FINANCE · Periodicals-class postage is paid at Washington, DC and at additional mailing offices. The English edition is printed at Cadmus Journal Services, Lancaster, PA. Postmaster:

FINANCE* ) DEVELOPMENT is published

quarterly in English, Arabic, Chinese,French, and Spanish by theInternational Monetary Fund,Washington, DC 20431, U.S.A.

Opinions expressed in articles andother materials are those of the authors;they do not necessarilyreflect IMF policy.

Ian S. McDonaldEDITOR-IN-CHIEF

Asimina CaminisSENIOR EDITOR

Paul GleasonASSISTANT EDITOR andADVERTISING MANAGER

Elisa DiehlASSISTANT EDITOR andBOOK REVIEW EDITOR

Luisa Menjivar-MacdonaldART EDITOR

June LavinSENIOR EDITORIAL ASSISTANT

Jessie HamiltonADMINISTRATIVE ASSISTANT

ADVISORS TO THE EDITORPeter AllumHugh BredenkampAdrienne CheastyEric CliftonNaheed KirmaniRoger KronenbergAnne McGuirkElizabeth MilneOrlando RoncesvallesJerald SchiffGarry Schinasi

Periodicals-class postage is paid atWashington, DC and at additionalmailing offices. The English edition isprinted at Cadmus Journal Services,Lancaster, PA. Postmaster: please sendchange of address to:

Finance & Development700 19th Street NWWashington, DC 20431 USATelephone: (202) 623-8300Fax Number: (202) 623-6149E-mail: [email protected] site: http://www.imf.org/fanddEnglish edition ISSN 0015-1947

© 1998 by the International MonetaryFund. All rights reserved. Requests forpermission to reproduce articles should besent to the Editor-in-Chief. Finance &Development will normally give permissionpromptly, and without asking a fee,when the intended reproduction is fornoncommercial purposes.

J ANUARY 1, 1999 will mark a historic moment in international monetary coopera-tion as the European Economic and Monetary Union (EMU), an area with an econ-omy equal to that of the United States, is born. On that date, 11 of the EuropeanUnion's 15 member countries will lock their exchange rates; the common currency,

the euro, will come into being; and national authorities will hand over the responsibil-ity for monetary and exchange rate policies to area-wide institutions. Several articles inthis issue of Finance & Development consider the wider significance of this epochal event. Inthe first, Michael C. Deppler discusses the implications of EMU for the IMF's sur-veillance activities. Some of the major policy challenges facing the euro area in main-taining growth and curbing unemployment are examined by Klaas Knot, DonoghMcDonald, and Karen Swiderski. The impact of EMU on the other economies ofWestern Europe and on other countries is analyzed by John Green and Phillip L.Swagel, while Alessandro Prati and Garry J. Schinasi address the issue of ensuringfinancial stability within EMU.

One of the most striking developments of recent years has been the growth of inter-national financial transactions and international capital flows. This has brought sub-stantial benefits, but it is not without risk. In their article, Barry Eichengreen andMichael Mussa emphasize that capital account and financial liberalization are inevitablefor countries that wish to participate in the open global economy and discuss ways inwhich this process can best be managed. Some of the key issues in capital account lib-eralization are examined by R. Barry Johnston, while Saleh M. Nsouli and MounirRached look at the steps that need to taken by the Southern Mediterranean countries inopening their capital account.

Since 1982, a series of financial crises have hit emerging markets, most recently thecrisis in East Asia. Donald J. Mathieson, Anthony Richards, and Sunil Sharma discusshow emerging markets can better manage the risks associated with integration into theinternational monetary system, while Daniel C. Hardy considers how to predict bank-ing crises.

The development of monetary policy instruments in Russia and some of the lessonsthat have been learned in the process are discussed by Tomas J.T. Balino. Charles Enochand Anne-Marie Guide take a look at the success countries have had with currencyboards and the circumstances in which such monetary arrangements are most likely tobe effective. And Harold James, Professor of Modern History at Princeton University,takes a fresh look at the evolution of IMF conditionality over the years and some of thenew challenges faced in this area.

Ian S. McDonaldEditor-in-Chief

New Finance & Development Advertising Representative

Finance & Devefopmenf has recently engaged IPC Enterprises Inc. as itsadvertising sales representative. All inquiries about advertising in Finance &Development should be directed to

Ms. Linda Marx, Director of Sales, IPC Enterprises Inc.,142 East 30th Street, New York, NY 10016, U.S.A.

(Telephone: (212) 252-0222, fax: (212) 252-1020,e-mail: [email protected]).

LETTERKBBBHHJBff/fiqBJ

©International Monetary Fund. Not for Redistribution

http://www.imf.org/fandd

FINANCE\5 DEVELOPMENT

EUROPEAN ECONOMIC AND MONETARY UNION

2 The Implications of EMU for IMF SurveillanceMichael C. Deppler

4 Policy Challenges for the Euro AreaKlaas Knot, Donogh McDonald, and Karen Swiderski

8 The Euro Area and the World EconomyJohn Green and Phillip L. Swagel

12 Ensuring Financial Stability in the Euro AreaAlessandro Prati and Garry J. Schinasi

CAPITAL ACCOUNT LIBERALIZATION

16 Capital Account Liberalization and the IMFBarry Eichengreen and Michael Mussa

20 Sequencing Capital Account LiberalizationR. Barry Johnston

24 Capital Account Liberalization in the Southern MediterraneanSaleh M. Nsouli and Mounir Rached

OTHER TOPICS

28 Financial Crises in Emerging MarketsDonald J. Mathieson, Anthony Richards, and Sunil Sharma

32 Are Banking Crises Predictable?Daniel C. Hardy

36 Monetary Policy in RussiaTomds J. T. Balino

40 Are Currency Boards a Cure for All Monetary Problems?Charles Enoch and Anne-Marie Guide

44 From Grandmotherliness to Governance: The Evolution ofIMF ConditionalityHarold James

Books48 Global Development Fifty Years after Bretton Woods:

Essays in Honour of Gerald K. Helleineredited by Roy Culpeper, Albert Berry, and Frances Stewart-Vinod Thomas and Sarwar Lateef

49 Has Globalization Gone Too Far?by Dani Rodrik-Lant Pritchett

50 Current Issues in Economic Development: An Asian Perspectiveedited by M.G. Quibria and J. Malcolm Dowling-Qflizar Hussain

51 Privatization and Entrepreneurship: The Managerial Challenge in Central andEastern Europeedited by Arieh A. Ullmann and Alfred Lewis—Raj M. Desai

52 Foundations of International Macroeconomicsby Maurice Obstfeld and Kenneth Rogoff-Peter Clark

54 Letters to the Editor

56 Index

A QUARTERLY PUBLICATION OF THE INTERNATIONAL MONETARY FUND DECEMBER 1998. Volume 35. Number 4


How will EMUaffect theway the IMFmonitorseconomicdevelopmentsin its Europeanmembers?

The Implications of EMUfor IMF SurveillanceMichael C. Deppler

HE THIRD and final stage ofEuropean Economic and MonetaryUnion (EMU) is set to begin onJanuary 1,1999, with the establish-

ment of a currency union encompassing 11 ofthe European Union's 15 member countries—Austria, Belgium, Finland, France, Germany,Ireland, Italy, Luxembourg, the Netherlands,Portugal, and Spain. On that date, these coun-tries will lock their exchange rates; the euro,their common currency, will be born; andnational authorities will hand over responsibil-ity for monetary and exchange rate policies toarea-wide institutions. With the economic sizeof the initial euro area rivaling that of theUnited States, EMU will represent an impor-tant milestone for both the European Unionand the international monetary system.

EMU will have important implications forthe IMF's work and procedures, including theIMF's "surveillance" activities—that is, itsregular monitoring of economic policies anddevelopments in member countries. The IMFhas a mandate, under Article IV, Section 3 ofits Articles of Agreement, to oversee theinternational monetary system and, to thatend, to exercise firm surveillance over mem-bers' exchange rate policies. Regular consul-tations with members—known as Article IVconsultations—represent the main vehiclefor IMF surveillance. They are supplementedby periodic multilateral surveillance exer-cises, including those conducted in connec-tion with the IMF's World Economic Outlook,which normally appears twice a year, and itsannual International Capital Markets report.In addition, the IMF addresses issues from aregional perspective, whenever appropriate.

Although EMU represents a major changefor the international monetary system, it hasevolved gradually, giving the IMF time toadapt its practices. The IMF's ExecutiveBoard has met regularly to consider the

EMU process as a whole, and EMU-relatedmatters have become increasingly prominentin Article IV consultations with individualcountries. Moreover, since the signing of theMaastricht Treaty in 1992, contacts betweenIMF staff and institutions in the EuropeanUnion have multiplied as the latter hasexpanded its surveillance over the economicpolicies of its member states and the conver-gence criteria established by the treaty haveplayed a growing role in policy formulationby national authorities. EMU issues havealso featured prominently in the IMF'sglobal surveillance exercises, including chap-ters on EMU in the editions of the WorldEconomic Outlook published in October1997 and October 1998.

Nonetheless, the advent of EMU willrequire further adaptation of the IMF's sur-veillance of euro area economies. Regulardiscussions with individual euro area mem-bers will remain a central element of IMFsurveillance. But discussions with theEuropean Central Bank (ECB), which beganoperations on June 1, 1998, will become avital complement to the IMF's discussionswith national authorities. The same wouldapply to the European Union's Council ofMinisters should it exercise its prerogativesregarding matters pertaining to the exchangeregime of the euro.

From an IMF surveillance perspective, thedistinguishing feature of EMU is the transferof monetary and exchange rate authorityfrom 11 member states to supranationalbodies even while authority for fiscal andstructural policies remains vested largely inthe national authorities. As a result, IMF sur-veillance will need to be conducted simulta-neously at both the national and thesupranational levels, keeping a weather eyeon the implications both of the combinedeffect of nationally determined policies for

A4 Finance & Development / December 1998

T


area-wide monetary and exchange rate policiesand, conversely, of area-wide monetary policyfor national policy requirements.

In the context of the Article IV surveillanceprocess, this means that discussions at the euroarea level will focus on monetary and exchangerate policies, but against the backdrop of an eval-uation of the combined fiscal and structural poli-cies of the members of EMU, which is needed toassess the coherence of the overall macroeco-nomic policy mix. For instance, a given area-widemonetary policy and set of national fiscal policiesmight be judged to prompt an inopportune (from an interna-tional or euro area point of view) rise in real interest rates orappreciation in the real effective exchange rate. Depending onthe circumstances, this might argue for an easier policy by theECB or for an across-the-board or selective tightening of fiscalpolicies at the national level. The principal change in the con-tent of consultations at the national level will be that discus-sions of monetary policy will focus on the impact of area-widemonetary policy on developments and policies at the nationallevel, as well as on the implementation of monetary policyoperations through the national central banks.

Michael C. Deppler isDirector of the IMF'sEuropean I Department.

In addition to adapting its Article IV con-sultations to the advent of EMU, the IMF willcontinue its regular surveillance of economicpolicies in the European Union. It will payparticular attention to the surveillance carriedout by EU institutions—notably by theCouncil of Ministers in the context of theStability and Growth Pact, as well as in con-nection with the Council's broad economicpolicy guidelines and employment policyguidelines—and to policies affecting trade andthe integration of EU markets.

The euro area faces a number of challenges ahead inensuring that its new policy framework operates smoothlyand in addressing the root causes of its high unemployment.The IMF is committed to discharging its surveillance respon-sibilities both to assist EMU participants and their commoninstitutions in making EMU a success and, more generally, tohelp achieve the hopes and aspirations of the internationalcommunity. E3B

O Economic Policy Management:Mastering the Analytical Instruments(4 Workshops from January to mid-February)

© Applying the Tools of MacroeconomicPolicy: Exchange Rates, Taxation andDecentralization(5 Workshops from March to mid-April)

© Firms, The Market and The State:Privatization, Redistribution and Poverty(5 Workshops from April to mid-June)

Q The Big Picture: LDCs, Transition Economiesand Emerging Markets in the Context ofGlobalization(6 Workshops from mid-June to July)

For further information, please contact: Philippe Messeant • CERDI • 65, boulevard Francois Mitterrand •63000 Clermont-Ferrand (France) • Telephone: 33.4.73.43.12.30 • Fax: 33.4.73.43.12.28E-mail: [email protected] .fr

Finance & Development / December 1998 **

ECONOMIC POLICY WORKSHOPSWorkshops organized every year from January to July

For mid-career policymakers from LDCs and Transition Economies(from ministries such as Planning or Finance, Agriculture, Health, Trade...).Particioants from oarastatals or the orivate sector are also most welcome.

With instructors drawn from the faculties of the CERDI and its academic part-ners (European and North-American), as well as major International Institutions

(World Bank, IMF, European Commission, United Nations Agencies, World Trade Organization, AidAgencies, Central Banks...).


Policy Challenges forthe Euro AreaImproved economic performance and a sound policy frame-work in the euro area provide a solid basis for EMU. Butimportant challenges remain: making an eclectic approach tomonetary policy transparent, strengthening public financesfurther, addressing the structural causes of high unemployment,and ensuring an appropriate mix of economic policies.

Klaas Knot, Donogh McDonald,

and Karen Swiderski

I N PREPARING for European Eco-nomic and Monetary Union (EMU),the euro area countries have consider-ably improved their macroeconomic

policy environment from what it was in theearly and mid-1990s (Chart 1). Inflation hasfallen to very low levels, fluctuating around anannual rate of 1'A percent during 1997-98,and there is little sign of inflationary pressurefor the period ahead. Public finances are in farbetter shape than earlier in the decade. Despitelackluster economic growth in the region, theaverage ratio of government deficit to GDPwas reduced by more than 3 percentage pointsbetween 1993 and 1997.

Chart 1Fiscal imbalances and inflation rates

in the euro area have droppedGeneral government fiscal balance Consumer price inflation

(percent)

To ensure that EMU runs smoothly—and,especially, to achieve and sustain the highergrowth rates that will be necessary to reducehigh unemployment rates—euro area coun-tries will need to consolidate and capitalizeon these achievements. This will requirecontinued prudent macroeconomic policiescombined with reforms aimed at makinglabor and product markets more flexible.Effective policy coordination will also beneeded, because fiscal and structural policieswill continue to be designed and imple-mented at the national level.

Monetary policyThe design and implementation of monetarypolicy in the euro area will be the preserve ofthe European System of Central Banks(ESCB), comprising the European CentralBank (ECB), which will make monetarypolicy decisions, and the national centralbanks, which will be primarily responsiblefor implementing those decisions.

The ESCB's mandate clearly gives priorityto price stability, which has been defined bythe ECB as inflation of less than 2 percent inthe euro area as a whole. The emphasis onprice stability is reinforced by the highdegree of independence given to the ESCBand other provisions that insulate decisionmakers from political pressures. In pursuinglow inflation, the ESCB will be continuingpolicies already firmly established by thenational central banks. Nonetheless, EMU

* Finance & Development / December 1998


represents a change in the monetary policy regime in twokey respects. First, the scope for national monetary andexchange rate policies—already constrained by the EuropeanMonetary System's exchange rate mechanism (ERM)—willdisappear. Second, the geographic orientation of monetarypolicy will broaden to take account of conditions throughoutthe euro area. Until now, because of the anchor role thedeutsche mark has played in the ERM, the DeutscheBundesbank's policies have been the predominant influenceon the monetary stance of euro area countries.

This change in regime will produce important challengesfor the ESCB. Some of these—such as putting in place thepertinent information and operational systems, or compilingarea-wide monetary statistics—are fairly straightforward, iftime consuming. Others are more complex. In particular, theECB will initially be faced with many uncertainties concern-ing the demand for financial assets and the mechanismsthrough which monetary policy influences the economy. Ofcourse, even long-established central banks such as the Bankof England, the Bundesbank, and the U.S. Federal ReserveSystem often have to deal with these problems. However, thedifficulties of assessing the impact of policies are likely to beeven greater for the ECB—at least initially—because of boththe restructuring of financial markets that will accompanythe introduction of the euro and the effects of the regimechange on product and labor markets.

In this environment, policymakers from diverse nationalbackgrounds will need to demonstrate early on that they areable to reach common positions based on a pan-Europeanperspective in a timely fashion. A workable, well-understoodmonetary framework will be essential. Approaches based onpast behavioral relationships of economic models—whetherin the context of monetary targeting or inflation targeting—are likely to be difficult to implement at the outset of EMU,owing to the substantial change in regime. Against thisbackground, the ECB recently agreed on a monetary policystrategy based on three main elements: (1) achieving pricestability, defined as an annual inflation rate of less than2 percent; (2) giving money a prominent role, by adopting areference value for the growth of a monetary aggregate; and(3) using a range of economic and financial indicators toreach a broadly based assessment of the outlook for futureprice developments. The euro area's subdued inflation ratewill help the ECB pursue this eclectic approach while itbuilds its reputation. Given the lags between policy actionand its effect on inflation, however, the ECB will need toexplain its strategy and the factors influencing its delibera-tions clearly and frequently, especially at the outset.

Fiscal policyThe budget of the European Union (EU), at a little more than1 percent of its member countries' GDP, is relatively small,and the fiscal policy role of EU institutions has consisted pri-marily of the surveillance of national fiscal policies. In thiscontext, the Stability and Growth Pact (SGP), which was

agreed in June 1997, enhances the fiscal policy framework forEMU. It does this, in part, by establishing more clearly thanthe Maastricht Treaty the principle that countries that fail tocorrect deficits judged to be excessive will be subject to finan-cial sanctions. Equally important, it strengthens the medium-term framework for surveillance. In the pact, countries havecommitted themselves to attaining medium-term budgetarypositions that are close to balance or in surplus, so that theywill be able to deal with the normal fluctuations of the busi-ness cycle while keeping the general government deficitbelow 3 percent of GDP. Countries will submit medium-term"stability" programs once a year to the EU's Council ofMinisters, on the basis of which the Council will assess thecompliance of countries' budgetary strategies with the SGP'smedium-term goal.

There has been considerable discussion as to whether bud-gets need to be balanced on a structural basis (that is, afterthe cyclical effects on the budget have been removed) inorder to provide countries with adequate room to deal withnormal cyclical fluctuations while keeping their deficits

"Policymakers . .. will need todemonstrate ... that they areable to reach common positionsbased on a pan-Europeanperspective in a timely fashion."below 3 percent of GDP. Using past output fluctuations as aguide, it seems that most countries would be able to achievethe latter two objectives by ensuring that the deficit does notexceed 1 percent of GDP at normal levels of capacity utiliza-tion. However, other considerations—including uncertain-ties regarding future output volatility, the desirability ofallowing adequate scope for discretionary countercyclicalpolicies, and the vulnerability of some countries' fiscal posi-tions to interest rate shocks or a reduction of EU structuralfunds, as well as the practical difficulties inherent in the mea-surement of output gaps—might argue for a more ambitiousgoal. Moreover, fiscal pressures associated with the aging ofthe euro area's population underline the benefits of substan-tially reducing the ratio of public sector debt to GDP, andthereby the burden of interest spending. There is consider-able uncertainty surrounding the estimates, but annual pen-sion and health spending in the euro area could rise by asmuch as 7 percentage points of GDP over the next 30 years ifcountries do not reform their current programs. Taking all ofthese considerations into account, the SGP's objective of fis-cal balance or a small surplus in the medium term appears tobe reasonable, though a more ambitious goal may be war-ranted in a few countries.

At one level, the task facing fiscal policymakers in the euroarea seems quite modest. The general government structural

Finance & Development / December 1998 "


deficit for the euro area in 1998 is estimated to be a little over 1percent of GDP. Moreover, interest spending for the area as awhole is projected to fall by 'A to 3A of 1 percent of GDP overthe next three to four years, reflecting declining debt levels andlower interest rates, with the second of these factors especiallysignificant in those countries that in the past experienced largedifferentials between their interest rates and those of Germanyand the other "core" members of the ERM. As a result, theadjustment needed to achieve a balanced structural fiscal posi-tion is much smaller than the estimated structural deficit for1998 would seem to indicate. This said, the improvementsneeded in the fiscal balance understate the extent of spendingreform required, in view of the need to reduce distortions inthe tax system, especially those that discourage the creation ofjobs or weaken incentives to work. Government revenues arehigh in the euro area—nearly one-half of GDP, on average.The rate of increase in non-interest spending will need to bemuch less than the trend growth of GDP if countries are toreduce taxes significantly while achieving the medium-termfiscal goal set out in the SGP.

There is a danger that fiscal policies will fall short ofrequirements. Despite stronger economic growth and fallinginterest burdens in the euro area, the fiscal position is pro-jected to improve only slightly in 1998, with a significantweakening in the primary structural balance (the structuralbalance net of interest payments). Admittedly, fiscal perfor-mance in 1998 should be seen in the context of the very largeadjustment made during 1996-97 and the need in somecountries to replace temporary measures adopted in 1997with more permanent measures. Nevertheless, there is a riskthat, now that they have qualified to be part of the euro area,governments will be tempted to loosen their fiscal belts. Thisoften happens during cyclical upswings because govern-ments confound improvements in actual fiscal positionswith improvements in the underlying position. It _is notunusual at cyclical peaks for output to exceed its cyclicallynormal level by 3-4 percent; at such levels of resource utiliza-tion, actual budgetary positions would, in most countries,have to be in surplus by about 2 percent of GDP to be consis-tent with balance at normal levels of capacity utilization.

Against this background, governments need to provideconvincing evidence in their policies in 1999 and theirmedium-term stability programs that their fiscal strategiesare in line with medium-term needs, emphasizing strictcontrol of spending based on forward-looking reforms.Inadequate fiscal policies would reduce the ECB's room formaneuver, resulting in higher interest rates than wouldotherwise be needed. Moreover, should the euro area econ-omy falter, countries could be forced to override the auto-matic stabilizers and to tighten policies in order to preventgeneral government deficits exceeding 3 percent of GDP.

Structural policiesEuropean labor markets have adapted poorly to changes inthe global economic environment over the past three

decades. Between 1970 and 1997, unemployment in the euroarea rose by 10 percent of the labor force, compared with noincrease in the United States and a much smaller increase inother industrial countries outside the euro area (Chart 2).The euro area's rate of job creation has compared even lessfavorably: its labor-force participation rate has remainedunchanged (despite rising female participation), while rateshave increased in other industrial countries.

To ensure the long-term success of EMU, structural poli-cies will need to effect a sea change in the performance ofEuropean labor markets. Structural reform is imperative notonly because of the economic waste and social costs associ-ated with high unemployment but also for a number ofother reasons. Without the possibility of recourse to theexchange rate instrument, flexible markets will become evenmore important in enabling countries to adjust to shocks,especially asymmetric ones. The persistence of high unem-ployment could erode public support for prudent macroeco-nomic policies, and many may question whether EMU wasworth the effort unless it sets the stage for a lasting andmarked reduction in euro area unemployment. Moreover,higher employment ratios—along with spending reformsand the reduction of public debt—will be needed to help off-set the fiscal impact of aging populations.

It is widely recognized that Europe's structural problemshave complex and wide-ranging sources—social benefit sys-tems that often provide inadequate incentives to work, taxsystems that distort incentives, excessive labor market regula-tions, and product market regulations and subsidies thatweaken competition. While the required mix of reformsvaries from country to country, most euro area countriesneed to address issues in each of these areas.

EMU members will need to resist pressures for wages andsocial benefits to converge across the euro area before pro-ductivity has converged. To counteract such tendencies, itwill be important to address factors that delink wage behav-ior from local labor market conditions, particularly policies

Chart 2The unemployment rate has increased more in

euro area than in other industrial countries(percent of labortorce)

8 Finance & Development / December 1998

Source: Organization for Economic Cooperation and Development.1 Australia, Canada, Japan, and New Zealand.


and institutions that give power to insiders—for example, rigid employment protectionlegislation and subsidies or market supportfor selected enterprises and sectors.

Although progress is being made withstructural reform, in most countries it hasnot been enough to make a noticeable dentin structural unemployment. To speed thisprocess, the EU has introduced labor marketsurveillance procedures, which provide forthe Council of Ministers to issue annualguidelines on employment policies andassess countries' performances in this con-text. Enhanced EU surveillance provides anopportunity to reinforce the case for reform.It should also aim to ensure that measuressuch as reductions in the work week areimplemented flexibly so as to guard against,among other things, raising employers' laborcosts. It will also be critical to maintain thepressure for reform, as there is a danger thatthe attention paid by national policymakersto the labor market may diminish as cyclicalunemployment declines.

Economic policy coordinationUnder EMU, monetary policy will be gearedto economic conditions in the euro area as awhole, but, in contrast with national mone-tary areas, primary responsibility for fiscaland structural responsibilities will be decen-tralized. One of the advantages of thisdecentralization is that individual countrieswill be able to adapt fiscal and structuralpolicies to their own specific problems.However, it will complicate the task ofachieving the appropriate policy mix at theeuro area level, because inadequate nationalpolicies, through their implications for thesingle monetary policy and the exchangerate for the euro, will inevitably havespillover effects on other countries. Forexample, where fiscal policies in more cycli-cally advanced countries are exacerbating,or providing an inadequate counterweightto, cyclical divergence, they will tend topush up the average interest rate for theentire euro area, to the detriment of lesscyclically advanced euro area countries.More flexible markets are another tool forattenuating cyclical divergences and theiradverse effects, especially on the less cycli-cally advanced countries, limiting the extentto which downward rigidities in wages ham-per adjustment.

The differences in cyclical positions in theeuro area on the eve of EMU are significant,with countries falling into two groups,although with considerable variation withineach group. For the group of more cycli-cally advanced countries—Austria, Finland,Ireland, Luxembourg, the Netherlands,Portugal, and Spain—which account forabout one-fourth of euro area GDP, the out-put gap in 1998 is estimated to be negligible,while the average gap for the less cyclicallyadvanced economies—Belgium, France,Germany, and Italy—is estimated at 2 per-cent. In a number of the cyclically advancedcountries, fiscal policy is not playing astrong enough role in restraining demand,while in most euro area countries, structuralpolicies are not yet adequate to temperdivergences. Fortunately, the more cyclicallyadvanced countries have only a moderateweight in the euro area economy and differ-ences in cyclical positions among the threelargest euro area countries currently are notlarge. Nevertheless, the current situationunderlines the potential for cyclical diver-gences, if not properly handled, to causestrains that would affect the economic per-formance of the entire area.

In a resolution adopted in December1997, the European Council (the namegiven to the regular summit meetings ofEU heads of state or government) empha-sized the need for enhanced policy coordi-nation. It envisaged that economic policycoordination would be effected throughthe various surveillance instruments pro-vided for in the Maastricht Treaty. In par-ticular, the European Council called for theEU's annual broad economic policy guide-lines to be developed into an effectiveinstrument for ensuring sustained conver-gence. The guidelines should be moreconcrete and country-specific than in thepast and should give more attention tostructural issues. The European Councilalso emphasized the importance of earlywarnings of fiscal policy concerns underthe SGP. In addition, it was agreed thatministers of the euro area countriescan meet as the Euro-11 group to discussissues related to the single currency; how-ever, EU surveillance decisions will betaken by the full Council of Ministers, inwhose deliberations all 15 EU membersparticipate. B39

Klaas Knot is anEconomist In theEuropean Union Unit ofthe IMF's European IDepartment.

Donogh McDonald is anAdvisor in the EuropeanUnion Unit of the IMF'sEuropean I Department.

Karen Swiderski is aSenior Economist in theEuropean Union Unit ofthe IMF's European IDepartment.

•as*Finance & Development / December 1998 ~ '


The Euro Area and theWorld EconomyEuropean monetary union will bring both new opportunitiesand fresh challenges for economic policies in countries outsidethe euro area.

