls chapter in this volume (Chapter 10).
These are analyzed in some detail in Branson (2001).
The European Monetary System, which actually involved extensive credit facilities in
order to secure the stability of intra-regional exchange rates, was in a state of political
disarray that made markets doubt the future of European monetary integration when
it was hit by the crisis. It was the lack of a cooperative response to speculative pressure
that forced its weak members to exit the system. See Volz (2006).
See, for example, Kuroda and Kawai (2003) and Nasution (2005).
See Ogawa and Shimizu, Chapter 5 in this volume.
This problem is more striking and difficult in the case of Singapore.
The 42.3 percent in SIC 9 for Philippines is probably unclassified SIC 68. See also
Myanmar.
REFERENCES
Branson, William H. (2001). Intermediate Exchange-rate Regimes for Groups of
Developing Countries. In Jorge Braga de Macedo, Daniel Cohen, and Helmut
Reisen (eds), Dont Fix, Dont Float. Paris: OECD.
Branson, William H. (2005). The Asian Crisis as a Bubble in Foreign Exchange
Markets. Manuscript, Princeton University, Princeton, NJ.
Kuroda, Haruhilo, and Masahiro Kawai (2003). Strengthening Regional Financial
Cooperation in East Asia. PRI Discussion Paper 03A-10, Policy Research
Institute, Ministry of Finance, Tokyo.
Mundell, Robert (1961). A Theory of Optimal Currency Areas. American
Economic Review 51: 65765.
Nasution, Anwar (2005). Monetary Cooperation in East Asia. Journal of Asian
Economics 16: 42242.

264
Radelet, Steven, and Jeffrey Sachs (1998). The Onset of the East Asian Financial
Crisis. NBER Working Paper no. 6680, National Bureau of Economic
Volz, Ulrich (2006). On the Feasibility of a Regional Exchange Rate System for
East Asia: Lessons of the 1992/93 EMS Crisis. Journal of Asian Economics 17:
110727.
World Bank (2000). East Asia Update, Special Focus: Poverty During Crisis and
Recovery. East Asia and Pacific Publications, September 18, pp. 16.
Young, Alwyn (1995). The Tyranny of Numbers: Confronting the Statistical
Realities of the East Asian Growth Experience. Quarterly Journal of Economics
110: 64180.

10.
Capital markets and exchange

rate stabilization in East Asia:
diversifying risk based on currency
baskets
Gunther Schnabl
10.1
INTRODUCTION: MORE EXCHANGE RATE

FLEXIBILITY IN EAST ASIA?
Before the Asian crisis of 199798, China, Hong Kong, Indonesia, Korea,
Malaysia, the Philippines, Singapore, Taiwan, and Thailand pursued a
common exchange rate peg to the US dollar. This (informal) East Asian
dollar standard (McKinnon and Schnabl 2004a) was beneficial for growth
in the region for several reasons. First, it ensured macroeconomic stability
by bringing their domestic inflation to US levels. Second, the joint peg to
the dollar provided low transaction costs not only for trade with the US
but also for intra-regional trade flows, which make up about 50 percent
of overall East Asian trade. Third, exchange rate stability provided low
transaction costs for short-term and long-term international and intraregional capital flows. After the onset of the East Asian crisis, the East
Asian dollar standard fell apart. While China, Hong Kong, Singapore, and
Taiwan kept their exchange rates rather stable against the dollar during the
crisis, the currencies of the crisis countriesIndonesia, Korea, Malaysia,
the Philippines and Thailanddepreciated sharply against the dollar,
and the depreciation of their currencies was accompanied by cumbersome
recessions.
The post-crisis policy recommendations for the exchange regimes in
East Asia have been of diverse natures. Associating exchange rate stability
against the dollar with overly low risk premia on volatile capital inflows,
the IMF recommended more exchange rate flexibility (Fischer 2001). In
contrast, McKinnon and Schnabl (2004a, 2004b) argued that exchange
rate stabilization against the dollar is fully rational, even post-crisis,
because of the high degree of dollarization of international and intraregional trade and capital flows. A third strand of literature has proposed
267

268
Currency baskets for East Asia?
maintaining the exchange rate pegs in the region, while at the same time
pegging to more than one anchor currency. According to Williamson
(2000 and 2005), currency basket arrangements would be beneficial as they
stabilize the nominal effective exchange rates. To maintain intra-regional
exchange rate stability, Ogawa and Ito (2002) have proposed coordinating
the currency baskets in the region.
While exchange rate volatility against the dollarin particular at high
frequenciesdeclined (close) to pre-crisis levels over the period from
after the East Asian crisis up to late 2004, over the year 2005 exchange
rate volatility against the dollar increased substantially. As will be argued
in this chapter, the rising exchange rate flexibility against the dollar may
be a result of higher exchange rate stability against the Japanese yen and
the euro reflecting the importance of Japan as a trading partner in the
region as well as the growing role of the euro as a (stable) international
currency.
10.2
THE RATIONALE FOR EXCHANGE RATE

STABILIZATION IN EAST ASIA
Despite policy recommendations to pursue more exchange rate flexibility,

fully flexible exchange rates such as the dollareuro rate are unlikely for the
East Asian currencies (except the Japanese yen) for several interdependent
reasons: transaction costs for international trade, macroeconomic stabilization, and the dollarization of international capital flows.
The idea of a positive effect of exchange rate stability on international
trade via lower transaction costs has found new support in a branch
of literature building on Frankel and Rose (2002). For East Asia, the
common dollar peg has been beneficial not only for trade with the US but
also (and particularly) for intra-regional trade, which is mostly invoiced
in US dollars.1 By jointly pegging to the dollar, the East Asian countries
also avoided beggar-thy-neighbor depreciation, thereby enhancing
macroeconomic stability in the region.
Macroeconomic stabilization is another important motivation for
exchange rate stabilization. By pegging the exchange rate to the dollar
or the euro, a small open economy can sustain a stable price level for two
reasons. First, undisciplined monetary policy that aims to finance public
expenditure via inflation tax is contained (Crockett and Goldstein 1976).
Second, even if public expenditure is sound, in a small open economy
nominal exchange rate fluctuations are likely to translate into fluctuations
of the domestic price level. Exchange rate stabilization is equivalent to the
stabilization of the domestic price level (McKinnon 1963).

Capital markets and exchange rate stabilization in East Asia
269
Besides international trade and macroeconomic stabilization, international capital flows have gained increasing importance as an explanatory
variable for exchange rate stabilization in countries that have underdeveloped capital markets in domestic currencies (like most emerging East Asian
economies). The incentive for exchange rate stabilization arises from the fact
that private and public agents in developing countries and emerging markets
find themselves unable to borrow or to lend in their domestic currencies.
Liability Dollarization
As put forward by Eichengreen and Hausmann (1999), because of a long
history of inflation and currency depreciation, banks and enterprises in
emerging markets and developing countries cannot use their currencies to
borrow abroad or to borrow long term, even domestically, a phenomenon
they have called original sin. International creditors lend in dollars or euros
and thereby shift the exchange rate risk of open positions in foreign debt to
the debtor countries. The consequence is a currency mismatch as projects
that are financed with foreign currency generate domestic currency.
If private banks are able to shift the foreign exchange risk of international borrowing to (private) domestic debtors by foreign currency lending,
the currency risk is transformed into default risk, as households and small
private enterprises tend to remain unhedged. The upshot is that liability
dollarization creates an incentive for exchange rate stabilization at both
high frequencies (i.e. daily exchange rate fluctuations) and low frequencies
(fluctuations of the exchange rate level over months, quarters, and years).
Short-term exchange rate stabilization
With domestic foreign exchange markets being small and illiquid, most
international short-term payment transactions in emerging markets and
developing countries are denominated in US dollars or euros. Domestic
currencies are generally not accepted as international means of payment.
In East Asia, the dollar has been used not only for goods and capital transactions with the US but also for invoicing intra-regional trade and capital
flows. In these underdeveloped capital markets, an active forward market
in foreign exchange against the dollar (or euro) is for the most part nonexistent. If hedging instruments are available, the cost of hedging dollar
liabilities (i.e. the premium on buying dollars forward with the domestic
currency) is high. The result is that the foreign exchange risk of short-term
capital transactions typically remains unhedged.
Monetary authorities can provide an informal hedge for private shortterm capital transactions by minimizing daily exchange rate fluctuations.
This provides compensation for the underdeveloped private market in

270
forward exchange. In practice, the monetary authorities establish formal or

informal limits to daily exchange rate fluctuations defined in terms of percent
exchange rate changes (Chmelarova and Schnabl 2006), allowing private
banks and enterprises to repay their short-term foreign currency liabilities
with minimal exchange rate risk. As shown in Figure 10.1, except for crisis
periods, exchange rate volatility against the dollar of the East Asian currencies has tended to be smaller than for freely floating currencies such as the
Japanese yen and the euro (before 1999, the German mark). In contrast,
the monetary authorities of large countries with highly developed capital
markets (mainly the US, the euro area, and Japan) can leave day-to-day
exchange rate fluctuations to market forces for two reasons. First, private
transactions with emerging markets and developing countries are mainly
invoiced in domestic currencies. Second, for transactions between these large
countries, well-developed domestic capital markets provide a broad variety
of instruments to hedge the foreign exchange exposures at low costs.
Long-term exchange rate fluctuations
At low frequencies, the rationale for exchange rate stabilization in debtor
countries originates in long-term liability dollarization. If net debt is
denominated in foreign currency, increasing long-term exchange rate stability is equivalent to reducing default risk on balance sheets (Eichengreen
and Hausmann 1999). In particular, a sharp domestic currency depreciation puts balance sheets at risk, possibly forcing indebted enterprises
and financial institutions into default. In the case of an appreciation, the
competitiveness and profit margins of the export industry are at risk. Even
low-frequency exchange rate fluctuations around a constant level may pose
a risk for the domestic economy, as larger uncertainty is reflected in higher
risk premiums on domestic interest rates.
To shield domestic enterprises and financial institutions against this
uncertainty, the authorities may control low-frequency exchange rate fluctuations to enhance the stability of the domestic financial system. In practice, monetary authorities in emerging markets and developing countries
use daily exchange rate changes as an intermediate target, smoothing
daily fluctuations in order to control long-term fluctuations of the exchange
rate level (Chmelarova and Schnabl 2006). As shown in Figure 10.2, most
East Asian currencies have fluctuated significantly less against the dollar
since the Asian crisis than the freely floating eurodollar exchange rate. In
contrast, large countries with highly developed capital markets can allow
for more long-term exchange rate flexibility for two reasons. First, with
international debt denominated in domestic currency, balance sheets are
not exposed to exchange rate fluctuations. Second, given the lower degree
of openness, export industries are less important for economic growth.

271

99
02
05
1.
1.
1.
.0
.0
.0
01
01
01
Thai baht
96
1.
.0
01
93
1.
.0
01
90
1.
Malaysian ringgit
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
Chinese yuan
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
Philippine peso
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Japanese yen
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
.0
Hong Kong dollar
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Euro (Deutsche mark )
Singapore dollar
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
Indonesian rupiah
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
96
99
02
05
1.
1.
1.
1.
.0
.0
.0
.0
01
01
01
01
Swiss franc
93
New Taiwan dollar
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
Korean won
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
05
1.
.0
01
02
1.
.0
01
99
1.
.0
01
96
1.
.0
01
93
1.
.0
01
90
1.
.0
01
4%
3%
2%
1%
0%
1%
2%
3%
4%
Exchange rate volatility against the US dollar of selected crisis and non-crisis currencies, 1990:012005:12
(daily)
Bloomberg. Volatility is measured in daily percentage changes against the dollar.
Figure 10.1
Source:
01
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
01
.0
4%
3%
2%
1%
0%
1%
2%
3%
4%
272

00
n
Ja
02
J
an
04
n
Ja
90
n
Ja
92
n
Ja
96
n
Ja
98
n
Ja
00
Taiwan dollar
n
Ja
n
Ja
98
n
Ja
00
n
Ja
02
n
Ja
04
Ja
80
90
100
110
120
90
90
90
Ja
80
East Asian exchange rates against the dollar, 1990:12005:12
Thai baht
n
Ja
04
80
96
an
80
n
Ja
02
90
an
90
94
00
100
n
Ja
an
100
92
98
110
n
Ja
an
Philippine peso
96
110
90
an
90
100
110
120
130
140
Ja
80
90
130
04
94
Ja
04
120
n
Ja
an
Ja
02
130
02
92
Ja
00
120
n
Ja
an
Ja
98
Hong Kong dollar
Ja
96
130
90
Ja
94
140
an
IMF: IFS, central bank of China. Index 1999.01 5 100.
Figure 10.2
Source:
n
Ja
Ja
92
140
94
98
Malaysian ringgit
an
90
140
96
80
80
an
90
90
n
Ja
100
100
Ja
110
94
Ja
110
an
n
Ja
120
92
n
Ja
130
n
Ja
120
90
Ja
Chinese yuan
n
Ja
130
n
Ja
140
Ja
140
n
Ja
04
80
02
80
00
90
90
98
100
100
100
96
110
110
110
94
120
120
120
92
130
130
130
90
140
140
140
Ja
92
92
92
n
Ja
Ja
Ja
Ja
Ja
94
94
94
Ja
98
Ja
96
n
Ja
98
Ja
00
00
Korean won
96
n
Ja
96
98
Euro
Ja
Ja
00
Singapore dollar
Ja
Ja
Ja
02
02
02
Ja
Ja
Ja
Ja
Ja
04
04
04
273
Asset Dollarization
While liability dollarization provides an important rationale for exchange
rate stabilization for internationally indebted countries, creditor countries
such as China and Russia are also at risk. Supported by the continuous
rise in the US current account deficit, an increasing number of countries
in East Asia, Latin America, the Middle East, and the Commonwealth of
Independent States (CIS) are running sustained current account surpluses.
In East Asia, while China, Japan, Singapore, and Taiwan exhibited current
account surpluses for most of 1990s, the crisis countriesIndonesia,
Korea, Malaysia, the Philippines, and Thailandreversed their current
account positions during or after the Asian crisis and have remained net
exporters since then. East Asia has become the most important creditor in
the international capital markets.
McKinnon and Schnabl (2004b) explain the rationale for exchange rate
stabilization in creditor countries that are not able to lend internationally
in their domestic currencies. Because of underdeveloped financial markets,
capital controls, or even simply the fact that the dollar seems to be a more
reliable store of value than the domestic currencies, private investors find
dollar assets more attractive than international claims denominated in
domestic currency. Conversely, the United States, as the largest debtor
country, is disinclined to incur debts denominated in foreign currencies; the
position of the US dollar as the worlds prominent international currency
allows US private and public agents to borrow internationally in domestic
currency. The exchange rate risk of international lending is shifted to the
creditors.
By fixing exchange rates at high frequencies, the authorities can hedge
the risk for private, short-term international lending. If capital markets are
underdeveloped, forward transactions by risk-averse East Asian traders
wanting to hedge their open positions in foreign exchange are difficult,
even for net international creditors. The authorities can provide an overall
hedge by minimizing exchange rate fluctuations on a daily basis. In contrast, day-to-day exchange rate fluctuations in Japan, the US and the euro
area are left to market forces.
At low frequencies, the motivation for exchange rate stabilization by international creditor countries can be linked to the perception of risk by private
and public holders of net foreign currency assets. Based on sustained current
account surpluses, Japan, China, Korea, and Taiwan (among other East
Asian countries) have accumulated substantial amounts of international
assets. Because of the asymmetric nature of the world currency system, the
greater part of these international assets can be assumed to be denominated
in US dollars (for instance, held in US government bonds). When East Asian

274
investors accumulate assets in US dollars, they become anxious about an

appreciation of the domestic currency against the dollar, as their savings
would lose value in terms of the domestic currency. This fear of appreciation may be compensated by higher interest rates in the debtor country as
suggested by the open interest rate parity. But, with varying interest rates in
the anchor country, the perception of risk may change. If the depreciation
pressure on the anchor currency is sustained, individual or institutional
holders of large dollar assets will increasingly be at risk.
This sets the stage for official foreign exchange intervention. Once
private investors decide to repatriate their international assetsor decide
to convert domestic dollar assets into domestic currency assets in highly
dollarized countries, such as Russiathe appreciation of the domestic currency will reduce the value of dollar assets on domestic balance sheets. For
example, for Japanese insurance companies, whose liabilities to annuity
holders are in yen but with a substantial share of their assets in dollars, the
reduction in the yen value of their dollar assets could wipe out their net
worth. In China, firms and households will hold dollar assets only if there
is a substantial business convenience in doing so or if the interest rate on
dollar assets is higher. The primary downside risk is that the yuan (yen)
could appreciate against the dollar and thus reduce the yuan (yen) value
of their dollar assets.
Depending on how sensitive domestic holders of dollar assets are to
this risk, periodic runs from dollars into yuan and yen could occur simply
based on rumors of appreciation. The reaction of monetary authorities will
depend on the exchange rate regime. In Japan, where the exchange rate is
principally flexible, foreign exchange intervention is restricted to periods
when the exchange rate level is perceived to be too high and detrimental to
growth (Hillebrand and Schnabl 2006). Despite its status as independently
floating, Japan has accumulated the largest stock of international assets
to date. China, which has been committed to a tight peg to the dollar, has
absorbed private capital inflows on a more regular basis. Note that any
rise in appreciation expectations will further add to the speed of reserve
accumulation.
10.3
CURRENCY DIVERSIFICATION IN
INTERNATIONAL RESERVES AND
CURRENCY BASKETS
Given this rationale for exchange rate stabilization, it seems unlikely that
East Asian countries will pursue fully flexible exchange rate strategies in
the short and medium term. Instead, they will tend to stabilize exchange


1.90
275
East Asia (against the dollar)

Emerging Europe (against the euro)
1.70
Percent
1.50
1.30
1.10
0.90
0.70
0.50
Jan 00
Jan 01
Jan 02
Jan 03
Jan 04
Jan 05
Source: IMF. Regional exchange rate volatility defined as arithmetic averages of 12

month rolling standard deviations.
Figure 10.3
Exchange rate volatility in East Asia and emerging Europe
rates in the form of tight or soft pegs based on smoothing daily exchange
rate fluctuations. McKinnon and Schnabl (2004a) have shown that, postcrisis, the East Asian emerging market countries have returned to their
pre-crisis exchange rate stability against the dollar, at least at high frequencies (i.e. day-to-day or week-to-week exchange rate fluctuations). Indeed,
measured as an arithmetic average of month-to-month percent exchange
rate changes of all East Asian countries, post-crisis exchange rate volatility
against the dollar has declined gradually through the year 2004.
The Risk of One-Sided Dollar Pegs
Although the dollar remains the most important anchor currency in the
region, Figure 10.3 also suggests that East Asian exchange rate volatility
against the dollar was on the rise in 2005. Chinn and Frankel (2005) argue
that the role of the dollar as an international currency may be challenged
by the euro, depending on the long-term inflation expectations for the US
economy. In East Asia, the degree of macroeconomic stabilization, which
is achieved via exchange rate pegs, hinges not only on domestic efforts to
keep the exchange rate stable against the anchor currency but also on the
monetary policy in the anchor country.

276
While low and stable inflation has been a crucial prerequisite for the role
of the dollar as the dominant international currency, the degree of price
stability has fluctuated over time. Since the late 1960s, the US dollar has
experienced several phases of rising inflation and sustained depreciation
pressure. During these periods a relatively loose US monetary policy has
been transmitted via reserve accumulation into rising inflationary pressure
in countries stabilizing their currencies against the US dollar. Back in the
late 1960s and early 1970s, an expansionary fiscal and monetary stance in
the US contributed to a world-wide increase in inflation, which finally triggered the breakdown of the Bretton Woods system. While the European
currencies were de-linked from the dollar in the early 1970s (thereafter
stabilizing their exchange rates against the German mark), most countries
outside of Europe, for instance those in East Asia, continued to peg their
currencies more or less tightly to the dollar. The international role of the
dollar was enhanced by its dominant role as an invoicing currency for
international trade, the deepness of US capital markets, and the lack of
alternative international currencies. This has led Dooley et al. (2004) to
argue that the United States is at the center of what they have called a
revived (informal) Bretton Woods system of fixed exchange rates.
Recently, however, an exceptionally loose fiscal and monetary stance
under the Bush administration has triggered a discussion about the impact
of fast reserve accumulation on the countries stabilizing their exchange rates
against the dollar. When the Federal Reserve kept the interest rate at historically low levels from 2001 to 2004, the dollar came under strong depreciation
pressure. As many countries continued to stabilize exchange rates against
the dollar they accumulated large amounts of dollar reserves. Figure 10.4
shows the substantial speed of reserve accumulation in East Asia, which
has accelerated in many countries since 2001, most dramatically in China,
Korea, Malaysia, and Japan.
The fast reserve accumulation has two main downsides. First, as under
fixed exchange rates the scope for sterilization of foreign exchange intervention is limited, many countries in East Asia, Latin America, and the
Middle East experienced fast monetary expansion. Although inflation has
been contained in most countries so far, the fast growth of monetary aggregates has contributed to surging stock and real estate prices. An eventual
burst of such bubbles may result in cumbersome crises like those experienced after the Asian crisis and painful post-bubble recessions like those
experienced in Japan since the early 1990s.
Second, for countries with sustained current account surpluses, rising
world inflation has a negative impact on the real value of export revenues
and international assets. If, as in the case of many commodity exporting
countries, export revenues are earned in dollars and spent on imported

277

Malaysia
Thailand
90 92 94 96 98 00 02 04
n n n n n n n n
Ja Ja Ja Ja Ja Ja Ja Ja
10
20
30
40
50
Hong Kong
90 92 94 96 98 00 02 04
n n n n n n n
Ja Ja Ja Ja Ja Ja Ja
Philippines
Japan
90 92 94 96 98 00 02 04
n- an- an- an- an- an- an- anJ J J J J J J
Ja
900
800
700
600
500
400
300
200
100
0
90 92 94 96 98 00 02 04
n n n n n n n n
18
16
14
12
10
8
6
4
2
0
Ja
20
40
60
80
100
120
140
Singapore
Germany
90 92 94 96 98 00 02 04
n- an- an- an- an- an- an- anJ J J J J J J
Ja
140
120
100
80
60
40
20
0
90 92 94 96 98 00 02 04
n an an an an an an an
J J J J J J J
Indonesia
90 92 94 96 98 00 02 04
n n n n n n n
Ja
120
110
100
90
80
70
60
50
40
30
20
10
0
Ja
40
36
32
28
24
20
16
12
8
4
0
Taiwan
90 92 94 96 98 00 02 04
n- n- n- n- n- n- n- nJ a J a Ja Ja Ja Ja Ja J a
US
10
20
30
40
50
60
Korea
90 92 94 96 98 00 02 04
n n n n n n n
90 92 94 96 98 00 02 04
n an an an an an an an
J J J J J J J
Ja
280
240
200
160
120
80
40
0
Ja
220
200
180
160
140
120
100
80
60
40
20
0
Official foreign exchange reserves of East Asian and G3 countries in billions of dollars, 1980:012005:6
(monthly)
IMF: IFS. Billions of dollars. Note different scales on the y-axis.
Figure 10.4
Source:
China
90 92 94 96 98 00 02 04
n n n n n n n n
80
70
60
50
40
30
20
10
0
90 92 94 96 98 00 02 04
n n n n n n n n
800
700
600
500
400
300
200
100
0
278
goods from the euro area, dollar depreciation against the euro erodes the
real purchasing power of dollar denominated earnings. For international
creditor countries, such as Japan, China, Russia, and Saudi Arabia, which
have accumulated large stocks of dollar denominated international assets,
the appreciation of domestic currencies reduces the value of these assets in
terms of domestic currencies (section 10.2).
The upshot is that, between 2001 and 2004, private and public investors
reacted differently to the sustained dollar depreciation. As the sharply
rising US current account deficit went along with rising current account
surpluses in countries stabilizing their exchange rates against the dollar,
private investors tended to convert dollar positions into domestic currencies, bringing their domestic currencies under appreciation pressure. In
contrast, the monetary authorities in many emerging markets in East Asia,
the Middle East, Latin America, and the CIS tended to resist this appreciation pressure via foreign exchange intervention. From the perspective of
the monetary authorities, this leaning against the wind in the build-up
of international assets is fully rational because it shields export industries
against appreciation and maintains the nominal value of the large stocks
of international dollar assets. Both factors contribute to macroeconomic
stability.
In the longer term, however, the monetary authorities of East Asian
countries may change their exchange rate targets. If they expect the depreciation of the dollar to continue, they may consider reducing dependency
on the dollar as an anchor and reserve currency (Chinn and Frankel 2005).
The current expectations about the future value of the US dollar hinge on
the expected macroeconomic policies in the US. During 2004 and 2005, the
interest rate increases of the Federal Reserve helped to sustain the value
of the dollar, which appreciated against most currencies during 2005. Yet
if the US fiscal deficit and the low private savings rate are expected to
continue, and if a burst of the current real estate bubble seems likely,
the future federal funds rate may be expected to be lower. The implication
would be a higher level of US inflation and further depreciation pressure
on the dollar.
Diversification of Risk
If central banks around the world that have in the past used the dollar as
the predominant anchor and reserve currency see a significant probability
of a sustained dollar decline, they will consider reducing their dependency
on the US currency. In contrast to former periods of dollar depreciation
such as in the 1970s, today the euro has become a viable competitor as
a pegging and reserve currency (Chinn and Frankel 2005, ECB 2005).

279
Although the integration of the European financial markets is still lagging

behind the US, deep and liquid euro capital markets provide a broad
variety of investment opportunities. In addition, the European Central
Bank may be perceived as more inflation-averse than the Federal Reserve,
strengthening the attractiveness of the euro as an international currency.2
Instead of solely pegging to the dollar, the East Asian countries may peg
to a basket of currencies for several reasons. Williamson (2000 and 2005)
has proposed that a currency basket for the East Asian countries should
reflect the direction of trade (rather than the currency denomination of
trade). The benefits would be nominal effective exchange rate stability3
and lower fluctuations of overall trade (Gudmundsson 2005). The weights
of such a currency basket would closely reflect the trade structure of the
respective countries, giving substantial weights to the dollar, the yen,
and the euro as well as to smaller currencies such as the Korean won, the
Malaysian ringgit, and the Chinese yuan.4
In particular, Japan is one of the most important trading partners of the
smaller East Asian countries. The strong depreciation of the Japanese yen
against the dollar after 1995 (when the smaller East Asian countries kept
their exchange rates tightly pegged to the dollar) contributed to the Asian
crisis (McKinnon and Schnabl 2003). For this reason, Kawai (2002) has
proposed that the yen should be given prominent weighting in the East
Asian currency baskets in order to maintain intra-East Asian exchange
rate stability. Ogawa and Ito (2002), reflecting the argument of McKinnon
and Schnabl (2004a) that the common peg of the East Asian currencies to
the dollar contributed to the stabilization of intra-regional exchange rates
and thereby fostered intra-regional trade, have proposed a coordination of
the weights of all East Asian currency baskets. This could be achieved by
Williamsons (2005) proposition that the dollar, yen, and euro should be
treated equally in the East Asian exchange rate strategies.
In addition to this purely trade-oriented approach, there is also a case
for pegging to a basket of currencies from the point of view of macroeconomic stabilization. As shown above, since 2001 the tight pegs to the dollar
in East Asia have contributed to a fast accumulation of foreign reserves.
With underdeveloped capital markets and partially free capital flows, the
scope for sterilization is limited. The respective expansion of the money
supply may be regarded as a threat to price and macroeconomic stability.
Given the need for exchange rate stabilization as outlined in section 10.2,
the domestic price level can be regarded as a function of the price level of
the anchor country. In the case of a currency basket, the domestic price
level would be a function of the price levels of all anchor countries. For
this reason, countries would be inclined to give a higher weight to currencies that are regarded as particularly stable. From this point of view, the

280
Japanese yen may not qualify as an anchor currency as long as the zero
interest rate policy and deflation continue.
Furthermore, the expectations about the longer-term stability of specific
anchor currencies will also have an impact on the choice of the currency
composition of foreign reserves. In the past, the foreign reserves of the
East Asian countries were widely considered to be denominated in US
dollars, as East Asia was widely dollarized and exchange rates were stabilized against the dollar. However, if the East Asian central banks expect
a further depreciation of the dollar, they may wish to diversify their portfolio of international currencies, giving a higher weight to the euro (or the
yen).
This can be achieved with the help of two strategies. First, while pegging
against the US dollar continues (causing foreign reserves to be accumulated in US dollars), dollar assets can be converted into euro assets.
Although the peg against one anchor currency would be compatible with
a diversification of reserves, the downside of this strategy is that dollar
sales would put further depreciation pressure on the dollar and therefore
would require additional foreign exchange intervention. This effect could
be avoided if the restructuring of the currency structure of official foreign
reserve holdings were to take place in times of dollar appreciation, as was
the case in 2005.
Although there is no need to give different currencies similar proportions
as anchor and reserve currencies, countries may strive to harmonize the
currency structure of the foreign assets with the weights of the currencies
in their intervention baskets. Gudmundsson (2005) argues that many
central banks use a minimum variance analysis to determine their reserve
compositions. This implies that reserve structures mirror intervention
basket structures in order to reduce nominal fluctuations of the value of the
international reserves. For instance, Russia had given the dollar a weight
of 60 percent in foreign reserves and 65 percent in the currency basket in
mid-2005, giving the euro weights of 33 percent and 35 percent, respectively
(Schnabl 2006).
To this end, the desire to diversify the currency denomination of international reserves may enhance the role of the euro in possible basket
strategies. This implies a causal relation between policy goals concerning
the reserve composition and the desired exchange rate target. Further note
that exchange rate stabilization based on basket strategies also would allow
full hedging of the foreign exchange risk of international payments flows,
as uncertainty only originates in the exchange rate fluctuations between
the dollar, the euro, and the yen. For these exchange rate fluctuations,
the highly developed capital markets in Japan, the US, and the euro area
provide sufficient tools for hedging the foreign exchange risk.

10.4
281
ESTIMATION OF BASKET STRUCTURES
Frankel and Wei (1994) have proposed an OLS estimation that allows for
the tracking of the structures of undisclosed currency baskets. The empirical analysis of the currency basket structures in East Asia proceeds in two
stages. First, we test for the basket structures before the Asian crisis, which
are expected to reveal a strong US dollar weight. Second, based on a rolling
window approach, possible changes in basket structures are identified,
with a special focus on the year 2005.
Following Frankel and Wei (1994), we use an outside currencythe
Swiss francas a numraire for measuring exchange rate volatility in the
East Asian currencies (except the yen). This volatility could then be partitioned into movements in major currencies against the Swiss franc. For
example, if changes in the Korean won against the Swiss franc are largely
explained by changes in the US dollar against the Swiss franc, the US
dollar has very high weight in the Korean currency basket. We regress the
exchange rates of each of the nine East Asian currencies on the US dollar,
the Japanese yen, and the euro5 with the Swiss franc as numraire:
eEastAsiancurrencySwissfranct 5 a1 1 a2eDollarSwissfranct
1 a3eYenSwissfranct 1 a4eEuroSwissfranct 1 ut
(10.1)
The multivariate OLS regression6 is based on first differences of logarithms

in the exchange rate e. The residuals are controlled for heteroscedasticity.
The daily data are compiled from Bloomberg. The a coefficients represent
the weights of the respective currencies in the currency basket. If the East
Asian currency is closely fixed to one of the major currencies appearing on
the right hand side of equation (10.1), the corresponding a coefficient will
be close to unity. If a coefficient is close to zero, there is no exchange rate
stabilization against that particular currency.
The Pre-Crisis Currency Baskets
First, we estimate the composition of East Asian countries currency
baskets for the pre-Asian crisis period, which starts in February 1994
when China unified its foreign exchange market and ends in May 1997
before the first major turbulence (869 observations). Table 10.1 reports the
results, showing the high weights of the dollar in the East Asian currency
baskets. The estimates for a2 are all close to unity, ranging from 0.82 for the
Singapore dollar up to 1.00 for the Chinese yuan, the Hong Kong dollar,
and the Indonesian rupiah. The correlation coefficients (R2) being close to

282
Table 10.1
Pre-Asian crisis East Asian currency basket structures

(02/01/9405/30/97)
Constant a1
Chinese
yuan
Hong Kong
dollar
Indonesian
rupiah
Korean
won
Malaysian
ringgit
Philippine
peso
Singapore
dollar
New
Taiwan
dollar
Thai baht
0.00
(1.15)
0.00
(0.30)
0.00
(3.19)
0.00
(1.42)
0.00
(1.48)
0.00
(0.34)
0.00
(1.32)
0.00
(0.84)
0.00
(0.61)
Yen a3
DM a4
R2
1.01*** 0.01
(1.48)
(158.63)
0.00
1.00***
(0.25)
(454.79)
1.00*** 0.01
(0.92)
(144.93)
0.06***
0.97***
(3.31)
(66.27)
0.09***
0.88***
(5.30)
(54.80)
0.02
0.97***
(0.74)
(43.34)
0.14***
0.82***
(4.83)
(34.37)
0.03**
0.98***
(1.38)
(57.30)
0.02
(1.70)
0.01
(1.36)
0.01
(0.85)
0.01
(0.29)
0.01
(0.45)
0.01
(0.45)
0.08***
(2.97)
0.01
(0.54)
0.97
0.01
(0.35)
0.95
Dollar a2
0.92***
(81.25)
0.08***
(5.17)
1.00
0.97
0.93
0.90
0.86
0.86
0.93
Notes: t-statistics in parentheses; ** significant at the 5% level; *** significant at the

1% level; 869 observations. White heteroskedasticity-consistent standard errors and
covariance.
Source:
Datastream. Daily data.
unity indicate that fluctuations of the East Asian currencies exchange rates
against the Swiss franc can almost fully be explained by fluctuations of the
dollar against the Swiss franc.
The results show that high dollar weights also can be achieved under a
downward crawling peg arrangement, as in Indonesia. Before the Asian
crisis, Indonesia let its currency crawl smoothly downward at 4 to 5
percent per year but kept the rupiah virtually fixed to the dollar on a
day-to-day basis. The a2 coefficients of the Korean won, the Philippine
peso, and the Taiwan dollar are very close to unity with lower, but
still large, t-statistics. For the Thai baht and the Malaysian ringgit, the
a2-coefficients are still close to 0.9, with some small weight for the yen as
measured by a3.
Singapore shows the lowest weight for the dollar (82 percent) and smaller
(but highly statistically significant) weights for the yen (14 percent) and the

283
German mark (8 percent). There is some evidence of small weights for the
yen in the pre-crisis East Asian currency baskets of Korea (6 percent),
Malaysia (9 percent), Singapore (14 percent), Taiwan (3 percent), and
Thailand (8 percent). However, except for Singapore there is no evidence
of exchange rate stabilization against the German mark. All in all, before
the Asian crisis East Asia can be characterized as adhering to an informal
dollar standard (McKinnon 2005).
Changing Currency Structures
As outlined in section 10.3, after the Asian crisis there were policy recommendations to increase the weights of the Japanese yen in the East Asian
currency baskets to minimize macroeconomic turbulence caused by bilateral exchange rate fluctuations against the Japanese yen. Since 2001 there
has also been a rationale for putting the euro into the currency basket, not
only because of trade linkages but also because of the high degree of monetary stability of the euro and the wish to diversify international reserves.
The year 2005 would have been an optimal time for a change of currency
basket structures favoring more euros over dollars as the dollar appreciated against the euro because of rising US interest rates.
Using rolling regressions, the country panels in Figure 10.5 summarize
the dollars weight in each East Asian currency basket since the early
1990s. Based on daily data, the rolling 130-day a2 coefficients are plotted
for each of the East Asian countries (except Japan). A window of 130
days corresponds to an observation period of six months (5 observations
per week). The first window starts on January 1, 1990 and ends on June
29, 1990. The a2 coefficients are calculated for the first period. Then the
window is shifted by one day and the coefficients are calculated again, up
to December 2005. A value of unity stands for a 100 percent weight of the
respective currency in the respective currency basket. If the coefficient rises
above 1, the estimation process is unstable.
Figure 10.5 shows the time path of the dollar weights in the East Asian
currency baskets. China and Hong Kong have a very stable dollar weight
of unity for the whole observation period. Officially shifting toward a
currency basket regime in July 2005, China seems to have decreased the
weight of the dollar in its basket slightly since then. In Malaysia, which
has allowed for more exchange rate flexibility since July 2005, the weight
of the dollar remains close to unity. For the other countries in the pre-crisis
period, the dollar weights are slightly more volatile but close to unity.
During the Asian crisis, the exchange rate stabilization against the dollar
broke down in Indonesia, Korea, Malaysia, the Philippines, and Thailand,
as is reflected in the sharp declines of the a2 coefficients.