John Green and Phillip L. Swage

T HE advent of European Economicand Monetary Union (EMU) andthe development of a single, morepowerful economic market will

have an impact both on the other economiesof Western Europe and on other countrieswith established trade and financial links toEurope, including countries that will linktheir own currencies to the euro. The tran-sition economies of Central and EasternEurope and the Baltics, the developing coun-tries of the Mediterranean basin, and coun-tries in Africa are most likely to be affectedby EMU. Countries in Asia and the WesternHemisphere and advanced economies out-

side Europe will also be affected, though to alesser extent.

Global environmentThe global environment has been supportive,until now, of the transition to a single cur-rency for 11 members of the European Unionand the achievement of their economicobjectives. Strong demand by advancedcountries for exports from the euro area,coupled with the depreciation of the curren-cies of euro area countries over the past threeyears, has stimulated recovery in the euroarea and helped to offset the effects of theAsian crisis (see chart). As the growth in

o0 Finance & Development/December 1998


Euro area and United States: Indicators ofrelative economic size

Output and trade1

demand from these areas is likely to slow, however, it willbecome more important for demand within Europe tobecome self-sustaining.

There are also challenges for EMU in the global economicenvironment.

« The crisis in Asia and other emerging market economiescould produce adverse spillover effects on the euro area andmake monetary policy more difficult to carry out.

. If the crisis deepens further, external demand couldweaken and dampen the current recovery in Europe, which,in turn, could dampen confidence and domestic demand.

• In addition to the portfolio shifts expected as investorsbecome accustomed to holding the euro, financial marketvolatility could add to uncertainty in assessing the eco-nomic indicators that the European Central Bank (ECB)will be monitoring.

• The weakness of the yen over the past several years andthe recent volatility in the yen-dollar exchange rate demon-strate how rapidly market exchange rates can deviate fromthose indicated by medium-term fundamentals; such devia-tions could adversely influence financial conditions andmonetary policy deliberations in the euro area.

• The Asian crisis, coupled with the possibility of abroader crisis in emerging markets, could influence thetransmission of policy to the real economy if, for example,commercial banks in the euro area have to make substantialprovisions for nonperforming loans.

The euroWhile the future behavior of the euro cannot be readily pre-dicted, it is likely to appreciate against the dollar and thepound sterling over the next few years, but to depreciateagainst the yen. Exchange rate movements along these linesmay have already begun. Cyclical factors—the UnitedKingdom and the United States are at more advanced stagesof recovery than the euro area, but the euro area is well aheadof Japan—are one element of this reasoning.

Measures of fundamental equilibrium exchange rates fur-ther support this assessment. For example, in evaluatingexchange rate levels, the IMF uses a macroeconomic balanceapproach that considers medium-term national savings andinvestment, which, in turn, determine the medium-termcurrent account balance. For the euro area, demographicfactors—such as the proportion of the population approach-ing retirement age—can be expected to raise national savingsin relation to investment, which would lead to a currentaccount surplus in the medium term. At the same time, coun-tries in the euro area recorded a medium-term surplus in 1997that was larger than a reasonable medium-term equilibrium.This would suggest that the euro is likely to appreciate, relativeto exchange rate levels of euro area members in the first half of1998, toward a level consistent with a smaller surplus.

Also important will be the extent to which investors seethe euro as a stable store of value, which will depend on thecredibility of the ECB. Such credibility is likely to be strong

Sources: International Monetary Fund, World Economic Outlook (Washington,October 1998); and Bank for International Settlements, 68th Annual Report(Basle, June 1998).1 For 1998, IMF staff projections. Trade shares are calculated through 1997 only.2 Cross-border claims and local claims in foreign currency of banks located in

industrial reporting countries. European Union currencies comprise the deutschemark and the French franc only.

because the new central bank has been given the indepen-dence to pursue price stability as a primary objective, andits governing body is composed mostly of central bankerswho are known "inflation hawks" dedicated to maintainingprice stability.

It is also likely that, once the single currency takes effect,the national central banks of the euro area will reduce theirinternational reserve holdings. Trade within the euro areawill be denominated in a single currency and will no longerneed to be backed by international reserves. Estimates of theEMU countries' resulting "surplus" of international reservesrange from $50 billion to $230 billion. Given that theseamounts are small in relation to the total stocks of the inter-national assets and liabilities of the United States, any down-ward pressure on the exchange rate between the euro and thedollar resulting from the sale of U.S. dollars is likely to besmall. Rebalancing of official reserves will, in any event, beswamped by shifts in private sector supply and demand foreuro-denominated assets.

Advanced economies outside fhe euro areaDevelopments in the euro area will have important implica-tions for those countries of the European Union that will nottake part in the first round of monetary union.

Finance & Development / December 1998 *

GDP at purchasingpower parity

(billion dollars; 1990 prices)Shares in world trade

(percent)

I Initfld States Euro area Dfivelnninn nnimtrifis Janan

Currency composition of international lending, 1997(percent)

International note andbond markets International bank lending 2

European Union currencies (Yen i U.S. dollar


• Greece has continued to make significant progresstoward macroeconomic convergence with the EMU countriesand has indicated its objective of entering EMU in 2001. Inthe period ahead, Greece will follow tight fiscal and monetarypolicies to ensure that it meets the Maastricht Treaty's criteriaand to maintain broad exchange rate stability between thedrachma and the euro.

• In 1997, Denmark, Sweden, and the United Kingdom allmet the Maastricht criteria for inflation and interest rateconvergence, and none had excessive deficits. They have indi-cated, however, that they do not intend to participate at pres-ent. Any decision on membership would require approval bya public referendum in each country, and public supportdoes not appear strong at this time.

• In the United Kingdom, the authorities have set fivebenchmarks—various macroeconomic and structural con-vergence indicators—to ensure that U.K. entry into mone-tary union is consistent with its central economic objectiveof high and stable growth and employment.

• All three countries face the challenge of adjusting theirexisting economic policy frameworks and stances to be com-patible with EMU membership. This is unlikely to be a prob-lem for Denmark, because the krone has been linked to thedeutsche mark within the exchange rate mechanism (ERM) ofthe European Monetary System. The Swedish krona and thepound sterling have floated and not participated in the ERMsince the crisis of 1992, but it is unclear whether this would

10 Finance & Development/ December 1998

Euro area and selected countries: trade linkages(exports to and imports from trading partners in 1996; percent of total trade and output)

Partner countriesTrade 1 Output 2

Developing DevelopingEuro Other and Euro Other andarea advanced transition area advanced transition Total

Euro area 51.0 30.8 18.2 11.7 7.1 4.2 22.9

Denmark 47.1 40.2 12.7 11.1 9.5 3.0 23.6Greece 57.5 21.4 21.1 8.4 3.1 3.1 14.6Sweden 44.5 43.1 12.3 13.0 12.6 3.6 29.1United Kingdom 49.4 34.9 15.7 11.0 7.8 3.5 22.3

Japan 11.3 54.7 34.1 0.9 4.5 2.8 8.2United States 13.8 53.6 32.6 1.3 5.0 3.1 9.4

Asia 12.5 67.7 19.8 2.5 13.3 3.9 19.7Africa 39.8 34.4 25.8 7.7 6.7 5.0 19.5

of whichCFA franc zone 48.1 23.7 28.2 12.3 6.1 7.2 25.5

Middle East and Europe 26.9 42.8 30.2 6.9 11.0 7.8 25.6of which

Central and Eastern 51.0 16.1 32.8 16.8 5.3 10.8 32.9Europe

Western Hemisphere 13.3 61.4 25.3 2.0 9.1 3.7 14.8

Source: International Monetary Fund, World economic Outlook (Washington, October 1998).1 Imports plus exports of goods from and to partner countries as percent of total imports plus exports.2 Tfie average of imports plus exports of goods from and to the partner countries as a percent of GDP.

constitute an obstacle to future participation inthe euro area.

• Norway and Switzerland, although theyare not members of the European Union andare therefore not eligible to participate, willbe directly affected by monetary unionowing to both geographic proximity andtheir strong trade and financial links with theeuro area.

• The advent of the single currency andeconomic development in the euro area arelikely to have a smaller direct impact on thenon-European advanced economies. ForCanada, the United States, and the advancedeconomies of Asia, trade with the euro arearepresents a relatively small share of eachcountry's total trade or GDP, and the directimpact of developments in the euro area islikely to be limited (see table). Indirect influ-ences could be more important, however,especially for the United States. A redirectionof demand for international reserves from

• dollars to euros would tend to reduce the U.S.current account deficit as a counterpart to thenegative impact on the U.S. capital account. Ashift from international reserves held indollars to euros would also redistribute someinternational seigniorage revenue from the

U.S. Federal Reserve to the ECB. Overall, EMU andEuropean integration can be expected to have a positiveeffect on the United States and other countries, as the singlecurrency and the integrated market will facilitate financialand business transactions.

Developing and transition countries

For developing and transition countries, the main impact ofthe unified European market is likely to come through tradeand financial linkages. More robust activity and high importdemand in the euro area will lead to increased exports from,and output in, developing and transition countries.Similarly, exchange rate arrangements, financial marketdevelopments, and capital flows will have implications fordeveloping and transition countries and their policies.

« An increase in the cyclical strength of the euro area willlead directly to increased exports for Europe's trading part-ners. In addition, the launch of the single currency is likely tolead to further progress in reforming labor markets andaddressing other structural problems and, consequently, tomedium-term output gains. Trade with the euro area makesup between 40 and 50 percent of total goods trade for Africaas a whole, with countries in North Africa and the CFA franczone at the higher end of this range.

• A number of emerging market countries with close ties toEurope currently link their currencies to the deutsche mark


or the French franc. These links are expectedto be shifted to the euro. Changing valuesbetween the euro and the dollar or yen couldaffect these countries' external competitive-ness when the currency or currency basket ofthe exchange rate target deviates from thetrade-based effective exchange rate. At thesame time, countries with dollar- or yen-denominated debt would be affected: anincrease in the value of the euro would bene-fit those countries that peg their currenciesto the euro, because it would decrease thedomestic currency cost of servicing theirdollar-denominated debt. Conversely, anydepreciation of the euro would increase thecost of such debt service.

• To offset these effects, countries maydecide to adjust their exchange rate regimesto reflect better the composition of theirtrade and financial links or to change theirdebt-management policies. Adjustments todebt management will be facilitated as theeuro becomes widely used in trade andfinancial markets and accounts for a largershare of debt securities. Countries that pegto the euro will be able to lower their dollarexposure and reduce the fluctuations in theirdollar-denominated debt payments.

EMU may also lower borrowing costs:• Deeper and more liquid capital markets

in Europe will lower borrowing costs, both forcountries in the euro area and for other coun-tries that raise euro-denominated funds.

• EMU will allow private institutions, suchas insurance companies and pension funds,in the euro area to shift some of their port-folios into emerging market investments.Because investments outside their homecountries but within the euro area willbe reclassified as domestic currency invest-ments, investors will find that EMU effec-tively eases constraints imposed by currencyexposure requirements.

• Emerging market economies could ben-efit from direct and portfolio capital inflowsif converging asset returns in Europe leadglobal investors to increase their emergingmarket holdings in order to diversify acrosscountries.

EMU may also create some financial risksfor emerging market countries:

• A successful EMU that raises productiv-ity and growth could make Europe moreattractive to investors and increase the costof capital for emerging market economies.

• Increased competitiveness of Europeanfinancial institutions and the greater depth offinancial markets in the euro area could leadfirms in developing and transition countriesto raise capital in euros rather than in theirdomestic currencies, thus challenging localcapital markets. This could, however, providean incentive for developing and transitioncountries to strengthen their financial inter-mediation and build sound banking systems.

Challenges of EU enlargementThe European Union is currently beingenlarged to include the transition countriesof the Baltics and Eastern Europe, togetherwith selected European countries in theMediterranean region. Countries that in-tend to join the European Union will needto show progress toward meeting theMaastricht criteria, although this is notrequired for accession, nor are new mem-bers expected automatically to join EMU.Potential EU members must overcome anumber of challenges. They need to pro-gress with privatization and to continue toreduce government involvement in theeconomy while dismantling monopolies,removing trade restraints, and developingflexible labor markets.

Six countries—Cyprus, the Czech Republic,Estonia, Hungary, Poland, and Slovenia—have received favorable opinions from theEuropean Commission on their membershipapplications, and accession negotiations areunder way. These six countries have alreadymade good progress toward meeting the fiscalguidelines of the Maastricht Treaty, but infla-tion rates in the five transition countries inthis group are still above those in the euroarea. Although progress on this front is impor-tant, it would seem more relevant to defineeconomic convergence in terms of a broaderset of structural and institutional require-ments. Full capital account liberalization isamong these and will itself require develop-ment of robust financial sectors and strongerregulatory and oversight capacities to helpensure stability. 1MB

John Green is a DeputyDivision Chief in theWorld Economic StudiesDivision of the IMF'sResearch Department.

Phillip L. Swagel is anEconomist in the WorldEconomic StudiesDivision of the IMF'sResearch Department.

This article draws from material contained in

International Monetary Fund, World Economic

Outlook (Washington, October 1998).

Finance & Development / December 1998 '*


Ensuring Financial Stabilityin the Euro AreaThe introduction of the euro is expected to spur the creation ofpan-European financial institutions and markets, with consid-erable benefits for consumers and investors. Challenges remainwithin the euro area institutional framework for the manage-ment of systemic risk and financial crises.

Alessandro P ro t i and Ga r ry J . Sc h i n a s i

ENSURING financial stability withinthe European Economic and Mon-etary Union (EMU) will be chal-lenging in the early years, in part

because there are a number of aspects ofEMU that might increase the potential forsystemic events. First, there is the possibilitythat TARGET (the Trans-European Auto-mated Real-Time Gross Settlement ExpressTransfer system)—the settlement systemslated to take effect on January 1, 1999 thatwill link the real-time gross settlement sys-tems of European Union countries—will beused less extensively than expected andmight therefore not reduce systemic riskas much as has been, anticipated. Second,as new pan-European financial marketsemerge, the growth of cross-border unse-cured interbank lending could increase therisk of contagion, at least until an EMU-wide repo (repurchase agreement) market iscreated and use of secured (collateralized)interbank credit lines becomes more wide-spread. Third, the introduction of the euro islikely to encourage further bank restructur-ing and consolidation, but in an environ-ment where it may be difficult to close banksand to reduce costs through downsizing.Inefficient and unprofitable institutions maytherefore continue to operate, engaging inincreasingly risky activities.

The possibility of heightened systemic riskmay not be apparent in the early daysof EMU because market integration andbank restructuring may occur slowly. Thelimited number of cross-border mergers of

European banks that have taken place thusfar, the gradual increase of competitivepressures in the retail sector, widespreadpublic ownership of banks, and the underde-veloped state of capital markets in manyEMU countries may provide some EMUcountries with more time for restructuringbanking systems. Decentralized arrange-ments for market surveillance and crisismanagement (based on home countrysupervision, for example) may be allowed tocontinue temporarily, providing some timefor adjustment. Eventually, however, pan-European capital markets and banking sys-tems are likely to develop, creating a need forcentralized mechanisms for financial surveil-lance, systemic risk management, and crisisresolution. Institutional arrangements inadvanced countries indicate that the centralbank may be a natural place to centralizesome of these functions, but such centraliza-tion would require simplification of thecurrent division of responsibilities—as man-dated by the Maastricht Treaty and EuropeanUnion legislation—between the EuropeanCentral Bank (ECB), the national centralbanks, national supervisors, and nationaltreasuries (Box 1).

The current frameworkAgainst this background, the thinking andplanning about crisis management arestill evolving. Whereas some understandingis likely to be reached before EMU takeseffect, important decisions have yet to bemade about how EMU countries would



resolve a bank liquidity crisis that occurred, for example, atthe fine line between monetary policy operations and liquid-ity support for systemically important private financial insti-tutions. The lack of clarity on how problems will be dealtwith reflects, in part, the "narrow" price stability mandate ofthe ECB spelled out in the Maastricht Treaty. The mandate ofthe ECB calls for it to focus on monetary policy and gives itonly a limited, peripheral role in banking supervision and noresponsibility for providing liquidity support to individualfinancial institutions. In accordance with the limited role thetreaty envisions for the ECB, the European MonetaryInstitute—the ECB's precursor, which was dissolved onJune 1, 1998—organized its work so that a clear separationhas been maintained between monetary policy operationsand the provision of liquidity for reasons not having to dowith the conduct of monetary policy. No institution in EMUhas been designated as a lender of last resort; thus, no centralinstitution holds the responsibility for providing, or coordi-nating the provision of, liquidity in a crisis.

It is unclear how a bank crisis would be handled underthe current institutional framework (which is composed ofthe Maastricht Treaty, the Statute of the European System ofCentral Banks, and the regulations and guidelines issued bythe European Monetary Institute), especially if the crisiswere to involve a pan-European bank for which severalcountries shared supervisory and regulatory responsibili-ties. The main issue is whether the European System ofCentral Banks or the national central banks have effectivemechanisms and understandings in place for taking action ifa particular financial institution is having difficulties infinancing payment instructions sent either across TARGETfor real-time settlement or across one of the alternative net-ting payment systems. It appears that EU supervisors havereached several understandings about how to deal with

cross-border crises and that discussions about the lender-of-last-resort function are under way, but no final decisionshave been taken.

In fact, practitioners and academics do not agree on a con-ceptual framework for dealing with the immediate conse-quences of a banking crisis. Some observers have arguedthat, to avoid moral hazard, central banks should use onlyopen market operations to deal with a liquidity crisis. Othershave argued that when a systemic event occurs in whichthere is little or no doubt about solvency—as was the casewith the Bank of New York's computer failure in 1985—thecentral bank should have the possibility of discounting assets(for example, loans or commercial paper) other than eligiblecollateral. There is a diversity of experience and practiceamong the major central banks. For example, in the UnitedKingdom and the United States, as well as in some otheradvanced countries, central banks have considerable discre-tion in deciding what kind of collateral to accept in excep-tional circumstances to provide liquidity to the bankingsystem. By contrast, Germany's Bundesbank has almost nodiscretion about what kind of collateral it can accept, andthere has been no instance in which uncollateralized inter-vention was necessary.

The German modelThe German system is an important benchmark for examin-ing how crisis management might take place within EMU,because the mandate of the European System of CentralBanks is similar in many respects to the Bundesbank's. TheBundesbank—like the ECB—has no explicit responsibilityfor safeguarding the stability of the financial system and doesnot act as a lender of last resort. Indeed, the German frame-work for dealing with crises seems to be constructed so as toavoid a role for the Bundesbank in providing funds in rescue

Box1

Prudential supervision and financial stabilityThe European System of Central Banks Statute and theMaastricht Treaty assign to the European System of CentralBanks only limited functions related to prudential supervisionand the stability of the financial system, but it does have anexplicit role in promoting the smooth functioning of the pay-ment system. The flow of supervisory information betweenthe ECB and the competent authorities is also regulated by theBCCI Directive (Directive 96/25/EC of June 29,1995).

The European Monetary Institute's 1997 annual reportprovided some clarification on how these provisions willbe implemented in EMU. In EU legislation relating to pru-dential supervision of credit institutions and the stability ofthe financial system, the ECB has the option of playing anadvisory role and must be consulted only on draft EU andnational legislation that influences the stability of financialinstitutions and markets.

To ensure effective interaction between the EuropeanSystem of Central Banks and national supervisory authori-ties, the Maastricht Treaty stipulates that the EuropeanSystem of Central Banks "shall contribute to the smoothconduct of policies pursued by the competent authoritiesrelating to the prudential supervision of credit institutionsand the stability of the financial system." The EuropeanSystem of Central Banks will not systematically receive super-visory information; its requests for such information will beconsidered by national banking supervisors, who will informit, on a case-by-case basis, in the event of a banking crisiswith systemic implications. The EU Council of Ministers,upon the initiative of the European Commission, may assignspecific tasks to the ECB related to prudential supervision;however, the European Monetary Institute's annual reportindicates that any transfer of supervisory powers fromnational authorities to the central bank is considered prema-ture at this stage.

Finance & Development / December 1998 13


Box 2

Lender-of-last-resort operationsThe question of which institutions in EMU will have theauthority and responsibility to act as lenders of last resortduring liquidity crises is still ambiguous. Whether or not theGoverning Council of the ECB will choose to maintain thisambiguity remains to be seen.

In a local liquidity crisis (that is, one affecting a largeinstitution located in an EMU country), the key issue iswhether the national central banks will be able, withoutauthorization from the ECB, to provide liquidity support totroubled institutions. Although the national central bankshave scope for such operations, the ECB's GoverningCouncil could, by a qualified majority vote, prohibit themfrom purchasing ineligible collateral from illiquid institu-tions. Indirect means of assisting banks experiencing liquid-ity difficulties are also open to national central banks—for example, they could swap some of the liquid assets intheir balance sheet for some of the troubled banks' illiquidassets and assume the credit risk on the latter, or they couldguarantee the troubled institutions—but the GoverningCouncil may issue guidelines prohibiting such on- and off-balance-sheet operations or specify that they require prior

authorization. If guidelines are so strict as to prevent thenational central banks from providing direct or indirectliquidity assistance to troubled banks, they may be able toprovide assistance by opening up the definition of eligibleTier II collateral to include a broader range of assets, but thiswould require approval by the ECB's Governing Council andcould delay resolution of a crisis.

In the event of a general liquidity crisis that would affectthe entire EMU—for example, gridlock in an EMU paymentsystem or TARGET—the ECB may need to provide liquidityto avert a systemwide crisis. Collateralized intraday credit andextraordinary open market operations may be sufficient toinject the necessary funds in some instances, but in othersthese measures may not suffice, owing to a lack of eligible col-lateral. At such times, the risk of a systemic crisis would behigh, forcing the European System of Central Banks to acceptineligible paper as collateral for payment system overdrafts oropen market operations. For example, when a general liquid-ity crisis occurred in the United States during the stock mar-ket crash of October 1987, the U.S. Federal Reserve Systemgave banks unrestricted access to its discount window so thatthey could keep their credit lines to brokers and securitieshouses open.

operations. The system, in effect, has three lines of defense:(1) supervision and regulation by an independent body;(2) short-term liquidity assistance from the LiquidityConsortium Bank (a specialized institution—30 percent ofwhose capital is held by the Bundesbank, with the remainderheld by all categories of German banks—that ensures thetimely settlement of domestic and external interbankpayments), combined with brokered market solutions; and(3) deposit insurance and, if necessary, injections of publicfunds. In practice, thanks to close cooperation with the inde-pendent supervisory authority and the Bundesbank, theLiquidity Consortium Bank has been able to identify solventinstitutions to which short-term liquidity assistance shouldbe provided. This close cooperation has also allowed theBundesbank to be involved in resolving problems by encour-aging strong banks with ample liquidity to purchase illiquidbut sound assets from troubled institutions in need of liq-uidity. Deposit insurance and public funds have been used todeal with insolvent institutions.

Is the German model applicable to EMU?There are a number of reasons why such a framework (threelines of defense, with no central bank funds) might not beapplicable in the event of a crisis within EMU. First, no insti-tution corresponding to the Liquidity Consortium Bankexists in other EMU countries, nor is one planned at theEMU level. Second, even if such institutions existed, theywould be inadequate in relation to the size and the cross-border systemic implications of a liquidity crisis involving a

major pan-European banking group unless they wereendowed with considerable resources and had much greateraccess to supervisory information than national supervisorsare likely to provide to the ECB. Third, the current agree-ment about sharing information between the ECB and thenational supervisors—which can be summarized by the for-mula of no real obligation, no real obstacle, and some under-standing—would probably not give the central bank thesame authority to broker a solution to a banking crisis at theEMU level as the Bundesbank has at the national level. TheECB could play such a role only if it were perceived to havethe same access to supervisory information at the EMU levelthat the Bundesbank has at the national level, or if it had theauthority to inspect counterparties in order to assess theircreditworthiness. Fourth, the German system has workedwell in an environment characterized by relatively under-developed capital markets and a large share of public owner-ship in the banking system, in which a crisis would unfoldin "slow motion," compared with the speed with which a cri-sis would probably spread through EMU-wide capital mar-kets and banking systems. Finally, in an integrated EMUbanking system with several EMU-wide institutions, the useof deposit insurance schemes and treasury funds would addto the time needed to determine how financial responsibili-ties should be shared among national authorities and coulddelay a decision about how to deal with a problem bank.

Under the current institutional framework, considerableuncertainty also remains about the scope that national centralbanks might have for providing emergency liquidity assistance



to troubled banks (Box 2). In all relevant crisissituations involving pan-European markets andinstitutions, the ECB appears to have to decideeither to inject extra funds into the system (inthe event of a general liquidity crisis) or to allownational central banks to intervene in a local liq-uidity crisis. Such intervention would requireintimate knowledge of counterparty institutions,however. Supervisory information would benecessary to assess the credit risk that such oper-ations would involve in the event that theEuropean System of Central Banks were forcedto accepl ineligible collateral. Moreover, the ECBwould be unable to rely on market assessmentsto distinguish between a liquidity crisis and asolvency crisis. (In most liquidity crises, the mar-kets would question the solvency of the institu-tion in difficulty because a solvent institutionwould have been able to borrow from the moneymarkets to meet its liquidity needs.)

Even if the ECB's involvement in the man-agement of liquidity crises is to be minimal—possibly only to authorize, or to refuse toauthorize, the national central banks to act aslenders of last resort—the current arrange-ments between national supervisors and theECB governing the exchange of supervisoryinformation seem to be too limited to allowwell-informed decisions during a fast-breakingcrisis. An arrangement in which the ECBdoes not have independent access to supervi-sory information on a systematic basis and inwhich banking supervisors will inform theEuropean System of Central Banks "on a case-by-case basis should a banking crisis arise"makes the ECB entirely dependent on nationalsupervisory authorities for the information itneeds to make sound decisions. In addition,the new framework is not clear about theunderstandings between the ECB, the 11 national centralbanks, the 11 supervisory authorities, and, possibly, the 11treasuries in EMU. In the event of a crisis involving aEuropean banking group, clarity and transparency about thesharing of information would greatly facilitate coordinationand management during the early stages of a financial prob-lem or crisis.

Constructive ambiguity?

The limited agreement on information sharing probablyreflects the fact that no clear lender-of-last-resort functionhas been attributed to the European System of Central Banksand that, at present, there does not seem to be a fully worked-out framework for crisis management in EMU. Currentunderstandings indicate that crises might have to bemanaged through ad hoc arrangements to do whatever is

Alessandro Prati is anEconomist in the CapitalMarkets and FinancialStudies Division of theIMF's ResearchDepartment.

Garry ]. Schinasi is Chiefof the Capital Marketsand Financial StudiesDivision of the IMF'sResearch Department.

necessary to avert systemic problems. The ideamay be that, in the event of a crisis, a nationalcentral bank or a national authority would finda way to provide liquidity support, and thencentral banks and supervisors would quietlypursue longer-term solutions, including find-ing buyers. (The role that national treasurieswould play in crisis management in EMU isanother open question. Whereas treasurersmay ultimately provide the funds for bankrescues, it is unlikely that they could be theimmediate source of liquidity.) Whereas thislack of transparency may be interpreted as"constructive ambiguity" intended to reducemoral hazard, the current understandings andarrangements within EMU would need signifi-cant further development before they would beworkable in an environment in which speed isincreasingly critical in the handling of financialand systemic crises. Some European authoritiesbelieve, however, that, once established, sucharrangements may well not be disclosed to thegeneral public because to do so would increasemoral hazard.

The current decentralized approach does notassign responsibility for supervising pan-European banks or for ensuring EMU-widefinancial market stability either to nationalcentral banks or to national governments. AsEuropean banking groups emerge, the ques-tions of whether national central banks couldadequately assess the risks of contagion andwhether the home country central bank of abank in difficulty could be easily identified willbecome increasingly relevant. In addition,decentralized lender-of-last-resort policies maycreate an uneven playing field and introducedifferent levels of moral hazard across EMU.At the same time, the ECB will be at the center

of European financial markets without the full set of toolsnecessary for independently assessing the creditworthinessof counterparties or a framework for rapidly providing sup-port to solvent but illiquid institutions. This is not likely tobe sustainable, and the ECB may be forced to assume a lead-ing and coordinating role in crisis management and bankingsupervision. 1MB

This article draws on Chapter 5 of International Monetary Fund, Inter-

national Capital Markets: Developments, Prospects, and Key Policy

Issues (Washington, October 1988), and a more detailed study by

Alessandro Prati and Garry I. Schinasi, "Financial Stability in EMU"

(Washington: International Monetary Fund, forthcoming in the IMF's

Working Paper series).



Capital Account Liberalizationand the IMFB a r r y Eichengreen and

Michael Mussa

Capital account liberalization may have substantial benefits, but recentexperience also underscores its risks. How should liberalization besequenced and managed to ensure that the benefits dominate?

T HE GROWTH of international financial transac-tions and international capital flows is one ofthe most far-reaching economic developments ofthe late twentieth century and one that is likely to

extend into the early twenty-first century. Net flows to devel-oping countries tripled, from roughly $50 billion a year in1987-89 to more than $150 billion in 1995-97, beforedeclining in the wake of the Asian crisis. Gross flows todeveloping countries and more generally have grown evenmore dramatically, rising by 1,200 percent between 1984-88and 1989-94. An increasing number of IMF membercountries have removed restrictions on capital accounttransactions in an effort to take advantage of the opportuni-ties afforded by this remarkable rise in internationalfinancial flows.

But these developments, as the official community hasacknowledged, raise important questions about the role ofthe IMF in financial liberalization. In September 1996, theInterim Committee (the committee of finance ministers andcentral bank governors that reviews IMF activities) requestedthe IMF Executive Board to analyze trends in internationalcapital markets and examine possible changes to the IMF's

Articles of Agreement so that the organization could betteraddress the issues raised by the growth of international capi-tal flows. In April 1997, the Interim Committee agreed thatthere would be benefits from amending the Articles to enablethe IMF to promote the orderly liberalization of capitalmovements. It reiterated this position in a statement issuedat the Annual Meetings of the World Bank and the IMF inHong Kong SAR the following September.

This idea that the IMF should actively promote theliberalization of capital flows has not gone unchallenged. Inthe wake of the Asian crisis, which has seen sharp reversals ofcapital flows for a number of countries, officials and aca-demics alike have questioned how desirable capital accountliberalization is and whether it is advisable to vest the IMFwith responsibility for promoting the orderly liberalizationof capital flows.

Growth of capital flowsPowerful forces have driven the rapid growth of interna-tional capital flows. Prominent among these are

• the removal of statutory restrictions on capital accounttransactions, which is a concomitant of economic liberalization



and deregulation in both industrial and devel-oping countries;

• macroeconomic stabilization and policyreform in the developing world, which havecreated a growing pool of commercial issuersof debt instruments;

• the multilateralization of trade, whichhas encouraged international financial trans-actions designed to hedge exposure to cur-rency and commercial risk; and

• the growth of derivative financialinstruments—such as swaps, options, andfutures—which has permitted internationalinvestors to assume some risks while limitingtheir exposure to others.

Above all, technology has played a role. Revolutionarychanges in information and communications technologieshave transformed the financial services industry worldwide.Computer links enable investors to access information onasset prices at minimal cost on a real-time basis, whileincreased computer power enables them rapidly to calculatecorrelations among asset prices and between asset prices andother variables. Improvements in communications technolo-gies enable investors to follow developments affecting for-eign countries and companies much more efficiently. At thesame time, new technologies make it increasingly difficult forgovernments to control either inward or outward interna-tional capital flows when they wish to do so. All this meansthat the liberalization of capital markets—and, with it, likelyincreases in the volume and the volatility of internationalcapital flows—is an ongoing and, to some extent, irreversibleprocess with far-reaching implications for the policies thatgovernments will find it feasible and desirable to follow.

It is important to recognize that financial innovation andliberalization are domestic, as well as international, phenom-ena. Not only have restrictions on international fina'ncialtransactions been relaxed, but regulations constraining theoperation of domestic financial markets have been removedas countries have moved away from policies of financialrepression. Domestic and international financial liberaliza-tion have generally gone hand in hand. Both respond tomany of the same incentives and pressures.

Costs and benefitsCapital mobility has important benefits. In particular, itcreates valuable opportunities for portfolio diversification,risk sharing, and intertemporal trade. By holding claimson—that is, lending to—foreign countries, households andfirms can protect themselves against the effects of distur-bances that impinge on the home country alone. Companiescan protect themselves against cost and productivity shocksin their home countries by investing in branch plants in sev-eral countries. Capital mobility can thereby enable investorsto achieve higher risk-adjusted rates of return. In turn,higher rates of return can encourage increases in saving

"It is importantto recognizethat financial

innovation andliberalizationare domestic,

as well asinternational,phenomena."

and investment that deliver faster rates ofgrowth.

At the same time, however, in a significantnumber of countries, financial liberalization,both domestic and international, appears tohave been associated with costly financialcrises. This association may be somewhatdeceptive, given that financial crises are com-plex events with multiple causes and haveoccurred in less liberalized as well as moreliberalized financial systems. Still, there havebeen enough cases where financial liber-alization, including capital account liber-alization, has played a significant role incrises to raise serious questions about

whether and under what conditions such liberalization—particularly capital account liberalization—will be beneficialrather than harmful.

At the theoretical level, the controversy over the benefits offinancial liberalization reflects diverging views on whetherliberal financial markets bring about an efficient allocation ofresources or are so distorted that the benefits they yield todirect participants too often are detrimental to the generalwelfare. Although a large "efficient markets" literature arguesthe first hypothesis, others insist that asymmetric informa-tion—a situation in which one party to a transaction has lessinformation than the other—pervades financial markets, andthat this greatly undermines their efficiency as mechanismsfor allocating resources. There is, moreover, good reason tothink that asymmetric information is particularly prevalentinternationally, because geography and cultural distancecomplicate the acquisition of information. While the revolu-tion in information and communications technologies—byreducing economic distance—has a profound effect in stimu-lating international financial transactions, it also leavesinternational markets—where information asymmetries areattenuated but not eliminated—particularly prone to thesharp investor reactions, unpredictable market movements,and financial crises that can occur when information isincomplete and financial markets behave erratically.

Role of policyThese developments make it critical to accompany financialliberalization with appropriate policies to limit excess volatil-ity and related problems and to contain their potentiallydamaging effects. As has long been recognized, sound macro-economic policies are essential for maintaining financialstability. A liberalized financial system is more demanding inthis respect than a repressed system in which large financialimbalances may be suppressed for long periods. Recent expe-rience, however, highlights the fact that macroeconomic sta-bility, while necessary, is not sufficient for financial stability,which also requires sound financial sector policies.

The first line of defense against financial risk must besound risk management by market participants themselves.



"The mostserious problemswith international

capital move-ments occur

when capitalflows out of a

country suddenly,precipitating

• • aa crisis.

Banks and nonbank financial intermedi-aries must manage their balance-sheet risksprudently. Corporate borrowers must rec-ognize and manage risks appropriately,which requires a strong system of corporategovernance. Any tendency to take on exces-sive risk can be contained through marketdiscipline facilitated by the adoption ofbest-practice accounting, auditing, and dis-closure standards. An appropriate environ-ment can be created by adopting policiesthat mandate proper accounting, auditing,and reporting rules and by taking care notto form a culture of implicit guarantees, sothat lenders will face significant capitallosses if they fail to assess credit risk prudent

When risk-management techniques are not well devel-oped, auditing and accounting practices leave much to bedesired, and other distortions interfere with the ability ofbanks and others to manage risk, prudential regulation hasan especially important role. The argument for prudentialregulation is reinforced when central banks and govern-ments backstop financial markets and provide a financialsafety net that can encourage banks and other market partic-ipants to take on excessive risk. A century and more of expe-rience points to the need, in most countries, for centralbanks to provide lender-of-last-resort services to preventilliquid financial markets from seizing up in periods of gen-eral distress. This backstopping function, though essential, isalso a source of moral hazard. The appropriate response fornational authorities is rigorous prudential supervision andregulation combined with careful design of the lender-of-last-resort facility to limit the scope and incentives for finan-cial market participants to take on excessive risk. Moregenerally, pursuing policies to develop a financial system thatrelies less heavily on banks and other intermediaries" andinvolves more direct risk bearing by ultimate investors canhelp to reduce the risks of costly crises and moral hazard.

Policies toward the capital accountThe most serious problems with- international capital move-ments occur when capital flows out of a country suddenly,precipitating a crisis. While sudden capital outflows can affectall forms of capital—debt, portfolio equity, and even directand real estate investment—the macroeconomic conse-quences are particularly serious when they involve debt, espe-cially sovereign debt and banking and financial system debt.Recent experience suggests, moreover, that short-term debtcan pose special problems for maintaining financial stability.

Like other risks, those posed by holdings of short-termdebt are best controlled at the source. The sovereign can andshould control its own borrowing. Banks and nonbank bor-rowers can and should avoid excessive dependence on short-term, foreign-currency-denominated debt. Banks shoulddevelop in-house models with which to manage risk, and the

national authorities can refer to these whencalculating risk weights for capital require-ments (as recommended by the 1996 MarketRisk Amendment to the Basle CapitalAccord). But where risk-management tech-niques are underdeveloped or significantfinancial market distortions exist, there is anargument for additional prudential mea-sures to identify and discourage excessiveshort-term, especially foreign-currency-denominated, borrowing that could jeopar-dize systemic stability.

Prudential regulations to contain therisks associated with capital flows have beendesigned and implemented in several ways.

Many countries have addressed the risk to the stability of thebanking system mainly by limiting banks' open net foreigncurrency positions (a net open position is the differencebetween unhedged foreign currency assets and liabilities,typically expressed as a percentage of the bank's capital base),while other countries (such as Chile and Colombia) havesought to discourage excessive foreign exposures of domesticcorporations and banks by taxing essentially all short-termcapital inflows. Some countries differentiate reserve require-ments for banks according to both the residency and cur-rency of denomination of deposits, while others differentiateaccording to currency of denomination but not residency.

There need not be a conflict between these policies and theobjective of capital account liberalization, defined as freedomfrom prohibitions on transactions in the capital and financialaccounts of the balance of payments. Indeed, the analogywith current account convertibility is direct. Article VIII ofthe IMF's Articles of Agreement, which defines currentaccount convertibility as freedom from restrictions on themaking of payments and transfers for current internationaltransactions and makes this an explicit objective of IMF pol-icy, does not proscribe the imposition of such price-basedrestrictions as import tariffs and taxes on underlying transac-tions. Correspondingly, capital account convertibility meansthe removal of foreign exchange and other controls, but notnecessarily all tax-like instruments imposed on the underly-ing transactions, which need not be viewed as incompatiblewith the desirable goal of capital account liberalization.

Exchange rate flexibility can also help to discourage exces-sive reliance on short-term foreign borrowing. There haverecently been a number of episodes in which an exchange ratepeg has been seen by both lenders and borrowers as a link in achain of implicit guarantees. In these circumstances, the highnominal interest rates characteristic of emerging markets canlead to very large short-term capital inflows. The exchangerisk associated with greater nominal exchange rate flexibilitycan play a useful, if limited, role in moderating the volume ofthese short-term flows. It can encourage banks and firms tohedge their short-term foreign exposures, which insulatesthem from the destabilizing effects of unexpectedly large



exchange rate movements. Greater exchangerate flexibility is no panacea, however; if intro-duced suddenly, without advance planning, andin a setting where banks and corporations haveheavy debts denominated in foreign currency,its effects can be destabilizing. But if the author-ities take advantage of a period of capitalinflows to introduce greater flexibility, so thatthe exchange rate begins its more flexible lifeby strengthening, the beneficial effects are likelyto dominate.

SequencingThe most important point to recognize in thesequencing of capital market liberalization isthe danger of precipitously removing restric-tions on capital account transactions beforemajor problems in the domestic financial sys-tem have been addressed. Among the problemsunder this heading are

« inadequate accounting, auditing, and dis-closure practices in the financial and corporatesectors that weaken market discipline;

« implicit government guarantees that encou-rage excessive, unsustainable capital inflows;and

• inadequate prudential supervision and reg-ulation of domestic financial institutions andmarkets, which open the way for corruption,connected lending, and gambling for redemp-tion (namely, the pursuit of high-return butlow-probability investments by institutionswith low or negative net worth).

Countries in which these problems are severebut that suddenly and fully open the capitalaccount run the risk of incurring a serious crisis.This implies that countries should liberalize thecapital account gradually, at the same time as theymake progress in eliminating these distortions.

In addition, liberalization that is limited toinflows to and through the banking systemcan pose considerable risks, even for a well-prepared and well-regulated system, if theseflows are substantial. Liberalization of capital inflows shouldthus proceed on as broad a front as possible, beginning withdirect investment inflows in order to avoid overloadingchannels more vulnerable to sudden reversals.

While foreign direct investment sometimes raises concernsabout foreign ownership and control, there is considerableevidence that economic benefits, including the transfer oftechnology and efficient business practices, are associatedwith such investment. Volatility in flows of direct investmentdoes not appear to generate the same acute problems offinancial crises as do sharp reversals of debt flows. For thisreason, liberalization of inward direct investment should

Barry Eichengreen is JohnL. Simpson Professor ofEconomics and PoliticalScience at the Universityof California, Berkeley.He was Senior PolicyAdvisor in the IMF'sResearch Departmentwhen the study on whichthis article is based wasprepared.

Michael Mussa is ChiefEconomist and EconomicCounsellor of the IMFand Director of itsResearch Department.

generally be an attractive component of abroader program of liberalization. Such liber-alization need not occur all at once; for coun-tries that face the prospect of large surges ofinward investment, a gradual approach may beadvisable. It generally makes little difference ifforeign investment is limited in some selectedsectors of the economy. The financial sector,however, is an important exception. Openingdomestic financial markets to participation byforeign (or multinational) financial institu-tions is an integral element of full capitalmarket liberalization. There can be importantbenefits, especially for smaller countries, fromthe diversification of risks that is made pos-sible when banks can operate across nationalboundaries.

ConclusionCapital account liberalization and financialliberalization more generally are inevitable forcountries that wish to take advantage of thesubstantial benefits from participating in theopen world economic system in today's ageof modern information and communicationstechnologies. As recent events have againdemonstrated, however, financial liberalizationalso has its dangers. As liberalized systemsafford opportunities for individuals, enter-prises, and financial institutions to undertakegreater and sometimes imprudent risks, theycreate the potential for systemic disturbances.There is no way to completely suppress thesedangers other than through draconian finan-cial repression, which is more damaging. Butthey can be limited considerably: soundmacroeconomic policies to contain aggregatefinancial imbalances and to ameliorate theeffects of financial disturbances can be com-bined with sound prudential policies designedto ensure proper private incentives for riskmanagement, especially in the financial sector.With these safeguards, orderly and properly

sequenced capital account liberalization and the broaderfinancial liberalization of which it is part are not onlyinevitable but clearly beneficial.

This article summarizes the conclusions of a study prepared by an IMF

staff team led by Messrs. Eichengreen and Mussa and comprising Giovanni

Dell'Arricia, Enrica Detragiache, Gian Maria Milesi-Ferretti, and Andrew

Tweedie. That longer study is published as IMF Occasional Paper 172,

Capital Account Liberalization: Theoretical and Practical Aspects

(Washington, 1998).

Finance & Development/ December 1998 19


SequencingCapital Account Liberalization

A country that is liberalizing its capital account faces the challenges of strengtheningfinancial institutions to ensure they are capable of operating in a more

market-oriented system and deciding how to achieve monetary aims and maintainmacroeconomic stability in a freer, more open environment.

R . Barry Johns ton

ALTHOUGH fears about the risks associated withcapital account liberalization have been reawak-ened by Asia's financial crisis, it is still a moreattractive option for many developing countries

than capital controls. First, there is much evidence thatcapital controls have not prevented outflows. Second, whencountries eliminate controls, they usually experiencestronger inflows, at least initially, as international investorsand residents who had placed their capital abroad react tothe improved investment environment. Capital inflows canimprove a country's balance of payments, smooth temporaryshocks to income and consumption, reduce borrowing costs,and spur economic growth.

Capital account liberalization is not without risk, however. Ifcapital inflows are not used efficiently, the markets may ques-tion their sustainability and the capacity of the recipient Coun-try to service its external debt. A loss of confidence couldtrigger reversals of capital flows and, in their wake, balance ofpayments difficulties and currency and banking crises.

Capital account liberalization is not an all-or-nothingaffair; there are as many ways to approach it as there arefinancial instruments and types of capital transactions (seebox). Capital flows can, for example, be intermediated by theinternational capital markets (when local nonfmancialagents are permitted to borrow or place funds abroad); bythe local capital markets (when nonresidents can access localfinancial markets and intermediaries); or a combination ofboth (when local financial intermediaries borrow or placefunds abroad). Capital controls can take various forms—including outright prohibitions, licensing and approval pro-cedures, and transaction taxes—each with a different effecton flows. A country may liberalize certain components of itscapital account while maintaining controls on others.

Although many of the challenges posed by capital accountliberalization are no different from those posed by the

liberalization of domestic financial systems, capital accountliberali/.alion adds an external dimension and urgency tofinancial sector reforms. Whether capital inflows are chan-neled through domestic intermediaries or compete withthem, the intermediaries will need to be strengthened, eitherto ensure the efficient use of the capital inflows or becausecompetitive pressures on—and the need to restructure—domestic financial institutions will increase. Moreover, capi-tal account liberalization may induce banks and corporationsto take on more foreign exchange risk.

Prudential regulationThe liberalization of capital flows can thus be viewed as oneaspect of financial sector liberalization, in which the role of theauthorities is, first and foremost, to establish an appropriateregulatory framework. A comprehensive liberalization of capi-tal transactions and transfers does not signify an abandonmentof all rules and regulations applying to foreign exchange trans-actions. Regulations may have to be strengthened in a numberof areas, including prudential regulations related to nonresi-dent and foreign exchange transactions and transfers.

Prudential measures are generally defined as officialactions (laws, regulations, and officially sanctioned policiesor procedures) that (1) promote the soundness of individualfinancial institutions by ensuring adequate risk managementand effective internal governance and by fostering marketdiscipline, and (2) protect investors against fraud and decep-tive practices and ensure that financial agents carry out theirfiduciary responsibilities. However, prudential measuresnormally apply to the domestic and foreign activities offinancial institutions only. With respect to investments thatare not intermediated through financial institutions, pru-dential regulations tend to be limited to the admission andtrading of investment instruments and the provision ofrelated fiduciary services in the domestic market; they are



rarely used to protect investors purchasing instruments inforeign markets.

International transactions may involve certain types ofrisk that are not present in domestic transactions, includ-ing transfer risk (a foreign debtor may not be able toobtain the foreign exchange necessary to service its exter-nal debt on time), sovereign risk (a government may failto service its external debt), and country risk (a substan-tial number of debtors in a particular country may havedifficulty servicing external debt because of macroeco-nomic or political instability or other reasons). In recog-nition of national differences in supervisory andaccounting practices, and in the enforceability of bank-ruptcy and other laws designed to protect investors, thestandards that apply for licensing foreign financial insti-tutions to operate in the domestic market or for authoriz-ing the listing and trading of foreign securities indomestic markets may need to be different from thosethat apply to domestic agents.

The scope and content of prudential measures andprocedures are changing dramatically worldwide, as tech-niques for identifying, measuring, and managing finan-cial risk are updated and as the need for globalharmonization of supervisory approaches grows. Variousinternational groups of supervisors are developing agrowing body of generally accepted principles, standards,guidelines, and best practices to provide guidance on theproper treatment of capital flows. Because the emphasisin risk management is shifting away from reliance onquantitative limits and toward greater oversight of insti-tutional capacity to manage risk and increased public dis-closure of information, the application of generallyaccepted best practices involves little, if any, restrictiveimpact on capital movements. Such generally acceptedpractices also highlight the importance of managing for-eign exchange exposures and liquidity.

In countries with weak or embryonic financial systems,both the pace of capital account liberalization and thedesign of prudential measures become more complex. Localfinancial institutions may have limited capacity to assess andmanage risks associated with large capital inflows, and regu-latory authorities may have limited supervisory capacity. Forprudential reasons, such countries may need to developfinancial institutions, markets, and instruments before theycan durably liberalize their capital account. Countries mayalso need to adopt international standards of corporateaccounting and the timely disclosure of information.Countries with weak financial systems or limited supervisorycapacity may have to adopt crudely designed measures toachieve legitimate prudential objectives, and these measuresmay have a restrictive impact on capital flows.

Particular attention may have to be paid to strengtheningthe banking sector. In many developing countries, banks arethe major financial intermediaries for capital flows (this wasthe case in Korea and Thailand, two of the countries hit hard

Types of capital transactions that may be subjectto controls

Capital and money markets

- Shares or other securities of a participating nature- Bonds or other debt securities- Money market instruments- Collective investment securities- Derivatives and other instruments

Controls on capital and money market transactions may apply to pur-chases made locally by nonresidents or sales or issues carried outabroad by residents (inflows), or to sales or issues carried out locally bynonresidents or purchases made abroad by residents (outflows).

Credit operations

- Commercial credits- Financial credits- Guarantees, sureties, and financial backup facilities

Controls may apply to inflows (credits provided to residents by non-residents) or outflows (credits provided by residents to nonresidents).

Direct investment and real estate transactions

Controls may apply to inward and outward direct investment, liquida-tion of investment, or purchases and sales of real estate made locally bynonresidents and purchases of real estate made abroad by residents.

Provisions specific to commercial banks

Controls may be applied to nonresident deposits and bank borrowingabroad (inflows) and to foreign loans and deposits (outflows).

Personal capital movements

Controls may be applied to deposits, loans, gifts, endowments, inheri-tances, legacies, and settlements of debts.

Provisions specific to institutional investorsControls may include limits on the purchase of securities issued bynonresidents.

by the East Asian crisis). Banks' interest rate and credit poli-cies can have a strong influence on the composition of capi-tal flows—for example, wide deposit-lending spreads mayencourage corporations to borrow overseas, while under-pricing of credit and maturity transformation risks may dis-tort the yield curve. There is ample evidence that allowingweak banks to expand their balance sheets will lead to bank-ing crises. Measures to strengthen banks include raisingcapital-adequacy ratios and loan-loss-provisioning require-ments; improving credit assessment, liquidity management,risk pricing, and bank management; increasing foreign par-ticipation in the banking sector; and conducting rigorouson-site inspections.

More direct measures may need to be taken to curb thegrowth of bank balance sheets (for example, placing limitson international borrowing by banks) in countries whosebanking sector weaknesses are underestimated by financialmarkets or in countries that are perceived to guarantee,



explicitly or implicitly, banks' overseas borrowing. In thesecases, it would be risky to liberalize banks' access to foreignfinance until systemic weaknesses have been addressed,appropriate supervisory standards adopted, and concernsabout moral hazard addressed. (The same concerns apply tocorporate borrowers believed by foreign lenders to be backedby government guarantees.)

Monetary and exchange rate policyWith greater freedom of capital movements, domestic short-term interest rates will increasingly be determined by foreignshort-term interest rates and expectations for short-termmovements in the exchange rate—that is, the covered-interest-rate-parity condition. Attempts by a country to setinterest rates and exchange rates that are inconsistent with theinterest-rate-parity condition could trigger sizable inflows oroutflows of short-term capital. Thus, the country's capacity touse monetary and exchange policies to achieve separatemacroeconomic targets will be increasingly constrained ascapital becomes more mobile. If monetary policy targetsinflation, the exchange rate cannot be used, for example, toachieve current account objectives (although fiscal policy canbe used to influence the savings-investment balance toachieve such objectives). Conversely, if the exchange rate istargeted to achieve current account objectives or if theexchange rate is fixed, monetary policy will not beautonomous enough to serve as a tool for achieving domesticstabilization or managing the consequences of short-termcapital inflows.

The initial policy response to strong capital inflows incountries with a pegged exchange rate has generally beensterilized intervention. However, such intervention involvesquasi-fiscal costs and is usually of limited effectivenessbecause it keeps domestic interest rates high, attracting fur-ther inflows. In some cases, pegged exchange rates have alsoled borrowers to underestimate the risks attached to loansdenominated in foreign currency. Therefore, as capitalaccount liberalization has progressed, a number of countrieshave moved toward greater exchange rate flexibility to dis-courage shorter-term capital inflows that are due solely tointerest rate differentials.

Some countries have adopted a strong nominal exchangerate anchor, subordinating monetary policy to the mainte-nance of the anchor, and allowed interest rates to adjust inresponse to capital movements. This has enabled them toavoid capital flow reversals that would have been provokedby uncertainty about the exchange rate. Because they cannotuse exchange rate policy to reduce short-term capitalinflows, they have paid greater attention to prudential stan-dards to reduce the risks associated with potential reversalsin capital flows.

Other countries have targeted the exchange rate (hence,monetary policy) to maintain competitiveness and relied onfiscal consolidation to achieve domestic stabilization and off-set the effects of large capital inflows. However, because fiscal

policy is fairly inflexible in the short term, the authoritieshave had to resort to other measures—capital controls, inparticular—to deal with volatile flows. Generally, such con-trols are effective primarily as temporary measures.

Monetary targets and instruments. Increased capital mobil-ity has implications for the design of monetary policy frame-works and the use of different monetary instruments. Underfixed or managed floating exchange rate regimes, the externalcounterpart of the money supply may become more volatileand the demand for domestically defined monetary aggre-gates, more sensitive to international interest rate differen-tials. As a result, it may be more difficult to identify amonetary aggregate that is stable enough to be used to predictthe evolution of other nominal economic variables. Capitalaccount liberalization thus reinforces the trend toward adopt-ing more eclectic monetary frameworks and giving moreweight to exchange rates in monetary assessments.

When capital is very mobile, the effectiveness of differentmonetary instruments is altered. For example, instrumentssuch as credit or interest rate ceilings or high nonremuneratedreserve requirements that impose high costs or administrativeconstraints on banks may be circumvented by disintermedia-tion through the capital account and lose some of their effec-tiveness. In contrast, monetary instruments that affect the costof money or credit in financial markets may be transmittedmore rapidly to credit and exchange markets, allowing thecentral bank of a country to influence the decisions of finan-cial institutions and markets that use the national currency inboth local and international operations.

Although a number of countries have reduced—or eveneliminated—reserve requirements, others still rely on themto influence supply and demand in the market for bankreserves (bank deposits with the central bank). However,nonremunerated reserve requirements act as a tax on thebanks that are subject to them and therefore encourage dis-intermediation to financial institutions and markets that arenot subject to such requirements. Although some countriesavoid this potential problem either by remunerating reserverequirements at rates that are close to market rates or byreducing or eliminating reserve requirements, others haveretained selected capital controls—for example, on the issueof certificates of deposit locally by nonresident banks—tosafeguard the effectiveness of their reserve requirements. Theexpectation is that limiting foreign banks' access to a particu-lar instrument would also limit disintermediation.

Open market operations play a core role in steering inter-est rates, managing liquidity in the market, and signaling thestance of monetary policy. In a context of free capital move-ments, the central bank is able, through open market opera-tions, to influence conditions in both the domestic and theexternal markets for the national currency. Standing facili-ties, rediscount quotas, and public sector deposits are fre-quently used to support open market operations,particularly to guide interest rate movements, and to trans-mit the central bank's message quickly and clearly to

99" Finance & Development / December 1998


markets. The adoption of indirect monetarycontrols has thus supported moves toward cap-ital account convertibility.