284

Thai baht
Dollars weight in East Asian currency baskets: 130-trading-day rolling regressions for a2, 1990:01
2005:12 (daily)
Bloomberg; 1 corresponds to 100%.
Figure 10.5
Source:
Philippine peso
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Indonesian rupiah
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Note: An a2-coefficient close to unity shows 100% weight for the dollar in the currency basket.
Taiwan dollar
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
Singapore dollar
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Malaysian ringgit
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Hong Kong dollar
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Korean won
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Chinese yuan
01.01.90 01.01.93 01.01.96 01.01.99 01.01.02 01.01.05
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0.4
0.3
0.3
0.3
0.2
0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.0
Taiwan dollar
.2
0
05
03
05
20
0.
20
4.
.0
03
20
1.
.0
03
.1
0.
20
05
20
03
.0
7.
4.
.0
03
03
20
05
20
1.
.0
03
.1
0.
20
05
20
7.
.0
03
20
03
03
.0
4.
20
1.
.0
03
Singapore dollar
.1
0.0
05
0.0
20
0.1
0.0
03
0.1
7.
0.2
0.1
05
0.2
05
0.3
0.2
05
0.3
05
0.4
0.3
05
0.5
0.4
05
0.5
0.4
05
Philippine peso
0.5
Source:
.2
0
.0
4
03
03
03
.0
1
.2
0
05
.1
0
.2
0
.2
0
05
.0
7
03
03
03
.0
4
.2
0
.2
0
.0
1
.2
0
.1
0
.2
0
.0
7
.2
0
03
03
.0
4
.2
0
.0
1
03
03
Malaysian ringgit
Korean won
.1
0
0.0
.2
0
0.0
03
0.1
0.0
05
0.1
.0
7
0.2
0.1
05
0.2
05
0.3
0.2
05
0.3
05
0.4
0.3
05
0.5
0.4
05
0.5
0.4
05
0.5
05
Indonesian rupiah
Hong Kong dollar
.0
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
03
.1
0.
20
05
Chinese yuan
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
03
.1
0.
20
05
0.5
0.4
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
03
.1
0.
20
05
0.5
0.4
03
0.5
285
Thai baht
Bloomberg. 1 corresponds to 100%.
Figure 10.6
Euros weight in East Asian currency baskets: 130-tradingday rolling regressions for a4, 2005:012005:12 (daily)
Post-crisis, Figure 10.5 suggests a declining trend for the weight of

the dollar in the currency baskets of Indonesia, Korea, the Philippines,
Singapore, Taiwan, and Thailand. Based on the Frankel and Wei (1994)
estimation, by 2005 the suggested weights of the dollar range from 45
percent in Korea up to 83 percent in the Philippines, while the weights of
the dollar remain high for China, Hong Kong, and Malaysia. This may
suggest an ongoing trend away from the dollar to the yen and the euro.
As the appreciation of the dollar in 2005 provided a particularly good
occasion to shift the structure of the currency baskets away from the dollar
toward the yen and the euro, the rolling weights of both currencies in the
East Asian currency baskets are estimated for this time period. Figure
10.6 shows the weights of the euro in 2005. While the German mark did
not have a significant weight in the East Asian currency baskets (except

0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.0
Indonesian rupiah
0.3
0.2
0.2
0.2
0.1
0.1
0.1
0.0
0.0
0.0
03
.0
1.
20
05
03
.0
4.
20
05
0.4
0.3
Source:
Taiwan dollar
05
20
20
4.
.0
03
20
1.
.0
03
20
0.
03
.1
20
.0
7.
20
4.
.0
03
03
20
1.
.0
03
Singapore dollar
0.
0.0
.1
0.0
03
0.1
0.0
05
0.1
03
/0
7/
05
03
/1
0/
05
0.2
0.1
03
/0
4/
05
0.2
03
/0
1/
05
0.3
0.2
05
0.4
0.3
05
0.4
0.3
05
0.4
05
0.5
05
Philippine peso
0.5
05
Malaysian ringgit
Korean won
0.5
20
03
.1
0.
20
05
0.4
0.3
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
0.4
03
.1
0.
20
05
0.5
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
0.5
03
.0
7.
20
05
03
.1
0.
20
05
Hong Kong dollar
0.5
7.
Chinese yuan
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
0.3
0.2
03
.1
0.
20
05
0.3
03
.0
1.
20
05
03
.0
4.
20
05
03
.0
7.
20
05
0.4
0.3
03
.0
7.
20
05
03
.1
0.
20
05
0.5
0.4
03
.0
1.
20
05
03
.0
4.
20
05
0.5
0.4
.0
0.5
03
.1
0.
20
05
03
286
Thai baht
Bloomberg. 1 corresponds to 100%.
Figure 10.7
Yens weight in East Asian currency baskets, 130-trading-day

rolling regressions for a3, 2005:012005:12 (daily)
for Singapore) before the Asian crisis, in 2005 the a4 coefficients increased
significantly for all countries except for China, Hong Kong, and Malaysia.
Although the coefficients are rather volatilewhich may be an indicator
for unstable estimationsthis may provide first evidence for a growing
role of the euro in East Asian exchange rate policies.
A similar trend is found for the Japanese yen (a3 coefficient). Except for
China, Hong Kong, and Malaysia, there are already significant weights for
the yen at the beginning of 2005; they seem to further increase during 2005
for most countries. The weights are highest for Korea and Taiwan, which
are important trading partners of Japan and which compete with Japan in
third markets because of a similar export structure. We also observe discretionary jumps of the weights around July 21, 2005, when China announced
its currency basket (although the Chinese exchange rate policy remains
dominated by the US dollar).7 (See Figure 10.7.)

287
100
80
Percent
60
July 21, 2005
40
20
euro
yen
dollar
0
03.01.2005
Source:
03.03.2005
03.05.2005
03.07.2005
03.09.2005
03.11.2005
Bloomberg. Arithmetic averages of the a-coefficients.
Figure 10.8
The changing East Asian currency structure in 2005
Finally, to obtain a comprehensive picture of the structure of currency

baskets in East Asia, we calculate arithmetic averages of the weights of
the dollar, the yen, and the euro in the currency baskets of the nine East
Asian countries (except Japan) during 2005. The result, shown in Figure
10.8, implies a rising weight for both the euro and the yen. The yen seems
to have received a higher weight on average than the euro. The official
announcement of the Chinese currency basket on July 21, 2005 seems to
have triggered an adjustment of the basket structures in other Asian countries, although the weights of the yen and the euro in the Chinese currency
basket still appear to be very small.
10.5
CONCLUSION
Before the 1997/98 East Asian crisis, the East Asian countries (with the
exception of Japan) pegged their currencies tightly to the dollar, forming
an informal dollar standard. As the motivations for pegging to the dollar
that is, macroeconomic stabilization, dollar denomination of international
and intra-regional trade, and capital flowsremained unchanged after the
crisis, the East Asian countries (except Japan) have maintained or returned
to their dollar pegs. However, the sustained depreciation pressure on the
US dollar, which can be linked to the rising US twin deficits, has led to

288
rising reserve accumulation in US dollars, which constitutes a risk for

macroeconomic stability. While the successive US interest rate increases
since 2004 may put a hold on this trend, expectations of a further decline of
the dollar may create an incentive to further diversify the risk of one-sided
dollar pegs based on basket strategies.
In East Asia, the yen and the euro may enter the currency baskets for
(partially) different reasons. As Japan is a more important trading partner
than the euro area and an important competitor in third markets, the yen
may enter the currency basket to equilibrate intra-Asian competitiveness.
Macroeconomic stabilization will matter less in the case of Japan because
of the still ongoing zero interest rate policy and deflation. In contrast, both
trade and competition in third markets with the euro area are less important, but the role of the euro as a macroeconomic anchor and international
store of value is growing. In the long term this may further enhance the role
of the euro in East Asia, both for exchange rate stabilization and as a store
of value for international reserves.
NOTES
1. For instance, Volz (Chapter 8, this volume) finds a strong trade-creating effect of similar
currency regimes in East Asia.
2. Schnabl (2006) shows how the Central Bank of Russia has promoted the role of the euro
as anchor, intervention, and reserve currency during 2005.
3. In contrast, for small countries that have one major trading partner, it makes sense to peg
to one currency instead of a basket.
4. In July 2005 China announced a currency basket reported to contain the dollar, the
yen and the euro, together with a substantial number of smaller currencies, such as the
Korean won, the Thai baht, the Malaysian ringgit, the Russian rouble, and the Canadian
dollar, among others.
5. Before January 1, 1999 the euro is represented by the German mark as the most important
currency of the European Monetary System.
6. Previous tests did not yield any evidence for any co-integrating vector between the four
exchange rates.
7. To control for distortions caused by the discretionary jump in the yuan/dollar exchange
rate on July 21 2005, this observation is removed from the sample.
REFERENCES
Chinn, Menzie, and Jeffrey Frankel (2005). Will the Euro Eventually Surpass the
Dollar as Leading International Reserve Currency? NBER Working Paper no.
11510, National Bureau of Economic Research, Cambridge, MA.
Chmelarova, Viera, and Gunther Schnabl (2006). Exchange Rate Management
in Developed and Underdeveloped Capital Markets. ECB Working Paper no.
636, European Central Bank, Frankfurt.

289
Crockett, Andrew, and Morris Goldstein (1976). Inflation under Fixed and
Flexible Exchange Rates. IMF Staff Papers 23: 50944.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2004). The Revived
Cambridge, MA.
ECB (2005). Review of the International Role of the Euro. European Central Bank:
Frankfurt.
Financial Fragility. NBER Working Paper no. 7418, National Bureau of
Economic Research, Cambridge, MA.
Fischer, Stanley (2001). Exchange Rate Regimes: Is the Bipolar View Correct?
Journal of Economic Perspectives 15: 324.
Frankel, Jeffrey, and Andrew Rose (2002). An Estimate of the Effect of Common
Currencies on Trade and Income. Quarterly Journal of Economics 117: 43766.
Rate Policies in East Asian Economies. In Takatoshi Ito and Anne Krueger
Gudmundsson, Mr (2005). The Role of the Effective Exchange Rate in Monetary
Frameworks: The International Experience. Mimeograph.
Hillebrand, Eric, and Gunther Schnabl (2006). A Structural Break in the Effects
of Foreign Exchange Intervention. Mimeograph.
Kawai, Masahiro (2002). Exchange Rate Arrangements in East Asia: Lessons from
the 1997/98 Currency Crisis. Monetary and Economic Studies 20: 167204.
McKinnon, Ronald (1963). Optimum Currency Areas. American Economic
Review 53: 71725.
McKinnon, Ronald (2005). Exchange Rates under the East Asian Dollar Standard:
Living with Conflicted Virtue. Cambridge, MA: MIT Press.
McKinnon, Ronald, and Gunther Schnabl (2003). Synchronized Business Cycles
in East Asia and Fluctuations in the Yen/Dollar Exchange Rate. The World
Economy 26: 106788.
McKinnon, Ronald, and Gunther Schnabl (2004a). The East Asian Dollar
Standard, Fear of Floating, and Original Sin. Review of Development Economics
8: 33160.
McKinnon, Ronald, and Gunther Schnabl (2004b). A Return to Exchange Rate
Stability in East Asia? Mitigating Conflicted Virtue. International Finance 7:
169201.
16: 31734.
Schnabl, Gunther (2006). The Russian Currency Basket. The Rising Role of the
Euro for Russias Exchange Rate Policies. Intereconomics 41: 13541.
Williamson, John (2000). Exchange Rate Regimes for Emerging Markets:
Reviving the Intermediate Option. Washington, DC: Institute for International
Economics.
Williamson, John (2005). A Currency Basket for East Asia, Not Just for China.
Policy Briefs in International Economics no. PB05-1, Institute for International
Economics, Washington, DC.

11.

John Williamson
11.1
INTRODUCTION
Two different types of currency basket have figured in recent discussions

on East Asian monetary arrangements. One idea, now being actively
promoted by the Asian Development Bank, involves a basket consisting
of East Asian currencies. Such a basket could be a fore-runner of an East
Asian currency, just as the basket ECU consisting of European currencies
was a fore-runner of the euro. It is also in principle possible that a basket
of this type could be used as a mechanism for organizing a joint float of
the East Asian currencies, just as in pre-euro days the EMU countries
could have held their currencies together with interventions motivated by
a currencys deviation from the basket ECU. However, it is a historical fact
that this is not how the European currencies maintained the margins of the
EMS. In practice, they defended bilateral margins against other members
of the EMS, without paying attention to the ECU basket.
A separate type of basket would be one to which all the members of a
regional bloc decided to peg their currencies. (In practice some members
of the bloc may float their currencies and merely use the basket as a numeraire in terms of which to assess whether their exchange rate is getting
too strong or too weak rather than pegging to it.) Such a basket need not
contain any regional currencies at all; indeed, I will argue subsequently that
there would be advantages to excluding all regional currencies (no matter
how important they are in regional trade). It is the design of this type of
basket that I propose to discuss in this chapter. This is not because I wish
to dismiss the importance or potential usefulness of a basket intended to
lead on to a regional currency, but simply because of a belief in the division
of labor. There are two distinct sets of questions raised by the two different
proposals, and we need to keep them strictly separate.
The purpose of pegging to a basket of major international currencies
is to stabilize the effective exchange rate against the shocks that would
Peterson Institute for International Economics: All rights reserved.
290
291
be imposed on a countrys macroeconomy by fluctuations in the value of

these currencies. Such fluctuations are completely exogenous to the policies
of most of the countries of East Asia but in practice have proven highly
destabilizing (such as in the run-up to the Asian crisis of 1997, to take the
most notorious example). Pegging to a common basket could also yield the
benefit of intra-regional stability, something that has been claimed as one
of the principal virtues of having all the regional currencies peg to the US
dollar (for example, by Ronald McKinnon 2002).
The first question taken up in this chapter is whether a basket should
reflect just trade or also some (or all) capital movements. The more strategic issues are pursued in subsequent sections. One is which currency areas
should use a basket, assuming that a common basket is indeed chosen. The
other is whether it would be better to use a common basket or a series of
baskets tailored to each countrys individual situation.
11.2
THE WEIGHTING SYSTEM
The argument for using just trade is that this is what determines a countrys effective exchange rate,1 and analysis of the optimal peg concluded
that it is the real effective exchange rate that is the most important influence on a countrys macroeconomic situation (both unemployment and
inflation).2 Differences between nominal and real effective exchange rates
can be accommodated by changes in central rates, so that the contribution
of the basket peg should be maintaining the nominal effective exchange
rate (NEER) constant in the face of shocks to exchange rates between
the basket currencies. Kawai and Takagi (2005) established that stability
of the real effective exchange rate is empirically important in achieving
macroeconomic stability in the East Asian countries.
Against this argument, it is sometimes pointed out that many other
transactions, notably capital account transactions, go through the foreign
exchange market. China has indicated that the weights in the currency
basket it now uses to assess its exchange rate are influenced by the origins
of FDI flows as well as of trade flows. Other economists have argued that
in many countries capital transactions are many times larger than the
value of trade. Personally I see little significance in what are undoubtedly
correct facts. On the contrary, I recall that when Chile moved from a peg
to the dollar to a peg to a basket of currencies it said that one motivation for the change was to introduce more noise into the dollarpeso
relationship so as to discourage short-term capital inflows (which came
overwhelmingly out of dollar holdings). Similarly, I do not see any point
in having the composition of the basket be influenced by the currencies in

292
which invoices are written. A Japanese exporter to an East Asian country

may well invoice in dollars, but if he changes the dollar price he quotes in
the light of changes in the dollaryen relationship, then stabilizing the real
effective exchange rate requires weighting the yen by Japanese trade rather
than by the number of invoices written in yen. Certainly I would want to
interpret trade to include all current account transactions rather than
just visible exports and imports, but beyond that I see no case for taking
other factors into account in determining the composition of the basket.
There are at least two subsidiary questions concerning the weights in
the basket, both of which I raised in Williamson (1982). The first question
is whether weights should be based on imports, exports, or total trade.
I concluded that the arguments favored using total trade weights rather
than giving either exports or imports special and differential treatment.
A second question is whether to use trade weights or elasticity weights.
I judged that the arguments favored using elasticity weights if these are
available, but of course they usually arent; however, trade weights should
be a reasonable proxy in most instances.
I therefore assume that if a currency basket is used to provide a currency peg or numeraire, it should be weighted by trade shares and not
influenced by the direction of capital flows or the currencies in which trade
is denominated.
11.3 THE CURRENCY AREAS THAT MIGHT ADOPT

A COMMON BASKET
The next topic that needs to be addressed is which currency areas might
be candidates for pegging to a common basket (or for using it as a numeraire) if it were created. Table 11.1 shows the direction of trade for the five
longstanding members of ASEAN (Indonesia, Malaysia, the Philippines,
Singapore, Thailand), the three big countries of North East Asia (China,
Japan, and Korea), the two smaller currency areas in the Greater China
region (Hong Kong and Taiwan), and one other large Asian country that
is sometimes now included in such analyses (India).
While the trade patterns of these currency areas are obviously different,
in most cases they are not dramatically different. All the currency areas
except Hong Kong and India show intra-regional trade within 10 percent
of the regional average (excluding Japan and India) of 55.5 percent. All
except Indonesia and India (and, trivially, Japan) trade with each of the
main currency blocs within a range of 10 percent of the regional average
(after excluding intra-regional trade). These considerations suggest at
least seven currency areas to be good candidates for using a common

Table 11.1
Direction of trade of East Asian economies in 2004

(percentage)
Direction United
of trade States
China
293
Exports
Imports
Total
Hong Kong Exports
Imports
Total
Indonesia Exports
Imports
Total
South
Exports
Korea
Imports
Total
Malaysia
Exports
Imports
Total
Philippines Exports
Imports
Total
Singapore Exports
Imports
Total
Taiwana
Exports
Imports
Total
Thailand
Exports
Imports
Total
Weighted
Exports
averageb
Imports
Total
Japan
Exports
Imports
Total
India
Exports
Imports
Total
22.8
7.7
15.2
17.0
5.3
11.1
13.5
5.7
9.6
17.8
12.7
15.3
18.8
14.6
16.7
17.5
16.0
16.7
13.0
12.7
12.9
18.0
13.2
15.6
15.9
7.6
11.8
18.3
9.5
13.9
22.7
14.0
18.4
18.4
7.0
12.7
Japan
EU
12.4
16.1
14.3
5.3
12.1
8.7
21.8
19.3
20.5
8.3
21.6
14.9
10.1
16.1
13.1
15.8
20.6
18.2
6.4
11.7
9.1
8.3
25.6
17.0
13.9
23.6
18.7
11.3
15.9
13.6
NA
NA
NA
3.5
3.5
3.5
18.1
12.4
15.3
14.0
8.0
11.0
14.3
12.1
13.2
13.8
10.8
12.3
12.6
12.1
12.3
15.5
8.8
12.1
14.5
13.5
14.0
11.3
9.9
10.6
14.7
9.9
12.3
14.9
10.6
12.7
15.8
12.7
14.3
22.6
23.1
22.9
Non-US
Western
hemisphere
4.6
4.8
4.7
2.7
2.0
2.3
2.1
2.5
2.3
6.3
3.4
4.8
1.8
1.6
1.7
1.9
1.8
1.8
2.1
1.4
1.7
NA
NA
NA
2.6
2.3
2.5
4.1
3.5
3.8
5.4
4.6
5.0
3.4
5.2
4.3
Rest of
NonJapan
East
Asiac
30.1
39.4
34.8
55.3
67.6
61.5
35.6
43.3
39.4
41.4
28.8
35.1
44.6
47.4
46.0
46.6
42.1
44.3
51.9
45.2
48.5
48.2
30.1
39.2
38.8
34.4
36.6
40.3
43.5
41.9
47.6
44.6
46.1
22.9
24.2
23.5
Rest of
worldd
12.1
19.5
15.8
5.8
4.9
5.3
12.7
17.1
14.9
12.5
22.7
17.6
12.2
8.1
10.2
2.8
10.8
6.8
12.1
15.5
13.8
NA
NA
NA
14.1
22.2
18.2
11.1
16.9
14.0
8.5
24.1
16.3
29.3
37.0
33.1

294
Table 11.1
(continued)
Notes:
a
Data for Taiwan are from the 2004 Statistical Year Book of the Republic of China, tables
122 and 123. Due to incomplete country detail the Non-US Western hemisphere cannot be
calculated; the Rest of the world is left out as a result. Incomplete country detail further
means that for Taiwan imports the EU consists of only Belgium, France, Germany,
Italy, Netherlands, Sweden, Switzerland (included as it has an FTA with the EU), and
Britain. For Taiwan exports the EU consists of Belgium, France, Germany, Ireland, Italy,
Netherlands, Spain, Sweden, and Britain. All data for 2003.
b
For the weighted average Taiwan has been excluded due to incomplete data. Due to inconsistent reporting by countries weights do not add up to exactly 100 percent, but they have
been deflated by the actual sum to yield a total of 100 percent.
c
Non-Japan East Asia is the aggregate in the IMF DOTS database for Developing Asia
(all Asia, excluding only Japan); subtracted where applicable are Afghanistan, Bangladesh,
Bhutan, Sri Lanka, India, Maldives, Nepal, Pakistan, Palau, and Timor. As the aggregated
data do not equal the sum of individual countries, Taiwan is assumed to be included in the
aggregate; in addition it contains where applicable Brunei, Myanmar, Cambodia, China,
Hong Kong, Indonesia, South Korea, Laos, Macau, Malaysia, Philippines, Singapore,
Thailand, Vietnam, and Asia Not Elsewhere Reported (NES).
d
The residual is calculated as 1 minus the other five categories.
Source:
IMF Directions of Trade Statistics May 2005.
basket: China, Malaysia, Philippines, Singapore, South Korea, Taiwan,

and Thailand.3 To these I would certainly want to add Hong Kong, since
adoption of a common basket peg would tend to stabilize relative exchange
rates within East Asia, meaning that the crucial variable is the distribution
of extra-regional trade, and this is fairly typical in the case of Hong Kong.
A more marginal case is Indonesia, whose trade with Japan is more than
10 percent above the regional average, but I shall assume in what follows
that Indonesia would adopt a common basket as well. On the other hand,
the Indian trade pattern is distinctly different from that of the East Asian
countries, so that if India were to adopt a basket of currencies to guide its
intervention policy it seems probable that it should adopt its own tailormade basket rather than a common East Asian one.
One remaining question is what should happen to the new members
of ASEAN, namely, Cambodia, Laos, Myanmar, and Vietnam. I have
not analyzed this issue, which at present is of a secondary order of
importance.
By far the most important issue is whether Japan should adopt a
common basket as the numeraire to guide its intervention policy (assuming
that it adopts a basket for that purpose at all). The reason for posing this
as an issue is not that the Japanese trade pattern is dramatically different
from that of other East Asian countries: once one has allowed for the fact
that Japan does not trade with itself, its trade pattern is in fact quite similar

295
to that of its neighbors. The reason is rather the fundamental fact that a
country cannot peg its currency to itself. If Japan were to peg to a common
basket (or use a common basket as numeraire), then the common basket
could not include the yen. All the other regional currencies would similarly
be excluded from the common basket, which would consist exclusively of
extra-regional currencies. If one limits the basket to currency areas with
which the region does at least 5 percent of its total trade (on the ground that
a string of small currencies complicates the system without much affecting
its overall behavior), then a common basket would consist exclusively of
dollar and euro. If the yen remained as volatile in terms of those currencies as it has been in the past, this would tend to destabilize the effective
exchange rates of all of the other countries that peg to the common basket
and thus reduce the value of the system to them.
How important would this additional volatility be? That is a question
that one can attempt to answer on the basis of a simulation using data
from past exchange-rate movements. Such a comparison is made in the
next section of this chapter. In parallel, an empirical examination is made
of whether a common basket peg would have been advantageous.
11.4
INDIVIDUAL-COUNTRY BASKETS VERSUS

COMMON BASKETS
The purpose of this section is to offer a comparison of the behavior of

the nominal effective exchange rate of the nine currency areas identified
in the last section (China, Hong Kong, Indonesia, Malaysia, Philippines,
Singapore, South Korea, Taiwan, and Thailand) under four different
hypothetical exchange-rate policies:
1.
2.
The first one considered is the actual policy that was followed, which
resulted in the variations in the nominal effective exchange rates in
most cases reported by the IMF in International Financial Statistics.
Table 11.3 presents, where available, the standard deviation of each
currency areas monthly NEER over the period 200004 as reported
by IFS. Data for Indonesia, South Korea, Taiwan, and Thailand were
not available from that source, but a substitute source or concept as
reported in note a to Table 11.3 was used.
The second policy supposition is that each currency area pegged its currency to a tailor-made basket representing its own individual trade composition. Each basket was composed of those currencies with which the
currency area conducted at least 5 percent of its total trade. The weight
of the dollar was supplemented by trade with the rest of the Western

296
3.
4.
hemisphere, the rest of non-Japan East Asia, and two-thirds of the rest
of the world, to reflect the fact that the former two regions and a large
number of rest of the world countries have traditionally pegged to, or
measured their exchange rates in terms of, the US dollar. Similarly, the
weight of the euro was supplemented by one-third of the trade with the
rest of the world, reflecting the fact that a number of other currencies
peg to the euro or else that their exchange rates tend to be influenced
by the euro. In the cases of Japan and other East Asian currencies, the
weights in the basket were simply set equal to the percentage of trade
with the respective currency areas. The resulting baskets are shown in
column 2 of Table 11.2. Table 11.3 then presents estimates of what the
standard deviations of variations in the NEERs would have been had
the currencies in question been pegged to those baskets.
The third supposition is that each currency area pegged its currency to a
common basket consisting of the three major international currencies,
namely, the dollar, the euro, and the yen. The dollars weight is again
supplemented by the trade of the nine-currency group with the rest of
the Western hemisphere and two-thirds of that with the rest of the world
(but not, of course, with the rest of non-Japan East Asia, since those
countries are assumed to be pegging to the basket). Similarly, the euros
weight is supplemented by one-third of trade with the rest of the world.
The weight of the yen in this basket is given by the percentage of trade
conducted in 1999 by the whole of the group of nine currency areas with
Japan. These weights add up to less than 100 percent, since they exclude
intra-East Asian trade. The weights are therefore blown up to make the
weights sum to 100 percent. The composition of the resulting basket is
shown in column 3 of Table 11.2. Estimates of the standard deviations
of variations in the monthly NEERs had the currencies been pegged to
such a basket are presented in Table 11.3.
The final supposition is that each currency area pegged its currency to
a common basket consisting of the dollar and the euro only. The relative weights of the dollar and the euro are exactly as in the previous
basket; they are calculated by blowing up the weights of column 3 for
these two currencies so as to make them sum to 100 percent when the
yen is no longer in the basket. The composition of the resulting basket
is shown in column 4 of Table 11.2. Estimates of the standard deviations of variations in the monthly NEERs had the currencies been
pegged to such a basket are presented in Table 11.3.
It can be seen that in the majority of cases (five out of nine) the shift
from countries actual policies to a peg to a tailor-made basket would have
reduced the variability of the effective exchange rate. The rather surprising

Table 11.2
297
Currency weights (percentage)
Country
Currency area/
country
Individual
country
weights
Weights for
3-currency
basket
Weights for
2-currency
basket
China
Euro-zone
Japan
United States
Hong Kong
South Korea
Euro-zone
Japan
United States
China
Euro-zone
Japan
United States
China
Malaysia
Singapore
South Korea
Euro-zone
Japan
United States
China
Hong Kong
Singapore
Euro-zone
Japan
United States
China
Hong Kong
Malaysia
Singapore
Euro-zone
Japan
United States
China
Malaysia
Euro-zone
Japan
United States
China
Euro-zone
Japan
18.6
14.5
47.2
12.2
7.4
10.2
9.1
35.6
45.1
16.4
21.0
35.0
9.0
5.6
8.2
4.9
13.6
13.1
46.6
8.4
4.6
13.7
12.7
18.5
39.0
9.6
6.9
5.0
8.2
16.0
9.3
49.5
9.5
15.8
16.4
15.0
50.6
17.9
15.3
16.7
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
46.6
39.1
30.0
23.4
39.1
Hong Kong
Indonesia
Malaysia
Philippines
Singapore
South Korea
Taiwana
60.9
60.9
60.9
60.9
60.9
60.9
60.9

298
Table 11.2
Country
(continued)
Currency area/
Individual
country
country weights
Thailand
Weights for
3-currency
basket
Weights for
2-currency
basket
United States
China
Hong Kong
Euro-zone
44.5
12.2
11.3
16.0
46.6
60.9
30.0
39.1
Japan
19.2
23.4
United States
44.9
46.6
China
8.2
Malaysia
5.8
Singapore
6.0
60.9
Note: a Taiwan data for 2003 are from the Statistical Yearbook of the Republic of China
2004. The euro zone includes only Belgium, France, Germany, Ireland, Italy, Netherlands,
and Spain. Two-thirds of the total of the Non-US Western hemisphere and rest of the
world is allocated to the US and one-third is allocated to the euro zone.
Source:
IMF DOTS Database May 2005.
fact is that this should not have been true in the other four cases. It is
conceivable that this is an artifact that arises from the measurement of the
effective exchange rate in column 2 purely on the basis of the major trading
partners rather than the more comprehensive base used by the IMF.
However, an economic explanation might also lie behind this paradoxical finding: perhaps the dollar peg (in some cases) or near-peg (in others)
reduced intra-regional exchange rate instability and this effect outweighed
the expected effect of the basket in reducing instability against outside currencies. Indeed, Ronald McKinnon (2002) has argued that an important
virtue of a common dollar peg for the region is exactly that it would in this
way reduce intra-regional exchange rate instability.
Column 3 shows that in every single case a shift to a common basket peg
would have reduced effective exchange rate instability as compared both
to actual historical experience and to a peg to a tailor-made individualcountry basket. This again has to be a reflection of the McKinnon effect:
a common pegand a common basket peg is as good as a common dollar
peg in this respecteliminates intra-regional exchange-rate instability. Of
course it is true that the benefit of a common basket peg is exaggerated in

Table 11.3
Country
China
Hong Kong
Indonesiaa
Malaysia
Philippines
Singapore
South Koreaa
Taiwana,d
Thailanda
299
Standard deviations of East Asian NEERs under different

baskets 200004
Actual
historical
experiencea
Individual
country pegb
$euroyen
basket pegc
$euro basket
pegd
5.21
4.18
6.35
5.29
9.55
2.54
3.32
4.74
2.92
3.49
1.73
5.61
2.77
12.68
1.62
4.42
4.92
3.87
1.58
1.56
3.85
1.58
5.41
1.25
2.01
2.51
1.90
2.11
2.08
5.51
2.11
7.65
1.51
2.17
3.34
2.54
Notes:
a
Standard deviations of period-end monthly nominal effective exchange rate are from IMF
IFS May 2005. Data for Indonesia and Taiwan are end-month from Thomson Datastream
Series JPMIDNB (Indonesia) and NTDTWER Taiwan. Data for South Korea and Thailand
end-month real effective exchange rates are from the Citibank CTERI Database. Jan 2000
5 100.
b
Individual country pegs are calculated as the standard deviation of the weighted average
(weights from Table 11.2, column 3) of individual country trade weights times the bilateral
trading partner currency/LCU. All bilateral exchange rates are set at Jan 2000 5 100.
c
The common basket peg is calculated using 1999 as the base year (USD 1 5 EUR 0.9363
and USD 1 5 JPY 113.91), with the weights in Table 11.2, column 4, yielding the USD, EUR,
JPY composition of USD 0.402, EUR 0.296, and JPY 32.1, respectively. From these the
LCU/USD, EUR, JPY rates implied by the basket composition are calculated, and these are
subsequently weighted by the implied (deflated to equal 100 percent) individual country total
trade weights for the US, the EU, and Japan to yield the variability under a common basket
peg. All bilateral exchange rates are set at Jan 2000 5 100.
d
The common basket peg is calculated using 1999 as the base year (USD 1 5 EUR 0.9363), with
the weights in Table 11.2, column 5, yielding the USD, EUR compositions of USD 0.560 and
EUR 0.412, respectively. From these the LCU/USD, EUR rates implied by the basket composition are calculated, and these are subsequently weighted by the implied (deflated to equal 100
percent) individual country total trade weights for the US and the EU to yield the variability
under a common basket peg. All bilateral exchange rates are set at Jan 2000 5 100.
NA 5 not applicable.
Source:
Authors calculations
this simulation, which assumes that intra-regional instability is reduced to

zero, something that would be true only if all the regional currencies were
to peg rigidly to the common basket. To the extent that countries operate
with bands, or even more if they float and merely use the common basket
as numeraire, there will still be some intra-regional instability even with
a common basket. But I see no reason to expect that a change in third

300
currency exchange rates will have a systematic effect in altering the position
of a currency relative to its central rate, so this seems as neutral an assumption as it is possible to make.
Column 4 goes on to ask what would have been the effect of substituting
a common dollareuro basket for the dollareuroyen basket of column
3. It is assumed that the yen would have continued to float in the same
way that it actually did, whereas it might conceivably be that if Japan had
adopted the common basket as numeraire this would have influenced the
yens value and made it less volatile. Neglecting that possibility, it can be
seen that omitting the yen from the basket to which the other East Asian
countries peg would in every case have led to increased instability. The
mean and median increases are around a third.
We already noted that Indonesia has a rather different trade pattern than
most of the other East Asian countries, so that the greater variability under
a common basket peg is not surprising. The other outlier is the Philippines.
Since Filipino trade is not so different from the average, this may seem
surprising. However, it is also notable that the Filipino effective exchange
rate was actually very unstable on the chosen measure, and this would have
been even more true with a peg to a tailor-made basket. The explanation
is that the Philippine peso was the only currency in East Asia with a trend
depreciation over the period in question. The measure of the variability of
the effective exchange rate used was the standard deviation of:
a wi (ci 2 i)
where wi is the (trade) weight of the ith currency area, ci is the exchange
rate between the local currency and the ith currency, and i is the average
exchange rate between the local currency and the ith currency over the
period 200004. Where there is a trend change in the value of the local
currency in terms of all other currencies, this formula tends to give large
measures of the effective exchange rate deviation at the beginning and
end of the period. One may in principle eliminate these by deflating all
exchange rates by a measure of the trend depreciation of the exchange
rate.
The first measure used was the IMFs NEER trend over the period
200004. The result of this deflation is shown in the second row of Table
11.4 (whose first row reproduces the result from Table 11.3). It can be
seen that this deflation does indeed reduce the standard deviation of the
NEER to a level in line with that in other countries, but that variability
would still have increased somewhat with a shift to a basket peg. It would
have increased further with a shift to a common basket peg, and a part of
this increase would have been reversed if the yen were excluded from the

Table 11.4
301
Standard deviation of the Philippine NEER, 200004
Basis
Actual
historical
experiencea
Reported in Table 11.3

Deflating by IMF
NEER
trend for 200004
Deflating by basket
trend
for period 200004
Individual $euroyen $euro basket

country pegb basket pegc
pegd
9.55
12.68
5.41
7.65
3.05
3.88
5.97
4.40
NA
NA
1.20
1.80
Notes:
a
Standard deviations of period-end monthly nominal effective exchange are rate from IMF
IFS May 2005.
b
Individual country pegs are calculated as the standard deviation of the weighted average
(weights from Table 11.2, column 3) of individual country trade weights times the bilateral
trading partner currency/LCU. All bilateral exchange rates are set at Jan 2000 5 100.
c
The common basket peg is calculated using 1999 as the base year (USD 1 5 EUR 0.9363
and USD 1 5 JPY 113.9), with the weights in Table 11.2, column 4, yielding the USD, EUR,
JPY composition of USD 0.402, EUR 0.296, and JPY 32.1, respectively. From these the
LCU/USD, EUR, JPY rates implied by the basket composition are calculated, and these are
subsequently weighted by the implied (deflated to equal 100 percent) individual country total
trade weights for the US, the EU, and Japan to yield the variability under a common basket
peg. All bilateral exchange rates are set at Jan 2000 5 100.
d
The common basket peg is calculated using 1999 as the base year (USD 1 5 EUR 0.9363),
with the weights in Table 11.2, column 5, yielding the USD, EUR composition of USD 0.560
and EUR 0.412, respectively. From these the LCU/USD, EUR rates implied by the basket
composition are calculated, and these are subsequently weighted by the implied (deflated to
equal 100 percent) individual country total trade weights for the US and the EU to yield the
variability under a common basket peg. All bilateral exchange rates are set at Jan 2000 5
100. NA 5 not applicable.
Source:
Authors calculations.
basket. Both of these results are contrary to what might be expected as well
as to the results of all of the other East Asian currencies. The explanation
of these paradoxical findings appears to be that there was a significant
divergence between the IMFs measure of the NEER and the value of the
Philippine peso in terms of the basket currencies in the later years of the
period, so that deflating the latter by the former gave rise to the same sort
of problem as that encountered without deflation.
An alternative is to deflate the exchange rates against the basket currencies by the trend depreciation of the Philippine peso in terms of the
average basket currency. Some of the results of this exercise are reported
in the third row of Table 11.4, although it hardly makes sense to deflate by

302
the basket currencies for the non-basket hypotheses. The measured variability of the basket solutions does indeed decline when that is done, but
one worries that this is approaching a tautology.
11.5
CONCLUDING REMARKS
The proposal for a number of East Asian countries to peg their currencies to a common basket comes out relatively well in this comparison.
Obviously the proposal can do nothing to lay to rest the concerns that
any system of pegging is vulnerable to provoking speculative runs: one has
to decide whether that is a risk worth taking in order to gain the benefits
of avoiding the instability and the danger that the market may generate
misalignments, inherent in floating. What the results suggest is that if a
country decides to peg its currency, then pegging to a basket can result
in significantly less instability than actual policy. Moreover, pegging to
a common basket to which other countries in the region also peg can
further reduce instability. If there is a desire to realize those benefits but
also to avoid the risk of pegging, then an appropriate policy may be to
adopt a managed float guided by the use of a common basket as numeraire. Unfortunately it seems impossible to make any convincing estimate
of the impact of this policy in the absence of a theory of floating rates,
another of how intervention is determined, and a third of how the rate
responds to interventionnone of which we presently have or are likely
to have any time in the foreseeable future. Hence the suggestion of what
an appropriate policy might be must remain speculative. But that is no
excuse for ignoring the results on the advantage of a common basket peg,
which are quite strong and, I would conjecture, robust. They point to a
need for collective action, such as the region seems increasingly interested
in developing.
NOTES
1. A countrys effective exchange rate is defined as its average (trade-weighted) exchange
rate.
2. See Williamson (1982) for a survey of this literature and the basis for this statement.
3. I acknowledge that there might be political problems in having both Mainland China and
Taiwan use the same basket, although I would note one advantage of any cooperation
taking this form: it could in practice be economically effective without imposing the need
for direct political dealings.

303
REFERENCES
Kawai, Masahiro, and Shinji Takagi (2005). Strategy for a Regional Exchange
Rate Arrangement in East Asia: Analysis, Review, and Proposal. Global
Economic Review 34: 2164.
McKinnon, Ronald (2002). After the Crisis, The East Asian Dollar Standard
Resurrected. In A.H.H. Tan (ed.), Monetary and Financial Management in the
21st Century. Singapore: World Scientific Publishing Co.
Williamson, John (1982). A Survey of the Literature on the Optimal Peg. Journal
of Development Economics 11: 3961.

12.

Masahiro Kawai1
12.1
INTRODUCTION
ASEAN13 finance ministers have been pursuing regional financial cooperation since the Asian financial crisis of 199798.2 Reflecting lessons
learned from the crisis, the aim has been to strengthen national financial
markets and to establish regional self-help mechanisms for crisis prevention and management. Their efforts have focused on regional economic
surveillance (i.e. the Economic Review and Policy Dialogue, ERPD),
regional short-term liquidity arrangements (i.e. the Chiang Mai Initiative,
CMI), and local-currency bond market development (i.e. the Asian Bond
Markets Initiative, ABMI). Asian central bank governors, participating
in the Executives Meeting of East Asia-Pacific Central Banks (EMEAP),3
have also made efforts to improve their policy dialogue and promote
Asian bond market development, for example through Asian Bond Funds
(ABF).
While there has been substantial progress in all these areas, little progress
has been made in monetary and exchange rate policy coordination. Given
the high and ever-rising degree of economic interdependence through
market-driven trade, investment, and financial flows, East Asian economies have found it increasingly important to maintain relatively stable
intraregional exchange rates while allowing sufficient flexibility against the
US dollar. A first, modest step in this direction could be the creation of
an Asian currency unit (ACU), which is an appropriately weighted index
of East Asian currencies designed to monitor the collective movement of
regional currencies against key external currenciessuch as the US dollar
and the euroand each component currencys movement relative to the
ACU regional benchmark. Such an index can also be used by markets for
varied purposes and, once certain conditions are met in the future, the
ACU could be used by the regions authorities for the purpose of monetary
transactions and exchange rate policy coordination. This possibility was
highlighted in recent statements at the East Asian Finance Ministers meetings in Hyderabad, India, held on the sidelines of the Asian Development
Bank (ADB) Annual Meeting.4
304

305
This chapter attempts to explain the objectives of creating such a regional

currency unit, clarify some technical issues to be resolved in constructing
it, provide the economic logic of the unit, and explore its potential role for
future exchange rate and monetary policy coordination in East Asia.
12.2
OBJECTIVES OF CREATING AN ASIAN

CURRENCY UNIT
The creation of an ACU was first proposed by Kuroda and Kawai (2002).5
An ACU can be useful in three ways:
as a statistical indicator summarizing the collective movement of

Asian currencies;
as a currency basket used by the market; and
as an official unit of account for monetary and exchange rate policy
coordination.
First, the immediate objective of creating an ACU is to use it as one of the

tools for monitoring foreign exchange market conditions. It represents the
collective movement of regional currencies against key external currencies,
such as the US dollar, the euro, and the British pound. Also, as a regional
benchmark, an ACU can help gauge the degree of divergence of each
component currency. A detailed analysis of divergence indicators offers
valuable information for identifying idiosyncratic problems in a particular currencys market and in helping pursue appropriate macroeconomic
policies to eliminate vulnerabilities that adversely affect the exchange rate.6
In this sense, the ACU is nothing more than a statistical indicator that is
useful for regional policy dialogue and national policymaking, but without
requiring automatic policy changes or market interventions. Thus, it may
be more appropriate to call the currency unit an ACU index.
Second, the ACU index, if constructed in a market-friendly way, can
be useful for financial markets in developing new, tradable instruments.
Asias futures exchanges may be interested in listing ACU futures as a
financial commodity, which can provide hedging instruments for traders
even in the absence of active onshore derivatives markets for some highly
regulated currencies. Capital markets may develop ACU-denominated
bonds; Asian governments, or any other creditworthy government in the
world, could issue sovereign or quasi-sovereign bonds in ACU; private
corporations of any nationality could issue corporate bonds in ACU; and
pension, provident and mutual funds and other institutional investors
could invest in ACU-denominated bonds.7 Both Asian and non-Asian

306
central banks could hold part of their foreign exchange reserves in ACU.
Even commercial banks could accept ACU deposits and make ACU loans.
Looking beyond the financial markets, exporters and importers may wish
to denominate cross-border trade in ACU.
Third, in the future, and at a more advanced stage of its development,
an ACU could play a significantly important official role, similar to
that played by the European currency unit (ECU) within the European
Monetary System (EMS) during the 1980s and 1990s.8 Following the
European path, East Asian authorities may decide in the years to come
to stabilize their exchange rates against the ACU basketas initially
intended under the Exchange Rate Mechanism (ERM). Official ACUs
may be created by a regional reserve pooling institution similar to the
European Monetary Cooperation Fund, and Asian central banks may be
authorized to use official ACUs to settle balances with other central banks.
Alternatively an ACU could be allowed to function as a regional parallel
currency, as part of a market-driven approach to East Asian monetary
unification (Eichengreen 2006). The ACU as a parallel currency would
compete against other national currencies in the region for use as a unit of
account, medium of exchange and store of value.
Against the backdrop of the evolution of monetary cooperation in Europe
and the creation of the euro, it is crucial to emphasize the significant difference between the ECU and the ACU index. The ECU was an official unit
of account from the mid-1970s, while the proposed ACU index is merely
a statistical indicatorat least until authorities decide otherwise. Initially,
the ACU index movement, and each component currencys divergence
from it, requires neither an automatic policy response nor any obligation
for exchange market intervention. In this sense, the ACU index is merely
an indicator that offers additional information in helping monitor currency
market developments. Once Asian monetary authorities decide to pursue
monetary and exchange rate policy coordination, however (whether along
the line of the European experience or by using a different approach), the
ACU may gain officialor even legal tenderstatus. But it will take years
or even decades before such a decision is made (if at all) given the various
impediments to the evolution of Asian monetary integration.9
12.3
CONSTRUCTING AN ACU INDEX
Technical Issues
In constructing a weighted index of Asian currencies, or ACU, several
technical issues need to be addressed:10

307
choice of fixed shares vs. fixed units of component currencies in the

basket;
choice of economic variables and indicators used to calculate
currency weights/units;
coverage of currencies included in the basket; and
frequency of periodical revisions of currency shares/units.
First, a decision must be made whether to use fixed currency shares or

fixed currency units in constructing an ACU. Given the ECU experience,
constructing the ACU as fixed units of component currencies is more
desirable. The fixed-units formula, as opposed to the fixed-shares formula,
can produce a smooth change in the ACU index when revising currency
units, adding a new currency to the basket, or eliminating an incumbent
currency from the basket. Under the fixed-units formula, effective weights
of individual currencies in the basket will fluctuate as their exchange rates
change.
Second, a decision has to be made as to what economic variables and
indicators should be used in calculating the appropriate numbers of currency units in a basket. How to choose these economic variables depends
on the ultimate objective for constructing the currency basket. In general,
economic size matters in any currency basket as it represents the relative importance of a country in the global and regional economy. If the
objective is to measure relative international price competitiveness of
Asian economies against the rest of the world or among Asian economies,
then a measure of trade activity will be the most important indicator. If
the objective is to measure the relative importance of individual national
financial markets, particularly for cross-border financial transactions and
capital flows, then financial market-related indicatorssize and liquidity
of financial marketsas well as some measures of overall capital account
convertibility, are important factors to consider.
As the objective is a combination of these two, I suggest three types of
size variables for consideration: (1) the size of the economy; (2) the size
of export and import trade activity; and (3) the size of financial market
activity, particularly for cross-border purposes. Nominal US dollarbased gross domestic product (GDP) converted at market exchange
ratesrather than PPP-based GDPis the best proxy for economic
size of an economy, as it captures the current monetary value of overall
national economic activityrather than measuring a countrys standard
of living. Total trade valueexports plus importsis an appropriate
measure of trade activity, though a decision has to be made on the
choice between total trade with the rest of the world and total trade with
regional partners.11 Total size of national financial markets adjusted for

308
the degrees of exchange and capital controls and/or cross-border capital

account transactionssuch as the value of international debt securities
issued by residentsare a good proxy for cross-border financial market
activity.12 These three economic variables can be used in combination in
constructing an ACU index.
Third, the question of which currencies to include in an ACU index
must be addressed. Currency baskets of various regional groups within
Asiasuch as East Asia, South Asia, Central Asia, and Oceaniamay
be considered separately. Including all Asian currencies in a single basket
would be too complex and diverse to be aggregated. The most promising
group thus far is a subset of East AsiaASEAN13as its members have
actively worked on monetary and financial cooperation since the Asian
financial crisis.13 Following the recent East Asia Summit, launched in Kuala
Lumpur in December 2005, an ACU index including India, Australia, and
New Zealand (in addition to ASEAN13) might also be considered. There
can be various combinations of currency coverage, and new currencies may
be added to any initial currency basket if deemed appropriate. The formula
should thus maintain an open-ended currency composition system.
Fourth, a decision must be made on the frequency of periodical revisions
of assigned currency units in the basket. This decision is important not
only for authorities of component currencies, but especially for markets, as
changes in weighting will affect interest rates calculated on the different currency components of the basket. This will impact the value of the various
financial instruments denominated in ACU. It is therefore important to set
transparent rules for periodic revisions of units to allow financial market
analysts to easily calculate basket interest rates. Given the dynamic pace of
transformation of many East Asian economies, the numbers of individual
currency units could be revised as often as every two to three years.
Existing Currency Baskets
Let us examine currently existing currency basket indexes in light of the
ACU index to be constructed from the perspective of the above considerations. We review currency weights used by the US Federal Reserve
(FRB), the European Central Bank (ECB), and the Bank of England
(BOE) in defining their nominal effective exchange rates (NEERs) as well
as those used by market participants in defining various baskets of bonds.
Table 12.1 presents a summary of several currency basket weights that are
adjusted so the sum of individual ASEAN13 currency weights equals 100
percent. It also includes several other emerging Asian economies (Hong
Kong, Macau, Taipei, China, and India), and Australia and New Zealand
for comparison.

309

China,
Peoples
Rep
Japan
Korea, Rep. of
Brunei
Darussalam
Cambodia
Indonesia
Laos PDR
Malaysia
Myanmar
Philippines
Singapore
Thailand
Vietnam
ASEAN13
Hong Kong
SAR
Macau
SAR
Economies
Table 12.1
3.2
5.0
2.4
7.0
4.5
100.0
8.3
38.7
12.4
28.9
11.7
2.5
6.1
2.4
5.9
4.1
100.0
5.9
26.9
ECB
38.5
FRB
38.8
10.3
27.6
BOE
6.0
11.2
6.0
100.0
14.7
Central Bank NEER
5.0
8.9
3.8
14.8
8.7
100.0
15.8
25.1
33.8
BLBG-JPM
ADXY
0.2
5.1
0.1
5.3
0.4
2.9
6.4
5.1
1.8
100.0
27.8
9.8
0.4
34.8
RIETI
AMU
Currency index
Summary of various currency basket weights
1.9
19.4
13.0
2.8
2.8
100.0
17.0
45.2
15.0
HSBC
ADBI
0.4
7.2
15.3
20.5
13.8
12.1
0.8
100.0
25.4
30.7
9.6
JACI
10.4
9.6
50.1
10.3
100.0
47.6
10.2
9.4
ELMI1
JPMorgan
Bond market index
7.3
13.1
6.1
22.3
12.1
100.0
22.4
25.3
13.8
ABF iBoxx
310

3.3
3.5
7.9
FRB
5.8
3.2
0.5
10.4
ECB
Central Bank NEER
(continued)
11.2
11.2
7.8
BOE
7.2
15.4
BLBG-JPM
ADXY
RIETI
AMU
Currency index
HSBC
ADBI
3.3
1.2
JACI
9.4
9.6
ELMI1
JPMorgan
Bond market index
ABF iBoxx
Note: FRB 5 US Federal Reserve Boards NEER (nominal effective exchange rate); ECB 5 European Central Banks NEER; BOE 5 Bank of
Englands NEER; BLBG-JPM ADXY 5 Bloomberg-JPMorgan Asia Currency Index (currency basket); RIETI AMU 5 Research Institute of
Economy, Trade and Industrys Asian Monetary Unit (currency basket); HSBC ADBI 5 HSBCs Asian USD Bond Index (USD-denominated
fixed-rate straight bonds); JACI 5 JPMorgan Asia Credit Index (USD-denominated bonds); ELMI1 5 JPMorgan Emerging Local Markets
Index Plus (local currency-denominated money market instruments); ABF iBoxx 5 Asian Bond Fund iBoxx Pan-Asia Index (local currencydenominated bonds).
Taipei,
China
India
Australia
New
Zealand
Economies
Table 12.1
311
The central banks construct NEERs using bilateral trade (export and
import) relationships as weights. In the case of the FRB, ASEAN13
accounts for 32.7 percent of the global USD NEER, while in the case of
the ECB and the BOE it accounts for only 22.0 percent and 11.6 percent of
their global NEERs, respectively. In these indexes, Japan and the Peoples
Republic of China (PRC) naturally account for the largest weights at
2739 percent; followed by Republic of Korea (Korea); Hong Kong;
Taipei, China; and Singapore. JPMorgan also calculates the BloombergJPMorgan Asia Currency Index (ADXY) as a measure of aggregate
emerging Asian currency movements (i.e. excluding the Japanese yen),
using extra-East Asian (including India) trade weights. In ADXY, the
pattern of important country weights is basically the same as those for
central bank NEERs. The Asian Monetary Unit (AMU) constructed by
Japans Research Institute of Economy, Trade and Industry uses trade
volume and PPP-based GDP to determine currency weights. As a result,
the weight assigned to China is the largest, followed by Japan, Korea, and
Singapore.14
Several regional bond indexes have been constructed by the private
sector. These include: (1) the Asian USD Bond Index (ADBI) calculated
by HSBC; (2) the Asia Credit Index (JACI) provided by JPMorgan; (3)
Emerging Local Markets Index Plus (ELMI1) calculated by JPMorgan;
and (4) the iBoxx Pan-Asia Index, developed by the International Index
Company Limited for the Asian Bond Fund (ABF) under the EMEAP
initiative.15 In these measures, country weights vary across indexes, but
large weights tend to be assigned to economies like Korea, Hong Kong,
and Singapore, while smaller weights are assigned to countries like China
despite the large economic size. The reason is that in bond market indexes,
factors like market size, liquidity, and openness are taken into account,
which make weights of economies with developed bond markets higher
and those with less developed markets lower.
These existing currency basket indexes suggest that depending on the
objective of creating such indexes the currency weights can vary considerably. For the construction of the ACU in the current East Asian context,
it is crucial to capture the importance of individual participating countries
(and hence currencies) for the regions cross-border activities in the real
and financial sectors. From this perspective, an ACU index should be
constructed by selecting the most appropriate economic variables and
indicatorssuch as GDP at market exchange rates, trade volume, and
financial market size and openness. Once the ACU is constructed, effective
currency weights can easily be altered, a new currency can be added, and
any component currency can be eliminated without causing discontinuous
shifts in the basket value.

312
12.4
ECONOMIC LOGIC OF INTRODUCING AN

ACU
If one of the ultimate objectives of Asian monetary and financial cooperation is the promotion of regional monetary integration and the eventual
creation of a monetary union, why is an Asian currency basket useful? An
ACU is a useful step towards monetary union because it facilitates creating
stable exchange rates among the regions currencies.
There are two ways to establish a stable intraregional exchange rate relationship. One is for each economy to peg the exchange rate to a common
key currency. The other is for regional economies to adopt collective
policy measures based on certain rulessuch as the Snake and the ERM
adopted by Europe in the 1970s and 1980s. As economic, particularly
structural, convergence is not sufficiently developed among East Asian
economies and their political relationships are not mature enough for a
tightly coordinated exchange rate arrangement, it is more realistic to select
a major currencylike the US dollar, the euro, the yen, or the yuanor
a basket of global or regional currencies as exchange rate anchor for each
countrys exchange rate stabilization. Here we argue that there is a strong
case for selecting a basket of Asian currencies, rather than a single currency,
as an anchor for exchange rate stabilization in East Asia.
Dollar, Yen, or Yuan for East Asia?
Until the Asian currency crisis of 1997, the general practice had been
to peg currencies to the US dollar and, by so doing, indirectly stabilize
intraregional values of East Asian currencies. But this dollar standard
system revealed its flaws during the crisis. For example, for countries with
strong economic ties to Japanthose beginning to satisfy optimum currency area (OCA) conditionspegging to the US dollar meant volatile yen
exchange rates because of large yendollar rate fluctuations. This created
large financial risks in managing their economies. In addition, many East
Asian economies do not satisfy OCA conditions with the US because of
structural differences and dissimilar supply shocks affecting the East Asian
and US economies.16
Pegging to the yen is therefore more reasonable, particularly given
Japans economic weight in East Asia and its multinationals networking capacity throughout the region. Continuing economic recovery from
Japans decade-long economic stagnation may also make the yen a strong
candidate to become an East Asian anchor. But the problem is that Tokyo
has not grown into a world-class financial center comparable to New
York or London, and the yen has not achieved full internationalization to

313
assume the sole anchor role in the region. The US dollar remains prominent enough in the regions finances to warrant shared status with the yen
as currency anchor.
Current economic trends suggest that China will surpass Japan in
economic size by 2020 and the EU and the US by 2050, barring growth
disruptions arising from (1) possible economic and financial crises, (2)
resource, energy, and environmental constraints, or (3) political and social
instability. The issue, then, is whether the yuan will become a currency that
matches or even exceeds the US dollar, euro, or yen. Thus, there is strong
potential for the yuan to become a key anchor currency in the region in
the long run because of Chinas rapid pace of economic growth and the
possibility for it having greater macroeconomic influences on other East
Asian economies. However, the yuan is unlikely to dominate at least for
several decades. First, for the yuan to become an international currency,
China must transform into a fully open economy with regard to trade,
investment, and finance. The country must allow international use and
holdings of its currency by liberalizing cross-border capital flows and
exchange controls. Full liberalization of the capital account requires complete transition to a market economy and the establishment of a sound and
resilient financial sectorwhich could take another 1020 years, perhaps
longer. Second, even though, in a best case scenario, Chinas per capita
income may rise to one-third to one-fifth of that in Japan, Europe, or the
US by 2050, it is doubtful whether China can sufficiently reduce poverty,
build income equality, or ensure smooth political transitionsnecessary
caveats for credible international currencies. Third, even if the yuanor
for that matter the yenbecomes an international currency, the question
remains whether it will grow into a leading international currency that can
match the US dollar. This would depend on how the US economy performs
in the future. Unless serious difficulties confront the American economy
breaking the US dollars law of inertia as the incumbent international
currencyit is difficult to imagine the yuan challenging the status quo.17
A Case for a Currency Basket
In principle, a relatively small countrys currencythe Korean won, Thai
baht, or any other East Asian currencycould also be chosen as an anchor
currency, but this would be unlikely because of the economys small size.
Thus, no single currencybe it the dollar, euro, yen, or yuanis a good
candidate as sole anchor currency for East Asia in the future. Rather, it
makes sense for regional and non-regional currencies together to fulfill
the requirements for an anchor. A currency basket composed of the US
dollar, euro, yen, yuan and, possibly, other regional currencies would

314
Table 12.2
Choice of an anchor currency in a game between Japan and

China
China
Japan
Note:
US dollar
yen
yuan
Thai baht
US dollar
yen
yuan
Thai baht
2, 2
0, 0
0, 0
0, 0
0, 0
3, 2
0, 0
0, 0
0, 0
0, 0
2, 3
0, 0
0, 0
0, 0
0, 0
1, 1
Payoffs are indicated in each cell in the order of Japan and China.
be desirable for East Asia. Although the yuan is not yet an international
currency, excluding it from the basket is inappropriate given Chinas
growing economic stature in East Asiaparticularly from a trade or competitiveness perspective. If East Asian economies stabilized their currencies
against similarly defined baskets, they could achieve both relatively stable
effective exchange rates and relatively stable intraregional exchange rates.
Essentially the currency basket arrangement can create a stable exchange
rate environment for the entire region.
This point can be illustrated by a theoretical game played by Japan
and China over the choice of an anchor currency (Table 12.2).18 In this
game, if Japan and China propose different currencies as regional anchor
for exchange rate stabilizationsay Japan chooses the yen and China
the yuanthere is no regionally consistent exchange rate arrangement,
so each countrys pay-off is zero. If they choose the same currency as the
regions common anchor, they both benefit from exchange rate stabilization, but the country whose currency is chosen may gain more than the
other (though this assumption is not crucial). If both countries choose the
US dollar, the pay-offs are 2 each; if both countries choose the Thai baht,
the pay-offs are 1 each; if they choose the yen, the pay-offs are 3 for Japan
and 2 for China; and if they choose the yuan, the pay-offs are 2 for Japan
and 3 for China. The pure-strategy Nash equilibrium is the simultaneous
choice of either the US dollar, the baht, the yen, or the yuan. Within the
context of ASEAN13composed of 13 authorities as the games playersat least 13 Nash equilibria will be obtained (or 14 or 15 depending on
whether the US dollar and/or the euro are included in the game). A mixed
strategy of these multiple equilibria may be realized by choosing a basket
of all these currencies, though some technical issues need to be resolved to
find appropriate weights attached to the currencies.

315
The resulting currency basket includes both regional and non-regional

currencies. The collection of regional currencies can be called the ACU.
If non-regional currencies are excluded from the games considerations as
candidates for East Asias anchor currency, then the authorities will choose
a basket comprising only regional currencies, that is, the ACU. Essentially
an ACU is useful because no single Asian currency is dominant as a key
currency in the region.
12.5
POTENTIAL ROLE OF AN ACU FOR

MONETARY AND EXCHANGE RATE POLICY
COORDINATION
Once a constructed ACU index fulfills the initial objectivesserving as a

statistical indicator for the purpose of currency market monitoring, and
as a market index for private (and some public) sector usethe next stage
could be for authorities to use the ACU for monetary and exchange rate
policy coordination. At this stage, authorities may (but do not have to) give
official statussuch as legal tenderto the currency basket.
Currently no consensus exists, even within ASEAN13, on a regional
exchange rate arrangement or on the potential role of an ACU. I offer
here a possible scenario on the development of such an arrangement and
the potential role of an ACU.19 In this scenario, monetary and exchange
rate policy coordination deepens in three stages, starting from loose policy
coordination to tight policy coordination, and then to complete policy
coordination. East Asian economies are already in the first stage of loose
policy coordination, including the strengthening of ongoing financial
cooperation initiativesthe ERPD, CMI, ABMI, and ABF under the
aegis of the ASEAN13 finance ministers and central bank governors.20
Authorities have started to study ways to further strengthen the reserve
pooling arrangement, going beyond the current CMItoward multilateralization together with enhanced regional economic surveillance.21 They
have made significant progress in developing local currency-denominated
bond marketsincluding conducting studies on Asian multi-currency
bond issuance and an Asian settlement system.22
At this stage, it would be best for East Asian economies to adopt policies
that stabilize exchange rates against a common basket of both regional and
external currencies (the US dollar, the euro, and the ACU) to ensure relative stability of their effective exchange rates and intraregional exchange
rates.23 An ACU index can serve as a useful tool in measuring the degree
of joint movement of East Asian currenciesfor example in the context
of the eventual unwinding of global payments imbalances. Once China

316
moves to a more flexible exchange rate regime, ACU index movements

and divergences of component currencies from the basket can provide
more meaningful information. The regional surveillance process can also
focus more intensively on exchange rate issues by using an ACU index and
divergence indicators.
The second stage may take one of two approachesor an appropriate
combination of the two. One is the parallel currency approach advocated by Eichengreen (2006), which allows the ACU to freely circulate as
legal tender alongside component national currencies.24 As the ACU is
increasingly used by the market as a unit of account, store of value, and
medium of exchange in East Asia, the role of national currencies may
diminish, and there will be greater willingness to shift the role of national
currencies over to the ACU, with conditions for monetary unification
naturally created. This option emphasizes the role of market forces, rather
than political commitment, in dictating transition to a monetary union,
which is the goal of the third stage.
Another is the launch of intraregional exchange-rate policy coordination
along the line of European monetary integration. As East Asian economies
become more integrated, achieve greater economic convergence, and
hence are better positioned to commit to meaningful policy coordination,
a common framework for intraregional exchange rate stabilization can
be developed. This can be done by establishing an Asian Snake or an
Asian Exchange Rate Mechanism. One way to do this, after a period
of gathering experience and building confidence, is for all participating
economiesJapan together with emerging East Asiato shift from stabilization of their exchange rates against a common basket of external and
East Asian currencies (comprising the US dollar, euro, and ACU) to a
formal exchange rate stabilization against a common basket of East Asian
currencies alone, that is, the ACU. Such an ACU-based exchange rate
stabilization scheme requires: (1) well-defined rules for currency market
interventions and monetary policy coordination, so as to establish a credible monetary anchor within East Asia; (2) a fully elaborated short-term
liquidity support arrangement capable of allowing central banks frequent
interventions in foreign exchange markets; and (3) fiscal policy rules
designed to lend credibility to the exchange-rate stabilization scheme.25
Whichever approach is taken, the second stage requires both market
acceptance of the ACU and significant policy initiatives. In this sense, an
appropriate combination of the two approaches is realistic and desirable.
This transitional stage requires greater institution buildingincluding a
regional currency and bond settlement system and a full-fledged reservepooling fund to issue or manage official ACU.
The final stage, feasible only in the very long runif everis to adopt

317
a single regional currency, a common monetary policy, and the delegation

of monetary policy-making authority to a supranational central bank. This
requires a significantly higher level of political commitment and a more
heightened sense of shared community among citizens and states of the
East Asian economies than currently exists.
12.6
THE WAY FORWARD
To support the ongoing process of market-driven integration, more systematic, coordinated institution building is clearly needed. While several
ACU indexes can be constructed for different groups of Asian economies,
ASEAN13 is a natural starting point because of its current, increasingly
active financial cooperation efforts. Once introduced and operative, the
ACU can become an important tool in regional economic surveillance,
help develop Asian financial marketsparticularly for local currencydenominated bondsand contribute to further monetary and financial
cooperation. It is clearly important for relevant finance ministers and
central bank governors to coordinate fully on these initiatives.
Nonetheless, its creation does not automatically guarantee the emergence of ACU-denominated bonds or the beginning of close monetary
and exchange rate policy coordination. East Asia has yet to achieve significant economic and structural convergence, resilient and open financial
marketsparticularly bond marketsand adequate regional institutional
arrangements to support policy coordination. Authorities have yet to assess
the pros and cons of enhanced exchange rate policy coordinationthe benefits of stable intraregional exchange rates versus the costs resulting from a
potential loss of sovereignty over national monetary policymaking.
The impetus toward regional exchange rate policy coordination may
come sooner rather than laterwhen the US dollar depreciates sharply
against East Asian (and other major) currencies in the unwinding process
of global payments imbalances. If East Asian economies must accept
currency appreciation against the dollar, they had better do so collectivelythereby ending the so-called Bretton Woods System IIwhile
maintaining intraregional rate stability, so that the costs of adjustment can
be spread among them and, hence, reduced for each of them. This type of
informal policy coordinationwhich requires Chinas shift toward greater
exchange rate flexibility and a faster pace of yuan appreciationhas the
potential to create an East Asian monetary zone, of the type observed in
Europe after the collapse of Bretton Woods System I in the early 1970s. In
this context, an ACU can play a useful role in boosting policy coordination
across East Asia.

318
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
9.
The author is grateful to Koichi Hamada and Ulrich Volz for their comments on an
earlier version of the chapter, to Giovanni Capannelli for providing data, and to Guy
Sacerdoti and Steve Green for their editorial assistance. The findings, interpretations,
and conclusions expressed in the chapter are entirely those of the author alone and do
not necessarily represent the views of the ADB, its executive directors, or the countries
they represent.
ASEAN13 includes the 10 ASEAN member countries (Brunei, Cambodia, Indonesia,
Lao PDR, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Vietnam)
plus China, Japan, and Korea.
EMEAP includes Australia, China, Hong Kong, Indonesia, Japan, Korea, Malaysia,
New Zealand, the Philippines, Singapore, and Thailand.
In Hyderabad the finance ministers of China, Japan, and Korea stated: We noted
the importance of sharing a long-term vision for financial integration in the region;
we agreed on further study of related issues, including the usefulness of regional currency units, through the ASEAN13 Finance Ministers Process (Joint Message, the
6th Trilateral [China, Japan, and Korea] Finance Ministers Meeting, May 4, 2006).
The ASEAN13 finance ministers also agreed on a research project, to be led by a
Japanese research institute, on exploring steps to create regional monetary units (Joint
Ministerial Statement of the 9th ASEAN13 Finance Ministers Meeting, May 4, 2006).
See also the earlier version of Kawai and Takagi (2005) as well as Mori et al. (2002).
For example, the depreciation of the Indonesian rupiah in August 2005 relative to other
East Asian currenciescalled the mini-currency crisissignaled that the economy
required monetary tightening to contain inflation and strong fiscal measures to reduce the
budget deficit. The authorities eventually took these actions and stabilized the financial
market. Another example is the recent appreciation of the Korean won and, to a lesser
extent, the Thai baht against the ACU, which indicate that the two economies international price competitiveness has declined relative to other East Asian economies.
See Ito and Park (2004), Dammers and McCauley (2006), and Ogawa and Shimizu
(Chapter 5, this volume) for the case for developing basket currency bonds.
The ECU was defined in 1974 by the members of the European Community (EC) as
a basket of participating currencies for EC accounting purposes. In 1975, the ECU
became a unit of account for the European Development Fund and, later, for the
European Investment Bank and for the EC budget. In 1979, European countries
launched the EMS to establish a formal, more systematic mechanism for achieving
intraregional exchange rate stability. To do this, they introduced (1) the Exchange Rate
Mechanism (ERM) that specified rules for currency market intervention and monetary
policy adjustment; (2) the very short-term financing facility that enabled central banks
to intervene in currency markets without disruptions; and (3) the ECU for purposes of
intraregional exchange rate stabilization. European central banks attempted to stabilize
intraregional exchange rates within a narrow fluctuation band using the ECU as central
parity. While European currencies remained free to fluctuate against the US dollar,
whenever the value of participating currencies in the ERM reached the upper or lower
limit of the fluctuation band, central banks were obliged to intervene in the currency
market to keep the value within band limits. However, in practice, the German mark
became a de facto anchor currency for exchange market interventions for non-German
central banks and the German Bundesbank was able to maintain relatively autonomous
monetary policy. Hence, the symmetrically designed ERM actually operated asymmetrically with the German mark functioning as a de facto key currency in Europe. The
ECU became the newly introduced European single currency, the euro, at the start of
the European Monetary Union in 1999.
For two or more countries to adopt and successfully maintain a common monetary and
exchange rate policy, it is necessary to achieve: (1) sufficient economic integrationfor
goods, services, capital and labor markets; (2) economic and structural convergence

10.
11.
12.
13.
14.
15.
319
across economies in terms of per capita incomes, market infrastructure and institutions,
financial market development and openness, and monetary policy practices; and (3)
political commitment to monetary integration. Though East Asia has already achieved
significant economic integration through trade and foreign direct investment, the region
is still weak in the other aspects. Anderson (2006) points out the importance of economic
convergence in the ratios of purchasing power parity (PPP) to market exchange rates
and argues that currently Asian economies have a significantly larger variation in these
ratios than do the 12 countries of the euro area.
See Girardin and Steinherr (2006a) for such technical discussions.
Other current account transactions such as remittances and investment income flows
may also be considered.
Capital account openness is an important factor in encouraging capital market participants to develop ACU-denominated bonds. One of the working groups under the Asian
Bond Markets Initiative (ABMI) is focusing on the possibilities of multi-currency bond
issuance.
Hong Kong may be added to this group as it is also an active participant of ASEAN13
finance and central bank deputies processes and a key member of the EMEAP.
See Ogawa and Shimizu (2005) and RIETI (2006) for AMU indexes.
The ABF initiative has been undertaken by 11 East Asian central banks under EMEAP
as a measure to strengthen the demand side of bond markets. ABF1was launched in
June 2003 to pool USD 1 billion of the regions foreign exchange reserves to invest in US
dollar-denominated sovereign and quasi-sovereign bonds issued by eight members
excluding Japan, Australia, and New Zealand. ABF2 was launched in December 2004
to invest USD 2 billion of foreign exchange reserves in eight country sub-funds, in local
currencies, and the Pan-Asian Bond Index Fund (PAIF), a listed open-ended bond
fund with investments across the region. See also Ma and Remolona, Chapter 4 in this
volume.
The ABF iBoxx Pan-Asia Index calculates the value of a bond fund investing in
local currency government and quasi-government bonds in eight emerging East Asian
markets. As the fund covers a variety of bond markets, a simple weighting by market
capitalization would distort the index in favor of large markets such as the PRC and
Korea and reduce the weight for smaller, but more developed, more liquid, and accessible markets such as Hong Kong and Singapore. Therefore, the weight of each economy
starts from an equal weighting baseline and is then adjusted by factors such as (a) local
bond market size (20 percent), (b) turnover ratio (20 percent), (c) sovereign local debt
rating (20 percent), and (d) market openness (40 percent). The ABF iBoxx Pan-Asia
Index uses the following methodology:
1. Local bond market size (S) is based on data from the Bank for International
Settlements (BIS) where available, or on a consolidated average poll of the iBoxx
Asian Index Committee.
2. Turnover ratio (T), considered a proxy for liquidity, is derived by comparing total
transaction size to market capitalization. Transaction size is obtained through the
consolidated annualized average polled from the iBoxx Asian Index Committee.
3. Sovereign local debt rating (R) is the best (highest) local currency long-term
rating from Fitch, Moodys, or S&P by converting the rating into a numerical
equivalent.
4. Market openness (O) is a qualitative measurement of relative market openness,
assessed on the basis of the legal and regulatory environment, the fiscal situation, market infrastructure, and back-office infrastructurecurrently countries are
grouped into the following categories: (a) Hong Kong and Singaporehighly open;
(b) Indonesia, Korea, Malaysia, Philippines, Thailandgenerally open; (c) China
relatively less open.
Given the relative weights of each factor, the following adjustment factor (AF) is calculated for each economy: AF 5 0.2 S 1 0.2 T 1 0.2 R 1 0.4 O. Each countrys weight

320
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.

is then calculated as the sum of the baseline weight (1/8) plus the adjustment factor: Wi
5 (1/8) 1 AFi. A technical committee and an oversight committee review the market
allocations annually. The ABF iBoxx index is rebalanced monthly.
Pegging to the euro would be more hazardous than pegging to the US dollar or yen
because of the lesser economic linkages between East Asia and Europe.
These arguments do not preclude the possibility that the yuan begins to function as an
important international currency for smaller neighboring countriessuch as Lao PDR,
Myanmar, and Mongolia. It is said that the yuan tends to be used and held in the northern part of Lao PDR and Myanmar and in the southern part of Mongolia. This could
grow as China continues its economic growth. Still, it would take a long time before
the yuan becomes a credible international currency from either a worldwide or regional
perspective.
This is an application of the battle of the sexes game-theoretic model, which is a
two-player coordination game. An example is the situation where a coupleKelly and
Chrisneed to decide whether they want to go to the football game (Kellys preference)
or the opera (Chriss preference), both hoping to go to the same place rather than different ones. See Hamada (2006) for the original application of such a game.
See Kawai (2006). Although there may be other scenarios, the one proposed here is
general enough to capture the substance of necessary elements of any other scenario.
However, views may differ across experts and policymakers with regard to the potential
role of an ACU. See also Girardin and Steinherr (2006b).
Though central bank governors are not formally involved in the ASEAN13 finance
ministers process, their deputies participate in the important working groups under
the finance ministers. See Kuroda and Kawai (2002), Bird and Rajan (2002), Montiel
(2004), Rajan and Siregar (2004), Girardin (2004) and Kawai (2006) for a review of
recent initiatives undertaken by ASEAN13 finance ministers.
Through the ASEAN13 ERPD process, finance ministers meet regularly to review
financial and economic issues affecting member countries. Eight members have set up
national surveillance units or their equivalents for economic and financial monitoring and are developing their own early warning systems. In May 2005, finance ministers
agreed to substantially strengthen the CMI by increasing bilateral currency swap size,
linking them to regional economic surveillance, raising the level of disbursement permitted without an International Monetary Fund (IMF) program from 10 to 20 percent,
and incorporating a collective decision-making mechanism for swap activation (a step
toward multilateralization). Total swap size reached USD 75 billion as of May 2006.
In May 2006, finance ministers tasked their deputies to further study various options
toward an advanced framework of the regional liquidity support arrangement (CMI
multilateralization or post-CMI).
The ABMI is marshalling technical expertise and building capacity for regional bond
market development. Current discussions are focusing on the creation of multi-currency
bonds, establishment of a regional credit guarantee mechanism, exploration of an Asian
settlement system, strengthening Asian credit rating agencies in conjunction with Basel
II implementation, and development of Asian Bond Standards.
See Williamson (2005 and Chapter 11, this volume) and Kawai (2006).
Eichengreen (2006) considers not only the parallel currency approach but also many
other potentially important issues on ACU by comparing them with the European ECU
experience. According to his findings, Europe should have developed the parallel currency approach but did not; instead adopting the ERM.
A practical policy in this approach would be to adopt a multi-track, multi-speed process
where economies ready for deeper policy coordination start the process, while others
begin readying themselves to join. A group of economies largely fulfilling optimum
currency area (OCA) conditions in East Asialike Japan and Korea; China, Hong
Kong, and Macau SAR; and Singapore, Malaysia, and Brunei Darussalamand with
sufficient political commitment, may wish to initiate sub-regional currency stabilization
schemes. This group would intensify monetary and exchange rate policy coordination
while expecting others to join later.