Discriminatory reserve requirements. Onemeasure frequently used to reduce capitalinflows, or to change their composition, is theimposition of differential reserve requirementson short-term borrowing from nonresidents.Although it is sometimes referred to as a"Chile-type" measure because it was adopted bythe Chilean authorities in 1991, it actually has along history of use in other countries.

Various empirical studies of such measureshave concluded that they have a temporaryeffect that can be eroded quite quickly, depend-ing on the scope of controls and the stage ofdevelopment of financial markets and instru-ments. Financial derivatives, as well as tradi-tional instruments not covered by exchangeregulations, can be used to circumvent controlson short-term capital. Moreover, if the differential reserverequirement is not applied to all short-term foreign capitalinflows, it may encourage larger short-term inflows throughinstruments not subject to the requirement. The danger isthat, to avoid controls, capital flows will be channeledthrough riskier, less regulated financial intermediaries;instruments and institutions may be created specifically tocircumvent the controls. Over time, Chile found it necessaryto extend its reserve requirement to all short-term capitalinflows. More recently, when faced with capital outflows,Chile eliminated its discriminatory reserve requirements.

SequencingTo maximize the benefits of capital account liberalizationwhile minimizing the risks, much thought must be given topace and sequencing. The conventional view of sequencingemphasizes the importance of achieving macroeconomicstability and developing domestic financial institutions, mar-kets, and instruments before liberalizing the capital account.According to this view, capital account liberalization shouldoccur late in a country's economic reform program. An alter-native view stresses constraints on reforms and the limitedcapacity of countries to reform themselves in the absence ofexternal pressures for reform; this view favors early capitalaccount liberalization, which can serve as a catalyst forbroader economic reforms and overcome vested interests'opposition to reforms. A middle view is that capital accountliberalization should be part of a concurrent, integrated, andcomprehensive approach to macroeconomic and structuralreform; in this view, the coordination of reforms in thedomestic and external sectors is the key issue.

The benefits, costs, and risks of each of these three strategieswill vary from country to country, depending on startingconditions and economic objectives. One approach would beto focus on the contribution each strategy would make to the

R. Barry Johnston is Chiefof the Exchange Regimeand Market OperationsDivision in the IMF'sMonetary and ExchangeAffairs Department.

broad objectives of improving efficiency inresource mobilization and allocation, and pro-moting financial and macroeconomic stability.Thus, to the extent that a specific reformimproves resource allocation and helps toachieve—or at least does not undermine—financial and macroeconomic stability, it shouldbe undertaken.

Reforms that increase diversity in thefinancial system (for example, by introducingnew technologies and instruments, as well asnew skills and risk-management capabilities),strengthen the capital structures of financialinstitutions, and promote competition wouldenhance efficiency. Liberalizing access to inter-national capital markets—when it leads to theadoption of new accounting and disclosurerequirements and increases incentives to reviseoutdated regulatory structures and ineffectivesupervisory arrangements—could improve

financial discipline. The introduction of new instruments forhedging and managing risks and the diversification of fund-ing sources and asset distribution would also strengthenfinancial systems. (Because hedging instruments have noreason to exist in the absence of price uncertainty, theirdevelopment will not occur until interest and exchange rateshave been freed.) Finally, in choosing which financial inter-mediaries to liberalize, countries should, in the interests ofefficiency and stability, consider those that are subject to themost developed regulatory framework, have the strongestgovernance, and pose the least risk of moral hazard.

Regulatory changes that might not enhance efficiencyinclude those that support existing monopolies and ineffi-cient financial structures—for example, by allowing only thedominant domestic financial institutions to access foreignsources of funding or by encouraging concentration, ratherthan diversification, of assets, funding sources, and risks.

Although certain rules about sequencing capital accountliberalization—for example, countries should liberalizelong-term flows before short-term flows, and foreign directinvestment before portfolio investment—have the appeal ofsimplicity, the fungibility of capital makes their practicalapplication difficult. The maintenance of restrictions on cer-tain types of capital transactions may serve primarily to buytime for the more fundamental restructuring of financialmarkets, the adoption of appropriate prudential standardsand supervisory arrangements, and the development of thenecessary indirect monetary instruments. B33

This article is based on R. Barry Johnston, 1998, "Sequencing Capital

Account Liberalizations and Financial Sector Reform," IMF Paper on

Policy Analysis and Assessment 98/8 (Washington: International Monetary

Fund).

Finance & Development / December 1998 *•


\v Capital AccountLiberalization in theSouthern MediterraneanSal eh A/1. N soul i and Mo u n i r Ra c h ed

The debate on capitalmobility has intensifiedin the wake of East Asia'srecent financial crisis,largely because of therisk of sudden reversals.The crisis demonstratesthe importance of astrong rnacroeconomicstance, sound institu-tions, and an orderlysequencing of reformsto maximize the benefitsand minimize the risksof capital accountliberalization.

oVER the past two de-cades, many countriesin the developing worldhave benefited from

relaxing restrictions on capital move-ments. In an increasingly globalizedworld, the Southern Mediterraneancountries of Algeria, Egypt, Israel,Jordan, Lebanon, Morocco, Syria,Tunisia, and Turkey could also reapconsiderable benefits from establish-ing full currency convertibility in anorderly manner.

A country is said to have achievedfull convertibility of its currencywhen residents and nonresidents areallowed to convert the currency, atprevailing exchange rates, into for-eign currencies and to use the latterfreely for international transactions.Full capital account convertibilitypaves the way for a more efficientallocation of savings and increases a

country's attractiveness to foreign investors.The elimination of controls on capital trans-actions gives businesses and individualsaccess to foreign financial markets andincreases the funding available for trade andinvestment. In addition, cross-border com-petition widens the choice of risk-adjustedrates of return open to investors, providesopportunities for portfolio diversification,and encourages domestic financial marketsto become more efficient.

Net private capital inflows to the SouthernMediterranean countries have picked up inrecent years, but they vary from country tocountry (Chart 1). Although many factorsaffect capital flows and growth, and it isdifficult to establish causality, it is worth not-ing that Lebanon, which has virtually no

exchange restrictions, has enjoyed massivenet inflows, while Syria, the country withmost restrictions, has received the smallestvolume of inflows. Generally, the countrieswith the most exchange restrictions have hadlower growth rates than those with more lib-eral exchange systems (Chart 2).

Exchange system restrictionsAll the Southern Mediterranean countrieslisted above except Egypt (for technical rea-sons) and Syria have established currentaccount convertibility by accepting the oblig-ations under Article VIII of the IMF'sArticles of Agreement. Egypt, Israel, Jordan,Lebanon, and Turkey have also achieved sub-stantial capital account convertibility, whileAlgeria, Morocco, Syria, and Tunisia still havesignificant restrictions.

Except for Israel and Turkey, the countriesin the first group above have virtually norestrictions on direct investment, while thosein the second group regulate either inward oroutward investment, or both. With regard tosecurities and money market instruments,fewer restrictions are imposed, in general, oninflows than on outflows, but the financialmarkets in most of the nine countries are notsufficiently developed to attract portfolioinvestment. Israel, Jordan, and Lebanonhave the most liberal codes for capital andmoney market transactions, with no restric-tions on either inflows or outflows. Althoughthe other Southern Mediterranean countriesgenerally do not prohibit foreign investmentin domestic securities, they impose restric-tions on outward flows, ranging from arequirement for prior approval to outrightprohibition. Issuance of securities by non-residents in domestic markets is unregulatedin five countries (Egypt, Israel, Jordan,



Chart 1

Private net capital flows, 1996-97 average(percent of GDP)

Lebanon, and Turkey), requires prior approval in Tunisia,and is prohibited in Syria. Only Tunisia regulates commercialborrowing from abroad, while Algeria, Lebanon, Syria, andTunisia have restrictions on lending abroad. All the countriespermit nonresidents to hold accounts in foreign and domes-tic currencies, but residents' accounts are subject to moreregulation than nonresidents' accounts and are fully convert-ible into foreign exchange only in Egypt, Israel, Jordan, andLebanon.

Present conditionsTo maximize the potential benefits and minimize the risks ofan open capital account, the Southern Mediterranean coun-tries must pursue an orderly approach to establishing andsustaining full convertibility.

Fiscal policy. A prudent fiscal policy is an important ele-ment in achieving and maintaining capital convertibility.Large fiscal deficits that require financing through moneycreation may destabilize the exchange rate and discourageboth foreign and domestic investment. Reliance on foreignloans with high interest rates creates debt-managementproblems, reduces creditworthiness, and weakens an econ-omy's ability to manage external shocks. During 1996-97, ofthe Southern Mediterranean countries that had alreadyachieved substantial capital account convertibility, onlyLebanon and Turkey had relatively large deficits. The fiscaldeficits of most of the countries that have significant restric-tions on capital movements (Morocco, Syria, and Tunisia),though manageable, still need to be reduced (Chart 3).Algeria has a surplus.

Monetary policy. A sound monetary policy that comple-ments and is facilitated by fiscal discipline is another criticalelement, because excess liquidity expansion will spill overinto the external sector. In 1996-97, monetary performancewas mixed in the countries with substantial capital convert-ibility. Monetary policy in Lebanon and Turkey appeared tobe too accommodating of fiscal policy, whereas it seemed tohave been more independent of fiscal policy in Egypt andIsrael, and Jordan had a highly restrained monetary policy.None of the countries with substantial capital restrictionsexperienced an excessive growth in domestic liquidity.

Exchange rate and reserves. A market-clearing exchangerate is essential to ensure external balance. Furthermore, toavoid wide exchange rate fluctuations, prudent macroeco-nomic policies need to be coupled with adequate interna-tional reserves. Of the Southern Mediterranean countrieswith substantial capital account convertibility, Egypt was theonly one to register a current account surplus in 1996-97,while Lebanon had a very large deficit. The deficit-to-GDPratios of the other three countries in this group were in thelow single digits. Of the countries with significant capitalrestrictions, two had current account surpluses and two hadrelatively modest deficits (Chart 4).

Israel and Turkey have flexible exchange rates, but theirreal effective exchange rates have risen slightly in recent

Charts

Central government balance, 1996-97 average(percent of GDP)

Chart 4

Current account balance, 1996-97 average(percent of GDP)

Sources: International Monetary Fund. World Economic Outlook: Interim Assessment(Washington, December 1997); and country authorities.

years. Egypt and Lebanon, which have practically anchoredtheir currencies to the U.S. dollar, witnessed sharp increasesin their real effective exchange rates. Among the countriesthat maintain substantial capital controls, Algeria and Syriaposted declines in their real effective rates, the former


Chart 2

Growth rates, 1996-97 average(percent change in real GDP)


because of an adjustment in the nominal exchange rate,the latter mainly because domestic prices rose more slowlythan prices in its trading partners. Morocco and Tunisiahave adopted a managed float that, in recent years, hasessentially stabilized their real effective exchange rates.All the countries had reserves well in excess of threemonths of imports.

Financial sector. An efficient and sound financial sector isan essential ingredient of capital account convertibility,enabling banks to invest capital inflows prudently andweather shocks. Efficiency requires market-based monetaryinstruments and a liberal regulatory framework. The sector'ssoundness depends on, among other things, effective bank-ing supervision and observance of prudential ratios.

All nine Southern Mediterranean countries have reformed,to varying degrees, their monetary instruments and bankingsystems. Countries that have already opened their capitalaccount rely more heavily on indirect instruments and onmarket-based interest rates. In Egypt, Israel, Lebanon, andTurkey, indirect instruments play an important role in deter-mining market interest rates. The treasury bill rate, deter-mined through auctions, and the discount rate have a stronginfluence on deposit and lending rates. The monetaryauthorities in these countries also rely, in part, on the sale andredemption of treasury bills to regulate bank reserves andcredit expansion. Interest rates in Jordan have been less flexi-ble, although the country has made some progress in usingindirect monetary instruments.

In contrast, in the countries that have substantial capitalrestrictions, deregulation of financial markets is a relativelyrecent phenomenon. Since 1994, Algeria has begun to relymore on indirect instruments, including auctions of treasurybills and repurchase agreements. In Morocco, reform of thefinancial sector has been in progress for nearly a decade, withmost interest rates being freed. Tunisia has liberalized-lend-ing rates for nonpriority sectors and recently freed mostinterest rates. Syria continues to allocate credit according toan annual plan, however, and to set interest rates.

Many of the Southern Mediterranean countries haveundertaken reforms of their banking systems, although datadeficiencies hamper analysis of bank performance in most ofthem. Furthermore, with few exceptions, data on prudentialindicators are either unpublished or unavailable. Nonetheless,it is possible to make some limited observations. In Egypt andJordan, progress has been made in increasing the capital baseof banks and meeting international prudential standards. Thecapitalization of Israeli banks is adequate, although their ratesof return are among the lowest in the region. Lebanon's bankshave regained profitability since its civil war ended, and theircapital-adequacy ratio has risen steadily.

A striking dichotomy exists in Turkey's banking system.The overall capital-adequacy ratio surpasses the minimum setby the Basle Committee on Banking Supervision, but statecommercial banks are undercapitalized. In addition, bankshave experienced persistent losses and liquidity problems

associated with direct lending at subsidized rates. In recentyears, private banks have strengthened their capitalization,but remain exposed to maturity mismatches of their assetsand liabilities and open net foreign exchange positions.

Morocco and Tunisia have well-established systems ofbank supervision and prudential standards. Moroccan banksare, on average, adequately, though unevenly, capitalized, andtheir loan-loss provisions are sufficient to counter a highratio of nonperforming loans. The profitability of Tunisia'sbanking sector, while high, has fluctuated considerably,reflecting the ongoing recapitalization by public and privatebanks. In Syria, banks remain undercapitalized.

Market orientation. A well-functioning price mechanismis essential to avoid distortions that reduce the efficiency ofresource allocation, affect capital flows adversely, and hindergrowth. Thus, subsidies, tax concessions, and price controlsneed to be phased out. All of the Southern Mediterraneancountries except Syria have taken measures to decontrolprices. Jordan and Lebanon have fairly liberal regulations onlabor markets, but rigidities remain in the other countries.Although several countries have made progress in privatizingand reforming, public enterprises, Algeria, Egypt, Morocco,Syria, Tunisia, and Turkey still have some way to go.

Tariffs that generate distortions between international anddomestic prices can also adversely affect resource allocationand capital flows. Several of the countries in the region haveliberalized trade by reducing tariffs and eliminating orreducing quantitative restrictions. Among the countries withsubstantial capital convertibility, Israel has the lowest averagetariff rate, at about 7 percent, and Egypt the highest, at28 percent. Among the others, Morocco has the lowest aver-age rate, at 20 percent, and Syria the highest, at 35 percent.

Meeting conditions for convertibilityThree conclusions can be drawn from this review. First,the Southern Mediterranean countries differ significantlyin the degree to which they restrict capital transactions.Second, some countries—even among those that havealready achieved substantial capital convertibility—need tostrengthen their economic and financial performance tosustain that convertibility. Third, among the countries withsignificant restrictions on capital transactions, Algeria,Morocco, and Tunisia need to reinforce their policies, andSyria will have to make a major effort to meet certain condi-tions before it can open its capital account.

Removal of restrictions. Restrictions that need to beremoved in some of the countries with substantial capitalaccount convertibility include those on foreign direct invest-ment (Israel and Turkey), residents' deposits abroad (Egyptand Israel), and resident and nonresident account convert-ibility (Turkey). The countries with substantial restrictionson capital transactions need to focus on liberalizing out-flows: Morocco and Tunisia have the fewest restrictions toremove, followed closely by Algeria, while Syria faces thegreatest challenges.



Policies. To sustain convertibility in countriesthat have already substantially liberalized theirexternal accounts and to remove remainingrestrictions in the others, actions need to betaken in various areas.

With regard to fiscal policy, Lebanon andTurkey need to take strong measures to addresstheir large fiscal deficits. Among the countrieswith substantial capital account restrictions,Morocco, Tunisia, and Syria will have to pursuefiscal reform.

Lebanon and Turkey need to review thecoherence of their monetary policies with otherpolicy instruments. These countries' fiscalproblems are reflected, in part, in the substan-tial positive differentials between their domes-tic deposit rates and the Eurodollar interest ratein London. For the countries with significantcontrols, the differential is generally smaller,given the constraints on capital flows; it isnotable that a high differential exists in Algeria,probably reflecting a determined effort to keepmonetary policy tight.

Turning to exchange rate policy, Egypt andLebanon, which have pegged their currenciesto the U.S. dollar in recent years, have wit-nessed large increases in their real effectiveexchange rates. While this indicator does notnecessarily suggest that the countries' curren-cies are overvalued, the steady upward trendunderscores the importance of carefully moni-toring exchange rate policy. The real effectiveexchange rates of countries with significantrestrictions have been stable. However, if thesecountries were to open their capital accounts, they wouldhave to review their exchange rate systems carefullyv

With regard to the financial sector, Syria is the only coun-try that has yet to adopt indirect monetary instruments. Asregards the soundness of banking systems in the region, thelack of data for several countries is a major weakness. Greatertransparency is essential to maintaining orderly capitalaccounts in countries with substantial capital convertibilityand to enabling the countries with significant restrictions tomove to capital convertibility in an orderly manner. Thecountries for which some data are available—Lebanon,Tunisia, and especially Syria—need to raise the capital-assetratios of their banking systems.

Turning to market orientation, Syria has the widest array ofcontrols to dismantle, followed, at a considerable distance, byAlgeria. Most of the countries still have much work ahead ofthem in privatizing and reforming their public enterprises—Algeria and Syria, in particular, and, to a lesser extent, Egypt,Morocco, and Tunisia. Labor markets still need to be liberal-ized in all the Southern Mediterranean countries exceptJordan and Lebanon. Finally, all the countries except Israel

Saleh M. Nsouli is DeputyDirector of the IMFInstitute.

Mounir Racked is aSenior Economist in theMiddle Eastern Divisionof the IMF Institute.

need to reduce their import tariffs signifi-cantly, and Syria needs to eliminate quantita-tive import restrictions.

Speed and sequencingTwo basic questions arise for countries that stillhave restrictions: (1) how fast should the capi-tal account be liberalized, and (2) in whatorder should the necessary steps be taken?

To achieve an orderly move to capitalaccount convertibility, each country has tofind, subject to various financial and structuralconstraints, the adjustment path that will max-imize its welfare over time. Syria's financialconstraints suggest that it would need to movefaster than Algeria, Morocco, and Tunisia toestablish capital account convertibility. Indeed,the latter three countries are servicing theirexternal debt in an orderly manner and haveadequate reserves; although Syria also has ade-quate reserves, it has experienced difficultiesservicing its external debt and has considerableexternal arrears. Morocco and Tunisia have thefewest structural constraints, Algeria has many,and Syria has the most. Thus, looking solely atstructural constraints, the move to full convert-ibility could be fastest in Morocco and Tunisia,followed by Algeria and, at a considerable dis-tance, by Syria. Given its financial constraints,however, Syria would need to launch a signifi-cant package of structural reforms immediatelywhile implementing supportive macroeco-nomic adjustment policies.

The appropriate sequencing of reforms willbe determined by each country's conditions. The followinggeneralizations can be made, subject to the caveat that an inte-grated and coherent package of policies should be in place.Syria would have to give priority to addressing the structuralconstraints on its incentives system, strengthening the bankingsector, and moving to indirect instruments of monetary con-trol, while putting in place a market-determined, unifiedexchange rate. Algeria would need to give priority to reform-ing its financial sector and privatizing its public enterprise sec-tor. The appropriate sequencing in Morocco and Tunisiawould involve strengthening and reforming their banking sys-tems and privatizing their public enterprises. For the SouthernMediterranean countries to be able to identify more specificmeasures that would enhance banking soundness and increasetransparency, they would all have to include the improvementof banking system data in their initial action packages. IMfl

This article is based on Saleh M. Nsouli and Mounir Racked, 1998,

"Capital Account Liberalization in the Southern Mediterranean Region,"

IMF Paper on Policy Analysis and Assessment 98/11 (Washington:

International Monetary Fund).



Finanin Emerging Markets

D o n o I d j . Mo t h i e s o n , Anthony Richards, and S u n / / S h a r m a

Since 1982,emergingmarkets havebeen rocked bythree majorfinancial crises.How can theymanage therisks associatedwith greaterintegration intothe interna-tional financialsystem?

S INCE the Thai baht first came underattack in July 1997, currencies andasset prices have plunged throughoutAsia, as capital has fled from countries

once favored by investors. The Asian crisis, likethe Latin American debt crisis of the 1980sand the Mexican crisis of 1994-95, has had abroad and devastating impact, not only on theeconomies of the affected countries but alsoon other developing countries believed to be"similarly situated." An examination of thesimilarities between the crises, as well as oftheir differences, sheds light on the Asiancountries' sudden fall from favor and suggestsactions that may enable them to weather suchstorms in the future.

SimilaritiesIn the months or years leading up to eachof the crises, capital inflows to emergingmarkets surged (see chart). Able to getfinancing in the international markets onincreasingly favorable terms, a number ofdeveloping countries built up massive sover-eign and private debt denominated in for-eign currencies—much' of it unhedged.

Between the first oil crisis of 1973 and theoutbreak of the debt crisis in 1982, netprivate capital flows to emerging marketsamounted to $165 billion, or about 1 percentof emerging markets' GDP over that period.For most of the 1970s, borrowers in emerg-ing markets were able to get syndicated inter-national loans at low—and even negative—real interest rates; these loans were denomi-nated in U.S. dollars and priced at spreadsover LIBOR (the London interbank offeredrate). Although the debts were hedged tosome degree by holdings of U.S. dollar-denominated reserves, fewer hedging instru-ments were available in the 1970s than today,leaving borrowers with large exposures tointerest rate and exchange rate movements.

Developing countries regained their accessto international financial markets in the

early 1990s. From 1990 to 1997, yield spreadson Brady bonds fell from an average of1,100 basis points over U.S. treasury bondswith comparable maturities to 350 points.Between 1991 and 1996, the average matu-rity on new Eurobond issues grew from4.4 years to 8 years. Net private capital flowsto emerging market countries soared to$1.04 trillion during 1990-96 (about 3 per-cent of their total GDP).

Despite the explosive growth of globalderivative products in the 1990s, unhedgedcurrency and interest rate exposures alsoplayed a central role in the Mexican and Asiancrises. Indeed, in some instances, govern-ments and private entities increased theirexchange rate exposures just before the crises.In 1994, the Mexican government shiftedfrom issuing peso-denominated debt (mainlyCetes) to issuing short-term debt securities(Tesobonos) with debt-service payments in-dexed to the U.S. dollar. The foreign exchangeexposure of nonfinancial corporations alsoplayed a key role in the Asian crisis. Domesticinterest rates in countries with a fixed orpegged exchange rate were higher than for-eign interest rates; as a result, many firmsfinanced their operations through securityissues and loans in foreign currency. Theyneglected to hedge these often large exposuresbecause domestic derivatives markets wereundeveloped and purchasing offshore hedg-ing products would have raised the cost ofborrowing abroad; moreover, their govern-ments had made a credible commitment to anexchange rate peg or a preannounced crawl.

Another common feature in all three criseswas the weak state of the financial systemsand regulatory regimes of the affected coun-tries. Both the controlled financial systems ofthe 1970s and the liberalized ones of the1990s had serious structural weaknesses. Inthe 1970s, many emerging markets main-tained tight constraints on external financialtransactions, directed credit allocation by


Crises


domestic institutions, and set ceilings on loan and depositinterest rates. Because bank operations tended to be limited toapproved or priority activities, there was little opportunity fordiversification, and a large share of banks' loan portfolios con-sisted of nonperforming loans. Moreover, banking controlsstifled the development of prudential supervisory systems.

In 1994-95, concerns about the health of Mexico's bank-ing system undermined the defense of the Mexican peso.From mid-1990 to mid-1992, 18 Mexican banks that hadbeen nationalized in 1982 were privatized. As interest rateswere freed, credit controls and lending restrictions removed,and compulsory liquidity ratios abolished, bank creditexpanded rapidly. By 1993, however, credit expansion hadslowed considerably because of concerns about the quality ofbanks' loan portfolios. After the peso was allowed to float, itsvalue dropped sharply and interest rates rose, contributing toa further deterioration of bank portfolios.

Although the Asian countries affected by the crisis of1997-98 had begun to make improvements in prudentialsupervision and regulation in the first half of the 1990s,imprudent lending continued, in part because of remaininginadequacies and the limited experience of financial institu-tions in pricing and managing risk. Private corporationsunderestimated the risk of domestic and foreign borrowing

and became highly leveraged. The weaknesses in banks' bal-ance sheets became apparent in 1996-97, when rising inter-est rates, depreciating currencies, collapsing real estate andequity prices, and the precarious situation of many corpora-tions led to a sharp deterioration in asset quality, provokingsome full-fledged banking crises.

All three crises took investors by surprise. Bank lendingincreased in 1981 to every country later obliged to restructureits debt, and bond and loan interest rate spreads were stable inthe first half of 1982. Similarly, Mexico's decision in December1994 to float the peso was unexpected, despite periods of tur-bulence for domestic interest rates, stock prices, and the peso-dollar exchange rate in the preceding 11 months. Most investorswere also surprised by the scope and intensity of the Asian cri-sis, in part because of the affected countries' strong record ofgrowth and stability and their cautious fiscal policies. Yieldspreads on bonds and syndicated loans declined for most Asianeconomies between 1995 and 1997, and no sovereign credit rat-ing was downgraded in 1996 or the first half of 1997. In themonths leading up to the outbreak of the crisis, Eurobondspreads for Indonesia, Malaysia, the Philippines, and Thailandfluctuated in relatively narrow ranges. Spreads did not spikeuntil the depth of the Korean predicament became known andspeculators attacked the Hong Kong dollar in October.

Surges and composition of private capital flows before crises

Source: International Monetary Fund, World Economic Outlook database.1 Aggregate flows to Argentina, Bolivia, Brazil, Chile, Colombia, Cote d'lvoire, Ecuador, Mexico, Morocco, Nigeria, Peru, the Philippines, Uruguay, and Venezuela.2 Aggregate flows to Indonesia, Korea, Malaysia, the Philippines, and Thailand.



Spillover effects were extensive in all three crises. In 1982,Mexico's debt-servicing difficulties soon spread to othercountries in Latin America as well as to countries in Asia andAfrica, as international bankers withdrew credits even fromcountries that had not demanded a rescheduling. Manycountries encountered liquidity problems; some decided tosuspend payments and renegotiate their credits. The subse-quent Mexican crisis of 1994-95 triggered turbulence in theforeign exchange and equity markets of the larger LatinAmerican countries, and Asian currencies and securitiesmarkets plummeted in January 1995, amid uncertaintyabout Mexico's ability to service its debt and the interna-tional community's willingness to provide a support pack-age. Similarly, the floating of the Thai baht in July 1997 led toa reassessment of prospects for other Asian countries.

Contagion was made more virulent by weak banking sys-tems. Banking crises broke out in nine heavily indebted LatinAmerican countries in 1982, when public and private enter-prises had difficulty servicing their debt. In 1994 and early1995, the volume of past-due loans held by Mexican banksincreased sharply, and bank deposits fell by 16 percent (morethan $7.5 billion) in Argentina. Following the depreciation ofThailand's baht in July 1997, pressure was put on other Asiancountries viewed by investors as having similar fundamen-tals, including overvalued currencies and banking systemsfacing potential problems with nonperforming loans. Avicious circle ensued. Believing that weak financial systemswould make it impossible for certain countries to sustainhigh interest rates, speculators pummeled their currencies.As the currencies plunged, the financial positions of bothnonfinancial corporations and banks deteriorated, and theproportion of nonperforming bank loans increased, arous-ing concern about the soundness of the banking systems andfurther undermining investor confidence.