321
REFERENCES
Anderson, Jonathan (2006). Still Not a Great Idea. UBS Asian Focus, May 12.
Bird, Graham, and Ramkishen S. Rajan (2002). The Evolving Asian Financia
Architecture. Princeton Essays in International Economics 226, Princeton
University, Princeton, NJ.
Dammers, Clifford, and Robert McCauley (2006). Basket Weaving: The Euro
Market Experience with Basket Currency Bonds. BIS Quarterly Review March:
7992.
Eichengreen, Barry (2006). The Parallel Currency Approach to Asian Monetary
Integration. AEA Papers and Proceedings 96: 4326.
Girardin, Eric (2004). Information Exchange, Surveillance Systems and Regional
Institutions in East Asia. In Asian Development Bank (ed.), Monetary and
Financial Integration in East Asia. The Way Ahead, Volume 1. Houndmills and
New York: Palgrave Macmillan, pp. 5395.
Girardin, Eric, and Alfred Steinherr (2006a). Lessons in the European Experience
for the Creation of an Asian Currency Unit (ACU). Mimeograph. Asian
Development Bank, Manila.
Girardin, Eric, and Alfred Steinherr (2006b). Properties, Requirements, and
Benefits of an Asian Currency Unit (ACU). Mimeograph. Asian Development
Bank, Manila.
Hamada, Koichi (2006). A Remark on the Political Economy of East Asia.
Mimeograph. A paper presented to the Young Leaders Forum 2006 East
Asian Integration, organized by Japanese-German Center Berlin (JDZB),
Reichenow.
Ito, Takatoshi, and Yung Chul Park (2004). Developing Asian Bond Markets.
Canberra: Asia Pacific Press.
Kawai, Masahiro (2006). Toward a Regional Exchange-rate Regime in East
Asia. A revised version (June) of the paper presented to the Bellagio Conference
on New Monetary and Exchange-rate Arrangements for East Asia held in
Bellagio on May 2326, 2006.
Kawai, Masahiro, and Shinji Takagi (2005). Towards Regional Monetary
Cooperation in East Asia: Lessons from Other Parts of the World. International
Journal of Finance and Economics 10: 97116.
Kuroda, Haruhiko, and Masahiro Kawai (2002). Strengthening Regional
Financial Cooperation in East Asia. Pacific Economic Papers no. PEP-332,
Australian National University, Canberra.
Montiel, Peter J. (2004). An Overview of Monetary and Financial Integration
in East Asia. In Asian Development Bank (ed.), Monetary and Financial
Integration in East Asia: The Way Ahead, Volume 1. Houndmills and New York:
Palgrave Macmillan, pp. 152.
Mori, Junichi, Naoyoshi Kinukawa, Hideki Nukaya, and Masashi Hashimoto
(2002). Integration of East Asian Economies and a Step by Step Approach
towards a Currency Basket Regime. IIMA Research Report no. 2, Institute for
International Monetary Affairs, Tokyo.
Ogawa, Eiji, and Junko Shimizu (2005). A Deviation Measurement for Coordinated
Exchange Rate Policies in East Asia. RIETI Discussion Paper Series 05-E-017,
Research Institute of Economy, Trade and Industry, Tokyo.
Rajan, Ramkishen, and Reza Siregar (2004). Centralized Reserve Pooling for
the ASEAN13 Countries. In Asian Development Bank (ed.), Monetary and

322
Financial Integration in East Asia. The Way Ahead, Volume 2. Houndmills and
New York: Palgrave Macmillan, pp. 285329.
RIETI (2006). AMU and AMU Deviation Indicators. Research Institute of
Economy, Trade and Industry, Tokyo.
Williamson, John (2005). A Currency Basket for East Asia, Not Just China.
Policy Briefs in International Economics no. PB05-1, Institute for International

13.
The illusion of precision and the

role of the renminbi in regional
integration
Yin-Wong Cheung, Menzie D. Chinn, and
Eiji Fujii1
13.1
INTRODUCTION
On July 21, 2005, China announced a long-anticipated revision to its

exchange rate regime.2 Against a backdrop of rising protectionist sentiment in the United States and increasingly acrimonious mutual recriminations over the benefits and costs of an open international financial system,
the move was warmly, albeit cautiously, welcomed. The wariness arises
from the uncertainty surrounding the exact nature of the new exchange
rate regime and how rapidly the Chinese authorities are willing to allow
the currency, the renminbi (RMB), to appreciate. So far, the increase in the
RMBs value against the dollar has been quite modestin the order of a
few percentage points.
The intensity of the debate regarding the degree of RMB misalignment
reflects the increasing role of China in the international arena and its rapid
pace of export growth and penetration into global markets. Much of the
pressure for RMB revaluation is driven by Chinas rapidly growing trade
surplus with the US (and more recently with the rest of the world) and its
accumulation of foreign reserves. On the other hand, Chinas domestic
problems and its role in the regional production process have received
limited attention in the debate.
While the theme of the current study is not to evaluate arguments
advanced for RMB revaluation, it should be noted that Chinas external
balances have notuntil quite recentlyconstituted a prima facie case
for RMB undervaluation. Economic theory, for instance, suggests that
the RMB value should be linked to the magnitude of the overall trade
balance instead of to the size of a given bilateral trade balance. Yet in 2004,
Chinas trade surplus relative to GDP was not particularly large compared
with, say, those recorded by, say, Japan and Germany. With respect to
325

326
The role of China
reserve accumulation, however, the recent Chinese experience has been

quite phenomenal. Nonetheless, official reserve holdings might be deemed
insufficient as concerns over Chinas contingent liabilities come to the fore.
For instance, one recent study estimates that, in 2005, Chinas total nonperforming loan liability, a key component of Chinas contingent liability,
stood at USD 900 billion, a figure higher than its reserve holdings.3
The extant discussion routinely emphasizes the impact of RMB valuation
on global trade relationships and global trade imbalances. We, however,
would like to emphasize the role of modern China in East Asia and, thus,
the implications of Chinas foreign exchange policy for the region. In this
regard, two observations are worth mentioning. First, the Asian economies
have tended to link their currencies to the US dollar via either a de facto or
de jure peg, even after the 1997 financial crisis. As has been observed on a
number of occasions, the policymakers in these economies tend to favor a
stable foreign exchange environment, often perceived to be conducive to
capital inflows and economic growth.4 To the extent that these currencies
were stabilized at low values, the East Asian exchange rate regimes are
often viewed as part of a mercantilist approach to economic development
(e.g., Dooley et al. 2003 and Chapter 6, this volume). Second, the evolution
of Chinas position in international and regional markets is closely related
to the trend toward the globalization of production. Specifically, declining
transportation costs and decreasing trade barriers have precipitated the
internationalization of the production process, a phenomenon variously
termed production fragmentation or vertical specialization (Arndt 1997, Yi
2003). Given these circumstances, China, with its incentive structure and its
abundant labor, has grown into a key production/manufacturing hub for
East Asia, mainly serving as the last segment in the international production
chain. A by-product of the manufacturing relocation process is an upsurge
in intra-regional trade between China and its neighboring economies.
The integration of China into the world economy has brought about
substantial adjustment in production and trade in developing economies,
especially those in East Asia. These economies, which beside their extensive
production and trade linkages with China possess both real and financial
sectors that are less sophisticated than those in developed countries, are
very susceptible to the (adverse) effects of RMB exchange rate volatility.
Thus, Chinas exchange rate policy has implications for both its domestic
economy and that of the wider Asian region.
In this chapter, we step back from the debate about the merits of one
exchange rate regime versus another5 and, indeed, do not take a stand on
how large a revaluationor devaluationis necessary (although our conclusions will inform the debate over what the appropriate actions might
be). Rather, we re-orient the discussion of currency misalignment back

Precision and the role of the renminbi in regional integration
327
toward theory and empirics; in particular, we want to focus on the difficulty

in defining and calculating the equilibrium (real) exchange rate in theory
and on quantifying the uncertainty surrounding the measurement of the
level of the equilibrium.
Why is such a reassessment necessary? We think it is necessary to review
the evidence and conclusions in the context of the various underlying
premises. That is, like the story lines in the 1950 Japanese crime drama
movie Rashomon, each analyst seems to have a different interpretation of
what constitutes misalignment.6 At the heart of the differences are contrasting ideas about what constitutes an equilibrium condition, to what
time frame the equilibrium pertains, and, consequently, what econometric
method to implement. Even when there is agreement on the fundamental
model, questions typically remain about the right variables to use.
The flux of exchange rate economics offers a hint of our main argument.
Since Meese and Rogoff published their seminal piece on the difficulties
inherent to empirical exchange rate modeling in 1983, a voluminous collection of studies echoing their conclusion has accumulated.7 Given the lack
of a commonly agreed theoretical model, assertions about the equilibrium
(and disequilibrium) level of an exchange rate should be guided by numerous caveats. A hasty decision on RMB policy based on not well-founded
evidence can do more harm than benefit to China and to its trading partners, especially the developing economies in the region.
13.2
THE EXTANT LITERATURE
A couple of surveys of the RMB misalignment literature have compared the estimates of the degree to which the RMB is misaligned. The
Government Accountability Office (GAO 2005) provides a comparison
of the academic and policy literature, while Cairns (2005b) briefly surveys
recent point estimates obtained by different analysts. Here, we review the
literature to focus on primarily theoretical papers and their economic and
econometric distinctions.
Most of these papers fall into familiar categories, either relying on some
form of relative purchasing power parity (PPP) or cost competitiveness
calculation, the modeling of deviations from absolute PPP, a composite
model incorporating several channels of effects (sometimes called behavioral equilibrium exchange rate models), or flow equilibrium models. Table
13.1 provides a typology of these approaches, further disaggregated by the
data dimension (cross section, time series, or both).8
The relative PPP comparisons are the easiest to make, in terms of calculations. Bilateral real exchange rates are easy to calculate, and there are

328
The role of China
Table 13.1
Time
series
Studies of the equilibrium exchange rate of the renminbi
BEER
Relative PPP,
Absolute
Balassacompetitiveness PPPincome Samuelson
relationship (with
productivity)
Macroeconomic
balance/external
balance
Wang (2004)
Cheung et
al.(2005a)
Bosworth
(2004)
Goldstein
(2004)
Wang (2004)
Bosworth
(2004)
Cross
section
Coudert and
Couharde
(2005)
Frankel
(2005)
Panel
Cairns
(2005b)
Cheung et
al. (2005a)
Cheung
et al.
(2005a)
Zhang
(2001)
Wang
(2004)
Funke
and
Rahn
(2005)
Cheung Coudert and

Couharde
et al.
(2005a) (2005)
Notes: Relative PPP indicates that the real exchange rate is calculated using price or cost
indices and that no determinants are accounted for. Absolute PPP indicates the use of
comparable price deflators to calculate the real exchange rate. Balassa-Samuelson (with
productivity) indicates that the real exchange rate (calculated using price indices) is modeled
as a function of sectoral productivity levels. BEER indicates composite models using
net foreign assets, relative tradable to nontradable price ratios, trade openness, or other
variables. Macroeconomic balance indicates cases where the equilibrium real exchange rate
is implicit in a normal current account (or combination of current account and persistent
capital inflows, for the external balance approach).
now a number of trade-weighted series that incorporate China. On the

other hand, relative PPP in levels requires the cointegration of the price
indices with the nominal exchange rate (or, equivalently, the stationarity
of the real exchange rate),9 but these conditions do not necessarily hold,
regardless of the deflator adopted in empirical analyses. Wang (2004)
reports interesting IMF estimates of unit labor cost deflated RMB. This
series has appreciated in real terms since 1997; of course, this comparison,
like all other comparisons based on indices, depends on selecting a year
that is deemed to represent equilibrium. Selecting a year before 1992 would
imply that the RMB has depreciated over time.

329
Bosworth (2004), Frankel (2005), Coudert and Couharde (2005), and

Cairns (2005b) estimate the relationship between the deviation from
absolute PPP and relative per capita income. All obtain similar results
regarding the relationship between the two variables, although Coudert
and Couharde fail to detect this link for the RMB.
Zhang (2001), Wang (2004), and Funke and Rahn (2005) implement
what could broadly be described as behavioral equilibrium exchange rate
(BEER) specifications.10 These models incorporate a variety of channels
through which the real exchange rate is affected. Since each author selects
different variables to include, the implied misalignments will necessarily
vary.
Other approaches center on flow equilibria, considering savings and
investment behavior and the resulting implied current account. The equilibrium exchange rate is derived from the implied medium term current
account using import and export elasticities. In the IMFs macroeconomic approach, the norms are estimated, in the spirit of Chinn and
Prasad (2003). Wang (2004) discusses the difficulties in using this approach
for China but does not present estimates of misalignment based on this
framework. Coudert and Couharde (2005) implement a similar approach.
Finally, the external balances approach relies on assessments of the persistent components of the balance of payments condition (Goldstein 2004,
Bosworth 2004). This last set of approaches is perhaps most useful for conducting short-term analyses. But the wide dispersion in implied misalignments reflects the difficulties in making judgments about what constitutes
persistent capital flows. For instance, Prasad and Wei (2005), examining
the composition of capital inflows into and out of China, argue that much
of the reserve accumulation that has occurred in recent years is a result of
speculative inflow; hence, the degree of misalignment is small.
Moreover, such judgments based on flow criteria must condition
their conclusions on the existence of effective capital controls. This is an
obviousand widely acknowledgedpoint (e.g., Holtz-Eakin 2005), but
one that bears repeating and, indeed, is a point that we return to at the end
of this chapter.
Two observations are of interest. First, as noted by Cairns (2005a), there
is an interesting relationship between the particular approach adopted by
a study and the degree of misalignment found.11 Analyses implementing
relative PPP and related approaches indicate the least misalignment. Those
adopting approaches focusing on the external accounts (either the current
account or the current account plus some persistent component of capital
flows) yield estimates that are in the intermediate range. Finally, studies
implementing an absolute PPP methodology result in the greatest degree
of estimated undervaluation.

330
The role of China
Second, while all these papers make reference to the difficulty of applying such approaches in the context of an economy ridden with capital
controls, state-owned banks,12 and large contingent liabilities, few have
attempted a closer examination of these issues.
The current chapter, and its working paper version (denoted as Cheung
et al. 2005a, in Table 13.1), contributes to this literature using the BalassaSamuelson approach (in which productivity differentials are used) and
implementing panel analyses of the PPPincome relationship, augmented
by variables motivated from the BEER and macroeconomic balance
literature.
13.3
A SIMPLE TIME SERIES APPROACH
Before turning to the more elaborate frameworks for evaluating the value
of the RMB, let us consider an approach often used in the aftermath of
the East Asian crisis of the mid-1990s, namely, indicators of exchange rate
overvaluation that are measured as deviations from a trend. Adopting this
approach in the case of China would not lead to a very satisfactory result.
Consider first what a simple examination of the bilateral real exchange rate
between the US and the RMB using this measurement would imply. Figure
13.1 depicts the official exchange rate series from 1986:1 to 2005:2, deflated
1.1
Official real
exchange
rate ($/RMB)
1.2
1.3
1.4
Trend
1.5
1.6
Trend for
adjusted
rate
1.7
1.8
Adjusted real
exchange rate
1.9
88
Figure 13.1
90
92
94
96
98
00
02
04
Real USD/RMB exchange rate, in logs (official and

adjusted) and trends

331
6.0
Trade weighted
value of RMB
5.6
5.2
27.5%
Overvaluation
4.8
4.4
4.0
80
Figure 13.2
82
84
86
88
90
92
94
96
98
00
02
04
Real trade-weighted value of RMB, in logs, and trend
by the CPIs of the US and China (higher values constitute a stronger

Chinese currency vis--vis the dollar). In line with expectations, in the years
since the East Asian crises, the RMB has experienced a downward decline
in value. Indeed, over the entire sample period, the RMB has experienced
a downward trend.
However, as is often the case with economies experiencing transitions
from controlled to partially decontrolled capital accounts and from dual to
unified exchange rate regimes, there is some dispute over what exchange rate
measure to use. In the Chinese case, it could be argued that, with a portion of
transactions taking place at swap rates, the 1994 mega-devaluation would
actually be better described as a unification of different rates of exchange.13
The import of this difference can be gleaned from the fact that the
imputed time trends then exhibit quite different behaviors and imply different results. Using the adjusted rate, one finds a modest undervaluation
of 4.8 percent in the second quarter of 2005. Using the official rate would
imply a slight, almost imperceptible, overvaluation of 1.4 percent.
A natural reaction would be to argue that simple bilateral comparisons are faulty. We would agree. However, appealing to trade-weighted
exchange rates would not necessarily clarify matters. Figure 13.2 depicts
the IMFs trade-weighted effective exchange rate index and a linear trend
estimated over the available sample period 19802005. One finds that a
simple trend (as used in the early warning literature) indicates that the
RMB is 25 percent overvalued.

332
The role of China

6.0
Trade weighted
RMB (CPI-deflated)
5.6
5.2
2.1%
Undervaluation
4.8
Trade weighted
RMB (PPI-deflated)
4.4
4.0
80
Sources:
82
84
86
88
90
92
94
96
98
00
02
04
IMF, Deutsche Bank, and authors calculations.
Figure 13.3
Real trade-weighted indices of RMB, in logs, and segmented

trend
A cursory glance at the data indicates that a simple trend will not do.
A test on residuals from recursive regression procedure applied to the
constant plus trend suggests a break with maximal probability in the third
quarter of 1986. Fitting a broken trendadmittedly an ad hoc procedureprovides a fairly good fit, as illustrated in Figure 13.3. In the second
portion of the sample, the estimated trend is essentially zero, a result that
is consistent with purchasing power parity.
Obviously, a more formal test for stationarity is necessary. Following
the methodology outlined in Chinn (2000a), we test for cointegration of
the nominal (trade-weighted) exchange rate and the relative price level.14
We find that there is evidence for cointegration of these two variables, with
the posited coefficients.15 This means that we can use this trend line as a
statistically valid indication of the mean value to which the real exchange
rate series reverts.
Interestingly, even here, the procedure indicates a very modest 2.1
percent undervaluation. These conclusions are not sensitive to the index.
Using Deutsche Banks PPI-deflated index, similar movements in the RMB
are detected.
Before we move on to the next section, we should point out that the data
in Figures 13.1 and 13.2 are indexes and do not give the exact exchange

333
value of the underlying currencies. To be sure, the conclusion regarding

overvaluation or undervaluation depends on how the trend line is drawn,
which in turn depends on the starting point of the sample.
13.4
THE PRODUCTIVITY APPROACH
The role of productivity is central to thinking about the evolution of the

Chinese currency. The standard point of reference in thinking about the
impact of productivity is the Balassa-Samuelson theory, which focuses
on the differential between traded and nontraded sectors (Balassa 1964,
Samuelson 1964). Interestingly, to our knowledge, nobody has attempted
to estimate the link between sectoral productivity estimates and the real
exchange rate for China, with the exception of Chinn (2000b).16
Nontradables, Productivity, and the Real Exchange Rate in Theory
Most investigations of the link between the relative price of nontradables
and the real exchange rate rely on the following construction. Let the log
aggregate price index p be given as a weighted average of log price indices
of traded (T) and nontraded (N) goods given by the relationship:
pt 5 (1 2 a) pTt 1 apN
t ,
(13.1)
where a is the share of nontraded goods in the price index. Suppose further
that the foreign countrys aggregate price index is similarly expressed by:
T*
N*
p*
t 5 (1 2 a*) p t 1 a*pt .
(13.2)
Then, the real exchange rate is given by:

qt ; (st 1 p*t 2 pt) 1 k,
(13.3)
where s is the log of the domestic currency price of foreign currency and
k is a constant accounting for the fact that the price levels are indices. In
other words, even though productivity is being accounted for, the very
fact that we only have price and productivity indices means that we can
only evaluate deviations from a relative PPP modified for productivity
differentials.
For a 5 a*, the following holds:
T
N*
T*
qt 5 (st 1 pTt 2 pTt *) 2 a [ ( pN
t 2 pt ) 2 ( p t 2 p t ) ] 1 k. (13.4)

334
The role of China
Although there are many alternative decompositions that can be undertaken, equation (13.4) is the most relevant, since most economic models
make reference to the second term as the determinant of the real exchange
rate, while the first is assumed to be zero by the application of purchasing
power parity to traded goods.
In order to move away from accounting identities, one requires a
model, such as the Balassa-Samuelson framework. The relative prices of
nontradables and tradables will be determined solely by productivity differentials, under the stringent conditions that capital is perfectly mobile
internationally and that factors of production are free to move between
sectors. Substituting out for relative prices yields:
|
N * T* T*
/q ) a t 2 aNt *] 1 k ,
qt 5 (st 1 pTt 2 pTt *) 2a [ (qN /qT ) aTt 2 aN
t ] 1a [ (q
(13.5)
where ai is total factor productivity in sector i (i 5 N, T ) and the qs are
parameters in the production functions.17
Most researchers have proceeded under the assumption that the first
term is I(0). This implies cointegrating relationships of the forms:
T
N*
T*
qt 5 2 a ( pN
t 2 pt ) 1 a ( p t 2 p t )
(13.6)
T*
N*
qt 5 2 a [ aTt 2 aN
t ] 1 a [at 2 a t ]
(13.7)
N*
qt 5 2 a [ aTt 2 aTt *] 1 a [ aN
t 2 a t ],
(13.79)
and
respectively (where the production functions in the tradable and nontradable sectors are assumed to be the same, so that the qs cancel out in equation (13.7) and the constants are suppressed). Equation (13.6) underpins
the analysis by Funke and Rahn (2005). Equation (13.79) provides the basis
for the empirical work in this section.18
Econometric Specification, Data, and Results
The cointegrating relationship is identified using dynamic OLS (DOLS;
Stock and Watson 1993). One lead and one lag of the right hand side variables are included. In a simple two-variable cointegrating relationship, the
estimated regression equation is:
11
q 5 b0 1 b1x 1 a giDxt1i 1 ut.
(13.8)
i5 21

335
Although this approach presupposes that there is only one long-run relationship, this requirement is not problematic, as in these extended samples
at most one cointegrating vector is usually detected.19 A deterministic trend
is also allowed in equation (13.8).
We take the two countries to be the US and China. In principle, it would
be preferable to consider China vs. the rest of the world. However, data
considerations, plus the fact that the misalignment debate revolves around
the USChina nexus, motivate us to adopt this perspective.
The data issues present the largest challenges. The straightforward calculations involve the exchange rate and the US variables, although, even in
the former instance, there are some calculations. For reasons discussed in
section 13.1, we do not rely on the official exchange rate in the years directly
leading up to 1994. Rather, the real RMB/USD rate is measured using the
nominal exchange rate, which is then adjusted, following Fernald et al.
(1999), as described above, and then deflated by the respective CPIs. US
data are derived from the Bureau of Labor Statistics and the Groningen
Growth and Development Center.
Once one has to determine the appropriate Chinese productivity
numbers, one enters a data quagmire. As is well known, even deciding on
the appropriate estimate of Chinese GDP can be a contentious matter (see
Rossi 2005). As demonstrated in Youngs (2003) dissection of Chinese
data, small changes in assumptions regarding the validity of the output
numbers and the deflators can radically alter the implied output per worker
and total factor productivity series substantially. Hence, all the estimates
provided in this section should be viewed as heroic in nature.
Forging ahead, we follow the method adopted in Chinn (2000b). Average
labor productivity, obtained by dividing real output in sector i by labor
employment in the same sector, is used as the proxy for sectoral total factor
productivity.20 The tradables sector is proxied by the manufacturing sector,
while the nontradables is proxied by the other sector. This latter sector is
defined as those sectors besides mining, manufacturing, and agriculture.
Two limitations of the data should be emphasized. First, since these
labor employment statistics are not adjusted for part-time workers, the
constructed sectoral productivity data are cross-checked using the manufacturing productivity reported by the World Banks World Development
Indicators for several countries. The productivity figures from these two
sources match quite well. Further, these series also match quite well for
both manufacturing versus tradables and other versus nontradables.
These outcomes serve to improve ones confidence that the proxies used
are not implausible.
Second, the proxy variable is labor productivity, rather than total factor
productivity (TFP) as suggested by the model. Canzoneri et al. (1996)

336
The role of China

2.0
1.6
Real manufacturing
output divided by
manufacturing
employment
1.2
0.8
Szirmai et al.
manufacturing
productivity
0.4
0.0
86
Figure 13.4
88
90
92
94
96
98
00
02
04
Measures of Chinese manufacturing productivity (in logs,

normalized to 1986 5 0)
have argued that use of labor productivity is to be preferred because it is

less likely to be tainted by misestimates of the capital stock. In any event,
there is little possibility of circumventing this problem. To our knowledge,
almost all calculations of East Asian TFP over long spans of time have
been conducted on an economy-wide basis, with a few exceptions, including Young (1995, 2003).
In addition to the productivity series calculated in the described
manner, we also relied on a productivity series obtained via a careful
analysis of the state-owned enterprise (SOE) and township and village
enterprise (TVE) output and employment figures reported by Szirmai et
al. (2005). The two Chinese manufacturing productivity series, in logs,
normalized to 198650, are shown in Figure 13.4. The productivity growth
rates are 9.5 percent and 7.1 percent, respectively, over the 19872003
period. These are more rapid than the US manufacturing growth rate of
4.6 percent over the same period. Figure 13.5 depicts the Chinese other
productivity series, which grows at a rate of 5.0 percent (compared to 2.0
percent for the US).
The DOLS results for estimations over the 19882002 period are
reported in Table 13.2.21 The estimates based on productivity numbers
calculated using output and employment figures are reported in columns
[1][4], while those based on the manufacturing estimates from Szirmai et
al. (2005) are reported in columns [5][8].

337
1.0
0.8
Real other
output divided
by other
employment
0.6
0.4
0.2
0.0
0.2
86
Figure 13.5
88
90
92
94
96
98
00
02
04
The measure of other productivity (in logs, normalized to

1986 5 0)
In column [1], estimates from the most basic specification, corresponding

to equation (79), indicate that each 1 percent increase in Chinese manufacturing productivity over US productivity results in a half percentage point
appreciation in the RMB against the dollar, in real terms. Increases in nontraded sector productivity depreciate the RMB, in line with the theoretical
prediction. However, the point estimate is somewhat large; in general,
when the productivity coefficients in the tradable and nontradable sectors
are similar, the coefficient should be about equal to the share of nontradables in the aggregate price index. On the other hand, the standard error is so
large that the 61 standard error bands encompass plausible values of a.
In column [2], the specification is augmented by a time trend. The point
estimate for the coefficient on traded sector productivity is now larger
in absolute value; however, neither the nontradables term nor the time
trend itself is statistically significant, casting doubt on the relevance of this
specification. Constraining the coefficients on tradables and nontradables
productivity to be the same yields a deterioration in the fit (the adjusted
R2 declines, while the standard error of regression increases). Augmenting
the constrained specification with a time trend produces estimates that are
correctly signed and ascribes a large portion of the secular movement in
the RMB3.5 percent per yearto the time trend. This is certainly an
undesirable result for prediction purposes, since the time trend is a proxy
for our ignorance.

338

(1)
Other
prod.
15
0.049
15
0.049
[0.527]
0.028
[0.022]
[0.516]
0.76
[2]
0.921
[0.231]
1.087
[1]
0.556
[0.184]
1.122
0.77
Official
deflator
Official
deflator
15
0.054
0.71
[0.088]
[3]
0.275
[0.088]
0.275
Official
deflatora
15
0.046
0.8
[0.224]
0.035
[0.009]
[4]
0.987
[0.224]
0.987
Official
deflatora
15
0.047
0.79
[0.748]
[5]
1.322
[0.550]
1.788
Szirmai
Mfg.
15
0.050
0.76
[0.724]
0.007
[0.014]
[6]
1.204
[0.616]
1.882
Szirmai
Mfg.
15
0.053
0.73
[0.242]
[7]
0.512
[0.242]
0.512
Szirmai
Mfg.a
15
0.052
0.73
[0.735]
0.010
[0.008]
[8]
1.145
[0.735]
1.145
Szirmai
Mfg.a
Notes: Official deflator refers to estimates obtained using productivity figures calculated using official deflator as reported by Holz (2005);
Szirmai refers to estimates obtained using manufacturing productivity numbers reported in Szirmai et al. (2005). All estimates use estimated
effective exchange rate as described in Fernald et al. (1999). Estimates obtained using dynamic OLS (DOLS) with one lead and lag of the right
hand side variables. Mfg. (Other) productivity is the differential labor productivity in the manufacturing (other) sectors. Pred. is the
predicted sign according to the Balassa-Samuelson hypothesis. SER is the standard error of regression. N is the number of observations.
a
Manufacturing and Other productivity are constrained to have equal and opposite signs.
Adjusted
R2
N
SER
Trend
()
Pred.
Balassa-Samuelson model of real exchange rate of renminbi, 198803
Mfg. prod.
Table 13.2
339
Using the Szirmai et al. (2005) manufacturing numbers produces interesting results. In column [5], the coefficients are correctly signed, albeit
somewhat large in absolute value. Interestingly, the point estimates are
not sensitive to the inclusion of a time trend (column [6]). Constraining
the coefficients on tradable and nontradable productivity to be equal
yields plausible estimates of a. Including a time trend, as in column [8],
produces more imprecisely estimated coefficients, while leaving the time
trend insignificant.
For reasons already alluded to, one may be dubious about these results.
An additional reason for skepticism is that the sample covers quite a
short period; using the DOLS approach results in a sample of only 15
observations. An obvious question is why we do not extend the sample
backward.
Two reasons guide our sample choice. First, the data are available on
a more or less consistent basis over this time period. Second, and perhaps
of even greater importance, it is not clear whether extending the data back
in time would be appropriate. The end of the second phase of economic
liberalization, which severed the link between firm management and the
governments economic plan objectives, roughly coincides with the beginning of the sample.
The data can be extended backward in time. Splicing data utilized in
Chinn (2000b) to the series discussed above, the DOLS regressions can
be re-estimated over the 19802004 period. This produces a sample of 22
observations (or 25, if simple OLS is implemented). The results are surprisingly similar to those reported in the first two columns of Table 13.2. The
elasticity of the real exchange rate with respect to the intercountry traded
sector productivity differential for the extended data set is 20.50, versus
0.56 for the truncated data used in Table 13.2, while the nontraded differential has an implausibly large impact of 2.2, versus the truncated data
sets 1.1. The pattern of estimates persists even with the inclusion of a time
trend.22
To sum up, it appears that, regardless of the measure of manufacturing
productivity used, the coefficient estimates point in the directions predicted
by the Balassa-Samuelson hypothesis.
Implied RMB Misalignment
In order to assess whether the RMB is misaligned, we take the long-run
coefficients from columns [1] and [5] of Table 13.2 and generate long-run
predicted values. One difficulty in conducting the assessment for the recent
period is that the sectoral output and employment data are available only
up to 2003. Indeed, the estimated manufacturing labor productivity data

340
The role of China

1.9
Real bilateral
exchange rate
(RMB/USD)
1.8
1.7
1.6
1.5
1.4
Estimated
productivity
Szirmai mfg.
productivity
1.3
88
Figure 13.6
90
92
94
96
98
00
02
04
Real bilateral (adjusted) exchange rate and predicted long

run rate, in logs, based on estimated productivity and Szirmai
et al. (2005) manufacturing productivity
from Szirmai et al. (2005) extend only up to 2002. We assume that, for the
latter, the productivity growth rates in 2003 and 2004 are the same as the
2002 rate of 9.1 percent, while, for the former, the 2004 rate equals the 2003
rate of 4.8 percent.
Figure 13.6 depicts the results (higher values of the exchange rate imply
weaker values of the RMB against the USD). Using the estimated productivity data, the RMB is only about 6.1 percent undervalued in 2004.
Interestingly, the greatest degree of undervaluation is in 1993 (about 30
percent), and it drops in 1994 (to 16 percent), despite that years devaluation. Using the Szirmai et al. (2005) data, the 2004 undervaluation is
negligible, at about 1.4 percent.
These counter-intuitive results suggest that something may be missing
from this approach. This framework assumes that the relative prices of
tradable and nontradable goods are determined solely by the relative
production prices. This assumption, in turn, relies on the assumption of
homothetic preferences across different per capita income levels. But this is
unlikely to be the case; hence, relative prices might be changing for reasons
apart from differing productivity trends. In fact, the argument that much
of the spectacular growth in Chinese income is a result of labor reallocation rather than rapid sectoral productivity growth is consistent with this
view.23

341
Another difficulty with this approach is that it relies on the relationship

holding over the sample period. If the entire sample period were one in
which the Chinese economy were adjusting toward a condition under which
the Balassa-Samuelson hypothesis would holdwithout actually achieving
that conditionthen this approach would be inappropriate. This is not a
difficulty specific to the current approach. It also occurs in cases in which one
is empirically validating purchasing power parity in levels but using price
indices. The limitation of such approaches, based on indices, motivates the
use of measurements with which price levels can directly be compared.
13.5
REAL EXCHANGE RATE AND INCOME

REGRESSION
In this section, we appeal to cross-country time-series evidence on the

determinants of the real exchange rate, where the real exchange rate is
measured in such a way that one can identify deviations from absolute
purchasing power parity.
We conduct the analysis in the following manner. First, we appeal to
the well-known cross-sectional relationship between the relative price level
and relative per capita income levels to determine whether the Chinese
currency is undervalued. Obviously, this approach is not novel; it has
been implemented recently by Coudert and Couharde (2005) and Frankel
(2005). However, we expand this approach along several directions. First,
we augment the approach by incorporating the time series dimension.24
Second, we explicitly characterize the uncertainty surrounding our determinations of currency misalignment.
Frankel (2005) exploits the well-known relationship, noted in Summers
and Heston (1991), between the real exchange rate and per capita
income, as recorded in the Penn World Tables. When the real exchange
value of a currency is expressed as the price in common currency terms
(International dollars) relative to the US price level, there is a positive,
monotonic relationship to the relative per capita income.
We amass a large data set encompassing up to 158 countries, over the
19752003 period. (Because some data are missing, the panel is unbalanced.)
Most of the data are drawn from the World Banks World Development
Indicators (WDI) and the Penn World Tables (PWT; Summers and Heston
1991). Since the PWT data end with 2000, we update the sample using the
WDI data up to 2003. Greater detail on the data used in this subsection and
elsewhere is reported in Appendix 13.1.
We estimate the relationship using a pooled time-series cross-section
regression, where all variables are expressed in terms relative to the US.