Debt restructuring has been a key element in the resolu-tion of all three crises. In the 1980s, the focus was on restruc-turing sovereign foreign currency obligations (in many cases,governments had either assumed or guaranteed the domesticbanks' foreign currency debt). Debt reschedulings, alongwith falling international interest rates, ultimately helped toreduce the "debt overhang" that was discouraging investmentin debtor countries. Although Mexico fully serviced its offi-cial domestic and foreign-currency-denominated obligationsduring the 1994-95 crisis, there was extensive restructuringof the nonfinancial sector's domestic bank loans as well as itsexternal bank and Eurobond obligations. In Asia, the restruc-turing process is still in its initial stages.

DifferencesIt is difficult to argue, strictly on the basis of macroeconomicfactors, that the Asian economies in 1996 were poised for thekind of turmoil they have experienced. Although there weresigns (rapidly growing domestic credit, real exchange rateovervaluation, declining stock markets, and a growing volumeof bank claims on the private sector) that policy corrections

might be needed, they did not presage the depth of the crisisthat would eventually engulf the region. Moreover, the macro-economic situation of the Asian countries in 1996 was, by andlarge, better than Mexico's in 1994—and economic funda-mentals were stronger in Mexico in 1994 than they had beenin the highly indebted Latin American countries in 1981.

The global economic environment in which the crisesunfolded was also different. On the eve of the debt crisis ofthe 1980s, the industrial countries were headed for a reces-sion: GDP growth had slowed dramatically, from an averagerate of 4 percent in 1978 to about 1 percent in 1981, damp-ening world trade. Developing countries that were not oilexporters experienced a decline in export growth and a dete-rioration of their terms of trade. At the same time, theirdebt-service payments rose sharply when the industrialcountries raised interest rates in the late 1970s in an effort tocheck inflationary pressures.

In general, the world economy has been more favorable foremerging markets in the 1990s. Inflation and nominal inter-est rates in the mature markets have been low and falling.The declines in asset yields in mature markets have madeemerging markets more attractive to investors and risk pre-miums have decreased in many asset markets, either becauseinvestors have developed a greater tolerance for risk orbecause they believe risks are diminishing. World tradeexpanded more than 6 percent a year during 1990-96.However, an increase in U.S. interest rates led to a more pes-simistic assessment of Mexico's prospects in 1994, and theupswing in the value of the U.S. dollar before the Asian crisisundermined the competitiveness of Asian countries whosecurrencies were pegged to the U.S. dollar.

As private capital flows surged during the 1990s, the rela-tive importance of official capital flows to emerging marketsdeclined sharply, from 49.5 percent of total capital flows in1970-81 to 9.5 percent in 1990-96. There was also a dra-matic change in the composition of private flows—the shareof foreign direct investment and portfolio flows increasedrelative to bank lending.

Another difference between the crises has been in theireffect on development strategies. Before the debt crisis of the1980s, many countries had pursued an import-substitutionstrategy behind high tariff walls, supporting the strategy withpolicies that set low (relative to inflation) interest rate ceilingson bank loans and deposits and directed bank loans to prior-ity sectors. Extensive capital controls were in place. Externalborrowing was typically undertaken by the public sector tohelp finance budget deficits. Such repressive systems discour-aged exports both directly (through taxes or limits on creditavailability) and indirectly (to the extent that exporters had touse expensive domestically produced goods). The 1980s pro-vided ample evidence of the shortcomings of the closed-economy, import-substitution model, and, by the beginningof the 1990s, many emerging market economies hadembraced a more outward orientation that included liberal-ization of external trade and financial transactions, fiscal



conservatism, structural reforms designed toincrease the flexibility of domestic goods andfactor markets, and an expanded role for theprivate sector.

Neither the Mexican crisis of 1994-95 northe Asian crisis has as yet produced a compa-rable change in development strategies.Indeed, after the Mexican crisis, many LatinAmerican countries strengthened their com-mitment to maintaining an open economy—one reason Mexico was able to regain itsaccess to global financial markets in less thana year. The Asian crisis has led, however, toa reexamination of the Asian model of de-velopment; highlighted the importance of aresilient, transparent, and well-regulatedfinancial system as a prerequisite for full cap-ital account liberalization; and demonstratedthat developing countries need better institu-tions to protect the vulnerable segments ofsociety and forge a durable consensus forglobal integration.

Lessons of the Asian crisisAlthough policymakers in emerging marketscan take certain steps—such as reducingexpectations of bailouts and improvingtransparency in government decision mak-ing and the operation of the banking andcorporate sectors—to help investors makeinformed choices, volatile capital flows arenot peculiar to emerging markets, and it isunrealistic to think that they will ever becompletely eliminated. It is therefore neces-sary to put in place institutions and policiesto manage and reduce the risks associatedwith them. Although strong macroeconomicfundamentals are necessary, they are not suf-ficient for averting all crises: a resilient finan-cial sector is required for coping with abruptchanges in asset prices and capital flows.Countries need effective regulatory andsupervisory controls, so that financial insti-tutions have the ability and incentives toprice and manage the risks associated withcapital flows. Market discipline—making itcostly for managers and owners to neglectthe health of their institutions—is also nec-essary. However, transparency, which is criti-cal to market discipline, is increasinglydifficult to achieve in a world where off-balance-sheet exposures are becoming largerand where the mechanisms for collectingdata are not up to the task of tracking newtypes of exposures.

The financial sector cannot be strength-ened overnight; policymakers therefore needto open up their financial systems in anorderly fashion. They may need to considerimposing temporary restrictions on certaintypes of inflows—for example, prudentialcontrols that increase the cost of external debt(particularly short-term debt). Although suchcontrols may lose their effectiveness overtime, they do slow inflows and thus buy timefor rectifying structural weaknesses. Pru-dential regulations limiting the volume ofinflows that can be intermediated through thebanking system may also be appropriate.

The Asian crisis has made clear that a weakbanking system combined with an open capi-tal account is an accident waiting to happen.Reliance on cross-border interbank funding,which can be quickly withdrawn, is theAchilles' heel of the international financialsystem. It may be possible to prevent exces-sive reliance on such funding by basing capi-tal requirements for banks on their liabilitiesas well as on their assets, or by imposingreserve requirements on interbank liabilities.Changing the weights given to different typesof risk may also be a way to raise capitalrequirements.

In addition to reinforcing lessons learnedfrom the two earlier crises, the Asian crisishas highlighted some new issues. First, theremay be a need to coordinate financial regula-tion and exchange rate policy so that coun-tries attempting to peg their exchange ratesalso strengthen prudential and reportingrequirements for financial institutions andcorporations. Second, because it will taketime to improve the supervisory and regula-tory capacity of many emerging markets,nontraditional supervisory measures maywarrant consideration—for example, limit-ing the safety net to a narrower group ofdeposit-taking institutions, allowing greaterinternational involvement in the bankingsystem, and restricting foreign borrowing bybanks and nonbanks. Third, because someborrowers will inevitably fail, it is necessaryto have efficient bankruptcy procedures toensure rapid resolution of situations thatcould otherwise trigger a crisis. I gill

This article is based on Chapter 3 of International

Monetary Fund, 1998, International Capital Markets:

Developments, Prospects, and Key Policy Issues

(Washington).

Donald ]. Mathieson isChief of the EmergingMarkets Studies Divisionof the IMF's ResearchDepartment.

Anthony Richards isan Economist in theEmerging Markets StudiesDivision of the IMF'sResearch Department.

Sunil Sharma, a SeniorEconomist in the IMFInstitute, was with theEmerging Markets StudiesDivision of the IMF'sResearch Departmentwhen this article waswritten.



Many countrieshave experiencedfinancial sectordistress at sometime. Althoughbank failures cancome as a surprise,information isoften availablethat can signal abanking system'svulnerability tocrisis.

Daniel C. Hardy

S TRAINS and disturbances anywherein an economy are likely to haverepercussions on the banking system.Because of the nature of the busi-

ness, banks are exposed to many potentialsources of danger: reliance on deposits manytimes larger than their capital, uncertainclaims on different sectors of the economy,assets that are longer term and less liquidthan liabilities, and, for banks involved ininternational transactions, assets and liabili-ties denominated in different currencies. Tothese variables should be added the possibil-ity that problems will originate within thebanking system itself, perhaps because of laxinternal controls or poor management.When a bank is facing possible bankruptcy,its owners and managers may take greaterrisks if they expect to avoid being heldaccountable. As a result, problems that canordinarily be contained may be magnified,and sooner or later the banking system willrun into difficulties. Furthermore, complexrelationships and mutual dependency typi-cally develop between banks and theirclients, as well as among banks, so that diffi-culties that are initially localized can spreadthroughout the banking sector and into theeconomy as a whole.

Ultimately, the institutional and structuralfeatures of an economy and, in particular, itsbanking sector will determine its susceptibil-ity to crisis. The government—including thecentral bank and the regulatory and supervi-sory authorities—plays an important rolein establishing the legal and institutionalframework. It must implement adequateprudential supervision and regulation, whichincludes requiring banks to address prob-lems as soon as they emerge. The govern-ment is also responsible for ensuring thataccounting and auditing practices meet cer-tain standards so that banks cannot maskproblems, such as a high proportion of non-performing loans, until they become unman-ageable. Accounting standards in enterprisesand nonbank financial institutions also needto be rigorous enough to ensure that theircreditworthiness can be assessed. Laxity inthese areas increases the likelihood of wide-spread banking sector distress.

Although these structural factors maymake it easier to identify which countries aremore likely to experience a banking crisiseventually, they provide little indication ofwhen one might occur. Under favorable cir-cumstances, a country with poor institu-tional arrangements can coast for a long time


BPMR . .

gfen|jffil^ uS H

BSiilH


without serious banking sector difficulties, but when theenvironment deteriorates, a crisis can emerge very rapidly.

History of banking crisesGiven these myriad vulnerabilities, it is not surprising thatbanking crises have a long history. The Great Depressionof the 1930s was exacerbated by bank failures in the UnitedStates and elsewhere. In recent decades, a large number ofcountries have experienced financial distress of varyingdegrees of severity, and some have suffered repeated boutsof distress.

In the early 1980s, the governments of several LatinAmerican countries, including Chile and Mexico, felt com-pelled to make up for losses in the banking system—forinstance, by buying substandard loans from the banks formore than their true worth—to preserve its solvency. Duringthe 1980s and 1990s, many African countries also had torestructure and recapitalize their banking systems, which inthe past had suffered large losses on loans to parastatal com-panies (companies at least 50 percent owned by the state)and on crop loans. In the late 1980s, the performance ofbanks in certain advanced industrial countries, particularlyin the Nordic countries, deteriorated to the point wheregovernments had to support some of the largest banks topreserve financial stability. In almost all the transitioncountries—those that transformed their economies from acommand system to a market-based system—major banksincurred large losses as a result of high and fluctuating infla-tion and the loss of traditional enterprise clients. Currentevents in East Asia have reminded the world once again ofhow rapidly and forcefully banking crises can erupt and ofhow difficult it is to anticipate the full ramifications of thesedramatic events. In all cases, banking crises resulted, at aminimum, in large losses of wealth and disruptions in thesupply of credit for investment and commerce. Resolving thecrises often involved large outlays of public funds.

These grave consequences underscore the value of predict-ing banking crises or, at least, identifying them rapidly andanalyzing events as they occur so as to be prepared.

Leading indicators of distressEven when a banking crisis appears to strike like a bolt fromthe blue, it will normally have had a long gestation period.Information is usually available that, when carefully sifted,can give a fair indication of the vulnerability of the bankingsystem to crisis. To be able to predict the timing of a bankingcrisis or to say, at least, when the risk of a crisis is high or low,one must identify reliable indicators and monitor them forchanges. A useful indicator is a variable that can be readilytracked and that behaves one way when the banking systemis not under pressure and a distinctly different way when thesystem is subjected to pressure. Ideally, the behavior of anindicator should provide a measure of the degree of risk orshould hint at the likely timing of the emergence of problemsin the banking system. Once bank distress is apparent,

developments in contemporaneous indicators can say some-thing about the severity and ramifications of the problem.

Variables that are useful as indicators do not necessarilystand in any simple causal relationship to banking crises.One indicator might measure a certain aspect of bankingsystem distress, such as bank losses. Another might capturethe evolution of a shock to the economy caused, say, bychanges in the terms of trade, which will affect the profitabil-ity of exporting industries and the level of government rev-enues and expenditures and, indirectly, the banking system.A third might capture some consequence of banks' incipientdifficulties, say, a widening spread between deposit and lend-ing rates. Often, fluctuations in indicator variables and theemergence of problems in the banking sector are both theproduct of some third, underlying influence. For example,banks may start rolling over loans to a loss-making industryand capitalizing interest in the hope of an eventual turn-around or bailout by the government. A rapid increase inloans to one sector may indicate that banks are resorting tosuch measures, which in the end are likely to result in largeand explicit bank losses.

Useful indicators can come from various sources andrelate to various'aspects of the economy. Some may comefrom the banking system itself and some from other sectors,while others may be macroeconomic.

Banking sector indicators. The most obvious indicatorsthat can be used to predict banking crises are those thatrelate directly to the soundness of the banking system. Itemsfrom banks' balance sheets or statements of revenue andexpenses may make clear when risks are increasing and, thus,when problems are emerging. These variables may even beavailable at the level of individual banks, where systemwidedistress often originates; the deterioration of individualinstitutions may not be apparent in aggregate data.

The primary direct indicator of banking sector soundnessand the likelihood of difficulties is the level of bank capitaliza-tion, that is, the amount by which a bank's assets exceed itsliabilities. Capital acts as a cushion against shocks and allowsa bank to continue honoring claims even when the value ofsome of its assets drops. The amount of capital that a bankshould hold depends primarily on the riskiness of its assets.Certain assets, such as loans to enterprises, are inherentlymore likely to become impaired than, for example, cash andreserves held with the central bank. A bank will clearly need ahigher level of capitalization if it lends mostly to industriesthat are subject to large fluctuations in output and profitabil-ity caused by external events or if it operates in an environ-menl of high and variable inflalion.

Changes in banks' capitalization, especially as revealed intheir profitability, can be as telling as their level of capitaliza-tion. A rapid erosion of banks' capital as they absorb mount-ing losses is both a signal and a component of bankingsystem distress. Even if banks continue to make a profit, arapid increase in the share of loans that are nonperforming orimpaired is a clear danger signal. Deteriorating loan quality



has been at the core of most systemic banking crises. Thelevel of nonperforming loans is thus a key indicator of themagnitude of banks' difficulties, even if banks themselvestend to be overoptimistic in their assessment of repaymentprospects.

Shifts in the structure of banks' balance sheets can also beinformative and may provide an earlier warning than willdata on losses that have already occurred. In some instances,banking crises seem to have been preceded by a rapidbuildup of loans to particular sectors. The commercial realestate sector is especially prone to cycles of rising prices,overinvestment, and heavy borrowing, followed by a slump.This phenomenon contributed to the savings and loan deba-cle in the United States in the 1980s and to various episodesof banking sector difficulties in the United Kingdom—forinstance, in 1973-75 and in the early 1990s.

Another significant indicator may be a rapid change in thematurity structure of banks' assets and liabilities, especially ifit is combined with differences in the currency denomina-tions of assets and liabilities at each maturity. Increasingreliance on short-term funding of relatively long-term assetsmakes a bank more vulnerable to changing attitudes towardeither the banking system as a whole or that particular insti-tution. Such a widening maturity mismatch may also implythat the difficulties a bank has recently experienced will soonaffect its ability to meet the claims of its creditors, thusspreading the contagion more rapidly.

These microeconomic, often bank-specific indicators canbe of great value, but, unfortunately, are not always available.The data that are available may be of poor quality, eitherbecause the institutional arrangements are not in place toproduce reliable data even in the best of circumstances orbecause bankers and their borrowers have a strong incentiveto present a rosy picture of their situation (especially whenthat situation is deteriorating). Hence, outside observers willhave to rely more on macroeconomic, aggregated data" or onprices available from the market, such as exchange rates.

Macroeconomic indicators. A number of recent studieshave explored whether macroeconomic data—such as thosetypically published in a central bank bulletin—can be usedas leading or coincident indicators of banking difficulties.The results generally suggest that these variables are indeedworth watching closely. Although they are far from fully reli-able, they do indicate when trouble may be brewing, andthey are mostly widely and rapidly available.

The evidence shows that certain macroeconomic variablestypically display a distinctive pattern both in the lead-up toan episode of banking system distress and while the episodeis unfolding. Overall, the pattern is that of a rapid end to aboom: after rising rapidly, real GDP, consumption, and,especially, investment start to decline; an acceleration ininflation is suddenly reversed; credit from the banking sys-tem to the private sector builds up rapidly, peaks, and thencontracts; real interest rates increase steadily; and the realeffective exchange rate appreciates and then depreciates. In

the lead-up to a crisis, banks often rely increasingly onforeign borrowing, which then dries up.

At least as many countries have experienced serious butcontained distress in their banking sectors as have sufferedfull-blown crises that put their solvency in doubt. For exam-ple, when the banking systems in the Nordic countries cameunder strain in the early 1990s, the problems in Denmarkwere spread among many small banks, while, in Sweden, sev-eral major banks received substantial government assistance.A comparison of experiences in the past two decades sug-gests that the indicators that are most useful in foretelling afull-blown banking crisis are not of equal value in signalingan episode of less systematic and less profound banking sec-tor distress. Declining output, an increase and then adecrease in inflation, and a fall in the real effective exchangerate tend to accompany all banking sector difficulties, what-ever their degree of severity. However, it appears that bank-ing crises are often preceded by an unsustainable increase ininvestment funded in large measure by an inflow of foreigncapital through the banking system. A marked characteristicof the onset of crisis is the contraction in banks' foreign bor-rowing. In contrast, a boom in consumption and bank lend-ing, together with rising real interest rates, often precedes amore limited episode of banking sector distress, whichemerges when this boom comes to an end.

These patterns suggest that certain external developments—in particular, heavy reliance on foreign borrowing—canmagnify the effect of a negative shock to the system and con-tribute to the development of a crisis. The causation, though,can also go in the other direction: a very large banking sys-tem crisis may itself precipitate an external crisis.

Variations in country experiencesThe general "boom and bust" pattern described above doesnot fit every case. The current East Asian crisis, for example,differs in important ways from the crises that have hit mosttransition countries and may also differ significantly fromrecent episodes of banking distress in Europe, Latin America,and countries (many of them in Africa) that rely on theexport of primary products.

In recent episodes of banking crisis or distress in East Asia,both the real effective exchange rate and banks' foreign bor-rowing displayed an exceptionally sharp boom and bust cyclejust before and during the onset of a crisis. In these cases, thecontemporaneous indicators are very telling, and macroeco-nomic behavior in the crisis period itself is different fromthat in previous periods. It would have been difficult, how-ever, to predict these events significantly in advance on thebasis of macroeconomic variables alone.

The pattern in countries that rely heavily on exports ofprimary products is different. The onset of banking sectordifficulties in these countries is typically preceded by a wors-ening of the terms of trade—that is, by declining commodityprices—which not only affects the profitability of some ofthe largest industries but also reduces government revenue,



weakens domestic demand, and threatensthe profitability of all enterprises. In thesecountries, the role of the terms of trade asa leading indicator makes banking problemsrelatively predictable, whereas fluctuations ininflation and domestic demand seem to bemuch less relevant to judging the probability ofa banking crisis.

ConclusionFor some time, the IMF and the World Bankhave been paying increasing attention to vari-ables that may signal incipient banking systemdistress, studying the institutional structure ofcountries' financial systems more carefully tospot where weaknesses might lurk and to lookfor ways to correct them. They gather and ana-lyze quantitative data on countries' banking systems for signs ofdeteriorating soundness and examine macroeconomic vari-ables in new ways to assess the implications of their behaviorfor the likelihood of banking sector difficulties.

The IMF and the World Bank certainly will do more. Theywill scrutinize recent events in Asia to identify what other

Daniel C. Hardy is aSenior Economist in theIMF's Middle EasternDepartment.

warning signs should have been heeded. Inmany countries, they will examine disaggre-gated data more closely and pay more attentionto accounting and data dissemination stan-dards. Ultimately, though, no indicator, or setof indicators, is wholly reliable as an instru-ment of prediction. Some bank failures willcontinue to come as a surprise. There will alsobe periods when a banking crisis appears to beimminent but does not occur, because of eitherskillful action by the government or sheer luck.Still harder to predict is the exact timing of theonset of a crisis, when a simmering problemboils over. This very uncertainty, however, rein-forces the need for vigilance and the prepara-tion of contingency measures to deal decisivelywith banking sector problems as they emerge.

Attempting to predict possible banking crises is the startingpoint, not only for preempting them but also for resolvingthem with the least cost to society. \t-WThis article is based on Daniel C. Hardy and Ceyla Pazarbasioglu, 1998,

"Leading Indicators of Banking Crises: Was Asia Different?" IMF Working

Paper 98/91 (Washington: International Monetary Fund).

You never know whoyou might meet inThe Region magazine

Milton Friedman

Alan Greenspan

James Buchanan

Jacob Frenkel

Carter Glass

Kenneth Arrow

Joseph Stiglitz

Anna Schwartz

Adam Smith

Carter Golembe

Robert Lucas

Alice Rivlin

Paul Volcker

Subscribe to The RegionPublished quarterly by the Federal Reserve Bank of Minneapolis,at no charge to subscribers.Write to: Federal Reserve Bank of Minneapolis, Public Affairs-R,P.O. Box 291, Minneapolis, Minnesota 55480-0291



Monetary Policy in RussiaOver the last few years, Russia has succeeded in developing thetools to carry out an effective monetary policy in a marketeconomy. What monetary policy instruments has the centralbank used to achieve this objective, and what lessons_has itlearned in the process?

To m a s J . T. B a I i n c

RUSSIA became independent at theend of 1991, when the SovietUnion dissolved into 15 states.Thus, it had not only to adapt

institutions to new political realities but alsoto transform a centrally planned economyinto a market economy. In the Soviet era,monetary policy simply accommodated theplan's decisions on resource allocation andpricing. In the emerging market economy,however, monetary policy had to be gearedtoward attaining price stability while allow-ing the market to play the major role in allo-cating resources.

Changing the role of monetary policyrequired a major effort. The public and offi-cials at many levels of government had to beeducated about what that role should be in amarket economy. Moreover, the monetaryinstitutions (central bank, commercial banks,and monetary instruments) had to be estab-lished or adapted at the same time that theeconomy had to be stabilized and had toadjust to market realities and the collapse ofthe Soviet state.

The Soviet Union's monetary policyMonetary policy had two main roles in theSoviet Union: ensuring the fulfillment ofthe economic plan and controlling house-holds' purchasing power. The economic plandefined how much of each good had to beproduced and set its price. It served as thebasis for the credit plan, which assigned ear-marked credits to each producer. The creditplan flows thus served as a tool to monitorthe economic plan's execution. Lending rateswere administratively fixed, and investmentfunds were allocated by the branch min-istries. Enterprises paid each other usingbank transfers and could use cash only topay wages and salaries. Their deposit bal-ances could be used only for the purposesspecified in the credit plan.

By controlling households' purchasingpower, the authorities sought to avoidqueues and shortages. Because cash was theonly form of payment outside the plan's con-trol, monetary policy focused on targetingthe amount of cash in circulation. A cashplan established how much currency the



Gosbank—the controlling institution in the Soviet Union'sbanking system—would issue and the sums to be allocatedto enterprises to pay wages and salaries. Households couldhold liquid funds either in cash or in savings deposits.Interest rates were low and were rarely changed. The govern-ment issued bonds sporadically, and the amounts placedwere small, owing to their low and uncertain yields and fre-quent suspension of debt servicing.

The Gosbank issued currency, cleared interenterprise pay-ments, and formulated and executed both the credit and thecash plans. It also financed the budget deficit. Thus, it was ahybrid institution, carrying out functions that in marketeconomies are split between the central bank and commer-cial banks. In addition, the Soviet banking system comprisedthe Savings Bank, which mobilized household savings, andseveral specialized banks. The Gosbank tightly controlledspecialized banks, setting ceilings on their credits and pro-viding most of their funding. Soviet financial arrangementsalso included a foreign exchange plan. The ruble was non-convertible. The exchange rate was set administratively, and asystem of subsidies and taxes offset differences betweenexport prices and domestic prices.

Further reforms took place in 1990 and 1991. The Gos-bank remained as the central bank of the Soviet Union, towhich the central banks of the various republics, includingRussia, were made subordinate. However, political develop-ments in 1991 made Russia's central bank practically inde-pendent of the Gosbank; there were now two de factomonetary authorities in Russia.

Monetary policy in the ruble areaPoorly designed monetary arrangements following the col-lapse of the Soviet Union impeded an effective monetarypolicy. The Gosbank disappeared, and monetary policy func-tions were vested in the central banks of the countries in theruble area. Russia and the other former republics (except theBaltics) agreed to maintain the ruble as their common cur-rency. The Central Bank of Russia became the sole issuer ofcash, but all the central banks could grant credit. Those cred-its increased ruble deposits with the central banks in thearea, which could be used for both interregional and intrare-gional trade. Thus, the central banks, seeking to collectseigniorage and promote economic growth in their jurisdic-tions, had an incentive to expand credit, while the resultinginflation spilled over to the whole ruble area.

Because payments between ruble area countries wereautomatically settled, the Central Bank of Russia could notcontrol them. Thus, the rate of growth of the Central Bank ofRussia's credits to other central banks in the ruble area (as apercentage of base money) jumped from 11 percent in thefirst quarter of 1992 to almost 50 percent in the next quarterat a time when prices were being liberalized. Several attemptsto coordinate monetary policy among central banks in theruble area failed. To deal with this problem, the Russian cen-tral bank centralized all interstate transactions in Moscow

and decided to settle them only to the extent that each othercountry in the ruble area had funds in its bilateral accountwith Russia. If a country ran a deficit, it had to negotiate a"technical credit" to cover it. As a result, the rate of growth ofinterstate credits fell sharply between the second quarter of1992 and the second quarter of 1993.

Fiscal deficits created serious monetary problems, owingto both their magnitude and the lack of public debt instru-ments, which forced the central bank to finance thesedeficits. Erratic policies failed to curb them. In the first quar-ter of 1992, the government tightened up on expenditures,

"Poorly designed monetaryarrangements following thecollapse of the Soviet Unionimpeded an effective monetarypolicy."

lowering both the deficit and its domestic financing. Duringthe rest of the year, however, expenditures rebounded, and sodid credit to the government.

Directed credits (those earmarked for specific enterprisesor sectors) were another major source of monetary expan-sion. Credit planning ended in 1991 and, with it, the aggre-gate ceilings on bank credits. Moreover, during 1992, Russia'scentral bank gave priority to restarting economic growthover fighting inflation. In this regard, it organized a clearingof interenterprise arrears, financing the bulk of those thatinsolvent enterprises were unable to cover when the clearingwas finished.

Despite those difficulties, the Central Bank of Russiabegan developing monetary instruments—directed creditsand reserve requirements on ruble deposits. However, thesewere mostly ineffective initially, chiefly owing to the centralbank's passive monetary stance. Required reserves wereblocked for one month in non-interest-bearing accountswith the central bank. In January 1992, the reserve ratioswere set at 15 percent for deposits with maturities of lessthan one year and at 10 percent for all other deposits.However, design shortcomings severely undermined theeffectiveness of reserve requirements: an insignificantpenalty for shortfalls, banks' ability to draw down theirreserves if they had a fall in deposits, and the rapid growth inforeign exchange deposits made the monetary impact ofchanges in reserve requirements unreliable.