342
The role of China
The results are reported in Table 13.3, for cases in which we measure
relative per capita income in terms of either market rates or PPP based
exchange rates. Furthermore, to examine the robustness of the results with
respect to different specifications, we report not only the pooled time-series
cross-section estimates (our preferred specification) but also fixed effects
and random effects models. 25
In all cases, the elasticity of the price level with respect to relative per
capita income is always around 0.220.33, which compares favorably with
Frankels (2005) 1990 and 2000 year cross-section estimates of 0.38 and
0.32, respectively.26 Interestingly, the elasticity estimate does not appear to
be sensitive to measurements of per capita income. In Tables 13.4 and 13.5,
the actual and resulting predicted and standard error bands are reported.
We make two observations about these misalignment estimates. First,
the RMB has been persistently undervalued by this criterion since the mid1980s, even in 1997 and 1998, when China was lauded for its refusal to
devalue its currency despite the threat to its competitive position.
Second, and perhaps most importantly, in 2003 the RMB was more than
one standard errorbut less than two standard errorsaway from the
predicted value, which in the present context is interpreted as the equilibrium value. In other words, by the standard statistical criterion that
applied economists commonly appeal to, the RMB is not undervalued (as
of 2003) in a statistically significant sense. Note that this result relies on an
agreement that we have identified the correct model; whether we estimate
the true specification adds another layer of uncertainty.
Figures 13.7 and 13.8 provide a graphical depiction of the actual versus
predicted values (for USD and PPP based per capita incomes), the prediction intervals, and how the RMB fits into the more general relationship.
The wide dispersion of observations in the scatter plots should give pause
to those who would make strong statements regarding the exact degree of
misalignment.
It is interesting to consider the path that the RMB has traced out in
these graphs. It starts out in the samples as overvalued, and over the next
three decades it moves toward the predicted equilibrium value and then
overshoots, so that, by 2003, it is substantially undervaluedby between
47 percent and 54 percent in level terms (greater in log terms) by these point
estimates.
Notice that the deviations from the conditional mean are persistent; this
has an important implication for interpreting the degree of uncertainty
surrounding these measures of misalignment. This suggests that deviations from the PPPrelative income relationship identified by the regression are persistent, or exhibit serial correlation. Frankel (2005) makes a
similar observation, noting that half of the deviation of the RMB from

343

3880
0.245***
(0.003)
0.023**
(0.009)
0.483
Pooled OLS
0.252***
(0.016)
0.041
(0.052)
0.601
Between
0.752
27.871***
0.330***
(0.032)
Fixed effects
(within)
USD-based GDP per capita
3.277*
0.276***
(0.012)
0.026
(0.043)
0.483
Random
effects
3880
0.294***
(0.006)
0.140***
(0.012)
0.338
Pooled OLS
0.297***
(0.028)
0.183***
(0.064)
0.399
Between
0.745
40.466***
0.217***
(0.032)
Fixed effects
(within)
PPP-based GDP per capita
Panel estimation results of the absolute PPP plus (ln(price) 5 c 1 b1*ln(GDPpc) 1 u)
1.410
0.250***
(0.017)
0.272***
(0.044)
0.338
Random
effects
Notes: Unbalanced panel of 174 countries 29 years (19752003). ***, **, and * indicate 1 percent, 5 percent, and 10 percent levels of
significance, respectively. Heteroskedasticity-robust standard errors are in parentheses. N is the number of observations.
Adjusted R2
F-test for
homo. c
Hausman
Chi-sq(1)
N
Constant
GDP p.c.
Table 13.3
344
The role of China
Table 13.4
Actual and predicted price levels for China by panel (pooled

OLS): USD-based GDP per capita
Year
Actual
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
0.245
0.344
0.354
0.308
0.272
0.287
0.486
0.661
0.716
0.872
1.041
1.174
1.233
1.161
1.114
1.331
1.406
1.385
1.313
1.555
1.443
1.399
1.405
1.453
1.490
1.510
1.522
1.530
1.517
Notes:
Predicted 12 std. err.11 std. err. 1 std. err. 2 std. err.

1.249
1.267
1.261
1.248
1.239
1.220
1.215
1.190
1.176
1.160
1.140
1.129
1.111
1.097
1.097
1.093
1.071
1.046
1.021
1.002
0.983
0.969
0.959
0.950
0.944
0.933
0.915
0.900
0.885
0.452
0.470
0.464
0.451
0.442
0.423
0.418
0.393
0.380
0.363
0.343
0.332
0.315
0.300
0.300
0.296
0.274
0.249
0.224
0.205
0.186
0.173
0.162
0.153
0.147
0.136
0.119
0.103
0.088
0.851
0.869
0.863
0.850
0.840
0.822
0.816
0.791
0.778
0.761
0.741
0.730
0.713
0.698
0.699
0.695
0.673
0.648
0.623
0.603
0.585
0.571
0.561
0.552
0.545
0.534
0.517
0.502
0.487
1.647
1.666
1.660
1.647
1.637
1.619
1.613
1.588
1.575
1.558
1.538
1.527
1.510
1.495
1.495
1.491
1.469
1.445
1.420
1.400
1.381
1.367
1.357
1.348
1.342
1.331
1.313
1.298
1.283
2.046
2.064
2.058
2.045
2.035
2.017
2.011
1.986
1.973
1.957
1.936
1.925
1.908
1.894
1.894
1.890
1.868
1.843
1.818
1.798
1.780
1.766
1.756
1.747
1.740
1.729
1.712
1.697
1.682
Values in log terms. Based on Table 13.3, USD pooled estimates.
the 1990 conditional mean exists in 2000. We estimate the autoregressive

coefficient in our sample at approximately 0.74 to 0.84 (based on USD and
PPP based per capita income figures, respectively) on an annual basis. A
simple, ad hoc adjustment based on the latter estimate suggests that the
standard error of the regression should be adjusted upward by a factor
equal to [ 1/ (1 2 r^ 2) ] 0.5 < 2. Figure 13.9 depicts the same data as presented in Figure 13.8, but with the standard errors adjusted to account for
the serial correlation. In this case, the actual value of the RMB is always

345
Table 13.5
Actual and predicted price levels for China by panel (pooled

OLS): PPP-adjusted GDP per capita
Year
Actual
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
0.245
0.344
0.354
0.308
0.272
0.287
0.486
0.661
0.716
0.872
1.041
1.174
1.233
1.161
1.114
1.331
1.406
1.385
1.313
1.555
1.443
1.399
1.405
1.453
1.490
1.510
1.522
1.530
1.517
Notes:
Predicted +2 std. err. +1 std. err. 1 std. err. 2 std. err.

1.176
1.199
1.190
1.175
1.164
1.141
1.134
1.101
1.090
1.068
1.044
1.033
1.010
0.990
0.994
0.991
0.964
0.935
0.906
0.883
0.854
0.837
0.825
0.810
0.802
0.786
0.765
0.750
0.735
0.274
0.297
0.289
0.274
0.263
0.240
0.232
0.200
0.188
0.167
0.143
0.131
0.109
0.089
0.092
0.089
0.063
0.034
0.005
0.019
0.048
0.064
0.076
0.091
0.099
0.115
0.136
0.151
0.166
0.725
0.748
0.740
0.724
0.713
0.691
0.683
0.650
0.639
0.617
0.593
0.582
0.560
0.539
0.543
0.540
0.514
0.485
0.455
0.432
0.403
0.386
0.374
0.359
0.351
0.336
0.315
0.299
0.285
1.627
1.649
1.641
1.626
1.615
1.592
1.584
1.552
1.541
1.519
1.495
1.483
1.461
1.441
1.444
1.441
1.415
1.386
1.357
1.333
1.304
1.288
1.275
1.261
1.253
1.237
1.216
1.201
1.186
2.078
2.100
2.092
2.077
2.066
2.043
2.035
2.003
1.991
1.969
1.946
1.934
1.912
1.891
1.895
1.892
1.866
1.837
1.807
1.784
1.755
1.738
1.726
1.711
1.703
1.688
1.667
1.651
1.636
Values in log terms. Based on Table 13.3, PPP pooled estimates.
within one standard error prediction interval surrounding the equilibrium

value.
13.6
IMPLICATIONS FOR FOREIGN EXCHANGE

POLICY
The key rationale for undertaking this study was our strongly held conviction that many strong policy recommendations were being made on the

346
The role of China

Relative price level
1.5
1
0.5
China 1975
0
0.5
1
1.5
2
China 2003
2.5
3
regression line
Figure 13.7
3.5
3
2
1
0
1
2
Relative per capita income in USD terms
1 standard error
2 standard error
Scatter plot of relative price level against relative per capita

income (in USD), conditional mean, and prediction intervals
1.5
1
0.5
China 1975
0
0.5
1
1.5
2
China 2003
2.5
3
5
3.5
0
Relative per capita income in PPP terms

regression line
Figure 13.8
1 standard error
2 standard error

income (in PPP terms), conditional mean, and prediction
intervals


347
3
2
China 1975
1
0
1
2
China 2003
3
4
0
Relative per capita income in PPP terms

regression line
Figure 13.9
1 standard error
2 standard error

income (in PPP terms), conditional mean, and serial
correlation adjusted prediction intervals
basis of weak empirical evidence or vague theoretical underpinnings. In

particular, we think that there needs to be some humility in recognizing
the limitations of our knowledge regarding the appropriate model and the
true nature of the relevant data.
In this study, contrary to the standard approach, we have documented
the extent of our ignorance. The results should give one pause for thought.
It may be that the Chinese currency is 20 to 40 percent undervalued, as
suggested by the macroeconomic balance approach. But that degree of
misalignment pertains to a view of what constitutes equilibrium that
obtains when the current account balance offsets the normal level of capital
inflows as judged by some usually arbitrary criterion. The normal level, in
turn, pertains to specific macroeconomic conditions that mightor might
notpersist.
The biggest uncertainty pertains to shocks that are difficult to model
but could nonetheless induce large changes to the equilibrium value of
the exchange ratenamely, the shocks that could spring from contingent liabilities and loopholes of capital controls. The mere act of quickly
revaluingor of moving to a relatively free floatmight in itself change
the equilibrium exchange rate if it triggers corporate defaults or causes
changes in the balance sheets of unhedged firms. The end of capital

348
The role of China
controls, either by fiat or by slow erosion, might also alter the equilibrium
exchange rate.
It is also important to note that the regime of currency stability has up
to this point served China well. The refusal of China to accede to competitive devaluation in the wake of the 1997 financial crisis helped facilitate the
recovery of the other East Asian economies. During the same period, the
intensity of integration between China and its neighboring economies, as
measured by intra-regional trade and financial transactions, was probably
encouraged by the stable value of the RMB.
Thus, one natural concern is whether the Chinese economy, given its
fragile financial systems and hidden domestic economic problems, is
capable of handling a floating RMB exchange rate without incurring a substantial domestic economic backlash that would cause substantial repercussions for the regional and even the global economies. Even though some
people perceive China to be a market economy, the reality is that China is
still intrinsically a transitional economy with a relatively primitive and inefficient banking and financial sector. It is still in the early stage of devising
and developing a prudential legal and regulatory framework that promotes
governance and financial stability. It is highly questionable whether the
current Chinese economy can withstand the potential financial instability
induced by full foreign exchange flexibility. A volatile RMB exchange rate
would most likely impede Chinas economic progress.
We also believe that Chinas Asian trading partners would suffer more
from its economic slowdown than they would benefit from a stronger RMB.
A volatile RMB is likely to impose extra costs on the integration between
China and its neighboring economies and hinder the cooperation between
these economies. With a growing role in an uncertain world, Chinas
foreign exchange policy can have unexpected (instead of expected) effects
on the world in general and on other regional economies in particular.
Thus, in designing her exchange rate policy, China should not only consider
the policys short-term and long-term impacts on the Chinese economy but
also take into account the policys implications for the region.
That being said, we do believe that China should pursue a policy of
gradually allowing the RMB to appreciate. Both the macroeconomic
balance and real exchange rateincome models suggest some undervaluation. One can think of these two models as bracketing the short run and
the long run. The productivity based model is perhaps relevant to the idea
of competitiveness over time, and even here there is some doubt. So the
implied direction of change is dependent on the horizon.
Nonetheless, in order to get to the medium run and the long run, one has
to get through the short run. The rapid accumulation of foreign exchange
reserves in 2005in contrast to that in the earlier yearscan be attributed

349
to a burgeoning current account surplus that has placed strains on the

Chinese economy. With the fragile banking system having difficulties with
assimilating this reserve accumulation, it makes sense for Chinese policymakers, in their own interests, to slow the pace of acquisition of dollar
assets. The actual value of the RMB-USD exchange rate is not itself so
crucial in this respect.
NOTES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
We thank Hoyt Bleakley, Henning Bohn, Robert Dekle, Mick Devereux, Michael
Frmmel, Reuven Glick, Linda Goldberg, Koichi Hamada, Randall Henning, Owen
Humpage, Gary Jefferson, Michael Klein, Inpyo Lee, Jaewoo Lee, Ron McKinnon,
Eiji Ogawa, Eswar Prasad, Andy Rose, Margot Schller, Ulrich Volz, Tom Willett,
and participants of the Federal Reserve Bank of San Francisco conference on External
Imbalances and Adjustment in the Pacific Basin, the HWWA/HWWI Conference
on East Asian Monetary and Financial Integration, and the 2006 ACEAS panel
at the ASSA Meetings for helpful comments. Dickson Tam and Jian Wang provided
excellent assistance in collecting data, while Hiro Ito and Michael Spencer provided
us with data. An earlier version of the chapter was circulated under the title Why the
Renminbi Might Be Overvalued (But Probably Isnt). Cheung, Chinn, and Fujii gratefully acknowledge the financial support of faculty research funds from the University
of California at Santa Cruz and the University of Wisconsin and of the Japan Center
for Economic Research grant, respectively. The views expressed are solely those of the
authors and do not necessarily represent the views and opinions of institutions with
which the authors are, or have been, associated.
We use the term China to refer to the Peoples Republic of China, exclusive of Hong
Kong, SAR, Macao, SAR, and Taiwan, ROC, sometimes referred to as Chinese Taipei
or Taipei, China.
Ernst & Young (2006). The subsequent retraction of this report by the company does
not negate the fact that there are substantial amounts of nonperforming loans, the estimation of which is surrounded by considerable uncertainty.
See Calvo and Reinhart (2000) for an explication of the fear of floating
phenomenon.
See among others, Eichengreen (2005), Goldstein (2004), Prasad et al. (2005), and
Williamson (2005).
For a review of the concepts of misalignment and the distinction between short run
and long run disequilibria, see articles in Hinkle and Montiel (1999). As Frankel
(2005) observes, there is a question about whether there is such a thing as an equilibrium exchange rate when there are two or more targets (e.g., internal and external
equilibrium).
Meese and Rogoff (1983). Cheung et al. (2005c, 2005d) present some recent evidence on
this issue.
Note that several authors rely on multiple approaches.
For a technical discussion, see Chinn (2000a).
Also known as BEERs, a composite of exchange rate models.
All the studies reviewed by Cairns imply undervaluation or no misalignment.
See Cheung et al. (2005b) for a description of how financial links between the rest of the
world and China are mediated by capital controls and the banking system.
See Fernald et al. (1999) for a discussion in the context of whether the 1994 devaluation caused the 199798 currency crises.
See Cheung and Lai (1993a) for a motivation for using the cointegration technique to
investigate the stationarity of real exchange rates.

350
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
The role of China

See Appendix 13.2 for detailed results.
Ceglowski and Golub (2005) do calculate relative unit labor costs. If one thought that
unit labor costs should be equalized, this would provide a basis for a calculation of
equilibrium exchange rates, based solely on the tradables (manufacturing) sector.
Note that if the production functions have the same form in the two sectors, then the qs
drop out.
Both equations have been exploited extensively. Equation (13.6) has been examined by
Kakkar and Ogaki (1999) for several exchange rates. Equation (13.7) has been estimated
by Hsieh (1982), Marston (1990), and most recently Choudhri and Khan (2004) and
Tang and Lee (2003).
Application of the Johansen maximum likelihood procedure indicates evidence of
cointegration at the 5 percent marginal significance level, using asymptotic critical
values. The procedure is implemented by including two lags of the first differences (i.e., a
VAR(3) in levels). The cointegration result holds irrespective of whether a deterministic
trend is allowed for in the cointegrating vector or not.
Unfortunately, we could not obtain long enough series of sector-specific deflators, so we
used the aggregate GDP deflator reported in Holz (2005) to deflate the sectoral output
reported in the ADBs Indicators of Developing Asian and Pacific Countries. The employment figures are also drawn from this source.
Since the DOLS procedure here uses one lead and lag of the right hand side variables, the
sample is truncated to 2002 even when 2003 data are available. The levels observations
for 2003 and 2004 are not used in the estimation procedure.
However, the results are not similar if the coefficients on the traded and nontraded
sectors are constrained to be equal and opposite. The impact of productivity differentials
disappears in this longer sample, unless a time trend is included. The sensitivity of the
results to the inclusion of time trends is another reason to focus on the shorter sample.
This argument is most closely associated with Young (2003), as well as Brandt et al.
(2005).
Coudert and Couharde (2005) implement the absolute PPP regression on a crosssection, while their panel estimation relies on estimating the relationship between the
relative price level to relative tradables to nontradables price indices.
Since the price levels being used are comparable across countries, in principle there is
no need to incorporate currency-specific constants as in fixed effects or random effects
regressions. In addition, fixed effects estimates are biased in the presence of serial correlation, which is obviously present in the data.
In addition to the obvious difference in the sample, our estimates differ from Frankels
in that we measure each countrys (logged) real GDP per capita in terms relative to
the US rather than in absolute terms. Hence, the resulting coefficient estimates are not
necessarily directly comparable.
REFERENCES
Arndt, Sven (1997). Globalization and the Open Economy. North American
Journal of Economics and Finance 8: 719.
Balassa, Bela (1964). The Purchasing Power Parity Doctrine: A Reappraisal.
Journal of Political Economy 72: 58496.
Beck, Thorsten, Asli Demirgc-Kunt, and Ross Levine (2000). A New Database
on Financial Development and Structure. Policy Research Paper no. 2147,
World Bank, Washington, DC.
Bosworth, Barry (2004). Valuing the Renminbi. Paper presented at the Tokyo
Club Research Meeting, February 910.
Brandt, Loren, Changtai Hsieh, and Xiaodong Zhu (2005). Growth and Structural

351
Transformation in China. In Loren Brandt, Gang Lin, and Thomas G. Rawski

(eds), Chinas Economy: Retrospect and Prospect. Asia Program Special Report
no. 129. Washington, DC: Woodrow Wilson International Center for Scholars.
Cairns, John (2005a). China: How Undervalued is the CNY? IDEAglobal
Economic Research, June 27.
Cairns, John (2005b). Fair Value on Global Currencies: An Assessment of
Valuation based on GDP and Absolute Price Levels. IDEAglobal Economic
Research, May 10.
Calvo, Guillermo, and Carmen Reinhart (2000). Fear of Floating. NBER
Working Paper no. 7993, National Bureau of Economic Research, Cambridge,
MA.
Canzoneri, Matthew, Robert Cumby, and Behzad Diba (1996). Relative Labor
Productivity and the Real Exchange Rate in the Long Run: Evidence for a Panel
of OECD Countries. Journal of International Economics 47: 24566.
Ceglowski, Janet, and Stephen Golub (2005). Just How Low Are Chinas Labor
Costs? Manuscript, Bryn Mawr College and Swarthmore College.
Cheung, Yin-Wong, Menzie Chinn, and Eiji Fujii (2005a). Why the Renminbi
Might be Overvalued (But Probably Isnt). Working paper, UCSC, Santa
Cruz.
Cheung, Yin-Wong, Menzie Chinn, and Eiji Fujii (2005b). Perspectives on
Financial Integration in the Chinese Economies. International Journal of
Finance and Economics 10: 11732.
Cheung, Yin-Wong, Menzie Chinn, and Antonio Garcia Pascual (2005c).
Empirical Exchange Rate Models of the Nineties: Are Any Fit to Survive?
Journal of International Money and Finance 24: 115075.
Cheung, Yin-Wong, Menzie Chinn, and Antonio Garcia Pascual (2005d).
What Do We Know About Recent Exchange Rate Models? In-Sample
Fit and Out-of-Sample Performance Evaluated. In Paul De Grauwe (ed.),
Exchange Rate Economics: Where Do We Stand? Cambridge, MA: MIT
Press, pp. 23976.
Cheung, Yin-Wong, and Kon. S. Lai (1993a). Long-run Purchasing Power Parity
during the Recent Float. Journal of International Economics 34: 18192.
Cheung, Yin-Wong, and Kon. S. Lai (1993b). Finite-sample Sizes of Johansens
Likelihood Ratio Tests for Cointegration. Oxford Bulletin of Economics and
Statistics 55: 31328.
Chinn, Menzie (2000a). Before the Fall: Were East Asian Currencies Overvalued?
Emerging Markets Review 1: 10126.
Chinn, Menzie (2000b). The Usual Suspects? Productivity and Demand Shocks
and Asia-Pacific Real Exchange Rates. Review of International Economics 8:
2043.
Chinn, Menzie, and Hiro Ito (2006). What Matters for Financial Development?
Capital Controls, Institutions and Interactions. Journal of Development
Economics 81: 16392. (Abbreviated version of NBER Working Paper no.
11370, National Bureau of Economic Research, Cambridge, MA.)
Chinn, Menzie, and Eswar Prasad (2003). Medium-term Determinants of Current
Accounts in Industrial and Developing Countries: An Empirical Exploration.
Journal of International Economics 59: 4776.
Choudhri, Ehsan, and Mohsin Khan (2004). Real Exchange Rates In Developing
Countries: Are Balassa-Samuelson Effects Present? IMF Working Paper no.
04/188, International Monetary Fund, Washington, DC.

352
The role of China
Coudert, Virginie, and Ccile Couharde (2005). Real Equilibrium Exchange Rate
in China. CEPII Working Paper no. 2005-01, Centre DEtudes Prospectives et
DInformations Internationales, Paris.
Dooley, Michael, David Folkerts-Landau, and Peter Garber (2003). An Essay on
the Revived Bretton Woods System. NBER Working Paper no. 9971, National
Eichengreen, Barry (2005). Is a Change in the Renminbi Exchange Rate in Chinas
Interest? Asian Economic Papers 4: 4075.
Ernst & Young (2006). Global Nonperforming Loan Report 2006. London: EYGM
Limited.
Fernald, John, Hali Edison, and Prakash Loungani (1999). Was China the First
Domino? Assessing Links between China and Other Asian Economies. Journal
of International Money and Finance 18: 51535.
Frankel, Jeffrey (2005). On the Renminbi: The Choice between Adjustment under
a Fixed Exchange Rate and Adjustment under a Flexible Rate. NBER Working
Paper no. 11274, National Bureau of Economic Research, Cambridge, MA.
Funke, Michael, and Jrg Rahn (2005). Just How Undervalued is the Chinese
Renminbi? World Economy 28: 46589.
Goldstein, Morris (2004). China and the Renminbi Exchange Rate. In C. Fred
Bergsten and John Williamson (eds), Dollar Adjustment: How Far? Against
What? Special Report no. 17. Washington, DC: Institute for International
Economics.
Government Accountability Office (2005). International Trade: Treasury
Assessments Have Not Found Currency Manipulation, but Concerns about
Exchange Rates Continue. Report to Congressional Committees GAO-05-351,
Government Accountability Office, Washington, DC.
Hinkle, Lawrence E., and Peter J. Montiel (1999). Exchange Rate Misalignment.
New York: Oxford University Press/World Bank.
Holtz-Eakin, Douglas (2005). Economic Relationships between the United States
and China. Statement of the CBO Director before the House Committee on
Ways and Means, April 14.
Holz, Carsten (2005). Chinas Economic Growth 19782025: What We Know
Today about Chinas Economic Growth Tomorrow. Manuscript, HKUST,
Hong Kong.
Horvath, Michael, and Mark Watson (1995). Testing for Cointegration when
Some of the Cointegrating Vectors Are Prespecified. Econometric Theory 11:
9841014.
Hsieh, David (1982). The Determination of the Real Exchange Rate: The
Productivity Approach. Journal of International Economics 12: 35562.
Kakkar, Vikas, and Masao Ogaki (1999). Real Exchange Rates and Nontradables:
A Relative Price Approach. Journal of Empirical Finance 6: 193215.
Marston, Richard (1990). Systematic Movements in Real Exchange Rates in the
G-5: Evidence on the Integration of Internal and External Markets. Journal of
Banking and Finance 14: 102344.
Meese, Richard A., and Kenneth Rogoff (1983). Empirical Exchange Rate
Models of the Seventies: Do They Fit Out of Sample? Journal of International
Economics 14: 324.
Prasad, Eswar, Thomas Rumbaugh, and Qing Wang (2005). Putting the Cart
Before the Horse? Capital Account Liberalization and Exchange Rate Flexibility

353
in China. IMF Policy Discussion Paper no. 05/01, International Monetary

Fund, Washington, DC.
Prasad, Eswar, and Shang-Jin Wei (2005). The Chinese Approach to Capital
Inflows: Patterns and Possible Explanations. NBER Working Paper no. 11306,
National Bureau of Economic Research, Cambridge, MA.
Rossi, Vanessa (2005). The Chinese Economy: Risky Reporting. International
Economics Programme Briefing Paper no. 05/03, Chatham House, London.
Samuelson, Paul (1964). Theoretical Notes on Trade Problems. Review of
Economics and Statistics 46: 14554.
Stock, James H., and Mark W. Watson (1993). A Simple Estimator of Cointegrating
Vectors in Higher Order Integrated Systems. Econometrica 61: 783820.
Summers, Robert, and Alan Heston (1991). The Penn World Table (Mark 5): An
Expanded Set of International Comparisons. Quarterly Journal of Economics
106: 32768.
Szirmai, Adam, Ruoen Ren, and Manyin Bai (2005). Chinese Manufacturing
Performance in Comparative Perspective, 19802002. Economic Growth
Center Discussion Paper no. 920, Yale University, New Haven.
Tang, Man-Keung, and Jaewoo Lee (2003). Does Productivity Growth Lead to
Appreciation of the Real Exchange Rate? IMF Working Paper no. 03/154,
International Monetary Fund, Washington, DC.
Wang, Tao (2004). Exchange Rate Dynamics. In Eswar Prasad (ed.), Chinas
Growth and Integration into the World Economy. Occasional Paper no. 232.
Washington, DC: IMF, pp. 2128.
Williamson, John (2005). A Currency Basket for East Asia, Not Just China.
Policy Briefs in International Economics PB05-1, Institute for International
Yi, Kei-Mu (2003). Can Vertical Specialization Explain the Growth of World
Trade? Journal of Political Economy 111: 53102.
Young, Alwyn (1995). The Tyranny of Numbers: Confronting the Statistical
Realities of the East Asian Growth Experience. Quarterly Journal of Economics
110: 64180.
Young, Alwyn (2003). Gold into Base Metals: Productivity Growth in the
Peoples Republic of China during the Reform Period. Journal of Political
Economy 111: 122061.
Zhang, Zhichao (2001). Real Exchange Rate Misalignment in China: An
Empirical Investigation. Journal of Comparative Economics 29: 8094.

354
The role of China
APPENDIX 13.1
DATA AND SOURCES
For Section 13.1

The nominal Renminbi exchange rate is the bilateral period average,
expressed against the USD (in USD/fcu; i.e. dollars per foreign currency
unit), obtained from the IMFs International Financial Statistics, and
from Hali Edison, for the adjusted exchange rates (Fernald et al. 1999).
The Consumer Price Indexes (CPIs) are drawn from the CEIC database,
extrapolated for 2004 and 2005 by using the CPI growth rates reported
in IFS. The CPI-deflated trade-weighted exchange rate is drawn from
IFS, while the Producer Price Index (PPI) deflated series was provided by
Michael Spencer at Deutsche Bank.
For Section 13.3
The nominal renminbi exchange rate source is described in section
13.1. Tradables and nontradables are proxied by manufacturing and by
other (which includes mostly services, construction, and transportation),
respectively.
For the United States, labor productivity in manufacturing: BLS (Foreign
Labor Statistics website) International Comparisons of Manufacturing
Productivity and Unit Labor Cost Trends, Supplementary Tables, 1950
2003, www.bls.gov/fls/prodsupptabletoc.htm. US productivity in 2004
calculated from output and hours from BLS website, accessed August 21.
Labor productivity in other for 19792002 from 60-Industry Database
(latest update February 2005), www.ggdc.net/dseries/60-industry.shtml
from Groningen Growth and Development Center. The 1995 chained
value added data and employment numbers for all industries above SIC 3.
Productivity growth rate for 200304 set at the 2002 rate.
For China, the basic productivity data are calculated as the ratio of
nominal sectoral output to sectoral employment, as reported in the Asian
Development Banks Key Indicators of Developing Asian and Pacific
Countries. The Chinese real output series are obtained by deflating using
the GDP price deflator (base year 2000), as reported in Holz (2005). The
alternative manufacturing productivity index is drawn from Szirmai et al.
(2005). The 2003 data assume 2002 productivity growth continues; 2004
figures for all Chinese data are extrapolated from 2003 growth rates.
Some additional results are also reported using a 19802004 sample.
These results are based on spliced series, where 198085 data on Chinese
productivity are drawn from the database used in Chinn (2000b).
Manufacturing productivity is that reported by the World Banks World

355
Development Indicators, while Other productivity is generated in the

method described using data from the ADB and the ILO. The effective
exchange rate is assumed to equal the official over the 198086 period.
For Section 13.4
The data for macroeconomic aggregates are mostly drawn from the World
Banks World Development Indicators. These include demographic variables, per capita income and government deficits. Relative price levels and
per capita income are drawn from the Penn World Tables (Summers and
Heston, 1991), as drawn from pwt.econ.upenn.edu/. Financial development indicators, including lending, stock and bond market capitalization,
are drawn from the Beck et al. (2000). The capital controls index is from
Chinn and Ito (2006). The (inverse) corruption index is drawn from the
International Country Risk Guide. Data for Taiwan are drawn from the
central bank of China, ICSEAD, and ADB, Key Indicators of Developing
Asian and Pacific Countries.

356
The role of China
APPENDIX 13.2
RELATIVE PURCHASING POWER

PARITY TESTS
Bilateral
Trade weighted
Panel A: Johansen cointegration tests

k
#,#[#,#]
Spec.
b1
b2
3
1,1[1,1]
w/o
1
1.503***
(0.125)
3
2,2[2,2]
w/trend
1
1.455***
(0.118)
2
1,0[0,0]
w/o
1
0.765*
(0.168)
2
2,0[2,0]
w/trend
1
0.828
(0.162)
Panel B: Horvath-Watson cointegration tests

Wald
Smpl
N
12.556
11.427
1987:42005:2 1987:42005:2 1987:12005:1 1987:12005:1
71
71
73
73
Notes: Panel A: k is lag in VAR specification. #,#[#,#] is the number of cointegrating

vectors according to a likelihood ratio test on the trace, maximal eigenvalue statistic, using
asymptotic [finite sample] critical values. Spec. indicates whether a deterministic trend is
assumed to be in the data (w/trend), or not (w/o). Finite sample critical values are
from Cheung and Lai (1993b). b1 are cointegrating vector coefficients. *(**)[***] denotes
significance at the 10 percent (5 percent)[1 percent] MSL for the null hypothesis of b2 5 1.
Panel B: Wald is the Wald test statistic for the joint null that the reversion coefficients in the
VECM representation are zero. Critical values are 9.72 (11.62)[15.41], from Horvath and
Watson (1995). Smpl is sample, N is the number of observations.

14.
Institutional and structural

problems of Chinas foreign
exchange market and the RMBs
role in East Asia
Zhang Jikang and Liang Yuanyuan1
14.1
INTRODUCTION
One big question mark with respect to East Asian monetary integration is
the Chinese currencys potential role in this process. In the 1980s and 1990s,
the Japanese government tried to promote the yen as the key regional currency. But the yen bloc, as it was often referred to, never came into being,
despite the yens status as one of the leading international currencies as well
as the existence of highly developed and liquid financial markets in Japan.2
Today, after years of deflation and economic weakness in Japan, the prospects of a yen bloc have diminished even further. The failure to establish
the yen as the regional anchor currency in East Asia is also a result of noneconomic factorsthat is, the other East Asian countries are not willing
to accept a Japanese leadership role in East Asia for political and historical
reasons. This puts the spotlight on the Chinese renminbi (RMB).
Several authors have predicted a growing orientation of East Asian
countries exchange rate policies toward that of China (see, for example,
Ho et al. 2005; and Hefeker and Nabor 2005). But there are many
uncertainties surrounding the RMBs prospects of becoming the leading
currency in the region, such as whether China will be able to sustain her
economic growth path and maintain political stability, and whether China
will be able to develop deep and liquid primary and secondary financial
markets. In this chapter, we focus on another aspect that is crucial for
developing key currency status, that is, the development of a fully functioning foreign exchange (FX) market. In particular, we investigate the
institutional and structural impediments facing the Chinese FX market,
taking into account the most recent reforms that accompanied the official
change of the Chinese exchange rate regime to a currency basket in July
2005.
357

358
The role of China
Our analysis shows that, while China has made considerable progress
in reforming its FX market, for the time being the RMB is far from being
prepared to take over the lead role in regional monetary integration and
to act as a regional anchor currency. The FX market is hampered by low
liquidity, high market concentration, limited transaction instruments, distorted market supply and demand, and passive intervention by the Peoples
Bank of China (PBOC).
This chapter is structured as follows: the following section gives a
brief historical overview of the development of Chinas foreign exchange
market, followed by an explanation of its current institutional structure in
section 14.3. Section 14.4 discusses structural features of the FX market,
and section 14.5 analyzes its problems. Section 14.6 discusses the recent
reform measures in detail. Section 14.7 summarizes and gives recommendations for further FX market reforms.
14.2
THE DEVELOPMENT OF CHINAS FOREIGN

EXCHANGE MARKET
Pre-1979: Strict Central Control

Before 1979, China had a highly centralized regime governing the supply,
demand, and allocation of foreign exchange. All FX earnings (mainly
export proceeds) had to be surrendered to the state-owned banks and the
PBOC. All FX expenditures (i.e. for imports or non-trade purposes) had
to be approved under the confines of the national FX plan, which leaned
toward the state sector.3 There was no market element in the formation of
the exchange rate, which was fixed to the British pound from 1952 and then
to a basket of international currencies from 1973.
197993: FX Retention and Swap
Liberalization of FX use began in 1979 with an earnings retention scheme
designed to encourage exports. Under this scheme, exporters were entitled
to retain a share of their FX earnings, initially with respect to exports
above some quota but from 1998 according to the full measure of exports.
From 1981 to 1984, exchange rates were set differentially for trade and
non-trade activities.4
The first sign of an actual FX market in China appeared in October
1980 when retained FX claims became transferable, first through the
swap service launched by the PBOC, then in provincial swap centers, and
finally in an integrated nationwide swap market.5 The late 1980s saw the

Chinas foreign exchange market and the RMBs role in East Asia
359
expansion of FX swap activity, with the number of market participants

increasing and swap exchange rates becoming more flexible. By the end
of 1993, there were 108 local swap centers and 18 markets joined to the
nationwide system. A mechanism for forming a market exchange rate had
thereby been established in coexistence with an official pegged rate.
The development of the swap market with its diversity of swap rates had
its own problems, including discrimination, rent-seeking, and unauthorized actions. Nevertheless, it acted as a transitional device, laying a foundation for the emergence of a true FX market in China. The 197994 period
also saw frequent adjustments of the official exchange rate, with a trend
devaluation of the RMB against US dollar.6 After the gradual devaluations
of the official rate, the central bank was well prepared to unify the exchange
rates and reform the exchange rate regime.
1994 and Post-1994: Compulsory Settlement on a Centralized Platform
The year 1994 was a turning point in Chinas FX reform. In that year, the
system of FX retention and submission that had existed for 15 years was
replaced with a compulsory settlement system under which foreign exchange
earners were obligated to sell their FX to state banks, while foreign exchange
users could buy it subject to certain conditions. On April 1, 1994, the single
managed floating exchange rate regime based on market supply and demand
was adopted. On April 4, 1994, the China Foreign Exchange Trading System
(CFETS) began operation, thus launching Chinas first unified national
inter-bank FX market. The governments FX management method was also
adjusted to rely more on systematic economic and legal measures, in contrast
with the former command approach. New rules governing the purchase of
FX by individuals for overseas visits, study, or other personal needs took
effect on April 1. These rules have been relaxed gradually over time, with the
upper limit on individual purchases raised to USD 8000 as of August 2005
from an initial USD 600 per person per visit in 1994. The success of the 1994
reform enabled conditional convertibility of the RMB under the current
account and brought a real FX market into existence.
Reform continued under the basic framework of the FX purchase and
sale system after 1994. In 1995, China added yen trading and ended the
circulation of foreign exchange certificates. In July 1996, the FX transactions of Foreign Invested Enterprises (FIEs) were integrated into the FX
purchase and sale system, allowing FIEs to buy foreign currency freely on
the inter-bank market. On November 27, 1996, China formally notified the
International Monetary Fund of the RMBs convertibility on the current
account. In 1997, qualified Chinese companies were allowed to open FX
settlement accounts to retain a portion of FX earnings from current account

360
The role of China
transactions. The PBOCs forward FX purchase and sale experiment was

launched in 1997.
The 199798 Asian financial crisis interrupted Chinas aggressive reform
timetable. A series of regulations were enacted and clarified by the PBOC
to strengthen the responsibilities of the State Administration of Foreign
Exchange (SAFE) with respect to falsely obtaining FX, failing to surrender FX, illegal arbitrage, and so on. The swap centers were closed on
December 1, 1998, with all FX transactions thereafter integrated into the
FX purchase and sale system.
In 2001, trading in dollar-denominated B shares on Chinas stock market
was opened to Chinese nationals (having formerly been limited to foreign
passport holders) with necessary currency exchange supported by the
CFETS. Trading in euros was introduced in April 2002. Then, in October
2002, all enterprises that qualified for conducting international business or
had regular FX incomes from current account transactions were allowed
to open foreign currency accounts for holding up to 20 percent of their
previous years FX income. Two-way trading was permitted from October
2003, with trading hours extended from half to full day.
Another landmark in the development of Chinas FX market was 2005.
Eight foreign currency pairs started trading in the inter-bank market in
May 2005. On July 21, the RMB was revalued with a 2 percent appreciation
against the US dollar, and the peg to the US dollar was replaced with a reference to a basket of currencies. The July reform was followed by successive
reform measures in August7 to enlarge the scope of forward FX purchase and
sale and swap between the RMB and foreign currencies to the retail market
between banks and their customers, to invite non-financial enterprises and
non-bank financial institutions to participate in the inter-bank market,
to add a quote-driven dealer market trading mechanism into the current
auction market, and to introduce inter-bank FX forward and swap trading.
In addition, a long expected market maker system for USDRMB trade was
finally announced in November 2005 based on the CFETS platform.8
14.3
THE CURRENT ORGANIZATION OF THE

FOREIGN EXCHANGE MARKET
Chinas FX market is composed of two parts, the inter-bank or wholesale

market and the retail market (see Figure 14.1). The major parties involved
in the FX market are as follows: (1) the CFETS, which functions as the
trading platform for the inter-bank market and is responsible for clearing the market and for providing the supervisory authorities with market
information; (2) the PBOC and SAFE, which function as regulatory

361
Regulatory Authority
PBOC
Authorize
Open Market
Operations
SAFE
Retail Market
Wholesale (Inter-Bank)
Market
Non-Financial
Enterprises*
CFETS
Trading Platform
Designated FX
Bank
Individuals
Non-bank Financial
Institutions
Enterprises
Note: * The new rules in August 2005 gave permission for non-financial enterprises to
trade directly in the inter-bank market, but the first actual market entry license did not
emerge until March 2006.
Figure 14.1
Organization of Chinas foreign exchange market
authorities (the PBOC authorized SAFE to regulate the inter-bank spot

and forward markets and regulates the retail market through SAFE); (3)
designated FX banks and other non-bank financial institutions9 authorized
by SAFE to engage in foreign exchange business; (4) enterprises that earn
and spend FX, whose activities were restricted within the scope of the retail
market until the FX market reforms in August 2005, which granted them
the right to trade directly in the wholesale market; and (5) individuals who
have FX trading needs.
The nature of the inter-bank FX market is for designated FX banks to
square their FX positions derived from retail FX business after maintaining an allowable FX working position.10 Before the August 2005 reform,
the inter-bank market had been a centralized, order-driven auction market
making use of organized exchange trading and an electronic broking
system, while retail transactions had been carried out in an over-the-counter

362
The role of China
(OTC) market. In May 2005, a new trading system was introduced on the
CFETS platform for the trading of eight foreign currency pairs, pioneering the use of market making and request for quote trading arrangements
before these were further applied to the inter-bank forward market and the
inter-bank spot market in later reforms. Currently, Chinas FX market is a
quasi-centralized hybrid market with a quote-driven dealer market system
facilitated by market makers going hand in hand with the original orderdriven auction market utilizing an electronic broking system.11
As an administrative unit of the PBOC, the CFETS is not only a trading
platform but also serves as an information and supervision platform. The
CFETS trading services mainly cover the foreign exchange market, the
bond market, and the bill market. In the foreign exchange market, the core
of the CFETS platform is a membership-based exchange with a nationwide
real-time electronic trading system. With an auction market trading
mechanism similar to that of an electronic broker, members make back-toback (anonymous) quotes on the trading platform through either on-site or
distant trading terminals.12 The electronic trading platform automatically
enables real-time matching of orders. The clearing function is integrated
into the CFETS platform, providing members with centralized, two-way
netting/clearing.
Since the introduction of eight foreign currency pairs and the August
2005 market reform, the CFETS trading platform has been updated to
incorporate new features, such as a market making system and request for
quote trading arrangements. We discuss the recent FX market reforms in
greater detail in section 14.6.
This market infrastructure was originally put in place to serve the needs
of the compulsory FX settlement system and facilitate the PBOCs absolute control of the market. But now it seems that CFETS must undergo
major transformation with respect to market-orientation, services offered,
technology, efficiency, and risk management. The possible direction of
reform will lean toward establishing parallel sub-trading systems targeting
different market segments. In the meantime, the CFETS should become
more independent of the central bank.13
14.4
STRUCTURAL FEATURES OF THE FX

MARKET
Market Size
Chinas inter-bank FX market is still small and shallow, even though
market size has seen a large increase in recent years. The average daily

Table 14.1
363
Foreign exchange market daily average turnover in selected

markets, 19952004 (USD billion/day)
China
Hong Kong
Japan
Singapore
United Kingdom
United States
1995
1998
2001
2004
0.26
90
161
105
464
244
0.21
79
136
139
637
351
0.30
67
147
101
504
254
0.83
102
199
125
753
461
Note: For markets other than Chinas, daily averages are for the month of April and
cover spot, forward, and swap transactions. For China, volume is based on the entire year
and pertains only to inter-bank spot transactions.
Source:
CFETS, BIS (2004).
turnover in Chinas inter-bank FX market is very low compared to that on

the worlds major markets (see Table 14.1). But Chinas market is growing
quickly, with daily turnover in 2004 up by 177 percent relative to 2001
and in 2001 by nearly half relative to 1998. By contrast, all major markets
except Japan experienced a decline in turnover between 1998 and 2001,
with recovery then following in 2004.14
The small scale of Chinas FX market is attributable in part to institutional factors. Restrictions on FX holdings for financial institutions and
non-financial enterprises, as well as for individuals, figure significantly
in stunting the development of the market. So, too, do capital control
measures that discourage cross-capital flows. Other structural features
also have a hand in limiting the market size. Chinas FX market has
been experiencing a high concentration of FX trading among a few large
banks that balance trades internally, turning to the inter-bank market
just once a day to readjust their reserves to the allowable level (see the
discussion later in this section, on market participants and concentration). Market entry requirements that restrict the pool of market participants, prohibition of two-way trading (both buying and selling) during
the same trading session until October 2003, and the limited scope of
products have all played a part in keeping the market both small and
shallow.
The recent dramatic growth in Chinas inter-bank market follows a sluggish start in the 1990s (see Figure 14.2). The years 1997 to 1999 actually
showed a downturn, from which the market did not recover until 2001.
This downturn is attributable in part to the adverse impact of the Asian

364
The role of China

USD bil/day
1.0
USD bil/yr
250
Daily average turnover
Annual turnover
200
0.8
150
0.6
100
0.4
50
0.2
0.0
0
94
Source:
95
96
97
98
99
00
01
02
03
04
China Money, various issues.
Figure 14.2
Annual and daily average turnover in Chinas inter-bank FX

market, 19942004
financial crisis but also to a broad-based inspection campaign carried out

by SAFE to ferret out the purchase of FX under false pretences, and other
illegal market activity. With the recovery of economic growth domestically, Chinas admission to WTO, and improvement in the foreign trade
and investment balance, the market picked up momentum, entering a
period of fast growth from 2001 to 2004, with an average growth rate of
50.32 percent. Market turnover reached a new height of USD 209 billion
in 2004, with a daily average of USD 830 million.
The rising turnover on Chinas inter-bank market coincided with rapid
growth in balance of payments flows (see Figure 14.3). Under the compulsory FX purchase and sale system, the inter-bank market functions solely
for banks to net out FX positions derived from retail trade. This retail trade
is in turn driven by bank customer activity captured in the current and
capital accounts. In recent years, short-term capital inflows have played
a major role in feeding the increase in supply on the FX market. Besides
economic fundamentals, this sharp increase in market turnover was driven
also by short-term factors on the supply side, including faster settlement
of export revenues as a result of the 3 percent reduction in the export tax
rebate in 2004, speculation on RMB appreciation, the reductions by banks
of their FX working positions to relieve the RMB demand pressure, the
repatriation of capital raised in stock listings by overseas listed companies,
and the conversion of foreign funds for investment in Chinas stock market
by qualified foreign institutional investors.

365
1400
1200
1000
800
600
400
200
0
Note:
Source:
94
95
96
97
98
99
00
01
02
03
Current account 2-way flow
Capital account 2-way flow
Inter-bank FX market turnover
PBOC net FX purchase
04
PBOC net purchase of FX is measured by the change in official reserves assets.

CFETS, PBOC Quarterly Statistical Bulletin.
Figure 14.3
FX market activity and balance of payment flows, 19942004

(USD billion)
From 1994 to 2004, spot trading in four currency pairs (RMB against
US dollar, RMB against HK dollar, RMB against Japanese yen, and RMB
against euro) constituted the whole inter-bank market turnover. But since
market segments were enriched after several market reforms in August
2005, the total market size now also covers trading in eight additional
foreign currency pairs (for a total of 12) as well as inter-bank forwards.
Figures 14.4 and 14.5 give an overview of the market activity in the eight
foreign currency pairs currently trading. It can be seen that, though
monthly volumes vary a lot, trading in these currency pairs is already an
important addition to the total turnover in Chinas FX market.
The inter-bank forwards started trading in August 2005, but, judging
by the number of transactions (see Table 14.2), the market turnover is still
lingering at a low level and can make little contribution to the total market
volume.
With more diversified trading products, Chinas inter-bank FX market
will be able to attract more trading interest in the long term. Market
size will also benefit from increasing volume brought by newly launched
market segments, such as foreign currency pairs and inter-bank forwards,
but so far this market still lacks depth and breath, and a substantial growth
in market size has yet to come.

366
The role of China
14.00
0.60
12.00
0.50
10.00
0.40
8.00
0.30
6.00
0.20
4.00
0.10
2.00
0.00
0.00
May 05
Jun 05
Jul 05
Aug 05
Total Turnover
Sep 05
Oct 05
Nov 05
Dec 05
Daily Turnover
Source: China Money, various issues.
Figure 14.4 Monthly and daily turnover in foreign currency pairs (USD
billion)
The July 2005 revaluation heralded reforms in the RMB exchange rate
regime. With more flexibility in the exchange rate formation mechanism
being introduced in the future, China will be expecting larger variations in
its FX market. A small and shallow market is certainly no boon to smoothing exchange rate variations. This may add to authorities concern about
allowing further flexibility in the exchange rate regime and cause delay in
the reform timetable; on the other hand, without a more flexible rate, a
truly deep market will be hard to achieve.
Market Segments
Chinas FX market is limited in product scope mainly to spot trading in
US dollars. For a long time since its establishment, the inter-bank market
only offered spot transactions between the RMB and the US dollar, the HK
dollar, the Japanese yen, and the euro, respectively. It was not until May
2005 that FX trading was expanded (from the original four) to include eight
new foreign currency pairs and not until August 2005 that inter-bank forwards were introduced.15 In the retail market, trading is also dominated by
spot transactions, with a small forward market and a nascent swap market.
The Bank of China (BOC, not to be confused with the PBOC) was the
first financial institution allowed to offer forwards, beginning in 1997,

Turnover in the
foreign currency
pairs from May 2005
to December 2005
367
52.13
Turnover in the
original spot market
in the first half of
2005
146.15
20
40
60
80
100
120
140
160
Note: As Chinas central bank stopped the practice of disclosing inter-bank turnover
figures in the spot trading market involving the RMB in the second half of 2005, a
comparison between these two segments across the same time range cannot be made.
Source: China Money, various issues.
Figure 14.5
Table 14.2
Turnover in Foreign Currency Pairs and the Original Spot

Market (USD billion)
Monthly number of transactions in the inter-bank forward
market
August 2005
September 2005
October 2005
November 2005
December 2005
Source:
8
23
38
94
114
with the other banks following suit after 2002.16 This market segment was
opened up to all financial institutions during the August 2005 reform, when
retail market FX swaps also started trading.
With the inception of forward transactions in 1997, the BOCs trading
volume rose to a peak of USD 11.5 billion in 2000 (see Figure 14.6). This
growth reflected a need for businesses to hedge against currency risk during
a period when the RMB was under pressure to depreciate in the wake of the

368
The role of China

2004
2003
2002
2001
2000
1999
1998
1997
0.00
5.00
10.00
15.00
20.00
Forward trading
Source: Data for the years 19972002 are from Bank of China Annual Reports, while data
for the years 200304 are estimations made by Zhang (2005).
Figure 14.6
Retail market forward FX trading (USD billion)
RMB/USD 100
875.00
870.00
865.00
860.00
855.00
850.00
845.00
840.00
835.00
830.00
825.00
820.00
815.00
810.00
805.00
800.00
Band narrowed after the

Asian Financial Crisis
Ja
n
Ju 94
Ja l 94
n
Ju 95
Ja l 95
n
Ju 96
Ja l 96
n
Ju 97
Ja l 97
n
Ju 98
Ja l 98
n
Ju 99
Ja l 99
n
Ju 00
Ja l 00
n
Ju 01
Ja l 01
n
Ju 02
Ja l 02
n
Ju 03
Ja l 03
n
Ju 04
Ja l 04
n
Ju 05
Ja l 05
n
06
July 2005 revaluation
Source:
Year
PBOC.
Figure 14.7
RMBUSD exchange rate, January 1994June 2006
Asian financial crisis. Rather than devalue, though, the PBOC responded
to the crisis by tightening the floating band. This led to a highly stable
relationship between the RMB and the US dollar (see Figure 14.7). Under
such conditions, the need to hedge risk diminished, and the forward market
contracted. In 2002, the BOCs forward trading volume declined by about

369
Spot
35%
Spot
Forward
Swap
53%
Swap
Forward
12%
Source:
BIS (2004).
Figure 14.8
Shares of FX trading by transaction type in global FX

markets, 2004
two-thirds from its peak, but the market has rebounded since 2003, when
more players entered the market.
Compared with the global FX market (see Figure 14.8), the lack of
product range in the Chinese market restricts overall growth in market
turnover and limits the functions of the market, especially the risk-hedging
function.
Chinas FX market is also limited by its trading concentration in the US
dollar. When the inter-bank market was established in 1994, only the US
dollar and the Hong Kong dollar were traded. The yen was added in 1995
and the euro in 2002. The US dollar, however, remains the overwhelmingly
dominant currency, accounting for 97.8 percent of total turnover in the
first half 2005.17
The dominance of the US dollar actually strengthened during the late
1990s and was little influenced by the introduction of the euro in 2002 (see
Figure 14.9). The high concentration in US dollar trading is not inconsistent with the important role the dollar plays in global trade and investment
as a vehicle currency (see Figure 14.10). Further, given the highly stable
RMBUSD exchange rate, conducting their affairs in dollars allows those
engaged in international business to minimize exchange risk. Having come
to take the stability of the exchange rate for granted, market participants do
not net out their open positions immediately but rather minimize transaction
costs by netting out positions internally. Therefore, US dollar domination in
the FX market is also a factor in the low level of overall market activity.
Great expectations had been attached to the introduction of new currency pairs, especially the EURRMB pair. In fact, however, the euro has
not come to play an important role in Chinas FX market either in terms
of increasing market turnover or in terms of influencing the formation

370
The role of China

%
USD bil
100
250
98
200
96
150
94
100
92
90
50
88
86
95
96
97
98
99
00
Turnover
Source:
01
02
03
04
05 1H
Share
Figure 14.9
Turnover and share of the US dollar in the inter-bank market,

1995first half 2005
Australian dollar
5.5%
Others
25.3%
Swiss franc
6.1%
Pound sterling
16.9%
US dollar
88.7%
Japanese yen
20.3%
Euro
37.2%
Note:
Source:
Total: 200%
Since each currency pair involves two currencies, the total sums up to 200 percent.
BIS (2004).
Figure 14.10
Currency composition in the global FX market, 2004
mechanism of the RMBUSD rate. The reason is that the RMBEUR

rate is determined indirectly by the EURUSD rate vis--vis the stable
RMBUSD rate. Since banks in China can obtain a better EURUSD rate
in the international market if they have open EURRMB positions, they
first convert euros to dollars in the international market and then trade the
RMBUSD position in the domestic inter-bank market instead of directly
trading EURRMB positions.

Table 14.3
CFETS members by institutional type
Wholly state-owned banks

Joint stock commercial banks
Policy banks
Urban commercial banks
Branch of commercial banks
Foreign banks
Trust and investment
companies
Rural credit cooperatives
Non-financial enterprises
Total
Source:
371
January 2003
March 2006
4
10
3
22
108
164
2
4
11
3
44
109
189
2
9
0
322
26
1
389
The US dollar domination in Chinas FX market reflects market participants dependence on the central bank to clear the market under the rigid
exchange rate regime. Lack of motivation to hedge two-way exchange risks
has also prevented participants from building up professional skills in FX
risk management and retarded the development of FX derivatives. This
may turn out to be one of the most important fragilities when an exit from
the current peg brings in more variations in the rates.
To sum up, Chinas inter-bank FX market is currently mainly a spot
market. Forward and swap trading have just recently emerged, with
volume trivial so far. Even with the introduction of eight foreign currency
pairs in 2005, the domination of RMBUSD trading is likely to continue
into the near future. The lack of diversity with respect to transaction types
restrains market turnover and limits liquidity.
Market Participants and Concentration
Though membership in CFETS reached 389 by March 2006 (see Table
14.3), market activity remained highly concentrated among a small number
of banks. This contrasts with the diverse body of market participants in
the global FX market, which includes dealers and non-financial entities as
well as banks and non-bank financial institutions. The global FX market
has seen trading between banks and other financial institutions seize an
increasing share of turnover (33 percent in 2004) (BIS 2004).
Chinas FX market is characterized by approaching monopoly, especially on the buy side. Although nearly half of CFETSs members are

372
The role of China
Table 14.4
Market concentration (in percent, as of December 2004)
C1
C2
C3
C4
C5
50.57
64.70
71.71
76.14
79.86
Note:
C1 indicates the market share of the market participant with largest trading volume
and C5 indicates the market share of the largest five participants.
Source:
Calculations based on trading volume figures in China Money.
%
100
USD bil
250
200
80
150
60
100
40
50
20
0
0
'95
'96
'97
'98
'99
'00
'01
'02
'03
'04
PBOC net purchase

Inter-bank market turnover
PBOC net purchase/market turnover
Sources:
Shu (2001) and Preeg (2003).
Figure 14.11
PBOC net purchase of FX and inter-bank market turnover,

19952004
foreign banks, their trading volume amounts to only a small portion of

the total. In 2004, the top five market participants,18 all of them domestic
banks, took away nearly 80 percent of market trading (see Table 14.4).
These large banks are usually net sellers in the market, with the rest of the
markets participants net buyers. The BOC itself is estimated to account
for more than half of net FX selling in 2002 (Xin 2003).
On the buy side, the PBOCs net purchase of FX accounted for an
average of 68 percent of the total inter-bank market turnover in the period
from 1995 to 2004 (Figure 14.11). In recent years, the PBOC has been
obliged to undertake massive buying in the face of heavy supply pressure
brought on by speculation on RMB appreciation and a relatively high
interest rate paid on RMB deposits. The result has been an increase in

373
official reserve assets to USD 610 billion at the end of 2004, up USD 324
billion in just two years.
14.5
PROBLEMS IN CHINAS FOREIGN EXCHANGE

MARKET
Distorted Market Supply and Demand

Foreign exchange market institutions and the choice of exchange rate
regime bear integrally on one another. A more flexible exchange rate
regime for China, capable of generating an exchange rate consistent
with balance of payments fundamentals, cannot be achieved without a
foreign exchange market in which supply and demand represent the true
preferences of market actors. On the other hand, Chinas trouble-ridden
exchange rate regime is itself to blame for many structural problems in the
FX market. Thus, beginning to fix the cracks in market supply and demand
not only would help tackle many structural deficiencies in the FX market
but also would be a crucial step toward disentangling the Gordian knot
of the RMB exchange rate regime and, in the medium run, developing the
RMB into a regional lead currency.
Three main problems exist with the market supply and demand for
foreign exchange in China. First, supply and demand do not express the
will of market participants but rather follow from their compliance with
distorting regulations. Under the FX purchase and sale system, exporters
and foreign investors must surrender at least 75 percent of their FX earnings to the designated FX banks, and these banks, in turn, must sell their
foreign currency receipts on the inter-bank FX market, subject to an allowable working position. On the supply side, then, neither the ultimate suppliers (i.e. exporters and investors) nor the designated banks that enter the
inter-bank market have the freedom to choose their preferred levels of FX
holdings. On the demand side, capital controls impede Chinese investment
overseas and restrict FX use by businesses and individuals.
Second, in recent years a pronounced excess dollar supply has emerged on
Chinas FX market. This excess supply follows from mounting surpluses in
the current account and strong inflows of foreign direct investment coupled
with an explosive growth in short-term capital flows. A growing part of
the excess supply of foreign currency must be absorbed by the PBOC. The
imbalance in Chinas FX market is aggravated by the compulsory supply
and impeded demand that is the product of the regulatory system. With
trends of a rising trade surplus, stable FDI inflows, and strengthening shortterm capital inflows powered in part by speculation on RMB appreciation

374
The role of China
4
3
2
1
0
1
2
3
4
1998
Source:
1999
2000
2001
2002
2003
2004
2005
2006
Federal Reserve Bank of St Louis and IFS.
Figure 14.12
Difference between RMB bank deposit rate and US federal

funds rate, 1998June 2006 (percentage points)
and in part by higher interest rates on RMB deposits (see Figure 14.12), the
gap between supply and demand is being driven ever wider.
Third, the supply and demand for FX are not fully reflected in the interbank market on the CFETS platform. This is because the large Chinese
banks transfer a part of their position-squaring operations to Hong Kong
or to the black market. Therefore, the excess supply of foreign exchange
on the inter-bank market is not reflective of overall market conditions. All
in all, the distorted market supply and demand compromise the markets
ability to yield a meaningful value for the exchange rate.
Passive Intervention of the PBOC
Surging excess supply of USD in the FX market has put pressure on the
RMB to appreciate. In order to maintain the pegged RMBUSD rate, the
PBOC has had to intervene by purchasing large quantities of US dollars
and then sterilizing these purchases through the issuance of central bank
bills in order to control the money supply. The PBOC was forced to play
the super market maker role in the FX market. So far, the PBOC has been
quite successful in keeping a tight rein on money supply growth through its
sterilization operations and in draining liquidity from the banking system
(see Figure 14.13); however, the PBOCs passive intervention could lead
to problems.
First, supply and demand imbalances under a compulsory FX settlement
system leave no leeway for the PBOC, which must absorb the appreciation

375
15%
10%
5%
0%
5%
10%
1999
2000
2001
2002
Bank reserves
Source:
2003
2004
2005
Official reserve assets
SAFE and IFS.
Figure 14.13
Growth in Chinas bank reserves and official reserve assets,

September 1999June 2006 (monthly rate of change)
pressure. Second, the commitment to maintain stability in the RMBUSD

rate results in high US dollar concentration in the FX market. Third, constant intervention in the FX market imposes high monitoring and administrative costs. Finally, PBOC sterilization operations could lead to a vicious
cycle, pushing interest rates higher only to attract greater speculative inflows
that must be purchased by the PBOC and in turn sterilized. As Figure 14.12
shows, the interest rate on RMB deposits has already been pushed above the
US Federal Funds rate. The downside of this kind of passive intervention is
that the PBOC loses independence in conducting monetary policy.
One way to go about relaxing the constraints on the PBOC would be to delegate the role of market maker to one of the state-owned commercial banks.
Any one of the Big Four could be considered for the role. This step could help
provide liquidity in the market and ease the pressure on the PBOC.
China has been aided in maintaining its currency peg in the face of sizeable current account surpluses by its use of capital controls. These controls
have helped protect against the kind of large-scale capital movements
that forced many Asian countries off their pegs during the Asian financial
crisis. With the gradual easing of capital controls in China, not only does
the currency face greater danger of one way bet speculative attacks but
independent monetary policy becomes more difficult to pursue.19 In the
past few years, expansionary fiscal policy has been the preferred approach

376
The role of China
to averting economic slowdown, but, with the Chinese economys continued robust growth cycle and a looming threat of inflation, monetary policy
could be vital to achieving a soft landing.
14.6
NEW REFORM MEASURES IN FX MARKET
Shortly before and after the July 21 RMB revaluation, several new reforms
(see section 14.2) were introduced in the FX market, mainly focusing on
three aspects: (1) broadening market access, (2) enriching market segments, and (3) transforming trading arrangements. Theoretically, these
new reform measures can improve on the market structures by increasing
liquidity and enhancing market functions. But doubts should be cast as to
whether these measures can fully deliver their intended effects and whether
this parallel and top-down reform methodology can work well in the
FX market.
Broadening of Market Access
As a result of the market reforms, there were 389 CFETS members as
of March 2006, an increase of 23 over June 2005. Among the new participants, Sinochem Corporation became the first non-financial enterprise
ever to enter the inter-bank market. Meanwhile, foreign banks continue to
show a strong interest in Chinas FX market.
Foreign banks participation: strategic thinking
The active participation of foreign banks in Chinas FX market is a result
of more long-term strategic concerns than of short-term profit concerns.
Even though foreign banks make up the largest share of banks in the
market, their share of trading volume is lackluster. The foreign banks have,
in fact, entered the market mainly to gain a foothold, with the potentially
growing international role of the RMB and the Chinese market in mind.
First of all, the RMB is still inconvertible, but a gradual loosening of
capital controls has already begun, allowing a certain limited volume of
cross-broader capital flows in both directions. Chinas strong foreign trade
sector and economic growth will ensure the RMB an increasingly central
role in East Asia and the international arena. The RMB will be one of the
most frequently traded currencies in the international FX market, and these
banks are counting on their early-on experience and familiarity with RMB
trading in Chinas FX market to give them a competitive edge in the future.
Second, foreign banks will receive full access to Chinas banking sector
by the end of 2006 as part of Chinas WTO commitment. There will exist

377
huge opportunities for the provision of new financial products and services
soon, many of them involving FX trading. Besides, gestures of goodwill
through helping develop Chinas financial market can also help foreign
banks maintain an amicable relationship with the Chinese authorities,
which is very important now that reforms in Chinas financial market
are making headway and bringing with them new potential sources of
revenue.
No matter what their motives are, foreign banks presence in Chinas
FX market will have great demonstration and spillover effects for domestic
institutions, and, with their rich experience of FX trading, sophisticated
FX risk management skills, and deep understanding of the market, they
will become a main pillar in new products trading.
Non-financial enterprises: latent needs of hedging against FX risks
With combined exports and imports standing at USD 1422.12 billion
in 2005,20 non-financial enterprises, especially companies engaged in
foreign trade, have every reason to become big players in the FX market.
However, the long-time stability of the exchange rate has caused inertia
among Chinese enterprises when it comes to hedging against FX risks. The
lack of FX risk management awareness and skills may prevent these enterprises from being deeply involved in the inter-bank market. Transaction
cost is another concern considering the fact that 54.7 percent of exports in
2005 were in the form of processing trade, which tends to have a relatively
low degree of risk exposure and small profit margins. Last but not least,
alternative ways of hedging against FX risks can also woo the enterprises
away from the domestic FX market. According to a survey conducted after
the July 21 reform (PBOC 2006), foreign trade companies in China have
responded to the exchange rate adjustment by providing a variety of tools
for hedging against greater exchange rate volatilities, with trade finance
(used by 31 percent of respondents) as their foremost choice. The use of
financial derivatives such as forwards and swaps is still less prevalent. In
2005, the value of forward FX sales was only 2 percent of total FX sales
in the spot market, while forward FX purchase was only 4 percent of total
FX purchases. The less-regulated offshore market for non-deliverable
forwards (NDF), in which some large enterprises have already been participating, is another source of competition for the Chinese onshore FX
market.
Overall, broadening market access to a wider range of institutions
would theoretically increase market liquidity and boost trading volume.
According to the market microstructure theory, the heterogeneity of
market participants in terms of their motives, analytical methods, and possession of private information and the resulting differences in expectations

378
The role of China

EURJPY
1%
USDHKD
26%
EURUSD
41%
USDCHF
1%
USDCAD
1%
USDJPY
23%
Source:
GBPUSD AUDUSD
3%
4%
Figure 14.14
Currency composition in foreign currency pair trading in

Chinas FX market (in percent), May 2005December 2005
is responsible for most of the trading volume and exchange rate volatility in the FX market.21 But, in reality, it may take a while for these new
participants to familiarize themselves with the rules and skills required
for using the FX market to hedge against emerging exchange rate risks.
How to encourage new players to enter the market and existing participants to trade in the market goes well beyond simply approving new
memberships.
Enrichment of Market Segments
Measures to enrich FX market segments mainly include adding the eight
foreign currency pairs in the spot market and establishing an inter-bank
forward market.22 This will enrich the original inter-bank market segments, which previously featured only spot transactions in four RMB
related currency pairs, enhance the market functions, and provide market
participants a new scope for investing their FX holdings as well as hedging
against FX risks.
Trading in foreign currency pairs is dominated by the three currency
pairs formed by the USD against the EUR, the HKD, and the JPY, respectively, while the EURUSD pair enjoys the deepest trading, with a market
share of 41 percent (Figure 14.14). Bidask spreads in trading among the
foreign currency pairs are tight, indicating strong competition among
market makers and quite high liquidity.

379
Foreign currency pair trading: an experiment ground

The most significant feature of the initiation of foreign currency pairs
trading is that it was launched in a market parallel to the original spot
market. The eight foreign currency pairs were not simply added to the
original inter-bank market but formed a new market of their own, making
use of a separate trading system on the CFETS platform. The addition
makes Chinas FX market more complete, incorporating both RMB
against foreign currency transactions and foreign currency against foreign
currency transactions. At the same time, the introduction of the foreign
currency pair trading through a parallel market has allowed Chinese
authorities to experiment with new FX market reforms without implementing them fully in the original RMB pairs market.
Instead of being restricted by the old market structures, this parallel
market experiments with brand new trading arrangements, such as market
making, and adopts international practice by introducing foreign banks
as market makers and setting up electronic trading systems with flexible
trading options. New structural reforms were thus experimented without
any changes being made to the FX market institutional frameworks, such
as the compulsory FX settlement and rigid exchange rate regime. As an
instant structural reform of trading arrangements in the original spot
market can be costly and risky insofar as the original market trades RMB
related currency pairs and thus cannot avoid the sensitive issue of the RMB
exchange rate, conducting experiments in a parallel market circumvents
these hurdles and reduces the reform costs. Besides, market activity in the
parallel market can provide decision makers with key information needed
to carry on further reforms.
Inter-bank forward market
Establishing the inter-bank forward market was seen as the key to making
Chinas FX market a truly functional market that could satisfy market
participants risk hedging needs. Yet, so far, the onshore forward market
has not lived up to expectations and has suffered from a low level of market
activity, one-way quotes, and a lack of independent pricing.
Market activity By the end of 2005, 64 banks had gained authorization
to trade in the inter-bank forwards market.23 Inter-bank forwards in four
currency pairs were offered by the CFETS in the new markets debut on
August 15, 2005, but by the end of 2005 only RMBUSD and RMBJPY
had been traded in the market and, not surprisingly, trading had been
dominated by the USD. Among the 277 market transactions in 2005,
263 involved RMBUSD forwards trading, while the other 14 involved
RMBJPY trading. Term structures are quite diversified, with seven

380
The role of China
different terms for USD forwards and four for JPY forwards. The most
liquid contracts are one-month and 12-month USD forwards.
Lack of liquidity and limited scale are common for a nascent market,
and the onshore market also has to face competition from the RMB NDF
market (see below), a market offering products with similar functions and
with longer history; however, one-way quotes and a lack of independent
pricing are the two largest impediments to greater market activity.
One-way quotes One-way quotes result from ubiquitous expectations
of RMB appreciation and the imbalance in the retail forward market.
As already noted, agents transact because they differ (Lyons 2001).
Homogeneous expectations, on the other hand, lead to one-way quotes
and difficulty in striking a deal. In the current market environment, it is
hard not to bet on the appreciation of the Chinese currency. The only heterogeneity comes from different expectations concerning the degree and
speed of appreciation, resulting in sporadic trades. The other reason for
one-way quotes originates in the retail market, where expectations of RMB
appreciation induced a sharp rise in FX forwards sales at the same time
that expectations of a US interest rate rise caused a discount in US dollar
forwards, which in turn led to a decline in FX forwards purchases. The
imbalance in the retail market was carried over to the inter-bank market
by banks, which, more often than not, inherit a long forward position in
the US dollar from their retail businesses.
Lack of independent pricing Since its establishment, quotes in the onshore
market have followed those in the RMB NDF market closely. Information
flows and the transmission of prices and volatility between two financial
markets trading similar products are normal. In the absence of capital
controls, arbitrage between two markets will also lead to convergence in
prices. But in the case of the RMB onshore forward market two things
have stood in the way of price convergence. First, there are capital controls
in place.24 Second, the onshore market and the NDF market target different market participants and therefore have different pricing mechanisms.
Major participants in the NDF market are multinationals and hedge funds.
Expectations, based on interest rate spreads or often purely on speculation, dominate the NDF pricing. On the other hand, the onshore market
primarily caters to banks and enterprises risk hedging needs. Pricing in the
onshore market should thus be based on covered interest parity and be more
rational than that in the NDF market. So the current similarity between the
two markets points to the onshore markets lack of independent pricing.
Actually, market participants in the onshore market largely base their
quotes on the prices in the NDF market instead of on interest parity.