During the ruble area period, exchange rates were flexible.The Central Bank of Russia began to intervene in theMoscow International Currency Exchange foreign exchangeauctions to smooth exchange rate fluctuations and allow theruble to depreciate gradually.



"Russia achieves monetaryindependenceIn July 1993, the problems of the ruble arealed Russia to introduce the Russian rubleand demonetize the pre-1993 rubles. Thismarked the end of the ruble area and thebeginning of Russia's full monetary inde-pendence. To make monetary policy effec-tive, however, Russia still had to reduce themonetary financing of the governmentdeficit, control refinancing, and developappropriate monetary instruments.

Financing the deficit. The size and vol-atility of the fiscal deficit underminedmonetary control. In 1993, expenditurecuts reduced the federal deficit from theequivalent of about 10 percent of GDP to6 percent. As a result, growth in centralbank credit to the government fell from almost 25 percent ofbase money in the last quarter of 1992 to less than 14 percentin the first quarter of 1993 and became negative the nextquarter. Higher expenditures and lower tax revenues in thesecond half of the year, however, increased the deficit andcentral bank credit to the government. The larger deficit in1994 required central bank financing equivalent to abouttwo times the stock of base money as of the end of 1993. Thecentral bank reacted, reducing interest subsidies and tighten-ing control over directed credits; nonetheless, its net domes-tic assets more than quadrupled during 1994.

The government more than halved the deficit in 1995.Moreover, the development of the market for governmentsecurities drastically reduced central bank financing of thebudget, to the equivalent of 91 percent of base money (as ofthe end of 1994). While lower tax revenues and higher inter-est expenditures raised the deficit during 1996, central bankcredit to the government expanded by the equivalent of lessthan 50 percent of base money. In 1997, weaknesses in rev-enues and in expenditure control raised the fiscal deficit tothe equivalent of 8'A percent of GDR

Control over refinancing. The Russian authorities tight-ened access and conditions regarding directed credits. As ofmid-1993, the interagency Commission on Credit Policyreceived the mandate to authorize all credit allocations of thecentral bank and to set and monitor quarterly credit ceilings.It moderated growth in directed credits, which were discon-tinued as of late 1994.

The Central Bank of Russia also made all its lending moreexpensive by making the interbank lending rate the basis forits lending rates. This encouraged banks to rely on the centralbank only as a lender of last resort, thus virtually eliminatingtheir overdrafts with the central bank.

Development of monetary instruments. The disappearanceof the ruble area, tighter control over the monetary financingof the fiscal deficit, and the elimination of directed credits

"Russia madestriking progress

in developinga range ofmonetary

instrumentsappropriate for amarket economy,

in the faceof severe

difficulties."

made it possible for the Central Bank ofRussia to implement monetary policywith indirect instruments and, in particu-lar, those that are market based. (For a dis-cussion of indirect monetary instruments,see Alexander, Balino, and Enoch, 1995.)The introduction of credit auctions inFebruary 1994 was a significant step inthat direction. Since the central bank set aminimum rate for the auctions thatexceeded interbank lending rates, how-ever, take-up was limited to banks thathad little or no access to interbank loans.Consequently, the credit auctions weregradually abandoned, and in early 1996the central bank introduced two Lombardfacilities (which provided short-termcredit, collateralized with government

securities, to banks): one in the form of an auction and theother in the form of a standing facility at a fixed (nonpenal)rate. These facilities had limited success. Demand at theLombard auctions was weak because of the high floor inter-est rate the central bank had set. The Lombard standing facil-ity failed to meet banks' emergency credit needs. Banks didnot know whether they needed an overnight credit until latein the day, but transferring the collateral required to borrowunder the facility took at least one day. In 1996, to deal withthis problem, the central bank allowed bank primary dealersto get uncollateralized overnight credit at a penal rate.

In 1995, the central bank added monetary instruments tomop up liquidity promptly at its initiative. It began to sellmore treasury bills than were necessary to meet the treasury'sfinancing needs. However, to have an instrument fully underits control, the central bank introduced deposit auctions:banks would offer to deposit funds with the central bank,each indicating the interest rate it wished to receive. Initially,though, the central bank set the rate in advance. This type ofauction met with little success and was discontinued in late1995. In 1996, the central bank changed the arrangements. Itnow monitors interbank market rates, and if they fall belowthe level it deems desirable, it offers overnight deposits toselected banks.

Although the market for treasury bills—issued by theministry of finance through the Central Bank of Russia—hasnot been fully used for monetary operations, it has enjoyedremarkable growth. An active secondary market in treasurybills developed over time. The central bank intervened in themarket, chiefly to avoid volatility in yields. In late 1996, itstarted repurchase operations (using treasury bills), which itsubsequently carried out twice a day.

In addition to developing market-based instruments, thecentral bank continued to use changes in reserve requirementsas a major monetary policy tool. It also improved the waythese requirements were computed. In particular, it extendedthem to foreign exchange deposits and defined the deposit



base as an average of daily balances. Sincerequired reserves have been unremunerated inRussia, extending requirements to foreign cur-rency deposits reduced the latter's attractiveness.Averaging reduces the scope for window-dress-ing operations. The central bank also got theauthority to transfer deposits from a bank's cor-respondent account, if necessary, to meet reserveshortfalls. Moreover, to be eligible to use 5 per-cent of their reserve holdings temporarily, bankshad to have fully complied with the reserverequirement over the preceding six months.These measures sharply improved compliancewith reserve requirements beginning in 1996.

Instrument coordination. Shortly after thedemise of the ruble area, the central bank intro-duced a monetary programming framework toset monetary policy goals, monitor execution,and coordinate its monetary instruments toensure attainment of those goals. Taking intoaccount the effects of various factors outside the direct con-trol of the central bank, the exercise has allowed the centralbank to project how much base money it has had to inject orabsorb to reach its goals.

The central bank has monitored the program's implemen-tation through a five-day liquidity management framework,introduced in 1995. It also reviews other information, suchas treasury-bill yields, interbank interest rates, and exchangerate movements. This monitoring allows the central bank todetermine the combination of instruments it needs to use toachieve the desired monetary stance.

Exchange rate policy. Since the ruble area was abandoned,the Central Bank of Russia has intervened regularly to pre-vent sharp fluctuations in the exchange rate. In 1993, largecapital inflows led to heavy intervention in order to prevent amajor nominal appreciation of the ruble. That interventionfueled monetary growth, however, and inflation accelerated.

In 1994, concerns about the expansionary fiscal policy ledthe public to shift its funds out of ruble-denominated assets.A foreign exchange crisis took place in October, leading to asharp depreciation of the ruble and a jump in inflation in thecorresponding quarter.

The 1994 crisis led to tighter fiscal and credit policies, andcentral bank purchases of foreign exchange became the mainsource of monetary growth in 1995. To stabilize exchangerate expectations, the central bank introduced an exchangerate band in July 1995. The band's shape was changed oneyear later to allow for a gradual depreciation of the ruble.The rate of exchange rate depreciation slowed steadily, from23.5 percent in 1995, to 16.5 percent in 1996, and to only6.7 percent in 1997.

Recent developmentsIn 1998, disappointing revenues and the absence of correc-tive measures led to fiscal imbalances that became difficult to

Tomas 1.T. Balino is Assistant

Director in the Monetary

Operations Division of the

IMF's Monetary and Exchange

Affairs Department.

finance in a context of international financialturmoil. In August, the authorities took drasticmeasures, which included a widening of theexchange rate band, a moratorium on repay-ment of foreign debt, a restructuring of domes-tic government debt, and a prohibition onforeigners investing in ruble-denominatedtreasury bills. While these measures are outsidethe scope of the issues discussed in this article,they have shown once again that fiscal disequi-libria can make it impossible to preservemacroeconomic stability, even if monetary pol-icy remains appropriately tight. In addition, thedisruption to Russia's market for governmentpaper will seriously jeopardize the future ofsuch paper, both for government financing andfor monetary policy.

ConclusionRussia made striking progress in developing a

range of monetary instruments appropriate for a marketeconomy, in the face of severe difficulties: the disintegration ofthe Soviet Union, the transformation from a centrally plannedeconomy to a market-oriented one, pressures to use creditpolicy as an engine of growth, and the ill-fated attempt tomaintain a poorly designed ruble area. In addition, inadequatelegal arrangements, lack of experience with the workings of amarket economy, and a payment system ill-suited to a marketeconomy further complicated the central bank's task.

Monetary instruments in Russia have evolved through aprocess of trial and error. Little could be achieved while themain objectives assigned to the central bank's credit policywere financing a large fiscal deficit and providing subsidizedcredits to certain sectors and enterprises. Bringing those ele-ments under control—an objective largely achieved after theforeign exchange crisis of late 1994—was a key element inallowing the central bank to move forward by developing amarket-oriented approach to monetary policy implementa-tion. This approach involved development of a set of indirectmonetary instruments: open market operations, a Lombardfacility, deposit auctions, and reserve requirements. There havebeen important institutional achievements. Unfortunately, thesharp deterioration in the macroeconomic environment andthe measures taken in August 1998 have resulted in majorsetbacks, illustrating once more the close interrelationshipbetween structural reforms and macroeconomic stability. 1MB

This article draws heavily on Tomas }. T. Balino, David Hoelscher, and

Jakob Harder, 1997, "Evolution oj Monetary Policy Instruments in Russia,"

IMF Working Paper 97/80 (Washington: International Monetary Fund).

Reference:

William E. Alexander, Tomas /. T. Balino, and Charles Enoch, 1 995, The

Adoption of Indirect Instruments of Monetary Policy, IMF Occasional

Paper 126 (Washington: International Monetary Fund).



Are Currency Boards aCure for All MonetaryProblems?Currency board arrangements may be coming back intofashion. What recent successes have countries had withcurrency boards and in what circumstances are they mostlikely to be effective?

Charles Enoch and Anne-Marie Guide

CURRENCY board arrangements,under which domestic currencycan be issued only to the extent,that it is fully covered by the cen-

tral bank's holdings of foreign exchange,were long generally dismissed as throwbacksto the colonial era. (See box for an explana-tion of what currency boards are and howthey work.) It was argued that such a rigid,rule-based arrangement was not well suited

to diversified economies in many of whichthe authorities had developed sophisticatedskills in monetary management. Instead,currency boards were seen as desirable and,indeed, workable only in very specialcircumstances, such as the small, openeconomies of city-states and small islands. InI960, 38 countries or territories were operat-ing under a currency board. By 1970, therewere 20 and, by the late 1980s, only 9.

The renewed interest in currency boardshas come in several waves, raising the num-ber of countries currently using them to 14(see table). Following the successful use ofa currency board to stabilize the economyin the aftermath of Argentina's hyperinfla-tion in 1991, additional attributes of cur-rency boards became evident as a resultof the successful efforts made by two transi-tion economies—Estonia and Lithuania—toachieve credibility quickly for their newlyestablished currencies. In 1997, a currencyboard was introduced to end the economicchaos in Bulgaria. In view of the favorableoutcome, and given that to date no currencyboard has had to be abandoned as a result ofa crisis, the discussion about potential candi-dates has recently been broadened further. Inearly 1998, there was serious discussion ofwhether a currency board would be anappropriate anchor to use in efforts to haltthe Indonesian currency crisis. Most recently,there have been calls to study the possibilityof stabilizing the Russian ruble under a cur-rency board.


Currency boards in operation

YearsCountry/region in operation Peg currency Special features

Antigua and 32 U.S. dollar Member of East Caribbean CentralBarbuda Bank (ECCB)

Argentina 6 U.S. dollar One-third of coverage can be in U.S.dollar-denominated government bonds

Bosnia and 1 deutsche markHerzegovina

Brunei Darussalam 30 Singapore dollarBulgaria 1 deutsche mark Excess coverage in banking department

to deal with banking sector weaknessesDjibouti 48 U.S. dollar Switched peg currency from French

franc to U.S. dollarDominica 32 U.S. dollar Member of ECCBEstonia 6 deutsche mark Excess coverage for domestic monetary

inverventionsGrenada 32 U.S. dollar Member of ECCBHong Kong SAR 14 U.S. dollarLithuania 4 U.S. dollar Central bank has the right to appreciate

the exchange rateSt. Kitts and Nevis 32 U.S. dollar Member of ECCBSt. Lucia 32 U.S. dollar Member of ECCBSt. Vincent and 32 U.S. dollar Member of ECCB

the Grenadines

Sources: Balifio and others (1997); and Ghosh. Guide, and Wolf (1998).


Currency boards and inflationUltimately, the relative merits of currency board arrange-ments and other forms of exchange rate pegs cannot beresolved by theory alone. Ghosh, Guide, and Wolf (1998) haveundertaken an empirical investigation, extending the existingliterature on inflation performance under fixed exchangerates. Given the empirically verified anti-inflationary capabil-ity of fixed exchange rate systems, it can be argued that insti-tuting a currency board arrangement makes sense only if theregime delivers even better inflation performance. By using adata set containing all IMF member countries over more thantwenty-five years, the study attempted to isolate the inherenteffects of a currency board arrangement regardless of themany country-specific challenges facing countries where sucharrangements are in operation—for example, hyperinflation(Argentina and Bulgaria), transition to a market economy(Bulgaria, Estonia, and Lithuania), volatile terms of trade(Eastern Caribbean Currency Board), post-conflict situations(Bosnia), or the presence of an international financial center(Hong Kong SAR).

Comparative statistics and more formal econometricanalysis confirm that, historically, currency board arrange-ments have done better than even other fixed exchange rateregimes. For example, the presence of a currency boardarrangement is found to lower annual inflation by about3.5 percentage points—the result of a "confidence effect" thatessentially arises from the faster growth of money demandmade possible by the greater institutional certainty asso-ciated with a currency board. In contrast to fears oftenraised by opponents of currency boards, Ghosh, Guide, and

Wolf (1998) did not find that existing currency boards hadany negative effects on growth.

Creating an operating environment

Even if economic arguments favor a currency board arrange-ment, its operational feasibility will depend on whether theattendant legal and institutional issues are effectivelyaddressed. Their importance should not be underestimated:although a currency board is a simple monetary arrange-ment, a range of important decisions must be made about itsspecific nature, including changes needed in the institutionalframework for financial management in the economy and,especially, in the legal environment in which central bankingis carried out. Unless these adjustments—which tend to bemore time-consuming than those involved in carrying outother exchange regime shifts and in many countries will haveto be resolved in full public view (for instance, in parliamen-tary debates)—are completed satisfactorily, a currency boardcannot be established in a way that will enable a country toachieve the necessary improvement in the credibility of itsmonetary policy.

The basic decisions that need to be made when a countryestablishes a currency board arrangement include choosing thepeg or anchor currency, setting the level of the peg, and deter-mining whether or not to include a "safety margin" for thefinancial sector. Changes are also required to the legal systemand the government's relations with the central bank.

• Obvious criteria to use in choosing an anchor include thestrength and international usability of a currency, which gen-erally rule out all but a handful of moneys. In fact, the 14

What is a currency board?

A currency board combines three elements: an exchange ratethat is fixed to an "anchor currency," automatic convertibility(that is, the right to exchange domestic currency at this fixedrate whenever desired), and a long-term commitment to thesystem, which is often set out directly in the central bank law.The main reason for countries to contemplate a currency boardis to pursue a visible anti-inflationary policy.

A currency board system can be credible only if the centralbank holds sufficient official foreign exchange reserves to atleast cover the entire narrow money supply. In this way, finan-cial markets and the public at large can be assured that everydomestic currency bill is backed by an equivalent amount offoreign currency in the official coffers. Demand for a "currencyboard currency" will therefore be higher than for currencieswithout a guarantee, because holders know that "rain or shine"their liquid money can be easily converted into a major foreigncurrency. In the event of a "testing of the system," a currencyboard's architects contend, automatic stabilizers will preventany major outflows of foreign currency. The mechanism worksthrough changes in money supply within the currency board

country—a contraction in the case of a flight into the anchorcurrency—which will lead to interest rate changes that, in turn,will induce investors to move funds. While this is essentially thesame mechanism that also operates under a fixed exchange rate,the exchange rate guarantee implied in the currency board rulesensures that the necessary interest rate changes and the atten-dant costs for the economy will be lower.

Economic credibility, low inflation, and lower interest ratesare the immediately obvious advantages of a currency board.But currency boards may prove limiting, especially for coun-tries that have weak banking systems or are prone to economicshocks. With a currency board in place, the central bank can nolonger be an unlimited lender of last resort to banks in financialtrouble. At most, it may make loans from an emergency fundthat is either set aside at the time the currency board is designedor, over time, funded from central bank profits. Another costcould be the national authorities' inability to use financial poli-cies, such as adjustments of domestic interest or exchange rates,to stimulate the economy; instead, under a currency board, eco-nomic adjustment will have to come by way of wage and priceadjustments, which can be both slower and more painful.



currency board arrangements currently in operation involvepegs to only three currencies: the U.S. dollar (10 countries),the deutsche mark (3 countries), and the Singapore dollar(1 country). In choosing from among this much narrowerfield, a country should carefully consider its current andprospective trade flows, as well as other economic links, withthe country issuing the currency to which a country's currencyis pegged. A country's choice can get more complicated if itseconomy is characterized by widespread "currency substitu-tion," where the currency used is not that of its major tradingpartner. In this case, or where the values of a country's tradewith two dominant currency blocs are roughly equal, a cur-rency basket is a theoretical option. In practice, however, allcountries that have introduced currency board arrangementsso far have opted for the simplicity of a single currency.

• Setting the exchange rate would appear straightforward,given that a currency board arrangement by definition has to

all, of the above-mentioned principles into the central banklaw. The law must define both the exchange rate and reserves,as well as specify the limited powers of the managing institu-tion under the system. It is sometimes argued that the rulesgoverning a currency board could be asymmetrical, permit-ting the central bank to appreciate the exchange rate butrequiring legal action before depreciation can be undertaken.For example, the rules governing Lithuania's currency boardcontain such provisions. The period needed to set up the nec-essary legal process will obviously differ across countries,depending on the availability of technically skilled lawyerswho can draw up a draft bill and the minimum parliamen-tary requirements for its enactment into law. In most coun-tries, the process will take time owing to parliamentarydiscussion about the merits of the proposed arrangement,which may itself require the relevant authorities to carry outan intensive information campaign.

"A sound legal basis is essential, because a currency board arrangementderives much of its credibility from the changes required in the centralbank law concerning exchange rate adjustments."

cover a monetary aggregate, usually the full amount ofreserve money but sometimes narrower definitions ofmoney. Yet the rate at which the central bank's availableinternational reserves cover the monetary aggregate in ques-tion varies depending on the exact definition of reservesused. Choosing the appropriate definition most likelyinvolves a trade-off: although a narrow definition of foreignreserves, such as "net reserves," might signal strong disciplineand possibly improve the credibility of the system, it mightalso require an up-front devaluation that would prove politi-cally and economically infeasible.

• In a "pure" currency board arrangement, the currencyboard has no margin to intervene as lender of last resort onbehalf of a bank in difficulties or to engage in open marketoperations. A country weighing the option of establishing acurrency board may, however, seek a "safety margin" of someexcess coverage, holding reserves of 100+x percent of themonetary base. In this case, interventions of up to x percent ofbase money would be possible without violating the currencyboard rules. While most operating arrangements do allow forsome form of limited intervention, the decision to include thisfeature should not be taken lightly. Room for maneuver in caseof unexpected difficulties is possible only at a more depreci-ated exchange rate than would have been necessary underother exchange arrangements. Intervention may also limit thetransparency—and, thereby, the credibility—of the system.

• A sound legal basis is essential, because a currency boardarrangement derives much of its credibility from the changesrequired in the central bank law concerning exchange rateadjustments. Countries seriously considering establishing acurrency board may therefore wish to incorporate some, or

• Finally, establishment of a currency board arrangementwill require the redefinition of the financial relationshipswithin the country's government. More often than not, theinitial inflationary impetus that is to be eliminated by movingto a currency board has been created through extensive centralbank financing of the government. Rules for a currency boardarrangement therefore need to prohibit new central bankloans to the government. What financial links there are to bebetween the central bank and the government, and how theseare to function—most important, how the central bank willhandle government deposits—will also need to be worked out.Although the central bank continues to handle governmentaccounts under some currency board arrangements, doing somay decrease the arrangement's transparency. Further diffi-culties may arise from the fact that government deposits arecallable at short notice, and consistency with currency boardarrangement rules can be achieved only if such accounts arefully covered by foreign reserve holdings. For these reasons,some economies with currency boards—most notably HongKong SAR—have moved all government accounts to commer-cial banks. Other economies with currency board arrange-ments, including some transition economies, have felt that thecommercial banking sector was not yet able to handle the gov-ernment's accounts and, hence, have opted to keep them withthe monetary authority. In this case, however, interest on theseaccounts can be paid only to the extent that the currencyboard has a flow income from its foreign reserve holdings thatexceeds its operating costs. In addition, transparency is likelyto be enhanced if the public debt management function, anauxiliary service provided by many central banks, is clearlyplaced outside the domain of the currency board, possibly by



creating an independent agency under theministry of finance.

Transition to a currency boardThe rules laid down in the new central banklaw will serve as guideposts for reorganizingthe central bank into a currency board. In anumber of countries that have recentlyadopted currency board arrangements, thishas involved setting up separate banking andissue departments, each with distinct func-tions and coming under the authority of dif-ferent deputy governors. Other countrieswith currency board arrangements, such asArgentina, have retained a unified structurefor the monetary authority. In either case, areorganization has to take place to allow easyidentification of the central bank's key activ-ities and to ensure that maintenance of therelevant currency cover ratios will be clearlyvisible. To that end, the currency boardarrangement will have to publish a well-defined set of statistics (including, forinstance, the balance sheet of the issuedepartment or statistics on selected assetsand liabilities included in that balance sheet)in a form, and according to a calendar, thatare consistent with the currency boardarrangement law.

Establishing a currency board arrange-ment will also generally involve reviewinghow the central bank will carry out its newcore functions, the most important of whichis reserve management. The added impor-tance of reserve management under a cur-rency board arrangement is obvious, giventhat the board's earnings from foreignexchange holdings will probably be its majorsource of income and because even a smallviolation of the cover requirement, whichcould arise from technical problems inreserve management, might cause serioustrouble for the arrangement.

Finally, conducting a review of the bank-ing sector and prudential standards anddeciding on the location of banking supervi-sion will generally also be necessary duringthe transition to a currency board arrange-ment. The review and, if required, a stream-lining of the banking sector are importantbecause of the elimination of, or reductionin, the central bank's ability to function as alender of last resort under such an arrange-ment. During this period, the authoritiesmay decide to transfer banking supervision,

which has often been carried out by the cen-tral bank, to an independent agency to avoidpossible circumvention of currency boardarrangement rules in case of banking sectordifficulties. If, for reasons of timing or orga-nization, banking supervision functions can-not be performed outside the central bank, ithas to demonstrate clearly that any support itprovides to banks in difficulty will not breachthe currency board arrangement rules.

Conclusion

Currency boards in many countries haveachieved impressive economic results, both inachieving lower inflation than other exchangerate regimes and in stabilizing expectationsafter prolonged hyperinflation. There havethus been calls for such arrangements to beestablished in a rather diverse group of othercountries, many of which are in crisis. Suchcalls should be viewed warily by nationalgovernments, for at least three reasons. First,the success stories largely reflect the experi-ences smaller countries have had with cur-rency boards, whose applicability to largercountries has yet to be fully demonstrated.Second, and equally important, the successfulestablishment of a currency board arrange-ment requires time for building consensus,as well as for careful planning and implemen-tation of important legal and institutionalchanges. Third, countries with one or severalweak banks may have to rehabilitate thembefore changing their monetary regimes.These prerequisites to establishing a currencyboard may, in many cases, be too involved andtake too much time to make it advisable for acountry to attempt to do so during a macro-economic crisis. IW]

Suggestions for further reading:

Tomds Balino, Charles Enoch, Alain Ize, Veerathi

Santiprabhob, and Peter Stella, 1997, Currency Board

Arrangements: Issues and Experiences, IMF

Occasional Paper 151 (Washington: International

Monetary Fund).

Charles Enoch and Anne-Marie Guide, 1997,

"Making a Currency Board Operational," IMF Paper

on Policy Analysis and Assessment 97/10 (Washington:

International Monetary Fund).

Atish R. Ghosh, Anne-Marie Guide, and Holger C.

Wolf, 1998, "Currency Boards: The Ultimate Fix?"

IMF Working Paper 98/8 (Washington: International

Monetary Fund).

Charles Enoch is anAssistant Director andChief of the BankingSupervision andRegulation Division of theIMF's Monetary andExchange AffairsDepartment.

Anne-Marie Guide is aSenior Economist in theMonetary and ExchangePolicy Review Division ofthe IMF's Monetary andExchange AffairsDepartment.



Guest Article

From randmolherlinessto GovernanceThe Evolution of IMF Conditionality

IMF conditionality has evolved steadily overthe years. It continues to evolve. Recentdevelopments have produced both a freshapproach and new challenges.

Harold James

ON FIRST sight, there is somethingunchanging, even eternal, aboutthe dilemmas presented by IMFconditionality. But that appear-

, ance is deceptive. During the 1990s, an olddebate took a new and highly political turn.There are two elements to the current debateabout conditionality, which comprises the pol-icy requirements that the IMF places on itsprogram lending. First, there is a long-standingand inherent difficulty in the implications oflending that is not strictly on market terms—that is, on loans made other than on the basisof trust or specific securities. That debateclearly affects not only the IMF but also lend-ing by any international financial institution.Second, a new element is provided by thedramatic political changes that followed thecollapse of communism and their coincidencewith a global communications revolution andgreatly increased financial interdependence(the bundle of issues popularly known as"globalization").

John Williamson's analysis in the early 1980sappears as apt today as it did then: "The tradi-tional criticisms, which initially stemmedprimarily from left-wing elements in borrow-ing countries, were that the IMF adopts adoctrinaire monetarist approach, that it isinsensitive to the individual situations of bor-rowing countries, that it imposes onerous con-ditions, that it is ideologically biased in favor offree markets and against socialism, and that itoverrides national sovereignty and perpetuatesdependency. Until recent years, the IMF didnot respond to such criticisms. Its aloofnessseems to have aggravated the critics ... andmisgivings spread even to the U.S. Congress."


OriginsBut such problems are not unique to the IMF's operations.They are inherent in any attempt to subject lending to a con-ditionality. The League of Nations programs for Hungaryand Austria in 1922 and 1923, for instance, raised exactly thesame issue, and the criticisms of them as excessively harshand intrusive on national sovereignty precisely prefigurelater debates. The external control imposed on politicallyfragile states emerging out of the postwar breakup of themultinational Hapsburg Empire was so extensive and toughthat it constituted a deterrent to embarking on similar pro-grams in other states. Instead, countries attempting to stabi-lize their currencies in the mid-1920s turned to the less"political" capital markets, with the result that, as a generalprinciple, the League's conditionality was counterproductive.A reaction against the experience of the League made someof the architects of the Bretton Woods system, particularlyJohn Maynard Keynes, desire a more automatic Fund. Butthe principle of conditionality—Keynes called it in a memo-rable phrase "being grandmotherly"—soon reasserted itselfin the lending of the new institution.