381
The NDF market is greatly exposed to speculation and large volatility. Thus, without independent and more rational pricing in the onshore
market, its risk-hedging functions will be limited, and it will be at a disadvantage in its competition with the NDF market. As mentioned above,
the RMB NDF market has 10 more years of history and a higher liquidity
than the onshore market. Considering the self-fulfilling quality of liquidity, the onshore market is already lagging behind. Lack of independent, let
alone competitive, pricing can be deadly to the nascent market in its race
to draw liquidity.
The underdeveloped money market in China may be the first to blame
for the lack of independent pricing in the forward market. So far, Chinas
money market remains inefficient, stifling it from becoming a market where
highly liquid products with a whole spectrum of maturities are traded. The
determination of interest rates is still regulated by the central bank. As a
result, no representative market interest rates for different maturities are
quoted, and the onshore forward market, in turn, does not have the interest
rate needed in order to apply the interest rate parity based pricing rules.25
Transformation of Trading Arrangements
After the market making system and the quote-driven dealer market
system were introduced on January 4, 2006, Chinas FX market became a
quasi-centralized hybrid market featuring the coexistence of a quote-driven
dealer market and an order-driven auction market utilizing an electronic
broking system. These two transformations in trading arrangements are
milestones in Chinas FX market reform and will have far-reaching influences on the market structure and exchange rate formation mechanism as
well as the central banks monetary policy.
Diversified market liquidity provision mechanism
Providing liquidity is the primary role of a market maker. Market makers
regularly quote both bid and ask prices and stand ready to buy and sell
at the quoted price. Market makers can absorb excess market supply or
demand by holding inventory and make a profit through bidask spreads.
The role of market makers is thus crucial to achieving market clearing and
providing liquidity. A market making system is especially important to
nascent markets, which oftenas is the case in Chinasuffer from low
liquidity or a one-sided market. With a mixed quote-driven dealer market
and order-driven auction market structure, liquidity can be provided
through multiple mechanisms, and market participants are given more
flexibility in choosing trading arrangements depending on the purposes
of the transaction, the type of counterparty, the level of transparency,

382
The role of China
the information they have, and the size of the order (Harris 2002). The
biggest problem in Chinas FX market structure is a lack of liquidity, and
the transformation of the trading arrangement, which brings a diversified
market liquidity provision mechanism and increases the probability of
deal-making, is the most important move among all the reform measures
from the point of view of boosting market liquidity.
Price discovery and the exchange rate formation mechanism
Another important role of the market maker is price discovery. The other
reform measures can indirectly have an impact on the exchange rate formation mechanism by changing FX market structures, but the introduction
of market makers and quote-driven dealer markets in the spot market
where the RMB is traded against foreign currency pairs has a more direct
and central role in the determination of the RMB exchange rate. After
the new trading arrangement was put in place, the RMB central rate went
from being determined by the previous days closing rate formed in the
inter-bank spot market to being determined by the weighted average of
the prices quoted by 15 market makers consisting of both local banks and
foreign banks. However, it should be duly observed that the price discovery function of market makers in Chinas FX market is still rather limited.
There still exists a narrow floating band prescribed by the central bank,
and regulators also regulate market makers net open FX position (both
gross aggregated position and single currency position) to limit their scope
of price discovery.26
The central banks control of the FX market
As mentioned earlier, before the market making system came into being,
the central bank was the largest de facto market maker in Chinas FX
market. Facing strong RMB appreciation pressure, the central bank
hadand still hasto frequently and passively absorb excess US dollar
supply in order to both clear the market and at the same time maintain
the US dollar peg. This leads to a variety of problems, including a lack of
independence and flexibility in administering monetary policy. Assigning
market making responsibilities to commercial banks gives the central bank
a way out of this vicious circle and allows the central bank more freedom
to tackle problems at home and to experiment with new open market
operations such as currency swaps. The growth rate of FX reserves is also
expected to fall with an easing of the pressure on RMB appreciation, so
that the central bank would no longer need to absorb huge US dollar sell
orders. The central banks adjustment to its FX market control mechanism
comes with the transformation in trading arrangements. Market makers
will inevitably take over at least part of the information processing and

383
price discovery functions. The Chinese authorities have clearly stated their
policy of not yielding to foreign political pressures by making any one-off
exchange rate adjustments.27 Thus, continuing to breathe more latitude
and more market forces into the exchange rate determination mechanism
and as a result allowing a certain amount of room for the gradual appreciation of the RMB is an expedient way of coping with the RMB appreciation
pressure.
14.7
SUMMARY AND IMPLICATIONS FOR THE

RMBS ROLE IN EAST ASIA
To sum up, Chinas FX market is characterized by low turnover; heavy concentration in spot trading, with the retail market in forwards just recently
opened up; nascent inter-bank forwards and retail market swap transactions; dominance by US dollar trading, with all other foreign currencies
accounting for only a tiny part of the market; a membership-based exchange
trading platform evolving toward a mixed trading mode incorporating
some features from the dealer market, with banks as the most prevalent type
of participants; a high degree of market concentration on both the sellers
side by the BOC and the buyers side by the PBOC; and its simple function
as the place for designated FX banks to net out open positions, with only a
limited capacity to serve investment or risk-hedging needs of FX users.
Among them, the most central problem is the lack of liquidity. Without
a deep market, all reform measures will only remain technical adjustments
with no profound impact on market functions. The recent policy initiatives addressed some of the structural problems in the FX market and are
expected to increase market liquidity, enhance market functions, and
influence exchange rate formation mechanisms. But the new reforms and
the newly established market have some of their own problems in need of
redress, which may reduce their effects with respect to correcting existing
structural failures. The new reforms in the FX market are carried out in a
top-down and parallel way. With a top-down reform, the market
has to undergo a transitional stage in which market participants are given
some time to adapt to the new trading rules, to invest in human resources
and technology, and to acquire the right market sense. In turn, it may
take a while before the market earns recognition and builds momentum in
accumulating liquidity. The new FX market reforms were also instituted in
a parallel way without completely changing the old system, which means
that the two institutional factors that most seriously constrain the development of Chinas FX market havent been completely removed and may
continue to take their toll. Nascent market segments such as the inter-bank

384
The role of China
forward market need to have more independent pricing in order to fully

realize their risk-hedging potential. In conclusion, certain structural problems in Chinas FX market such as low liquidity will linger and continue
to afflict the market, but, with the new reforms coming into effect after
overcoming their own problems and with the institutional restrictions to
be phased out in the future, market structures can expect great improvement in the future.
The fact that these institutional constraints and the structures in the FX
market are inter-related presents a dilemma for moving forward, as liberalizing on one front is difficult as long as the status quo is maintained on the
other. Specifically, the rigidity of the exchange rate regime imposes limits
on the growth and diversification of the FX market. But at the same time,
allowing more flexibility in the exchange rate requires a broader, more
diversified, competitive, and efficient market platform on which real forces
of supply and demand can determine the RMB value of foreign currency.
A profound change in one institutional pillar or the other is required to
break loose from the status quo.
Both the rigid exchange rate and the underdevelopment of the Chinese
FX market considerably limit the RMBs prospects for becoming a
regional lead currency anytime soon. Despite the enormous size of the
Chinese economy and its growing role in world and regional trade, the
RMB will not be able to act as key or anchor currency for an East Asian
RMB bloc unless China develops deep and liquid financial markets and
unless the RMB becomes fully convertible. For the latter to happen, continued FX market reform is crucial.
In our view, key elements of FX market development must include: (1)
further stimulating market demand by lowering market entry requirements, removing the real trade background rule in FX trading, and
allowing a certain amount of speculation; (2) encouraging enterprises
and banks to build up their skills and resources in FX risk management
and to speed up reforms in the banking sector; (3) improving the financial
environment by speeding up development in the money market, in the
market determination of interest rates, and in the credit rating system; (4)
phasing out the central banks control in the FX market and prudentially
and gradually opening capital accounts; and, finally, (5) introducing more
flexibility in the exchange rate regime.
NOTES
1.
An earlier version of this chapter was presented at the HWWA/HWWI conference on

East Asian Monetary and Financial Integration in Hamburg, Germany, December
1516, 2005 as well as at the Hamburg Institutes International Finance Seminar on

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
385
December 13, 2005. We would like to thank participants at these meetings, especially
our discussant Alvaro Santos Rivera, for very useful comments and suggestions. All
remaining errors are those of the authors alone. Some parts of this chapter draw from
Zhang and Liang (2006).
On the yen bloc see Kwan (2001).
The FX plan was formulated by the State Planning Commission in consultation with
the Ministries of Trade and Finance and the PBOC. See Zhang (2003).
The exchange rate for trade activities was set at 2.8 RMB/USD, while the official rate
still stood at 1.5 RMB/USD. This practice was abolished on January 1, 1985.
Despite the name, the swap market provided spot transactions only.
These adjustments were frequent, small, and slow, with mixed appreciation and depreciations, but on the whole the RMB had been devalued against the US dollar from the
early 1979 level of 1.50 RMB/USD to 5.72 RMB/USD at the end of 1993. Six large
official devaluations took place in 1981, 1984, 1985, 1986, 1989, and 1990.
Details of these policies can be seen in the Notice of the Peoples Bank of China on
Accelerating the Development of the Foreign Exchange Market which came out on
August 8, 2005 on the central banks website: www.pbc.gov.cn.
Details can be found on the State Administration of Foreign Exchanges website: www.
safe.gov.cn.
Non-financial enterprises were not allowed access until the August 2005 reform.
This allowable FX working position has to be verified and approved by SAFE.
The market structure part of this chapter focuses primarily on the inter-bank market.
Distance connections can be realized through Digital Divide Network, frame-relay, or
dial-up.
For a discussion on the need for reform of the CFETS see Zhang and Liang (2006).
According to the Bank for International Settlements (BIS), the main factors driving
the fall in turnover were the introduction of the euro, the growing share of electronic
broking in the spot inter-bank market, consolidation in the banking industry, and global
concentration in the corporate sector.
The eight foreign currency pairs include EURUSD, AUDUSD, GBPUSD, USD
CHF, USDHKD, USDCAD, USDJPY and EURJPY.
Presently, forward contracts exist for eight currencies (US dollar, Hong Kong dollar,
euro, yen, pound, Swiss franc, Australian dollar, and Canadian dollar) and 14 different
terms (from seven days to 12 months).
The PBOC stopped the disclosure of FX market turnover data in the second half of
2005.
They are the BOC, the Agriculture Bank of China, the Industrial & Commercial Bank of
China, the China Construction Bank, and the China International Trust & Investment
Corporation (CITIC). The first four are commonly referred to as the Big Four.
The impossible trinity holds that, with free capital mobility, a fixed exchange rate and
independent monetary policy cannot be realized simultaneously.
Data from the Ministry of Commerce of the Peoples Republic of China website: http://
zhs.mofcom.gov.cn/tongji.shtml.
See Sarno and Taylor (2001).
Other measures, such as opening the retail market forward FX settlement business to
all foreign exchange designated banks and introducing swap transactions, will not be
discussed here.
Among them, 48 are foreign banks, while 16 are domestic banks.
Ma et al. (2004) note that there is a large difference between the onshore interest rate and
NDF market implied interest rate, reflecting the existence of strict capital controls.
The only market interest rate in China is the seven-day repo rate.
Market makers cannot hold short positions in the US dollar.
This was clearly stated when Premier Wen addressed the press after the fourth session
of the 10th National Peoples Congress. See Wen (2006).

386
The role of China
REFERENCES
BIS (2004). Triennial Central Bank Survey of FX and Derivatives Market Activity in
2004. Basel: Bank for International Settlements.
China Money (various issues): www.chinamoney.com.cn.
Harris, Lawrence (2002). Trading and Exchanges: Market Microstructure for
Practitioners. Oxford: Oxford University Press.
Hefeker, Carsten, and Andreas Nabor (2005). Chinas Role in East Asian
Monetary Integration. International Journal of Finance and Economics 10:
15766.
Ho, Corinne, Guonan Ma, and Robert N. McCauley (2005). Trading Asian
Currencies. BIS Quarterly Review March: 4958.
Kwan, Chi Hung (2001). Yen Block. Toward Economic Integration in Asia.
Washington, DC: Brookings Institution Press.
Lyons, Richard K. (2001). The Microstructure Approach to Exchange Rates.
Cambridge, MA: MIT Press.
Ma, Guonan, Corrinni Ho, and Robert N. McCauley (2004). The Market for
Non-deliverable Forwards in Asian Currencies. BIS Quarterly Review June:
8194.
PBOC (2006). Survey of Enterprises Use of FX Risk Hedging Tools. www.pbc.
gov.cn/.
Preeg, Ernest H. (2003). Chinese Currency Manipulation and the U.S. Trade
Deficit. Statement before the USChina Economic and Security Review
Commission, September 25.
Sarno, Lucio, and Mark P. Taylor (2001). The Microstructure of the Foreign
Exchange Market: A Selective Survey of the Literature. Princeton Studies in
International Economics no. 89, Princeton University, Princeton, NJ.
Shu Youdong (2001). The Appreciation Pressure on RMB and the Releasing
Measures. Zhong Guo Huo Bi Shi Chang (China Money) November: 346.
Wen Jiabao (2006). Speech at the Press Conference of the 4th Session of the 10th
National Peoples Congress, March 14. http://news.xinhuanet.com/misc/200603/14/content_4301180.htm.
Xin Wang (2003). Problems in Chinas FX Market. International Economic
Review 1112: 16.
Zhang Guang Ping (2005). RMB Derivatives, a Market that Shouldnt be
Missed. China Business News, http://gp.stock.163.com.
Zhang Jikang, and Liang Yuanyuan (2006). Institutional and Structural Problems
of Chinas Foreign Exchange Market. China: An International Journal 4:
6085.
Zhang Zhichao (2003). Harmonious Development of Foreign Exchange Market
and Liberalisation of Capital Controls in China. Paper presented at the
International Seminar on Developing Chinas Foreign Exchange Market,
Shanghai, December.

Index
ABMI see Asian Bond Markets
Initiative (ABMI)
absorptive capacity 142
ACU see Asian Currency Unit (ACU)
ADBI see Asian USD Bond Index
(ADBI)
Adenauer, Konrad 42
ADXY see Bloomberg-JPMorgan Asia
Currency Index (ADXY)
AFAS see ASEAN Framework
Agreement on Services (AFAS)
AFTA see Asian Free Trade Area
(AFTA)
AIA see ASEAN Investment Area
(AIA)
AICO see ASEAN Industrial
Cooperation (AICO) Scheme
Alesina, Alberto F. 71
Alexander, Volbert 217
Allen, Franklin 82
allocative efficiency 1956
Amato, Jeffrey D. 102
AMU see Asian Monetary Unit
(AMU)
AMU Deviation Indicators 1056
Andersen, Lene 48
Anderson, Jonathan 319
APEC see Asian Pacific Economic
Cooperation (APEC)
arbitrage 89, 1478, 150, 154, 188, 380
Argentina 207, 215
Arndt, Sven 326
ASEAN Agreement for the Promotion
and Protection of Investments 38
ASEAN Economic Community
attracting FDI to 34, 37
best practices adopted in 34
developing institutions in 356
end goal of Vision 2020 13, 21, 22
free flow of services in 345, 37
labor mobility in 22, 34
lessons from EU 34, 2432, 367
rationale for 22
tariff harmonization in 324
ASEAN Framework Agreement on
Services (AFAS) 35, 37
ASEAN Industrial Cooperation
(AICO) Scheme 38
ASEAN Investment Area (AIA) 19,
22, 34, 37
ASEAN region
building the ASEAN Economic
Community 326
diversity in economic development
in 25
evolution in a regional context
1623
exchange rate policy coordination
with China 223, 241, 242, 245,
251
implicit coordination of exchange
rate policies 259, 260
monetary policy consistent with
223, 25662
REER indexes 2546
weights for currency baskets 236,
238, 2514
exchange rate stabilization with
cooperation 2358
exchange rate stabilization without
sustainable underlying policies
22435
investment, saving, current
account and REERs since
1990 22635
FDI flows to 14, 26, 27, 34, 37
institutional environment facing 24,
356
international economic environment
facing 245
lessons from EU 34, 2432, 367
member countries 101, 135, 215
openness of countries in 256, 324
power structure within 213
387

388
Index
surveillance process 1
trade patterns 6, 1416, 257, 197,
198, 223, 236, 23842
currency baskets and REER
indexes based on 2516
by sector 2425, 2467
by sector and market 24551
ASEAN Vision 2020 1819, 21, 22, 33
Asia Credit Index (JACI) 30910, 311
Asian Bond Fund I 23, 81, 91, 105,
207, 319
Asian Bond Fund II 4, 23, 81, 87, 90,
914, 96, 97, 105, 207, 319
further developments arising from
101
incentives to reduce impediments in
94, 98100
learning by doing element in 87,
934, 100101
Asian Bond Market Initiative (ABMI)
21, 23, 28, 81, 91, 105, 304, 315
Asian bond markets see regional bond
markets, Asian
Asian bonds, definition of 82
Asian Central Bank, creation of 317
Asian Cooperation Dialogue (ACD) 81
Asian crisis (199798)
capital outflows during 478
and Chinas foreign exchange market
360, 3634, 3678
currency basket structures before
and after 2817
current account balances before and
after 29, 17071, 204, 2734
debate on exchange rate policy
triggered by 159
and development of local currency
bond markets 87, 88, 207
dollar pegs temporarily lifted during
and after 210, 267
domestic currency interest rates
prior to 1723
exchange rate volatility before,
during and after 1638, 2246,
2678, 27072
inflation rates prior to 188
investment, saving, current account
and REERs before, during and
after 22635
Miyazawa Plan initiated during 23
original sin as one of causes of 204

and poverty 234, 235
real GDP before, during and after
226
regionalist trend triggered by 1, 3,
17, 22, 30, 36, 41, 61, 104, 195
swap transactions not used prior to
126
Asian Currency Unit (ACU) 1, 3, 7,
312, 208, 213, 30417
constructing ACU index 30611
currencies to be included 308,
31315
economic variables and indicators
for calculating currency
weights 3078
existing currency baskets 30811
fixed currency shares vs. fixed
currency units 307
frequency of periodical revisions
of currency shares/units 308
definition 304
economic logic of introducing
31215
case for a currency basket 31315
reasons for not pegging to a
common key currency
31213
objectives 3056
potential role for monetary
and exchange rate policy
coordination 31517
the way forward 317
see also Asian Monetary Unit
(AMU)
Asian Development Bank (ADB) 1, 16,
23, 91, 208, 290, 304, 354, 355
Asian Development Fund 6, 222, 237,
262
Asian Free Trade Area (AFTA) 1722,
25, 26, 33, 37
Asian miracle 235, 242
Asian Monetary Fund (AMF) 17, 23,
61
Asian Monetary Unit (AMU) 107
bonds denominated in
previous research on 1056
risks for bond issuers 45, 1313
risks for foreign investors 106,
11624, 133

Index
currency weights in 30910, 311
as a portfolio fund 112
see also Asian Currency Unit (ACU)
Asian Pacific Economic Cooperation
(APEC) 13, 17, 22, 23, 29, 81, 90
members 101
Asian USD Bond Index (ADBI)
30910, 311
asset dollarization 2734
Association of Southeast Nations
(ASEAN) see ASEAN region
Australia, East Asian trade with 239,
240, 241, 248, 249, 250, 252, 253
Australian dollar
share in global foreign exchange
market 370
weight in currency baskets 252, 253,
310
Austria, monetary policy decisions in
209, 210
Austro-Hungarian empire 75
automobiles, trade in 22, 32
Axelrod, Robert 78
Balassa, Bela 333
Balassa-Samuelson approach 187, 328,
330, 33341
Baldwin, Richard E. 82, 215
Bank for International Settlements
(BIS) 89, 102, 119, 363, 369, 370,
371
bank lending, characteristics of 64, 65
Bank of China (BOC) 3669, 372
Bank of England 216, 30811
Bank of Japan 73, 178, 182
Bank of Korea 73
bankruptcy 65, 234
Barre Report 43, 47, 57
barriers to monetary and financial
integration 2
Barro, Robert J. 71
Battellino, Ric 90, 101
Battle of the Sexes 78, 320
Beck, Thorsten 355
BEER models see behavioral
equilibrium exchange rate (BEER)
models
behavioral equilibrium exchange rate
(BEER) models 327, 328, 329, 330
Beijing 62
389
Belgium, monetary policy decisions

in 209
BelgiumLuxembourg Economic
Union 209
Bnassy-Qur, Agnes 135
Benelux, role in EMU negotiations
534, 56
Berger, Helge 215, 217
Bernanke, Ben 187
best practices, adoption of 34
Betz, Karl 195
bidask spreads 12431, 132, 134, 378,
381
bilateral trade integration 200202
Bird, Graham 320
BIS see Bank for International
Settlements (BIS)
Bisignano, Joseph 189
Blanchard, Olivier 157
Bloomberg 93, 127, 130, 132, 162, 164,
165, 167, 168, 182, 271, 281, 284,
285, 286, 287
Bloomberg-JPMorgan Asia Currency
Index (ADXY) 30910, 311
BOC see Bank of China (BOC)
bond financing, characteristics of 645
bond markets see currency basket
denominated bonds; regional
bond markets, Asian; regional
bond markets, European
Bordo, Michael D. 203, 207, 216
Bosworth, Barry 328, 329
Brandt, Loren 350
Branson, William H. 233, 263
Brazil 207, 215
Bretton Woods System 5, 28, 43, 47,
68, 75, 185, 187, 208, 276, 317
Bretton Woods II view 5, 9, 141,
14556, 276, 317
British pound
Chinese exchange rate fixed to 358
currency basket of 30811
market 370
Broughton, J.M. 82
Brunei
AFTA membership 22
trade patterns 14, 15, 197, 198
by sector 242, 243, 244, 2467
see also Brunei dollar

390
Index
Brunei dollar
exchange rate volatility against US
dollar 260
weight in currency baskets 309
bubbles 142, 144, 224, 233, 276, 278
Bundesbank 50, 52, 76
discount rate 20810
business cycles 72
Cairns, John 327, 328, 329
Calvo, Guillermo A. 173, 216, 349
Cambodia
AFTA membership 22
interest rates 203
monthly changes in reserves 261
openness of 25, 33
by sector 243, 244, 2467
transition period for 34
see also Cambodian riel
Cambodian riel
currency basket of 210, 211, 294
dollar 260
pegged to US dollar 210
Canzoneri, Matthew 3356
capital-account countries 186
capital controls 89, 142, 146, 148, 273
by China 89, 97, 313, 329, 330,
3478, 363, 373, 375, 376, 380
effectiveness of 212, 329
removal of 967, 99, 148, 313,
3478, 375, 376
Carare, Alina 216
Cardon, Pierre 102
Cargill, Thomas F. 82
CDS indices see credit default swap
(CDS) indices
CEIC Asia Database 200201
Ceglowski, Janet 350
central bank of China 161, 181, 272, 355
see also Peoples Bank of China
(PBOC)
centreperiphery model 185
CFETS see China Foreign Exchange
Trading System (CFETS)
Chaebols 64
Cheung, Hon 94
Cheung, Yin-Wong 328, 330, 349, 356

Chiang Mai Initiative (CMI) 1, 6, 20,
23, 41, 612, 81, 104, 213, 222,
2367, 238, 304, 315
China
admission to WTO 364, 376
balance of payment flows 3645
capital controls imposed by 89, 97,
313, 329, 330, 3478, 363, 373,
375, 376, 380
CPI index for 258
current account balance 170, 171,
175, 179, 180, 204, 273, 325,
3489, 364, 365, 373
emergence of Chinese economy 61
FDI flows to 175, 179, 185, 291,
373
foreign exchange reserves of 169,
179, 180, 204, 273, 274, 276,
277, 278, 3256, 329, 3489,
3724, 375, 382
foreign exchange risk facing 111
implicit coordination of exchange
rate policy with ASEAN 259,
260
interest rates 179, 182, 183, 184,
2034, 212, 372, 374, 375, 380,
381
labor productivity in 33541
liquid dollar-denominated assets of
175
monetary policy consistent with
exchange rate coordination with
ASEAN 223, 25662
nominal effective exchange rates 259,
299
REER indexes 2546
relations with Japan 2, 3, 48, 55, 62,
79, 80
role in fostering regional cooperation
1, 4, 78, 17, 48, 55, 578,
612, 326, 376, 3834
total non-performing loan liability
326
trade patterns 6, 15, 197, 198, 223,
236, 23842, 326, 377
currency basket and REER
indexes based on 2516, 293,
294, 297, 299

Index
by sector 2425, 2467
by sector and market 24551
weighted CPI index for trading
partners 257
see also Chinese foreign exchange
market; Chinese renminbi,
Peoples Bank of China
ChinaASEAN Free Trade Agreement
2021, 23
China Foreign Exchange Trading
System (CFETS) 359, 360, 362,
363, 365, 379
members 3712, 376
China Money 366, 367, 370, 371, 372,
378
Chinas foreign exchange market
35784
current organization of 36062
development of 35860
pre-1979: strict central control 358
197993: FX retention and swap
3589
1994 and post-1994: compulsory
settlement on a centralized
platform 35960
new reform measures in 37683
broadening of market access 376
central banks control of FX
market 3823
diversified market liquidity
provision mechanism 3812
enrichment of market segments
37881
foreign banks participation 3767
non-financial enterprises 3778
price discovery and the exchange
rate formation mechanism
382
transformation of trading
arrangements 381
problems in 3736
distorted market supply and
demand 3734
passive intervention of PBOC
3746, 3823
structural features of 36273
market participants and
concentration 3713
market segments 36671
market size 3626
391
summary and implications for

RMBs role in East Asia 3834
Chinese Bond Index Fund 101
Chinese renminbi (RMB)
bonds denominated in 91, 113, 114,
117, 125
currency basket of 7, 162, 163, 210,
211, 223, 236, 238, 241, 242,
2514, 256, 2817 passim, 291,
293, 294, 297, 299, 357, 358, 360
equilibrium exchange rate of
32549
extant literature on 32730
implications for foreign exchange
policy 3459
productivity approach 33341
real exchange rate and income
regression 3415, 346, 347,
356
simple time series approach
33033
euroRMB exchange rate 143,
1489, 150, 154, 155, 370
299
pegged to US dollar 5, 111, 113, 141,
161, 165, 179, 210, 211, 259,
267, 274, 3745, 382
as proxy for currencies of emerging
Asia 141
REER indexes 2546
role in East Asia 78, 376, 3834
as regional anchor currency 313,
314, 3578, 384
single managed floating exchange
rate regime (1994) 359
USD forward trading against
37980
USDRMB exchange rate 5, 1415
passim, 14850, 154, 155, 156,
161, 1648 passim, 175, 179,
210, 211, 259, 260, 267, 271,
272, 274, 359, 360, 36770
passim, 3745
determination mechanism 3823
equilibrium value of 32549
298, 309, 311
yen forward trading against
37980

392
Index
see also Chinas foreign exchange

market
Chinese stock market 360
Chinese yuan see Chinese renminbi
Chinn, Menzie D. 216, 275, 278, 329,
332, 333, 335, 339, 349, 354, 355
Chmelarova, Viera 270
Choudhri, Ehsan 350
CIS see Commonwealth of
Independent States (CIS)
Click, Reid W. 39
Cline, William 159, 190
cognitive dissonance 79
Cohen, Benjamin J. 203, 205, 206
cointegration tests 332, 356
Cold War 24, 50
Committee on the Global Financial
System 101
Common Agricultural Policy (CAP)
14, 26, 289, 43, 57
common citizenship, desire for 49
Common External Tariff 26, 32
common land border 200, 210, 202
Commonwealth of Independent States
(CIS) 75
community method approach 51
conflicted virtue 5, 6, 73, 1736, 179,
183, 185, 187, 189, 196, 204, 207
Confucius 80
contagion 48, 104, 134, 189, 215,
23031, 2356, 256, 262
Corden, W. Max 68
corporate bonds 101, 305
Corsetti, Giancarlo 215
Coudert, Virginie 328, 329, 341
Couharde, Ccile 328, 329, 341
CPI index
CPI indices for core ASEAN and
China + Hong Kong 257, 258
weighted CPI indices for trading
partners, core ASEAN and
China + Hong Kong 257
crawling peg arrangement 282
credit default swap (CDS) indices 92
credit guarantees 91, 105
credit rating agencies 91, 105
Crockett, Andrew 268
currency basket denominated bonds 67
East Asian currencies
previous research on 1056
regional bond market indexes 94,

99, 30910, 311
11624, 133
international currencies (euro/dollar/
yen)
liquidity of 13031, 134
risks for bond issuers 105, 1068,
10912
currency baskets
bonds issued in see currency basket
denominated bonds
composition of 67, 16063, 201,
21012, 27980, 288, 290,
30611, 31315
estimation of undisclosed basket
structures 2817
weighted by financial market
activity 307, 311
weighted by GDP size 307, 311
weighted by trade composition
236, 238, 2516, 279,
291302, 3078, 311
currency areas that might adopt a
common basket 223, 236, 238,
241, 242, 254, 256, 2925
ECU as 3031, 107, 208, 237, 256,
290, 306
effect on USDeuro exchange rate
156
effect on USDRMB exchange rate
156
individual-country vs common
baskets 295302
reasons for pegging to 27980, 288,
29091, 302, 31315
and stability of nominal effective
exchange rates 7, 268, 279,
295302
see also Asian Currency Unit (ACU);
Asian Monetary Unit (AMU)
currency competition 203
currency mismatch 62, 172, 174, 196,
204, 269
Dammers, Clifford 318
de Gasperi, Alcide 42
De Haan, Jacob 217
default risk on balance sheets 270

Index
deflation, threat of 1889
delinquent loans 65
Delors, Jacques 45, 54
Delors Report 45, 47, 52, 54, 55, 57
Denmark
exemption from third stage EMU
negotiations 47, 57
monetary policy decisions in 209,
210
Despres, Emile 185
Deutsche Bank 151, 332, 354
developing nations, financial problems
faced by 2034
disintegration of monetary unions 75
distance, effect of, on trade 200, 201,
202
division of labor 290
dollar, US see US dollar
dollar-denominated B shares 360
domestic debt securities outstanding in
Asia 88, 89
Dooley, Michael 157, 158, 185, 276,
326
Dowd, Kevin 206
Duisenberg, Wim 81
Duisenberg plan 53
Dylan, Bob 208, 210
dynamic OLS (DOLS) 334, 336, 338,
339
Dyson, Kenneth 51, 52
EAEG see East Asian Economic
Grouping (EAEG)
EAFTA see East Asian Free Trade
Area (EAFTA)
earnings retention scheme for exports
358
EASG see East Asian Study Group
(EASG)
East Asia Summit (Kuala Lumpur,
2005) 308
East Asian bond markets see regional
bond markets, Asian
East Asian dollar standard 5, 169,
201, 210, 213, 267, 274, 283, 287,
312
East Asian Economic Grouping
(EAEG) 22
East Asian Free Trade Area (EAFTA)
20
393
East Asian Study Group (EASG)

1920, 21
East Asian Vision Group (EAVG) 19,
20
economic development, diversity in
25
economic game 63, 77
connected with security game 789
Economic Review and Policy Dialogue
(ERPD) 304, 315
economies of scale 67, 126
ECOTECH 17
ECU see European Currency Unit
(ECU)
Edwards, Sebastian 212
Eichengreen, Barry 70, 172, 199, 204,
206, 213, 215, 269, 270, 306, 316,
349
Einstein, Albert 36
elasticities model of the balance of
trade 1878
elasticity weights for currency baskets
292
electronic broking system 361, 362,
379, 381
electronics and transport equipment,
trade in 27
ELMI+ see Emerging Local Markets
Index Plus (ELMI+)
EMEAP see Executives Meeting of
East Asia-Pacific Central Banks
(EMEAP)
Emerging Local Markets Index Plus
(ELMI+) 30910, 311
Emerson, Michael 199
energy, trade in 27
Engel, Charles 71
entrept trade 241, 242
equity financing, characteristics of
645
Ernst & Young 349
ERPD see Economic Review and
Policy Dialogue (ERPD)
euro
bidask spreads for trades between
East Asian currencies and 126,
127, 12830, 131, 132
major currencies and 1267,
130, 131, 132

394
Index
bonds denominated in 110, 1734

increasing importance of 2789,
280, 288
introduction of (1999) 31, 41, 46, 55,
76, 185, 208
as proxy for currencies of non-US
market economies 141
RMBeuro exchange rate 143,
1489, 150, 154, 155, 370
RMB spot trading against 365
share in Chinas foreign exchange
market 36970
market 370
USDeuro exchange rate 1425,
14757, 1637 passim, 168,
186, 268, 270, 271, 272, 370
USD spot trading against 378
163, 201, 252, 253, 279, 280,
281, 283, 2856, 287, 288, 296,
297, 298, 299
yen spot trading against 378
euro zone 689, 70, 71, 73, 76
trade intensity index between East
Asian countries and 108
Eurobond markets 31, 1734
Euroland
Asian savings absorbed by 147, 149,
151, 153, 155
assumptions concerning 142
150
definition 141
expected rate of return on capital in
1478
improved outlook for growth in 143,
153
interest rates in 143, 14651 passim,
153, 154, 155, 157
weaker outlook for growth in 1534
European Central Bank (ECB) 46, 50,
52, 76, 81, 167, 278, 279
currency basket weights used by
30811
Executive Board 210
Governing Council 210
European Coal and Steel Community
(ECSC) 81
European Commission 45, 51, 52,

545
European Community 38, 44, 318
European Council 29, 456
European Court of Justice 52
European Currency Unit (ECU) 1, 3,
7, 3031, 44, 46, 107, 208, 237,
256, 290, 306
European Monetary Cooperation
Fund 43, 445, 306
European Monetary Institute (EMI)
46
European Monetary System (EMS) 28,
3031, 37, 445, 47, 756, 107,
196, 208, 256, 288, 290, 306
European Payments Union 14, 28
European Snake 28, 434, 45, 47, 312
European Union 4
capital markets in 279
Common Agricultural Policy (CAP)
14, 26, 289, 43, 57
determination of monetary policy
decisions in 20810
economic cooperation with ASEAN
23
euro introduced (1999) 31, 41, 46,
55, 76, 185, 208
exchange rate cooperation in 289,
3031, 435, 47, 756, 237,
256, 306, 312
FDI in ASEAN region 14
financial services integration in 35
industrial policies in 33
lessons for East Asian countries 24,
2432, 367, 469, 558
monetary integration in 23, 4,
2831, 37, 4158, 689, 712,
756, 196, 199, 207, 20810,
306
political integration in 81
Single European Act 27, 29, 33, 34,
35, 37, 45
Single Market Program 17, 45
Stability and Growth Pact 50, 216
Structural Funds 57
trade patterns 14, 15, 16, 712, 198,
239, 240, 248, 249, 250, 252,
253, 293, 294, 311
exchange rate cooperation see regional
exchange rate cooperation

Index
Exchange Rate Mechanism (ERM) 44,
47, 50, 756, 306, 312
exchange rate spillover effects 196,
1979
exchange-traded funds (ETF) 95, 101
Executives Meeting of East AsiaPacific Central Banks (EMEAP)
23, 81, 87, 90, 91, 92, 94, 96, 97,
98, 100, 105, 304, 311
expected rate of return on capital 1478
export-driven development policy 5,
141, 142, 185, 186, 188, 2056
external balances approach 328, 329,
330, 347
externalities 67, 69
FDI see foreign direct investment
(FDI) flows
FDI intensity 200
Featherstone, Kevin 51, 52
Feldstein, Martin 73, 76, 82
Fernald, John 335, 349, 354
Fernndez-Arias, Eduardo 215
Festinger, Leon 79
financial institution failures 234
financial intermediation
carried out abroad 87, 88, 185
negotiable and market forms of 645
financial maturity, lack of 203
financial services sector, integration
in 35
Finland, monetary policy decisions in
209
First Banking Directive 29
First World War 207
fiscal institutions, development of 205
Fischer, Stanley 159, 267
fixed exchange rate regime 68, 6970,
71, 73
see also Bretton Woods System
Flandreau, Marc 203
floating exchange rate regime 68, 70, 73
flow equilibrium models 327, 328, 329
Folk theorem 77
Folkerts-Landau, David 185
foreign direct investment (FDI) flows
to ASEAN region 14, 26, 27, 34, 37
to China 175, 179, 185, 291, 373
intra-regional in East Asia 197,
200201, 202
395
weights in currency baskets

influenced by 2912
foreign exchange market in China see
foreign exchange reserves
accumulation of 1678, 169, 1789,
18081, 186, 1889, 204,
2734, 2768, 2878, 3256,
329, 3489, 3724, 375, 382
in ACU 3056
conversion of dollar assets into euro
assets 280
Foreign Invested Enterprises (FIEs),
foreign exchange transactions of
359
France
210
role in EMU negotiations 513
see also French franc
Frankel, Jeffrey A. 16, 28, 71, 73, 159,
160, 195, 200, 201, 211, 212, 215,
216, 268, 275, 278, 281, 285, 328,
329, 341, 342, 349
Fratzscher, Marcel 212
free trade, definition of 17
French franc 43, 47, 82
Fritz, Barbara 205, 215
FTA-plus arrangement 32
Fukuda, Shin-ichi 70
functional approach toward integration
42
Funke, Michael 328, 329, 334
G3 currency basket 105
Gale, Douglas 82
GAO see Government Accountability
Office (GAO)
Garber, Peter 185
Gartzke, Eric 74
GATT/WTO 24, 26, 33
see also World Trade Organisation
(WTO)
German Mark
appreciation of 43
Benelux currencies and Danish
krone tied to 44
credibility and reputation of 50, 76
exchange rate between D-Mark and
O-Mark 69

396
Index

dollar 163, 164, 165, 166, 270
weight in currency baskets 2823,
2856
Germany
Bundesbank discount rate 20810
foreign exchange reserves 169, 277
208, 210, 212
reunification of 4950, 69, 76
role in EMU negotiations 4951, 81
see also German Mark
Giavazzi, Francesco 207
Giradin, Eric 319, 320
Glick, Reuven 71, 75, 215
Global Financial Data 209, 225, 254,
255, 257, 258, 259, 260
global financial system 3, 478, 62
globalization of production 326
gold standard 68, 207
Goldstein, Morris 215, 268, 328, 329,
349
Golub, Stephen 350
Goodhard, Charles 208
Goto, Junichi 72
Gourinchas, Pierre Olivier 158
Government Accountability Office
(GAO) 327
government interventions in bond
markets, justification for 667
Goyal, Rishi 178, 187, 189
gravity model 71, 200
Greece, monetary policy decisions in
208, 209
Greenaway, David 206
Groningen Growth and Development
Center 335, 354
gross capital formation, ASEAN, by
country, since 1990 227, 228, 229,
230, 231, 234, 235
gross domestic product (GDP)
ASEAN, by country, since 1990 226,
227, 228, 229, 235
size of
effect on currency status 206, 207
effect on trade 200, 201, 202
weights in currency baskets related
to 307, 311
Gudmundsson, Mr 279, 280
Gyntelberg, Jacob 102
Habsburg Empire 69
Hague Summit (1969) 43, 44, 55
Hamada, Koichi 68, 70, 72, 73, 82, 320
Hard ECU (HECU) 52
harmonization, definition of 75
Harris, Lawrence 382
Hartmann, Philipp 126
Hausmann, Ricardo 172, 204, 269, 270
Hawk Dove game 63, 779
Hefeker, Carsten 357
Hernndez, Leonardo 166
Heston, Alan 341, 355
Hiemenz, Ulrich 38
hierarchy of currencies 2034
Hillebrand, Eric 173, 189, 274
Hinkle, Lawrence E. 349
Ho, Corinne 357
Holtz-Eakin, Douglas 329
Holz, Carsten 350, 354
home bias in investment preferences
144, 145
home country principle 35
Hong Kong
current account balance 171, 181
foreign exchange reserves 169, 181,
277
foreign exchange risk facing 11011
interest rates 182, 183, 184, 212
nominal effective exchange rates 299
trade patterns 198, 242
currency baskets based on 292,
293, 294, 297, 299
see also Hong Kong dollar
Hong Kong Bond Index Fund 93
Hong Kong dollar
major currencies and 127, 130
125
211, 281, 282, 283, 284, 285,
286, 292, 293, 294, 297, 299
dollar 164, 165, 166, 167, 168,
267, 271, 272
nominal effective exchange rates
299
pegged to US dollar 111, 113, 161,
165, 210, 211, 212, 267

Index
309, 311
Hong Kong Stock Exchange 94, 97
Horvath, Michael 356
Hsieh, David 350
Hugo, Victor 42
Hume, David 215
Hurricane Katrina 143, 155
Hyderabad, East Asian Finance
Ministers meetings in 304
iBoxx Pan Asia Index 94, 99, 30910,
311
IIC see International Index Company
(IIC)
ILO see International Labor
Organization (ILO)
impediments index 989
in-kind transactions 95, 96
India, trade patterns of 292, 293, 294
Indian rupee 310
Indonesia
capital controls imposed by 89
CPI index for 257, 258
180, 204, 227, 230, 232, 235, 273
external debt 227, 232
FDI in 27
financial institution collapses in 234
204, 277
foreign exchange risk facing 109,
110, 111, 132, 133
GDP 226, 227
gross capital formation 227, 231, 234
interest rates 1723, 183, 184, 203
299
REER movements since 1990 233,
234, 235
savings 227, 230, 231
trade patterns 14, 15, 27, 197, 198,
239, 240
253, 254, 255, 292, 293, 294,
297, 299, 300
by sector 242, 243, 244, 2467
397
by sector and market 245, 248,

249, 250
partners 257
see also Indonesian rupiah
Indonesian rupiah
bidask spreads for trade between
major currencies and 127, 129,
131
125
currency basket of 67, 162, 163,
21012, 251, 252, 253, 2816
passim, 293, 294, 297, 299, 300
dollar 1648 passim, 224, 225,
260, 267, 271
299
pegged to US dollar 161, 267
REER indexes 254, 255
infant industry argument 33
inflation rates
in ASEAN countries 29, 30
dollar pegs and 188, 267, 268,
2758
in Europe 187
harmonization of 75
in Japan 187
and monetary policies 210
and nominal exchange rates 2568,
268, 328
political stability and 205
restrictions on 30
in United States 187, 2758
information costs 69, 72, 76, 126
institutional environment 24, 356, 48,
66, 205
inter-bank foreign exchange market,
Chinese 359, 36062
market participants and
concentration 3713
market segments 36671
enrichment of 37881
non-financial members entering
376
size of 3626
supply and demand for FX not fully
reflected in 374

398
Index
inter-bank forward markets 361, 362,

365, 3669, 371, 378
problems in 37981
inter-bank spot markets 361, 362, 365,
366, 367, 369, 371, 378, 379, 382
interest rate parity pricing rules
38081
interest rates
Bundesbank discount rate 20810
Cambodia 203
changes in currency basket weights
affecting 308
China 179, 182, 183, 184, 2034,
212, 372, 374, 375, 380, 381
currency competition and 2034
divergence within East Asian
countries 29, 30, 2045
Euroland 143, 14651 passim, 153,
154, 155, 157
interest rate policy under floating
exchange rates 212
harmonization within EU 75, 76
Hong Kong 182, 183, 184, 212
Indonesia 1723, 183, 184, 203
Japan 11920, 1768, 182, 184, 187,
212, 280, 288
Korea 1723, 182, 183, 184
Laos 203
Malaysia 1723, 182, 183, 184,
2034
Philippines 1723, 1824, 2034
risk premia on 1723, 17684, 270,
274
Singapore 182, 183, 184
Taiwan 182, 183, 184
Thailand 1723, 182, 183, 184,
2034
United States 142, 143, 14552
passim, 154, 155, 157, 168,
1768, 179, 1824, 212, 276,
278, 283, 288
Intergovernmental Conferences (IGCs)
45
international capital flows, as incentive
for exchange rate stabilization
26974, 287
international currency, requirements
for acceptance as 2046, 31213,
357, 384
International Finance Corporation 91
International Index Company (IIC) 92,

98, 99, 100
International Labor Organization
(ILO) 355
International Monetary Fund (IMF) 3,
17, 23, 68, 159, 267, 300301, 320,
328, 329, 331, 332, 359
Annual Report on Exchange
Arrangements and Exchange
Restrictions 211
Direction of Trade Statistics (DOTS)
15, 136, 198, 200, 233, 239, 240,
254, 255, 257, 258, 259, 294, 298
International Financial Statistics
(IFS) 161, 169, 171, 181, 182,
183, 201, 261, 272, 277, 295, 354
World Economic Outlook (WEO) 171
intra-regional FDI 197, 200201, 202
intra-regional trade
East Asia 1416, 241, 267, 292
trade creating effects of monetary
integration 197202, 268,
279
European Union 14, 71
inventory holding costs 126
Italy
210
withdrawal from ERM 47
Ito, Hiro 355
Ito, Takatoshi 82, 105, 135, 217, 279,
318
JACI see Asia Credit Index (JACI)
Japan
173, 175, 176, 178, 179, 180,
204, 273
foreign exchange reserves 169,
1789, 180, 204, 273, 274, 276,
277, 278
interest rates 11920, 1768, 182,
184, 187, 212, 280, 288
net international investment position
175, 176
recovery from East Asian crisis 61
relations with China 2, 3, 48, 55, 62,
79, 80

Index
1, 4, 17, 48, 55, 578, 612
trade patterns 14, 15, 71, 108, 168,
197, 198, 239, 240, 248, 249, 250
252, 253, 292, 293, 2945,
296
see also Japanese yen
JapanASEAN Economic Partnership
21, 23
JapanSingapore Economic
Partnership Agreement (JSEPA)
17, 21
Japanese Ministry of Finance 175, 176
Japanese yen
127, 12830, 131, 132
130, 131, 132
risks for bond issuers 110
116, 117, 11920, 121, 124,
125
weight in AMU 11619, 121, 124,
1334
2945, 296
declining in importance 48
euro spot trading against 378
origin of 80
as regional anchor currency 31213,
314, 357
returns to AMU denominated bonds
evaluated in terms of 119, 121,
122, 124, 1334
returns to local currency
denominated bonds evaluated in
terms of 112, 116, 11718, 133
RMB forward trading against
37980
market 370
USDyen exchange rate 1638
passim, 173, 176, 178, 260, 270,
271, 274, 279, 312
399

163, 201, 251, 252, 253, 27983,
286, 287, 288, 295301 passim,
309, 311
Jeanne, Olivier 205
Jiang, Guorong 90, 101
Jonas, Eva 82
judicial system 36
Kakkar, Vikas 350
Kawai, Masahiro 16, 38, 104, 105, 107,
135, 197, 215, 263, 279, 291, 305,
318, 320
Kenen, Peter 75
key currency status, requirements for
2046, 31213, 357, 384
Khan, Mohsin 350
Kim, Chung-Jin 216
Kim, Heungchong 59
Kobe Research Project 55
Kohl, Helmut 4950, 81
Korea
capital controls imposed by 89, 97
179, 180, 204, 273
financial institution collapses 234
180, 204, 273, 276, 277
110, 112, 132, 133
interest rates 1723, 182, 183, 184
299
1, 4, 556, 578, 612
tax reforms in 97
trade patterns 197, 198, 239, 240,
241, 248, 249, 250
253, 293, 294, 297, 299
see also Korean won
Korean won
121, 125
163, 21012, 2826 passim,
293, 294, 297, 299

400
Index

271, 272
299
279, 297, 309, 311
Kregel, Jan 205
Kuroda, Haruhiko 61, 104, 263, 305,
320
Kwan, Chi Hung 385
labor mobility 22, 34, 45
Lai, Kon S. 349, 356
Lao kip
dollar 260
Laos
AFTA membership 22
interest rates 203
openness of 25
by sector 242, 243, 244, 2467
see also Lao kip
Latin American Free Trade Area 14
learning by doing 87, 934, 100101
Lee, Jaewoo 350
Lee, Jong-Wha 216
Leung, Julia 92
liability dollarization 26972
Liang Yuanyuan 385
liquidity
in Asian bond markets 12431, 132,
134
in Chinas foreign exchange market
3812
liquidity trap 174, 184, 187
local currency issuance by
multinational corporations 91
long-term exchange rate fluctuations
270, 272
Luxembourg, monetary policy
decisions in 209, 210
Lyons, Richard K. 380
Ma, Guonan 89, 102, 385

Maastricht Treaty (1992) 456, 47, 55
convergence criteria 46
Macau pataca, weight in currency
baskets 309
Makowiak, Bartosz 215
macroeconomic stability, policies
to maintain 267, 268, 275, 278,
27980, 2878, 291
Maes, Ivo 53, 54
Mahathir bin Mohamad 22
Malaysia
CPI index for 258
179, 180, 204, 227, 230, 232,
235, 273
FDI flows to 27
180, 204, 276, 277
foreign exchange risk facing 110, 111
GDP 226, 228
gross capital formation 227, 231,
234
interest rates 1723, 182, 183, 184,
2034
299
235
savings 227, 230, 231
tax reforms in 97
trade patterns 14, 15, 27, 197, 198,
239, 240, 241
254, 255, 293, 294, 297, 299
by sector 243, 244, 2467
249, 250
partners 257
see also Malaysian ringgit
Malaysia Bond Index Fund 93
Malaysian ringgit
131

Index
116, 117, 11920, 125
211, 252, 253, 2826 passim,
293, 294, 297, 299
260, 267, 271, 272
299
pegged to US dollar 111, 113, 161,
165, 210, 211, 259, 267
298, 309
managed float 290, 302
Mansfield, Edward 74, 82
market impediments 8991
removal of 94, 96100
market making 956, 360, 362, 374,
375, 379, 3813
market microstructure theory 3778
market openness factor 989, 100
market size
3626
portfolio allocation determined by
98
market weights, determination of 94,
98100
Marston, Richard 350
Matsuyama, Kiminori 206
maturity mismatch 62, 172, 204
McCauley, Robert N. 88, 89, 90, 101,
318
McKinnon, Ronald 71, 73, 159, 160,
163, 166, 168, 170, 172, 173, 175,
176, 178, 186, 187, 188, 189, 190,
199, 204, 210, 267, 268, 273, 275,
279, 283, 291, 298
Meese, Richard A. 327
MERCOSUR 32
minimum variance analysis 280
Minsky, Hyman 205
Mitterand, Franois 49, 51
Miyazawa Plan 23
monetary compensation amounts
(MCAs) 289
monetary integration
ACUs role in 306, 31617
401
benefits and costs of 6774

economic and conceptual aspects
6771
empirical appraisals 712
political economy aspects 724
case for, in East Asia 56, 30,
195214, 22263
degrees of 679
and development of regional capital
markets 2028, 3056
economist approach to 5051,
523, 54
in EU 23, 4, 2831, 37, 4158,
689, 712, 756, 196, 199,
207, 20810, 306
international political conflicts and
746, 8081
game theory applied to 7780
monetarist approach to 51, 54
and recovery of monetary policy
influence 20813
roles of East Asian states in 556,
578
trade creating effects of 197202,
268, 279
monetary policy
ACUs role in coordinating 306,
31617
effective coordination in ASEAN
and China 223, 25662
loose in United States 2756
monetary integration and recovery
of influence over 20813
monetary regimes, credibility of 205
money, functions of 2023
Monnet, Jean 42
Montiel, Peter J. 166, 320, 349
moral hazard 104, 159
Mori, Junichi 318
multinational corporations 91, 197, 380
Mundell, Robert A. 69, 223, 238, 262
Murase, Tetsuji 55, 56
Myanmar
AFTA membership 22
openness of 25
by sector 243, 244, 2467
see also Myanmar kyat

402
Index
Myanmar kyat
dollar 260
Myerson, Roger B. 77, 78, 79, 80
Nabor, Andreas 357
NAFTA see North American Free
Trade Agreement (NAFTA)
Nash equilibrium 77, 314
national jurisdictions, legal
accommodation of 97
NATO see North Atlantic Treaty
Organization (NATO)
Naya, Seiji F. 16, 21, 38
negative risk premium 17684
nested games 63, 77, 78
net investment income payments, US
1512
Netherlands, monetary policy
decisions in 209
network externalities 206
New Miyazawa Plan 17
New Zealand dollar 310
Nielsen, A. 82
Nitsch, Manfred 203, 215
correlations in monthly movements
in 223, 2569
currency baskets and stability of 7,
268, 279, 295302
currency weights used by central
banks in defining 30811
inflation rates and 2568, 268,
328
non-deliverable forwards (NDF)
market 377, 38081
non-tariff barriers 37
North American Free Trade
Agreement (NAFTA) 15, 17,
712, 215
North Atlantic Treaty Organization
(NATO) 50
Norway, monetary policy decisions in
208, 209
OCA see optimum currency area
(OCA)
OECD see Organization for Economic

Cooperation and Development
(OECD)
Ogaki, Masao 350
Ogawa, Eiji 1056, 107, 135, 136, 217,
279, 319
Ohno, Kenichi 176, 178, 188
oigashi (roll-over loan) 65
oil exporters, Asian savings consumed
by 143, 1556
Okinawa islands 62
one-way quotes 380
Oneal, John R. 74
optimum currency area (OCA) 195,
223, 262, 312, 320
order-driven auction market 361, 362,
381
order processing costs 125, 126
Organization for Economic
Cooperation and Development
(OECD) 35
original sin 6, 1723, 185, 196, 204,
206, 207, 269
Orphanides, Athanasios 216
Pacific Economic Cooperation Council
(PECC) 9091
Pagano, Marco 207
PAIF see Pan-Asian Bond Index Fund
(PAIF)
Pan-Asian Bond Index Fund (PAIF)
92, 93, 946, 97, 319
101
incentives to reduce impediments in
98100
parallel currency approach 213, 306, 316
Park, Sung-Hoon 59
Park, Yung Chul 105, 318
participating dealer model 95
PBOC see Peoples Bank of China
(PBOC)
PECC see Pacific Economic
Cooperation Council (PECC)
Penn World Tables 341, 355
Peoples Bank of China (PBOC) 8, 73,
212, 358, 36061, 362, 365, 368,
372, 377
passive intervention of 3746,
3823

Index
Petri, Peter A. 107
Philippine Bond Index Fund 93
Philippine peso
117, 125
163, 210, 211, 252, 253, 2826
passim, 293, 294, 297, 299,
300301
267, 271, 272
299
pegged to US dollar 161, 210, 267
Philippines
CPI index for 258
180, 204, 228, 230, 232, 235, 273
204, 277
111, 132
GDP 226, 228
234
interest rates 1723, 1824, 2034
299
234, 235
savings 228, 230, 231
tariffs in 33
tax reforms in 97
trade patterns 14, 15, 27, 197, 198,
239, 240, 241, 242
254, 255, 293, 294, 297, 299,
300301
by sector 242, 243, 244, 2467
259, 250
403

partners 257
see also Philippine peso
Pill, Huw 159, 173
Plummer, Michael G. 21, 32, 38, 39
policy assignment, theory of 69
policy externalities 30
political conflicts, relationship between
economic interdependence and
746, 8081
game theory applied to 63, 7780
political stability, requirement for 205
Pollins, Brian 74, 82
Pomfret, Richard 29
population, effect on trade 2012
portfolio balance model 150
portfolio effect of holding AMU bonds
106, 119, 1224, 133
portfolio management 65, 87, 122
Portugal, monetary policy decisions
in 209
poverty rates 234, 235
Prasad, Eswar 329, 349
Preeg, Ernest H. 372
price discovery 382
pricespecie flow mechanism 68
primary-based exports 267
primary market, transactions in 95
principal component analysis 72
prisoners dilemma 63, 77, 78, 79
product testing 34
production fragmentation 326
productivity approach 328, 330,
33341
pseudo-exchange rate union 68, 69
public goods 69, 90, 189, 199
purchasing power parity (PPP)
absolute PPPincome relationship
327, 328, 329, 330, 3415, 346,
347
relative comparisons of 3278, 329
pyramid-climbing 204
quote-driven dealer market 362, 381,
382
Radelet, Steven 234
Rahn, Jrg 328, 329, 334
Rajan, Ramkishen 320
Rashomon 327

404
Index
real effective exchange rates (REERs)

gains from stabilizing 2358, 291
movements in core ASEAN
countries since 1990 2314
relationship between current
account balances and 235
nominal components of see nominal
effective exchange rates
weights for currency baskets of
ASEAN and China 236, 238,
2514
movements for REER indexes
using 2546
regional bond market indexes, private
sector 94, 99
currency weights used in 30910,
311
regional bond markets, Asian 1, 4, 61,
62, 87101
Asian Bond Fund I 23, 81, 91, 105,
207, 319
Asian Bond Fund II 4, 23, 81, 87,
90, 914, 96, 97, 105, 207, 319
101
incentives to reduce impediments
in 94, 98100
learning by doing element in 87,
934, 100101
Asian Bond Market Initiative
(ABMI) 21, 23, 28, 81, 91, 105,
304, 315
currency denomination in 10435,
2367, 305
economic and conceptual issues
645
incentives to further reduce
impediments 98100
indexes for 94, 99, 30910, 311
liquidity in 12431, 132, 134
market impediments already reduced
968
overview of local currency bond
markets 8890, 1045
Pan-Asian Bond Index Fund (PAIF)
92, 93, 946, 97, 319
101
incentives to reduce impediments
in 98100
policy coordinating, reserve sharing

and 2368
political economy aspects 657
regional cooperative initiatives
and the ABF project 9091
shocks and development of 207
regional bond markets, European 31,
1734
regional capital markets,
development of 2028, 3056
regional exchange rate cooperation 1
case for, in East Asia 6, 1979,
22263
in EU 289, 3031, 435, 47,
756, 237, 256, 306, 312
role of Chinese currency in 78,
376, 3834
types of 68
see also Asian Currency Unit
(ACU); Asian Monetary Unit
(AMU); currency baskets
Reinhart, Carmen M. 173, 216, 349
Remolona, Eli M. 101, 102
renminbi see Chinese renminbi
request for quote trading arrangements
362
Research Institute of Economy, Trade
and Industry (RIETI) 30910,
311
reserve money, correlations in monthly
changes in 223, 259, 261
reserve sharing 2368, 262, 306, 315,
316
see also Chiang Mai Initiative (CMI)
Reszat, Beate 47
retail foreign exchange market, Chinese
36062
imbalance in 380
market segments 366, 367
Rey, Helene 158
rice, trade in 22, 27
Riese, Hajo 204, 205
RIETI see Research Institute of
Economy, Trade and Industry
(RIETI)
Rogoff, Kenneth 327
Rose, Andrew K. 28, 71, 73, 195, 199,
200, 215, 268
Rossi, Vanessa 335
Roy, Tobias 195

Index
Royama, Shoichi 82
Russett, Bruce M. 74
Russia 207, 273, 274, 278, 280
Ryou, Jai-Won 81, 82
Sachs, Jeffrey 234
SAFE see State Administration of
Foreign Exchange (SAFE)
Sakakibara, Eisuke 55
Samuelson, Paul 333
Sarno, Lucio 385
Saudi Arabia 278
savings
ASEAN, by country, since 1990 227,
228, 229, 230, 231
Asian 142, 143, 1456
absorbed by Euroland 147, 149,
151, 153, 155
absorbed by US 142, 144, 147,
149, 151, 153, 155, 157
consumed by oil exporters 143,
1556
virtuously high 174, 175
US rate of 142, 144, 145, 1845,
186, 188, 278
Scandinavian Monetary Union 82
Schelkle, Waltraud 195, 203
Schnabel, Gunther 160, 166, 168, 172,
173, 186, 188, 189, 190, 210, 267,
270, 273, 274, 275, 279, 280, 288
Schuman, Robert 42
Second World War 16, 43, 52, 53, 56,
81
secondary market, transactions in
956
securitized debt instruments 91
security game 63, 778
connected with economic game
789
services, free movement of 34, 37, 45
Sharpe ratio 112, 116, 119, 120, 121,
133, 134
Sheng, Andrew 101
Shimizu, Junko 105, 106, 107, 136, 319
shocks 143, 155, 207, 223, 29091, 312,
347
short-term exchange rate stabilization
26970, 271
Short-Term Monetary Support Facility
445
405
Shu Youdong 372

Singapore
CPI index for 257, 258
175, 180, 204, 228, 232, 235, 273
204, 277
110, 111
GDP 226, 228
gross capital formation 229, 231, 234
299
PAIF domiciled in 97
REER movements since 1990
2323, 235
savings 229, 230, 231
tariffs in 32
trade patterns 14, 15, 27, 197, 198,
239, 240, 241
254, 255, 256, 293, 294, 297,
299
by sector 243, 244, 2467
249, 250
partners 257
see also Singapore dollar
Singapore Bond Index Fund 93
Singapore dollar
131, 132
11920, 121, 125
21012, 252, 253, 2816
passim, 293, 294, 297, 299
267, 271, 272
299
REER indexes 254, 255, 256

406
Index

309, 311
Single European Act 27, 29, 33, 34, 35,
37, 45
single local currencies, bonds
denominated in 105
risks for bond issuers 10911
11718, 124, 125, 133, 134
Single Market Program 17, 45
Sinochem Corporation 376
Siregar, Reza 320
size of economy
effect on currency status 206, 207
effect on trade 200, 201, 202
weights in currency baskets related
to 307, 311
Snake, European 28, 434, 45, 47,
312
sovereign credit rating, portfolio
allocation determined by 98
Soviet Union 69
Spaak Report 53
Spain, monetary policy decisions in
209, 210
speculative pressures, exchange rate
coordination to combat 236
Stability and Growth Pact 50, 216
standards, harmonization of 34
State Administration of Foreign
Exchange (SAFE) 36061, 364
state-owned banks 358
state-owned enterprises (SOEs) 336
Steinherr, Alfred 319, 320
Stock, James H. 334
stock markets 30, 360
Stoltenberg, Gerhard 51
Story, Jonathan 39
subsidiarity principle 36
substitutability of assets 142, 146, 148,
150, 205
Summers, Robert 341, 355
swap markets 3589, 360, 366, 367,
369, 371
Sweden, monetary policy decisions in
209, 210
Swiss franc
dollar 163, 164, 165, 166, 167,
168, 271, 282

market 370
used as numraire 16063, 201, 281
Switzerland
monetary policy decisions in 209
see also Swiss franc
Szirmai, Adam 336, 338, 339, 340, 354
Taipei, China see Taiwan
Taiwan
179, 180, 204, 273
180, 204, 273, 277
110, 111
299
trade patterns 239, 240, 241, 248,
249, 250
253, 293, 294, 2978, 299
see also Taiwan dollar
Taiwan dollar
bidask spread for trade between
US dollar and 127, 127
2826 passim, 293, 294, 2978,
299
271, 272
299
310, 311
Takagi, Shinji 66, 135, 291, 318
Takeuchi, Atsushi 89
Tamamura, Chiharu 27
Tang, Man-Keung 350
tariff harmonization 324
taxes 90, 97, 99, 205
Taylor, Alan M. 75
Taylor, Mark P. 385
Taylor rule 208
Tchaidze, Robert 216
technical assistance 91

Index
Thai baht
131, 132
125
contagion caused by depreciation of
23031, 2356
163, 21012, 253, 254, 2826
passim, 293, 294, 298, 299
267, 271, 272
299
Thailand
CPI index for 258
180, 204, 229, 230, 232, 235, 273
external debt 229, 230, 232
financial institution collapses in 234
180, 204, 277
110, 111, 132
GDP 226, 229
231, 234
interest rates 1723, 182, 183, 184,
2034
299
235
savings 229, 230, 231
tax reforms in 97
trade patterns 14, 15, 27, 197, 198,
239, 240
253, 254, 255, 293, 294, 298,
299
by sector 242, 243, 244, 246
by sector and market 248, 249, 250
407

partners 257
see also Thai baht
Thailand Bond Index Fund 93
Thatcher, Margaret 52, 54
Tindermans Report 53
TIPS see Treasury inflation-protected
securities (TIPS)
township and village enterprises
(TVEs) 336
trade-account countries 186
trade-creating effects of monetary
integration 197202, 268, 279
trade diversion 25, 27, 32, 71
trade intensity
index of 72, 1078
11112, 133
measurement in gravity model 200
trade liberalization 245, 27
trade weights, currency baskets based
on 1067, 110, 111, 133
transaction-based cost 1267
transaction costs
in Chinas FX market 369, 377
from currency conversions 45, 69,
126, 206, 267
economic cooperation to reduce 22,
27, 34, 69, 72, 76, 267
for international trade 268, 377
investors decisions influenced by
135
regional bond index to reduce 94
transportation costs 326
Treasury inflation-protected securities
(TIPS) 145
Treaty of Rome 28, 29, 37
Turner, Philip 215
turnover ratio, portfolio allocation
determined by 98
twin crisis 234
UEMOA see West African Economic
and Monetary Union (UEMOA)
UNCTAD Foreign Direct Investment
Statistics 200
underwriting 64
United Kingdom
exemption from third stage EMU
negotiations 47, 57

408
Index

216
opt-out from ERM 44, 47
role in EMU negotiations 523
trade patterns 311
see also British pound
United States
Asian savings absorbed by 142, 144,
147, 149, 151, 153, 155, 157
144, 145, 1467, 149, 15051,
152, 171, 173, 174, 1845, 186,
188, 189, 273, 278, 287
exogenous shocks to economy 143,
155
expected rate of return on capital
1478
FDI in China 185
foreign exchange reserves 169, 277
improved outlook for growth in 143,
154
inflation rate 187, 2758
passim, 154, 155, 157, 168,
1768, 179, 1824, 212, 276,
278, 283, 288
labor productivity in 336
210, 212
net investment income payments
1512
savings rate 142, 144, 145, 1845,
186, 188, 278
trade patterns 14, 71, 108, 168, 198,
239, 240, 241, 242, 24553, 293,
311
see also US dollar
Urata, Shujiro 197, 215
Uruguay Round 33
US Bureau of Labor Statistics 335
US dollar
127, 12830, 131, 132
130, 131, 132
ABFI limited to 912, 105
liquidity of 131, 134

correlations of monthly changes
of USD exchange rates for
ASEAN + 3 countries 259, 260
East Asian currencies pegged to
emancipation from 89, 156,
2789
monetary independence lost
through 196, 21013, 375,
382
reasons for continuance of
5, 14157, 15989, 267,
26874, 287, 291, 326
risks associated with 10911, 113,
116, 123, 124, 133, 2758,
2878, 312
see also Bretton Woods II view;
East Asian dollar standard
euro spot trading against 378
euroUSD exchange rate 1425,
14757, 1637 passim, 186,
268, 270, 271, 272, 370
exchange rate volatility against
1638, 2246, 259, 260, 2678,
27072, 275, 282
HKD spot trading against 378
as key currency 16870, 269, 273,
276
monthly USD exchange rates for
core ASEAN countries 224,
225
returns to AMU denominated bonds
evaluated in terms of 11921,
124, 133, 134
returns to local currency
denominated bonds evaluated in
terms of 11216, 124, 125, 133
RMB forward trading against
37980
RMBUSD exchange rate 5, 1415
passim, 14850, 154, 155, 156,
161, 1648 passim, 175, 179,
210, 211, 259, 260, 267, 271,
272, 274, 359, 360, 36770
passim, 3745
determination mechanism 3823
equilibrium value of 32549

Index
share in Chinas foreign exchange
market 36971, 375
market 370
weight in currency baskets 67,
16063, 201, 21012, 251, 252,
253, 27985, 287, 2956, 297,
298, 299
yen spot trading against 378
yenUSD exchange rate 1638
passim, 173, 176, 178, 260, 270,
271, 274, 279, 312
US Federal Reserve, currency basket
weights used by 30811
Verdun, Amy 53, 54
vertical specialization 326
Very Short-Term Financing (VSTF)
Facility 45
Vietnam
AFTA membership 22
openness of 25, 334
by sector 242, 243, 244, 2467
see also Vietnamese dong
Vietnamese dong
dollar 260
international acceptance of 207
Vision 2020 1819, 21, 22, 33
Volz, Ulrich 59, 73, 201, 214, 217, 263,
288
von Furstenberg, George M. 217
Wald test statistic 356
Walter, Ingo 39
Wang, Seok-Dong 48
409
Wang, Tao 328, 329

Wang, Yunjong 59
wars, effects of 207
Watson, Mark W. 334, 352
weak currency status 203
Wei, Shang-Jin 160, 201, 211, 281, 285,
329
Wen Jiabao 385
Werner Report 28, 43, 44, 47, 53, 55,
57, 75
West African Economic and Monetary
Union (UEMOA) 82
Williamson, John 135, 217, 268, 279,
292, 302, 320, 349
withholding taxes 90, 97, 99
World Bank
East Asia Update 234, 254
local currency bonds issued by
91
World Development Indicators 226,
229, 231, 232, 233, 255, 257,
258, 259, 335, 341, 3545
World Trade Organization (WTO) 17,
245, 26, 33, 364, 376
World War I 207
World War II 16, 43, 52, 53, 56, 81
WTO see World Trade Organization
(WTO)
Xin Wang 372
Yasukuni Shrine 62
yen see Japanese yen
Yi, Kei-Mu 326
Young, Alwyn 263, 335, 336, 350
yuan see Chinese renminbi
Zaibatsu 64
Zhang Guang Ping 368
Zhang Jikang 385
Zhang Zhichao 328, 329, 385



Hamada Koichi 2009 Towards Monetary and Financial Integration in East Asia

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Hamada Koichi 2009 Towards Monetary and Financial Integration in East Asia

Hochgeladen von

Copyright:

Verfügbare Formate

Koichi Hamada, Beate Reszat and Ulrich Volz 2009

Chapter 11 The Peterson Institute for International Economics 2009

ISBN 978 1 84844 363 1

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Introduction: prospects for monetary and financial integration in

THE POLITICAL ECONOMY OF REGIONAL

1 The ASEAN economic community and the European experience

DEVELOPING BOND MARKETS IN EAST ASIA

Learning by doing in market reform: lessons from a regional

EXCHANGE RATE POLICIES IN EAST ASIA:

East Asias role in the revived Bretton Woods system

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Three cases for monetary integration in East Asia

THE ROLE OF CHINA

The illusion of precision and the role of the renminbi in

CURRENCY BASKETS FOR EAST ASIA?

Capital markets and exchange rate stabilization in East Asia:

TOWARD MONETARY INTEGRATION IN EAST

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Liang Yuanyuan, Research Associate, Center for International Finance

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

community as reflected in the Rome Treaty of 1957 and the creation of

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Introduction: prospects for monetary

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Towards monetary and financial integration in East Asia

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

monetary cooperation and financial integration in East Asia. European

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Towards monetary and financial integration in East Asia

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Towards monetary and financial integration in East Asia

by estimating the effect of exchange rate stability on trade in East Asia, he

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

yen weights in the currency baskets of Indonesia, Korea, the Philippines,

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Towards monetary and financial integration in East Asia

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

monetary integration was the collapse of the Bretton Woods system. If we

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

The ASEAN economic community

In November 2002, it was proposed at the Association of Southeast Asian

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

The political economy of regional integration

INTRA-REGIONAL ASEAN CONTEXT

As of 2004, intra-regional trade in ASEAN came to somewhat less than

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

World ASEAN- ASEAN- CH (%)

ASEAN exports to selected partners: 2004

IMF, Direction of Trade Statistics, CD-Rom (April 2000).

The political economy of regional integration

EVOLUTION OF ASEAN ECONOMIC

There have been many excellent surveys of regional economic integration

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

The ASEAN economic community and the European experience

Koichi Hamada, Beate Reszat and Ulrich Volz - 9781848443631

The political economy of regional integration