For the IMF, conditionality became an increasingly sensi-tive issue in the 1960s and, above all, in the 1970s for the fol-lowing reasons. First, because quotas were not raised in linewith the dramatic expansion of world trade (see chart onfacing page), higher levels of lending in relation to quotaswere required, with consequently increased conditionality.Second, the expansion of capital markets, which had beencompletely unanticipated at the time of the Bretton Woodsconference of 1944, offered an alternative source of capital.The result was that conditionality applied only to somedebtor countries, and the concept of countries "graduatingfrom" the IMF became increasingly popular. Here, however,the skittishness of markets soon produced some unpleasantsurprises. Before the outbreak of the 1982 debt crisis, manyfinance ministers and bankers had considerable confidencethat the IMF was irrelevant to all except the poorest coun-tries. Similar beliefs gripped the markets before the 1997outbreak of the Asian crisis. Third, conditionality becamemore complex in order to avoid unintended consequences inprograms. Previously, for instance, because of the pressureexerted by powerful political and civil service lobbies, fiscalconditions had often led to big cuts in government invest-ment but very little reduction in government consumption.As a result, economic prospects worsened. Programs there-fore began to specify elements in public spending—publicsector pay guidelines, investment levels, and the like. Such anexpansion of activities inevitably brought the IMF into thepolitical domain.

GuidelinesThese problems were only partially addressed in theguidelines on conditionality, which were approved by theIMF Executive Board in 1979. The performance criteria

specified in IMF programs should be as few as possible:Section 9 of the guidelines stated that they would be"normally confined to (i) macroeconomic variables, and(ii) those necessary to implement specific provisions of theArticles [the IMF's Articles of Agreement, or charter] or poli-cies adopted under them." Performance criteria would relateto other variables "only in exceptional cases when they areessential for the effectiveness of the member's programbecause of their macroeconomic impact." In practice, how-ever, the concept of a macroeconomic impact is both quitevague and quite inclusive.

The elaboration of additional ("exceptional") detailsintended to ensure that the macroeconomic criteria wouldbe observed inevitably drew the IMF into domestic politicaldebates. The shape of IMF programs emerged in discussionsbetween IMF officials and national civil servants and politi-cians. Without the latter's cooperation and involvement, acountry's program stood no chance of being realized, but

"The shape of IMF programsemerged in discussions betweenIMF officials and national civilservants and politicians."

officials often found it hard to convince their compatriots ofthis. On occasion, therefore, they used an argument aboutexternal pressure as a way to shelter programs from criticismor domestic political debate.

The extent to which such an exercise in shifting politicalresponsibility increases or decreases domestic political stabil-ity is contentious. Examples can be found for both sides ofthe argument—on the one hand, to show (hat dependenceon external discipline strengthens state authority, and, on theother, to demonstrate how it may erode politically responsi-ble behavior. In the post-Second World War reconstructionof Germany and Japan, which was a conceptual model forIMF activities during 1956-63 when Per Jacobsson wasManaging Director, there is no doubt that blaming the alliedpolicy and the military authorities for the economic difficul-ties that initially accompanied liberalization took the strainoff weak and vulnerable political structures and greatly facili-tated economic revival. Both the German currency reform of1948 and the Japanese financial reform of 1949 (the DodgePlan) were initially unpopular, and their substantial benefitsbecame apparent only after a considerable delay. In themeantime, it was helpful for economic reformers to havethe occupation authorities as a kind of benevolent dictatorimposing political stability. Weak governments like to be ableto reduce the domestic pressure applied by interest groupsand political parties by pointing to the need to respond to analternative pressure coming from the outside. In the course of



the 1960s, the IMF became accustomed tobeing used in this way as an external whip-ping boy or scapegoat. In the 1970s, evenquite large and powerful industrial coun-tries, such as the United Kingdom, saw thevalue of the IMF in this regard.

Changing role of IMFThis view was sustainable as long as the IMFremained—as it had been conceived atBretton Woods—an institution that would not interfere withnational sovereignty. C. David Finch, former Director of thethen Exchange and Trade Relations Department, expressedthis concept succinctly. The IMF, he said, "has not beenestablished to give guidance on social and political priorities,nor has its voting system been designed to give it the moralauthority to oversee priorities of a noneconomic nature. Itsfunctions have to be kept narrowly technical if it is to beeffective in the exercise of its role as a promoter of the adjust-ment process. For this purpose, the Fund has to accept thatthe authorities of a country are the sole judges of its socialand political priorities."

In the 1990s, this view of the IMF and its role changeddramatically. In large part, this was a consequence of reflec-tions on the collapse of communism and on the linksbetween political and economic reform. In the 1980s, manypolitical scientists believed that economic reform was moreeasily achieved by authoritarian regimes. The experience ofCentral Europe, in particular, completely reversed the gen-eral understanding of the link between economic liberaliza-tion and political democratization. In the new picture, only acountry whose government was sustained by a deep reserveof legitimacy would be able to bear the pains associated withadjustment.

This change had repercussions for the concept of condi-tionality. If there was less room for a benevolent authority inimposing economic reform, this would also mean question-ing the traditional role assigned to the IMF. Instead, the issueof "ownership" became central.

New consensusThe collapse of the centrally planned economies or (in thecase of China) their movement toward the market was thelast stage in creating a new consensus about economic policy,frequently but misleadingly referred to as the "Washingtonconsensus." The consequence has been an increasing homo-geneity of political outlook, as well as of the economic order.Indeed, one key insight is that the two are linked: that eco-nomic efficiency depends on a functioning civil society, onthe rule of law, and on respect for private property.

The post-cold war world has a quite different politics.There is no longer a lineup of East versus West, in which pro-Western regimes automatically obtain support, regardless oftheir levels of efficiency and competence and probity. Rather,the international community is adopting a much more

interventionist stance in which the logic thatassociates economic and political change istaken more seriously. The result has beenthe forcing of a much quicker pace of eco-nomic reform in some countries (for exam-ple, Egypt, which until the early 1990slargely resisted attempts to liberalize); thedisintegration of the political order inothers (the collapse and defeat of Mobutu'sZaire); and the descent into the status of

international pariah for others. The striking change in thisarea is that there is no longer an acceptance of domestic polit-ical inefficiency, corruption, or oppression.

Governance issuesThe most visible product of the new political environment isthe concern of the Bretton Woods institutions with "gover-nance." In August 1997, a new set of guidelines promulgatedby the IMF's Executive Board instructed the staff that, in pol-icy advice, the IMF "has assisted its member countries in cre-ating systems that limit the scope for ad hoc decisionmaking, for rent seeking, for undesirable preferential treat-ment of individuals or organizations." The IMF suggestedthat "it is legitimate to seek information about the politicalsituation in member countries as an essential element injudging the prospects for policy implementation." At thesame time, these guidelines also preserved the nonpoliticalvision of Bretton Woods, requiring the IMF's judgments notto be influenced "by the nature of the political regime of acountry." In particular, recognizing an obvious danger, theyspecify that "the IMF should not act on behalf of a membercountry in influencing another country's political orienta-tion or behavior."

The IMF's interest in governance was already reflected in anumber of very high profile decisions in 1996-97. Con-ditionality has come to the fore in each of four completelynew areas. First, military spending had never been a topic ofexplicit discussion by the IMF in the era of the cold war.Since 1993, however, it has been discussed in the IMF's WorldEconomic Outlook reports as a major problem of misalloca-tion of resources. In a number of cases, notably those ofPakistan and Romania, it became a central element in IMFdiscussions. Second, corruption is explicitly addressed: inAfrica, but also in Indonesia. Third, so also is democracyaddressed, although there is no reference to democracy inthe IMF's Articles of Agreement (unlike those of theEuropean Bank for Reconstruction and Development).Fourth, especially in response to the Asian crisis, a critiquedeveloped of a feature that had previously been regarded as alinchpin of Asia's economic success—the concept of "trust,"or of "strong informal networks"—and that was now rela-beled and condemned as "crony capitalism." This criticismwas linked to the attack on corruption, and "a stable andtransparent regulatory environment for private sector activ-ity" was laid out as the solution.

™ Finance & Development / December 1998

"The newapproach will

produce greaterglobal

prosperity andstability/'


There had been some consideration of humanrights issues in the past: in Poland, whose mem-bership application was held up in the 1980s afterthe imposition of martial law and the internmentof political dissenters; or, more discreetly andsubtly, in South Africa in the 1980s, whereapartheid was attacked as an inefficient laborpractice. But the scale of the discussion of politi-cal issues in the mid- and late 1990s is novel. Thegradual extension of the IMF into these areasis an immediate result of the new consensusabout economic practice and of a new worldpolitical order that it has helped to produce. Butit reflects something more profound—a realiza-tion increasingly shared throughout the worldthat the world economy, and world institutions,can be a better guarantee of rights and of pros-perity than some governments, which may becorrupt, rent-seeking, and militaristic. Economicreform and the removal of corrupt governmentsare preconditions both for the effective operationof markets and for greater social justice. Indeed,these two results, far from being contradictory assome critics imagine, are complementary.

Harold James is Professorof Modern History atPrinceton University andthe author of a number ofbooks, including TheInternational MonetarySystem Since BrettonWoods (New York:Oxford University Pressand InternationalMonetary Fund, 1996).

Fresh challengesThe new approach will produce greater global prosperity andstability. By helping to provide markets with better informa-tion, ensuring greater transparency, and limiting the irra-tional destructiveness of financial crises, the IMF canhelp markets operate more efficiently. But questions ariseconcerning the degree to which the IMF can be "even-handed" in its treatment of all its members. One of the mostfundamental issues is the political counterpart to the criti-cism expressed by Paul Volcker, former Chairman of the US.Federal Reserve System, of IMF economic programs: "Whenthe Fund consults with a poor and weak country, the countrygets in line. When it consults with a big and strong country,the Fund gets in line. When the big countries are in conflict,the Fund gets out of the line of fire." Addressing the issues ofmilitary expenditure, corruption, and undemocratic prac-tices is easier for international institutions in the cases ofsmall countries, or even politically isolated countries. But itis likely to be hard and controversial in large states withsubstantial military and economic potential—for instance,China or Russia. Discussion of such issues inevitably playsa major role in domestic politics. In Russia, this kind ofcriticism of international institutions is made by oppositionpoliticians such as Grigory Yavlinksy. They explain the prob-lems and failures of Russian reform programs by an unwill-ingness of the international community to go far enough inattacking corruption and in imposing reform from the out-side. In other cases, conditionality will be interpreted as ablatant attempt to impose Western values in the hopeof restraining or even crippling potential competitors (a

criticism frequently voiced, for example, byMahathir Mohamad, the Prime Minister ofMalaysia).

Second, there is the question of the IMF'sinstitutional capacity for implementation. Somerecent programs and statements also go intosuch issues of economic organization as thedismantling of cartels, the improvement ofaccounting practices, and banking supervision.On the one hand, it is easy to see the macroeco-nomic effects of the organizational or structuralflaws criticized by the IMF. On the other hand,correcting them takes the IMF into completelynew areas in which it has no previous experi-ence. It is clearly experienced in fiscal affairs andin advising on central bank policy, but not inwide-ranging reforms of the financial sector orin accountancy. The detailed reorganization ofcorporate balance sheets in order to ensuregreater transparency—which is incidentally alsoa problem in many industrial countries—is aless appropriate task for international institu-tions than for private sector consultants andaccountants. The gains, after all, will directlybenefit the companies undertaking the reforms.

Third, and most fundamentally, this process of adding newexpectations could create a dangerous momentum of its own.Part of the recent discussion in the U.S. Congress on an IMFquota increase involved the issue of whether to integrate envi-ronmental and labor standards into IMF programs. Many ofthe IMF's member countries rightly feel that economic reformprograms must be responsive to social and humanitarian con-cerns. But the amplitude of such an agenda may produce anexpectations trap. The more the IMF is seen to extend its man-date, the more it will be expected to undertake, and, inevitably,the greater the challenge it will face in trying to live up to thedemands. The IMF will need to resist institutional overstretch:to ensure that its mandate is limited, clearly defined, and sub-ject to realistic assessment of results. 1MB

References:

lohn Williamson, ed., 1983, IMF Conditionality (Washington: Institute

for International Economics).

C. David Finch, 1983, "Adjustment Policies and Conditionality," in

IMF Conditionality, ed. by John Williamson (Washington: Institute for

International Economics).

Michel Camdessus, 1998, "The IMF and Good Governance," address to

Transparency International, Paris, January 21.

Louis W. Pauly, 1996, "The League of Nations and the Foreshadowing

of the International Monetary fund," Essays in International Finance,

No. 201 (Princeton University: International Finance Section, December).

Paul A. Volcker and Toyoo Gyohten, 1992, Changing Fortunes: The

World's Money and the Threat to American Leadership (New York:

Times Books).



T

Roy Culpeper, AlbertBerry, and FrancesStewart (editors)

GlobalDevelopmentFifty Years afterBretton WoodsEssays in Honour ofGerald K. Helleiner

Macmillan Press Ltd.London, 1997, viii + 400 pp.£52.50 (cloth), £26.00(paper). St. Martin's Press,New York, 1997, viii +400 pp., $65.00 (cloth),$22.95 (paper).

HIS COLLECTION ofessays is the outcomeof a colloquium heldin Ottawa in June

1994 to honor the well-knownCanadian international de-velopment economist, GeraldHelleiner, and to commemoratethe fiftieth anniversary of thecreation of the World Bank andthe IMF. There is consistencybetween these two objectives—Professor Helleiner has devoteda great deal of intellectual effortto ensuring that the BrettonWoods institutions serve theirmember countries' aspirations.

The essays cover issues rang-ing from trade, aid, private capi-tal flows, and foreign directinvestment to reform of theBretton Woods twins. The bookdoes not pretend to provide adiversity of opinion. With oneor two exceptions, the authorsshare a distrust of the prevailingparadigm of neoclassical eco-nomics, believing for the most

part that market mechanisms are "too often inefficient" aswell as "not able to bring about socially desirable or equitableoutcomes." Moreover, they are critical of the Bank and theIMF, arguing persuasively for change. The authors—withone exception—believe, however, that these institutions canbe reformed. The exception, Keith Griffin, would go as faras replacing the World Bank and development assistanceas we know it with a system of international taxation andtransfers. The essays are well written, balanced, and some-times provocative, making this volume an easy and stimulat-ing read.

The book's unifying argument is that market-based, neo-classical approaches are inadequate for meeting the chal-lenges of development. The argument has several strands.Some emphasize market inadequacy, noting that marketmodels don't fully explain development. A stronger variant isthat market-based approaches do harm. Finally, there is theview that the focus on the markets misses the essential point.According to Roberto Frenkel, distinct market orientationsin Latin America prior to the debt crises suggest the role ofexternal factors in producing the crises. Benno Ndulu con-cludes that, domestically, it is the ownership of policyreform, improved governance, and development capacitythat will determine whether changes for the better will stick.

The critique of the Bretton Woods institutions is concen-trated in thought-provoking essays by Gustav Ranis andTony Killick. Ranis attempts a comprehensive assessment of

the World Bank's role, arguing that it has emerged as adominant, self-confident leader in lending, data collection,research, and policy advice. This has led to what Ranis calls"an essential lack of self-restraint" that has contributed to itsthree main weaknesses: a tendency to "fill every vacuum,"rather than relying on partners; the pressure to lend; and ahighly centralized decision-making structure.

Killick confines himself to the IMF's relations with low-income countries. His assessment is that IMF-supportedprograms appear to have a limited impact on key macroeco-nomic variables other than the real effective exchange rate,largely because of slippage in program implementation.Killick attributes this weak impact to the borrower's limitedsense of program ownership. Because governments typicallycome to the IMF when they are already in a crisis, "staff donot have the time to ensure that the government is fully onboard, just as the government will not have the time (evenwhen it has the desire) to undertake the consultations andpublic information necessary for consensus building." Whilethe programs may not be effective, Killick rightly observes,the IMF now has less difficulty persuading governments ofthe wisdom of following its advice, given the increasingacceptance of the "Washington consensus."

One view, shared by Ranis, is that the IMF should leavestabilization and adjustment programs in low-income coun-tries to the Bank. Killick rejects this view, arguing that thecurrent collusive oligopoly facing low-income countrieswould be preferable to a monopoly, particularly because theBank may suffer from some of the same problems and arro-gance that are attributed to the IMF. Moreover, if the Bank isto take selectivity seriously, it would be inappropriate toexpand its mandate.

A second view is that the IMF should design programsbased on a cost-minimization framework, assuming aminimum required growth rate, rather than letting thegrowth rate be a residual. The IMF now pays much moreattention to the impact of its programs on growth. But, hereagain, the advice is not easy to implement in a world ofdeclining aid flows and donors' reluctance to foot the billsimplicit in IMF programs.

Killick's analysis fails to focus on Bank-IMF relations. Inbest-practice situations, the two institutions complementeach other, with the Bank's longer-term development focusintegrated into the IMF's short-term stabilization programs.How to make this work consistently in favor of low-incomecountries is a challenge the two institutions must address.

The value of these writings in honor of Professor Helleineris in placing human development, participatory methods,and institution building at the core of development priori-ties. Unfortunately, the considerable intellectual strength ofthe authors is directed more at what is missing from market-oriented approaches than at confronting the need for a posi-tive, complementary agenda for action.

Vinod Thomas and Sarwar Lateef


DfV^|/CDwUrO


D ANI RODRIK'S book addresses impor-tant economic and social problemsbased on a careful review of the evi-dence, while offering some intriguing

and original contributions. Every reader will comeaway with some gems from this concise yet readablebook. I have four personal favorites.

First is Rodrik's excellent review of the literatureon whether globalization has contributed to wageinequality, particularly in the United States. This lit-erature is tricky. Trade economists acknowledge thattheir workhorse Heckscher-Ohlin-Vanek model ofcomparative advantage—according to which tradedepends on factor abundance—would predict awidening gap between the wages for skilled andunskilled workers in industrial countries as tradeexpands with poorer countries. Most economistsagree, however, that it is difficult to attribute a sig-nificant fraction of increasing wage inequality in theUnited States to expanded trade with developingcountries. Standing almost alone against this con-sensus, Professor Adrian Wood of the LondonSchool of Economics has made the case that previ-ous estimates of trade effects on inequality aremethodologically flawed and that these effects are,in fact, quite large. Rodrik provides a fair summaryof the evidence on all sides.

Second, Rodrik brings together the results of hisown original research on the link between tradeopenness and the extent of government spending. Heargues that individuals in more open economiesdemand—and get, when political institutions areresponsive—more insurance from their governmentsthan do individuals in relatively closed economies.This is an important point: if greater exposure toexternal economic shocks increases risks for individ-uals, and, given that private insurance for individuals'incomes typically does not exist, then increasedopenness can be expected to lead to greater govern-ment involvement. In fact, one might expect tradeeconomists to be leading the charge for social spend-ing to mitigate risk because explicit insurance forindividuals is vastly preferable to the indirect insur-ance that governments often provide by protectingdeclining firms and industries.

A third useful and original contribution isRodrik's discussion of the notion that, rather thanmerely shifting the demand for unskilled labor upor down, globalization might have increased theelasticity of the demand for labor, so that the quan-tity of labor demanded is more responsive to wagechanges. A change in the demand elasticity of laborhas interesting implications both for the incidenceof various labor regulations and for explaining thewidening income gaps between individuals.

Fourth, I like Rodrik's discussion of"fair" trade, introducing philosopherMichael Walzer's notion of "blockedexchanges"—societal restrictions on whatkinds of markets are allowed. I think that,in developing countries, most agitationfor trade restrictions based on "social"concerns—like workplace safety, childlabor, or rights to join unions—is basedon the self-interest of labor in the richercountries. Rodrik quotes a representativefrom the AFL-CIO (American Federationof Labor and Congress of IndustrialOrganizations) saying that "labor costsshould be removed from the equation" ofinternational competitiveness. Becausejob compensation is a package of moneywages and other elements—such as work-ing hours, unpleasantness, safety, andpersonal satisfaction—if workers cancompete by offering to work for lowerwages, it does not make much sense toargue that they cannot compete by accepting lesspleasant working conditions. Workers might actuallyprefer higher money wages and worse working con-ditions to the realistic alternative of better workingconditions and lower wages. Rodrik's discussion ofblocked exchanges is reminiscent of the legal doc-trine of "unconscionability" in limiting the freedomof contract and makes an interesting contribution tothe debate.

Rodrik makes a number of sensible proposals forboth economists and policymakers. To economists,he recommends more concern for social problems,more openness to the idea that some problemsmight be exacerbated by globalization, and moremodesty. Given the enormity of the social problemsconfronting us and how little we understandthe relationship between these problems andglobalization, following these recommendationsshould be easy.

In his concluding chapter, Rodrik recommendsthat governments "strike a balance between opennessand domestic needs," "do not neglect social insur-ance," "do not use competitiveness as an excuse fordomestic reform," and "do not abuse 'fairness' claimsin trade." All of these are sensible proposals, butRodrik, though optimistic, is no naif. He concludesby quoting Albert Hirschman to the effect that mak-ing progress happen requires "political entrepreneur-ship, imagination, patience here, impatience there,and other varieties of virtu and fortuna."

Lant Pritchett

Dani Rodrik

Has GlobalizationGone Too Far?Institute for InternationalEconomics, Washington, DC,1997, xi +128 pp., $20.95(cloth), $17.95 (paper).



BOOKS

M.G. Quibria andJ. Malcolm Dowling(editors)

Current Issues inEconomicDevelopmentAn Asian Perspective

Oxford University Press,New York, 1997, xix +432 pp., $65 (cloth).

CURRENT Issues in EconomicDevelopment: An Asian Per-spective is an interestingconference volume covering

diverse economic development issuespresented by leading experts in thefield. It contains many topics that wouldnot generally be found in a standarddevelopment economics textbook. Theissues range from governance, democ-racy, ethics, and corruption to tradeliberalization, agricultural development,the environment, and population.

While the volume has a broad andconceptual focus, selected chapters alsohighlight the successful experiences ofAsian countries (the Asian miracle),identifying their common patterns andany distinguishing characteristics thatmay have been largely responsible fortheir rapid growth during recent years.

Because the book was published in early 1997, itdoes not include the events associated with the 1997Asian crisis.

What factors are important for economic develop-ment, especially in Asia? A number of authors empha-size that adaptation of Western, market-basedeconomic systems may not be sufficient for rapiddevelopment. While Paul Streeten argues for govern-ment intervention, Basant Kapur takes a nonmaterial,

"The events of the past decadehave shown that the links betweenfinancial factors and developmentshould be further explored."

or philosophical, perspective on development.Similarly, Jean-Jacques Laffont focuses on the impor-tance of credible or "enabling" regulations to stimu-late private sector activity. Robert Barro, in anempirical chapter that covers 100 countries during1960-90, focuses only on economic growth vari-ables—such as per capita GDP growth—and finds lit-tle evidence of a link between democracy—proxied byan indicator of political rights—and growth. In con-trast to several other authors, Rodney Falvey and T.N.Srinivasan conclude that export orientation, trade lib-eralization, and openness are critical for growth.

While praising the systems of East Asia, Streetenstates: "It is now generally acknowledged that 'get-ting prices right' has not been the principal, andcertainly not a sufficient, recipe for the success ofEast Asian countries, although their government

interventions have been 'market-friendly,' and themarkets have been 'people-friendly'. " For transitioneconomies, the author compares the long-termeffects of reform to the short-term ones, gradualismto shock therapy, and flexible economic systems toinflexible ones. His policy prescriptions for goodgovernance include a long list of measures thatrange from social service provision to the mainte-nance of macroeconomic stability. However, hischapter does not address issues related to policytrade-offs and constraints or how such governmentspending will be financed.

In another interesting chapter, Kapur highlightsthe conflict between mainstream neoclassical eco-nomics and the role of ethics, culture, religion, andstrong family values. He suggests that includingthese variables is likely to alter the predictions ofmodels based exclusively on optimization behavior.The formidable ties between industry and govern-ment observed in Asian economies are stronglyinfluenced by religion and culture. In criticizing self-interested behavior and the "me-first" mentality,Kapur argues that this phenomenon may have beenresponsible for the decline in technological competi-tiveness in the United States. However, it is now clearthat the decline in U.S. competitiveness during the1980s was reversed in the 1990s, when numerouscompanies streamlined their operations and signifi-cantly enhanced their competitiveness worldwide.

Because economies pass through cycles and phasesthroughout their history, it is useful to analyze theirsuccesses and failures in terms of paradigms. Forinstance, the prevalence of culture and family bondsthat partly explain the strong ties between bankingand industry in most Asian economies—such asKorea and Thailand—may have contributed to easyaccess to credit and rapid economic growth duringthe 1980s and early 1990s, but excessive incentivesfor moral hazard eventually resulted in bankruptciesand closures of numerous financial institutions inthese countries. Moreover, given the recent experi-ence of Asian countries, it has become clear that it isnot that easy to maintain nontransparent and family-controlled financial systems in a world characterizedby massive capital flows.

The chapters are well written and address a newgeneration of issues in development literature. Theissues the book discusses, such as governance andways to deal with corruption, are increasingly beingincluded as part of policy recommendations. Theevents of the past decade have shown that the linksbetween financial factors and development shouldbe further explored. Overall, this volume will serveas a very valuable reference for policymakers.

Qaizar Hussain

SO Finance & Development / December 1998


W HEN the countries of the formerEastern bloc began the monumen-tal task of making private propertythe basis for productive relations,

the writing on the subject focused almost exclu-sively on policy issues or on expectations of firm-level adjustments. Several monographs and shortstudies have also attempted to quantify changes inenterprise behavior and performance after privati-zation. But less is known of the precise managerialdynamics that this adjustment entailed.

This anthology attempts to fill the gap. It adds toa growing but, as yet, thin literature on concreteexamples of management responses to privatiza-tion, covering a complex topic from diverse per-spectives. The "challenge" in the book's subtitle, onesoon realizes, is left in the singular for a reason: in ashort time, formerly centrally planned enterpriseswere forced to stop producing according to annualplans that dictated product lines, investments,wages, and prices and had to start manufacturingaccording to customer preferences, on the basis ofrelative prices. For an "embedded" corporate cul-ture, reared on indifference toward customers, qual-ity, and performance, this represented no less than afundamental shift in practically every facet of man-agerial behavior.

The 21 contributions to the volume are evenlydivided between conceptual papers and empiricalpapers based on survey research or case studies.Several of the conceptual papers offer some intrigu-ing, provocative analyses of managerial changes notlikely to be examined in modern business schools.Two examples are a semiotic perspective on changesin business language and symbolism by DeviJankowicz; and a cultural analysis of the socialembeddedness of capitalism by Michael Mauws andNelson Phillips that, predictably, concludes thatcapitalism represents a peculiar constellation of"historically contingent, culturally specific" insti-tutions that cannot be replicated in the post-communist countries of Eastern Europe and theformer Soviet Union.

Other conceptual papers are more narrowlyfocused on specific questions, such as the rolethat mergers and acquisitions have played in pri-vatizations in post-communist, developing, andindustrializing countries (Julio de Castro andNikolaus Uhlenbeck); whether a suitable system ofpatent protection can be implemented in Easternand Central Europe (Scott Erickson); whetherstrategic partnerships between Western and EasternEuropean firms can be mutually beneficial (FritzKroger); and whether consumption practicesof Western and Eastern European consumers

will converge following the transition(Christoph Melchers).

More compelling are the empirical con-tributions, which are based on surveyresearch, case studies of individual firms,or a combination of both, coveringBulgaria, the Czech Republic, Hungary,Poland, Romania, and Russia. The casestudies offer some fascinating insights intothe problems of standardizing accountingpractices (David Lutz and James Davis),the evolution of the "idea" of entre-preneurship from one associated withcriminal activity to one necessary for eco-nomic success (Robert Lynch and ValeriMakoukha), the positive impact of reli-gious beliefs on entrepreneurship (PatrickMarx), and changes in banking practices(Patrick Arens and Keith Brouthers).

Some arguments regarding managerial changesin this lengthy volume tend to overuse such con-cepts as "social habit" and "attitude," which are notdefined independently of a concrete result. The bestof these papers, however, avoid these more indeter-minate propositions and instead attempt to explainhow different managerial practices—in differentcountries or at different times—resulted in varyingorganizational performances and how differentcompanies' approaches to reinventing themselves ina market environment evolved. Frank Hefner andDouglas Woodward outline the circumstances inwhich foreign investment can transfer and spreadentrepreneurial skills to local partners. In a chapterthat should be particularly useful to studentsofv strategic management, case histories of fourHungarian enterprises suggest that those firms thatrelied on "old-boy" networks for marketing andsales in the initial transition period may have beenmore successful at reorienting themselves. A relatedpoint—that prereform organizational endowmentsmatter—is effectively made by Karen Newman andStanley Nollen, who examine managerial responsesto privatization in a Czech engineering firm.

In short, Ullmann and Lewis provide an unusualcombination of articles based on a variety ofapproaches to analyzing the managerial imperativesof transition to market. The volume's conclusion—that only a small portion of firms have effectively"recreated" themselves during the transformation—lends support to a standard claim in economicsociology—namely, that firms exist in a socializedweb of interpersonal relations. To succeed, firm-level changes must therefore be preceded by changesin these networks.

Raj M. Desai

Arieh A. Ullmann andAlfred Lewis (editors)

Privatization andEntrepreneurshipThe Managerial Challenge InCentral and Eastern Europe

International Business Press,New York, 1997, xix +378 pp., $49.95 (cloth).



BOOKS

T

Maurice Obstfeld andKenneth Rogoff

Foundations ofInternationalMacroeconomicsMIT Press, Cambridge,Massachusetts, 1996, xxiii +804 pp., $59.95 (cloth).

HIS book is a tour de force. Asrevealed by its title, it providesthe foundations for a system-atic, integrated treatment of

the main issues in international macro-economics. But the book has much moreto offer than analytic rigor. The authorsadopt a refreshingly chatty style and sup-ply a lot of the detail that is missing inmost academic journal articles. Along theway, they give numerous real-world exam-ples to illustrate their theories and developa number of useful diagrams and analyti-cal devices. The book is designed to beused as a second-year graduate textbook,for which it is well suited.

The organizing principle of the bookis that an open economy permits borrow-ing from, and lending to, the rest of theworld. Such transactions are necessarilyintertemporal—that is, they involve trad-

ing resources today for resources tomorrow. Thefocus is therefore on the determinants of the currentaccount, which reflects the difference between for-ward-looking saving and investment decisions. Thisexchange, which is both international and intertem-poral, allows consumers to smooth out their con-sumption over time; unbalanced current accountpositions may therefore serve a useful function andare not necessarily problematic. The intertemporalapproach to the current account was developed inthe 1980s and early 1990s, with contributions byObstfeld and Rogoff.

Obstfeld and Rogoff are largely successful in devel-oping a rigorous, integrated treatment of many top-ics, ranging from familiar exchange rate issues—suchas speculative bubbles, target zones, and speculativeattacks—to more specialized areas—such as thehome bias (the extremely large share of equity wealthof residents held at home) and equity premium (thelarge difference between the rate of return on equitiescompared with government bonds) puzzles. Theyalso analyze the impact of government deficits on thecurrent account and provide a comprehensive treat-ment of Ricardian equivalence—the idea that anincrease in the public debt will have little or no effecton real output and employment because taxpayerswill save more in anticipation of future higher taxesto pay the higher interest expense on the debt.Surprisingly, however, their discussion of the U.S. fis-cal expansion in the early 1980s and German unifica-tion in the early 1990s is not really informed by theintertemporal approach.

It is also odd that they do not cover the Dutch dis-ease (when a boom in a country's natural resource

sector produces a decline in other industries), a topicthat lends itself to the analytic framework. But theydo provide a masterful exposition of imperfectionsin international capital markets, covering a widerange of issues related to the debt crisis, such as sov-ereign risk and moral hazard. It will be interesting tosee how they apply the intertemporal approach tothe Asian crisis in a second edition of their book.

What sets this book apart, however, is the fullyintegrated sticky-price model based on dynamicoptimizing behavior that the authors offer as analternative to the Mundell-Fleming model (a modelthat extends the IS-LM framework to an open econ-omy, but ignores supply-side and intertemporal con-siderations)—the traditional workhorse for policyanalysis. The authors point out the obvious lack ofmicrofoundations of this standard approach,although they provide a lucid exposition of theMundell-Fleming-Dornbusch overshooting model—an extension of the Mundell-Fleming model—recognizing that it "remains so influential as to war-rant discussion in any serious treatment ofinternational monetary theory." They assert thatperhaps the Mundell-Fleming model's most contro-versial implication is that an unanticipated mone-tary contraction leads to a temporary decline inoutput. This assertion is surprising, given that therecent literature on monetary policy shocks appearsto support this aspect of the model.

Some of the results of the sticky-price model aresimilar to those from the Mundell-Fleming approach.For example, a surprise increase in the money stockcauses a country's currency to depreciate, temporarilyraising domestic income so that the country runs acurrent account surplus via the usual intertemporalconsumption-smoothing channel. With nontradcdgoods added, there is even overshooting, as in theDornbusch extension. However, the results relating togovernment spending appear anomalous: a perma-nent rise in world government spending leads to a fallin the short-term real interest rate, and a temporaryrise has no impact on the real rate. These results maymake sense within the specifications of the authors'model, but it is not clear how helpful they are for pol-icy analysis.

In summary, the authors have much to say aboutreal-world economic developments that is illumi-nating and sensible. It will take some time, however,before the new intertemporal paradigm replaces theexisting Mundell-Fleming-Dornbusch model. Thisprobably reflects an intertemporal adjustment onthe part of economists who view the world throughthe prism of the old approach.

Peter Clark



Richard H.R. Harper

Inside the IMFAn Ethnography of Documents, Technology, and Organizational Action

Academic Press, San Diego, California, 1998, xii + 303 pp., $75 (cloth).

T HIS book draws on extensive research during whichthe author had access to IMF staff at all levels. Itdescribes the particular nature of the day-to-day

demands placed on the institution and how work is orga-nized and carried out to meet its various responsibilities. Onthis basis, the author develops important insights into theways in which information technologies could be used toimprove productivity while analyzing the constraints thatthe nature of the staff's work (and its working culture) placeson the use of these new technologies. For the outside readerinterested in how organizational cultures and technologywork together, the book highlights how much organizationalinformation is needed if new technologies are to be appliedin an effective way.

This book is not a typical treatment of the IMF. Readersmay assume from its title that the book deals with howeconomic policies and issues are dealt with in the IMF.Rather, the book is about a completely different subject—organizational behavior—and, as a case study of an interna-tional bureaucracy and how it deals with complex issuesunder pressure, the book will be of particular interest tostudents of organizational behavior.

Brian C. Stuart

Ian G. Heggie and Piers Vickers

Commercial Management andFinancing of RoadsWorld Bank, Washington, DC, 1998, viii + 158 pp., $20 (paper).

T HE purpose of this book is to assist governmentsto commercialize their roads, that is, to "bring roadsinto the marketplace, put them on a fee-for-service

basis, and manage them like a business." Although writtenfor officials in developing countries, the lessons of this bookcan be applied everywhere, because roads in all countries arestill run as if they were public parks or social services, withlittle regard for commercial considerations. While notanswering all questions of commercial management andfinancing, this book sheds light on many of them. Itsstrength lies in sound analysis, clear writing, and mountainsof data.

This book is an expanded, updated, and internationalversion of World Bank Technical Paper No. 275 in the AfricaTechnical Series and includes examples from industrial,developing, and transition economies. An article based onthe earlier paper, "Commercializing Africa's Roads," byRupert Pennant-Rea and Ian Heggie, appeared in theDecember 1995 issue of Finance & Development.

Gabriel Roth

THE STRUGGLEFOR ACCOUNTABILITY

The World Bank, NGOs, and Grassroots Movementsedited by Jonathan A. Fox and L. David Brown

"This book offers a sound and thorough study of NGO campaigns around

the world and provides a critical appraisal of the greening and increased

transparency of the World Bank. The authors deliver one of the few careful

and systematic evaluations on this highly emotional and polemical topic."

— Peter M. Haas, University of Massachusetts at Amherst

"The Struggle for Accountability provides the first comprehensive evaluation of

the debates, struggles, setbacks, and limited victories of World Bank officials

and their activist critics. This body of evidence provides a valuable window into

a complex set of relationships that has real relevance to today's efforts to link

local realities to global policy formulation and reform."

— Raymond C. Offenheiser, President, Oxfam America


DOOI^CDvJwrOfCTflBBi

^^^^^^^ •M IIIU _ .i.

To order call 800-356-0343 (US & Canada)or (617) 625-8569.Prices higher outsideU.S. and subject tochange without notice.

http://mitpress.mit.edu 584 pp., 10 illus. $30 paper

^^HliT^I

HnI^^RRI


http://mitpress.mit.edu

Asian crisis

Some of the issues raised in "The AsianCrisis: Causes and Cures" (Finance &Development, June 1998) need greateremphasis. On both a macro and amicro (corporate) basis, there werecurrency and maturity mismatchesbetween borrowings and loans.Because of the availability of cheapcapital and overly optimistic estimatesof growth prospects, East Asian banksborrowed large sums in foreign cur-rency and liberally made loans indomestic currency largely for specula-tive investments. While the banks' for-eign borrowings tended to be shortterm, their loans were largely for long-term investment.

Between 1993 and June 1997, Thai-land's external debt escalated from $20billion to $69 billion, and almostthree-fourths of the debt had a matu-rity of one year. The volume of foreignbank debt amounted to about 45 per-cent of GDP. Banks made a large num-ber of long-term loans to the propertyand real estate sectors. When real estateand share values plunged, banks wereleft with a large volume of non-performing assets. Financial disasters,bank failures, and bankruptciesoccurred. As it became obvious thatloans might not be repaid, foreignbanks ceased to make fresh creditsavailable and recalled old loans, aggra-vating the crisis.

Developing countries may learn anumber of lessons from the East Asiandebacle. The magnitude of externalborrowing must be limited to a coun-try's capacity to use the loans produc-tively and generate surpluses to repayand service them in foreign currency.Mismatches between short-term bor-rowing and long-term lending must beavoided. The financial and bankingsector must be properly regulated, andcapital-adequacy standards and pru-dential lending norms observed bybanks and other financial institutions.Foreign portfolio investment shouldbe monitored and excessive inflowschecked. Such investments are the firstto be withdrawn, and sudden capital

outflows can destabilize a currency.IMF packages incorporating structuralreforms and stabilization measures areoften tough to implement and mayresult in recession and unemployment.Prevention is therefore better than acure, and developing countries woulddo well to avoid currency crises.

S. S. Kothari

Calcutta, India

"The Asian Crisis: Causes and Cures"(Finance & Development, June 1998)provides an analytic overview of thefinancial turmoil that has engulfedseveral Asian countries since July 1997.

The article recognizes only indirectlyand in very subdued terms the twobasic factors contributing to the crisis:the political corruption of semi-authoritarian regimes and unsoundfinancial systems. The two are linked—especially in Indonesia and SouthKorea—with rampant bribery amongbusiness groups, as pointed out byFrank Vogl of Transparency Inter-national in his article, "The Supply Sideof Global Bribery," in the same issue.

The article raises another seriousissue in connection with the prospectsfor recovery in 1999-2000—the cur-rent ailments of the Japanese economy.The continuing decline of the yenin terms of the dollar reflects theunsoundness of the financial sector inJapan, whose banks have an estimatedbad debt totaling about $650 billion.

N. A. Sarma

Hyderabad, India

Criteria for financial aidIn their article, "Aid Spurs Growth in aSound Policy Environment," (Finance6- Development, December 1997),Craig Burnside and David Dollarquote Nicholas Eberstadt of theAmerican Enterprise Institute as say-ing, in his testimony before the U.S.Senate Foreign Relations Committee,that financial aid to Third World gov-ernments for development is "posi-tively destructive." I disagree. The factthat countries such as Zambia main-tain poor policies despite the aid they

receive does not indicate that aid is initself destructive.

In the 1980s, Ghana's economic situ-ation was bad. As a result, Ghanaiansbegan to seek refuge in neighboringcountries. Following several years ofeconomic hardship, the country startedto recover economically after receivingfinancial aid from foreign donors.Subsequently, during the 1990s, Ghanaadopted sound economic policies,which eventually led to economicimprovement and reduction of pov-erty. Bolivia and Honduras are otherexamples of countries that improvedtheir policies after receiving aid.

Donors should establish criteria forproviding financial aid to developingcountries. These criteria shouldinclude sound economic policies andgood national leadership to ensure thataid is put to productive uses. Countriesthat cannot implement good reformprograms should not receive help,including technical assistance.

Sam OniahAnambra State, Nigeria

CorrectionsIn the September 1998 issue, a confer-

ence panelist in the photograph on

page 8 was incorrectly identified as

Nieves R. Confesor. That panelist was

Slow Yue Chia.

In the same issue, in the review of Isher

Judge Ahluwalia and I.M.D. Little's book,

India's Economic Reforms and Develop-

ment: Essays for Manmohan Singh, one

of the papers cited on page 51 was

incorrectly attributed, owing to an editori-

al error, to Mrs. Ahluwalia; the author is

Montek Ahluwalia.

CreditsPhotographs: cover, contents page, andpages 32, 36 and 38, Padraic Hughes; bookphotos, Pedro Marquez; authors' photos,IMF Photo Unit. Illustrations: page 8, MarkRobinson; contents page and pages 16, 20,24, and 40, Massoud Etemadi.

54Finance & Development / December 1998

LETTERS


International Monetary Fund • Publication Services • Box FD498 • 700 19th Street, N.W. • Washington,D.C. 20431 U.S.A.'Telephone: (202) 623-7430 • Telefax: (202) 623-7201 •E-mail: [email protected]


B i / | j f I d'l 11 M 9 |T^J /111 jm

World Economic Outlook: A Survey by the Staffof the tateiftftft^^ 7- , '

The World Economic Outlook—published twice a year(May and October) in English, French, Spanish, andArabic—presents IMF staff economists' projections andanalyses of global economic developments in the near andmedium terms. The October 1998 edition offers an in-depthdiscussion of policy responses to the financial turbulencethat began in East Asia in 1997 and then spread to Russiaand some Latin American economies, and its impact onthe world economy. l.t also explores issues related to theupcoming launch of the euro on January 1, 1999. Eachedition of the World Economic Outlook now includes anindex of some of the topics covered in recent years.

ISSN 0256-6877$36.00 (academic rate: $25.00); paper.1998 (Oct.). ISBN 1-55775-773-9. Stock #WEO-298

Visit IMF publications athttp://www.imf.org

fs^ptftf- jSer^oa tit mailed, phoped,te&ed,"W"0HW$ert,!'f>f 9S@ 'include AMEX,

:":^ ,-.©r Ma rP»!d tlftbet, expiration date,ai«i signature oo iiJtjcweditcard orders.

JiiliMMilMsaBglFor in-depth coverage of economic trends worldwide, decision makers in business, finance, and government in more than180 countries read Finance & Development, which is published four times a year (March, June, September, and December).

Professional category

12

3

4

5

6

7

8

9

10

11

12

13

99

College student

Central bank

Finance ministry

Commercial bank

Other financial institution

Planning agency

International or regional organization

Public library

Faculty—college or university

Other government agency

Nongovernmental organization

News agency

Private enterprise

Other

International Monetary FundPublication ServicesBox FD498Washington, D.C. 20431 USATel.:(202) 623-7430Fax: (202) 623-7201E-mail: [email protected]

Name.

Organization,(First) (Last)

Title

Street/P.O.Box.

State/Province.City.

Country. Postal code/Zip code

Air delivery (US$20 per year)Surface mail (complimentary)

Air delivery payment options (all orders must be prepaid).

Check or money order enclosed in US$ (Please make checks in US$ payable to IMF Publication Services)

Please charge US$_ .to my

American Express MasterCard VISA Expiration datemonth year

Acct. #

Signature(required on all credit card orders)

Fax#

Arabic Chinese English French Spanish

Language edition preferred


http://www.imf.org

ArticlesMartha Ainsworth, Setting GovernmentPriorities in Preventing HIV/AIDS, March

Tomas J.T. Balifio, Monetary Policy in Russia,December

James M. Boughton, Harry Dexter White andthe International Monetary Fund, September

Sharmini Coorey, Mauro Mecagni, and ErikOfferdal, Achieving Low Inflation in TransitionEconomies: The Role of Relative Price Adjustment,March

Michael C. Deppler, The Implications of EMUfor IMF Surveillance, December

Barry Eichengreen and Michael Mussa, CapitalAccount Liberalization and the IMF, December

Charles Enoch and Anne-Marie Guide, AreCurrency Hoards a Cure for All MonetaryProblems? December

Charles A. Enoch, Paul L. Hilbers, and ArtoKovanen, Monetary Operations in the EuropeanEconomic and Monetary Union, June

Julian Exeter and Steven Fries, The Post-Commnntst Transition: Patterns and Prospects,September

Stanley Fischer, The Asian Crisis and theChanging Role of the IMF, June

Philip Gerson, Poverty and Economic Policy inthe Philippines, September

Cheryl W. Gray and Daniel Kaufmann,Corruption and Development, March

Dale Gray, Energy Tax Reform in Russia andOther Former Soviet Union Countries, September

John Green and Phillip L. Swagel, The Euro Areaand the World Economy, December

Manuel Guitian, The Challenge oj ManagingGlobal Capital Flows, June

Sanjeev Gupta, Benedict Clements, and ErwinTiongson, Public Spending on HumanDevelopment, September

Daniel C. Hardy, Are Hanking Crises Predictable?December

Peter S. Heller, Aging in the Asian TigerEconomies, June

IMF Area Department Directors, How Has theAsian Crisis Affected Other Regions? September

IMF Staff, The Asian Crisis: Causes and Cures,June

IMF Staff, Currency Crises: The Role of MonetaryPolicy, March

IMF Staff, Mitigating the Social Costs of theAsian Crisis, September

IMF Staff, Should Equity He a Goal o\ EconomicPolicy? September

Harold James, From Grandniotherliness toGovernance: The Evolution of IMFConditionality, December

R. Barry Johnston, Sequencing Capital AccountLiberalization, December

Oussama Kanaan, Uncertainty Deters PrivateInvestment in the West Bank and Gaza Strip,June

Robert Klitgaard, International CooperationAgainst Corruption, March

Klaas Knot, Donogh McDonald, and KarenSwiderski, Policy Challenges for the Euro Area,December

Patrick Lenain, Ten Years of Transition: AProgress Report, September

Ian Lienert, Civil Service Reform in Africa:Mixed Results After 10 Years, June

John Lipsky, Asia's Crisis: A Market Perspective,June

Paul R. Masson, Miguel A. Savastano, and SunilSharma, Can Inflation Targeting Be a Frameworkfor Monetary Policy in Developing Countries?March

Donald J. Mathieson, Anthony Richards, andSunil Sharma, Financial Crises in EmergingMarkets, December

Paolo Mauro, Corruption: Causes, Consequences,and Agenda for Further Research, March

Ashoka Mody and Michael Walton, Building onEast Asia's Infrastructure Foundations, June

Saleh M. Nsouli and Mounir Rached, CapitalAccount Liberalization in the SouthernMediterranean, December

Guillermo Ortiz Martinez, What Lessons Doesthe Mexican Crisis Hold for Recovery in Asia?June

Mead Over, Coping with the Impact of AIDS,March

Stefano Pagiola, John Kellenberg, Lars Vidaeus,and Jitendra Srivastava, MainstreamingBiodiversity in Agricultural Development, March

Michael Potashnik and Joanne Capper,Distance Education: Growth and Diversity,March

Alessandro Prati and Garry J. Schinasi,Ensuring Financial Stability in the Euro Area,December

Lant Pritchett and Daniel Kaufmann, CivilLiberties, Democracy, and the Performance ofGovernment Projects, March

Christoph B. Rosenberg and Tapio O.Saavalainen, Dealing with Azerbaijan's OilBoom, Septcmlier

Jan Aart Scholte, The IMF Meets Civil Society,September

Lyn Squire, Confronting AIDS, March

Vito Tanzi, Economic Policy and Equity:Conference Participants Agree on Key Issues,September

George S. Tavlas, The International Use ofCurrencies: The U.S. Dollar and the Euro, June

Frank Vogl, The Supply Side of Global Bribery,June

Book ReviewsIsher Judge Ahluwalia and I.M.D. Little, editors,India's Economic Reforms and Development:Essays for Manmohan Singh, reviewed byMohsin S. Khan, September

Alan S. Blinder, Central Banking in Theory andPractice, reviewed by Manuel Guitian, June

Catherine Caufield, Masters of Illusion: TheWorld Bank and the Poverty of Nations, reviewedby Paul Streeten, March

Bimal Chakraborty, The United Nations and theThird World: Shifting Paradigms, reviewed bySanjeev Gupta, June

Roy Culpeper, Titans or Behemoths? TheMultilateral Development Banks, Volume 5,reviewed by Robert Picciotto, June

Roy Culpeper, Albert Berry, and FrancesStewart, editors, Global Development Fifty YearsAfter Bretton Woods: Essays in Honour of GeraldK. Helleiner, reviewed by Vinod Thomas andSarwar Lateef, December

Marcello De Cecco, Lorenzo Pecchi, andGustavo Piga, editors, Managing Public Debt:Index-Linked Bonds in Theory and Practice,reviewed by Graeme Wheeler, September

Kimberly Ann Elliott, editor, Corruption and theGlobal Economy, reviewed by Daniel Kaufmann,March

Richard H.R. Harper, Inside the IMF: AnEthnography of Documents, Technology, andOrganizational Action, reviewed by Brian C.Stuart, December

Ian G. Heggie and Piers Vickers, CommercialManagement and Financing of Roads, reviewedby Gabriel Roth, December

Devesh Kapur, John P. Lewis, and RichardWebb, The World Bank: Its First Half Century,reviewed by James M. Boughton, June

Ronald I. McKinnon and Kenichi Ohno, Dollarand Yen: Resolving Economic Conflict between theUnited States and Japan, reviewed by George S.Tavlas, March

Maurice Obstfeld and Kenneth Rogoff,Foundations of International Macroeconomics,reviewed by Peter Clark, December

Hiroyuki Odagiri and Akira Goto, Technologyand Industrial Development in /apart: BuildingCapabilities by Learning, Innovation, and PublicPolicy, reviewed by Robert Dekle, March

Louis W. Pauly, Who Elected the Bankers?Surveillance and Control in the World Economy,reviewed by Joaquin Ferran, September

M.G. Quibria and J. Malcolm Dowling, editors,Current Issues in Economic Development: AnAsian Perspective, reviewed by Qaizar Hus.sain,December

H.N. Ray, The World Bank: A Third World View,reviewed by Paul Streeten, March

Dani Rodrik, Has Globalization Gone Too Far?reviewed by Lant Pritchett, December

Anne C.M. Salda, Historical Dictionary of theWorld Bank, reviewed by James Feather, March

Arieh A. Ullmann and Alfred Lewis, editors,Privatization and Entrepreneurship: TheManagerial Challenge in Central and EasternEurope, reviewed by Raj M. Desai, December

Horst Ungerer, A Concise History of EuropeanMonetary Integration: From EPU to EMU,reviewed by Charles Enoch, June

Daniel Yergin and Joseph Stanislaw, TheCommanding Heights: The Battle BetweenGovernment and the Marketplace That IsRemaking the Modern World, reviewed by ClaireIJuksila, September

56Finance & Development / December 1998

INDEX 1998–VOLUME 35


International Monetary FundN E W P U B L I C A T I O N S

International Capital Markets:Developments, Prospects, and KeyPolicy Issuesby a staff team led by Charles Adams, Donald J.Mathieson, Garry Schinasi. and Bankim ChadhaThis year's International Capital Markets report foeuscson financial market behavior during the Asian crisis,policy lessons for dealing with volatility in capital flows,banking sector developments, initiatives in bankingsystem supervision and regulation, and management ofsystemic risk in the European Economic and MonetaryUnion.

English.$25.00 (academic rale: $20.00). (paper)1998. 196 pp. ISBN 1-55775-770-4. Slock #WEO-698

Trade Liberali/alion inIMF-Supported Programsby a slciIT learn led by Robert SharerThis study assesses trade liberalization in programssupported by the IMF. It reviews multiyear arrangementsin the 1990s and six detailed ease studies and discussesthe main economic factors affecting trade policy targets.

English.$25.00 (academic rate: $20.00). (paper)1998. 48 pp. ISBN 1-55775-707-0. Slock #VVKO-1897

Toward a Frameworklor Financial Stabilityby a staff team led by David Volkcrts-Landauand Carl-Johan IJndgrenThis study outlines the broad principles andcharacteristics of stable and sound f inancial systems.It examines the IMF's role in promoting financial stabilityand provides insight into how such a framework fits intothe IMF's surveillance of its members' f inanc ia l sectors.

English.$25.00 (academic rate: $20.00). (paper)1998. 81 pp. ISBN 1-55775-706-2. Stock #VVi;0-1697

Capital Account Liberalization:Theoretical and Practical Aspectsby a staff learn led by Harry Eichengreen andMichael Mussa wilh Giovanni Dell'Ariccia,Enrica Dclragiache, Gian Maria Milesi-Ferrelli,and Andrew l\vecdieOccasional Paper No. 172Capital account liberalization is an inevitable step forcountries wishing to realize the benefits of the globalizedeconomy. This paper reviews the theories behind capitalaccount liberalization and examines the risks associatedwith free capital flows. The authors conclude that thedangers can be limited through a combination of soundmacroeconomic and prudential policies and highlight theimportance of an orderly sequencing of reform, tailoredto each country's specific circumstances.

English.$18.00 (academic rate: $15.00). (paper)1998. 61 pp. ISBN 1-55775-777-1. Slock #S172KA.

Hedge Funds and Financial MarketDynamicsby a staff learn led by Barry Eichcngrcen andDonald J. Malhicson, with Bankim Chadha, AnneJansen, Laura Kodres, and Sunil Sharma.Occasional Paper No. 166Hedge funds are collective investment vehicles, oftenorganized as private partnerships and resident offshorefor tax and regulatory purposes. Their legal slatus placesfew restrictions on their portfolios and transactions,leaving their managers free to use short sales, derivativesecurities, and leverage to raise returns and cushionrisk. This paper considers the role of hedge funds infinancial market dynamics, wilh particular reference tothe Asian crisis.

English.$18.00 (academic rate: $15.00). (paper)1998. 75 pp. ISBN 1-55775-736-4. Slock #S166UA.

Visit the IMF at http://www.imf.org

International Monetary FundPublication ServicesBox FD498700 19th Street, N.W.Washington, D.C. 20431 U.S.A.

Telephone: (202) 623-7430Telefax: (202) 623-7201E-mail: [email protected] orders may he mailed, phoned, faxed, or e-mailed.Please include AMEX, VISA, or MasterCard number,expiration date, and signature on all credit card orders.


http://www.imf.org

upport Children's Art and Creativity

part of the ICAF Millennium Program the America 2000 mural was created49 talented 8- to 12- year-olds from across the United States on September 4,98 on The National Mall in Washington, D.C.

In July 1999, talented children from more than 60 countries will join togetherin Washington, D.C. to create the World 2000 mural.

(Actual Size 16 ft. x 24 /

Your organization could be a part of this historic event to involve children inthe celebration of the new millennium through art.

Contact us today to become an ICAF National Partner or Corporate Sponsor.

International Child Art Foundation, I 0s»04«99 MRB 1350 Connecticut Avenue, NW, Washin

Fax: (202) 530-1080, E-mail: [email protected], Http//:www.ieaf.org

http//:www.ieaf.org