T H E W O R L D B A N K
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
Telephone: 202-477-1234
Facsimile: 202-477-6391
Telex: MCI 64145 WORLDBANK
MCI 248423 WORLDBANK
World Wide Web: http://www.worldbank.org/
E-mail: [email protected]
ISBN-0-8213-4291-6
Financial SectorReform
Financial SectorReform
W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N TW O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T
™xHSKIMBy342916zv":%:-:*:,
Fi
na
nc
ia
l
Se
ct
or
R
ef
or
mT
HE
W
OR
LD
B
AN
K
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
ADB - Asian Development BankAFR - AfricaAPL - Adaptable Program LoanARPP - Annual Review of Portfolio
PerformanceBCEAO - West African Central BankBSP - Bangko Sentral ng PilipinasCAMEL - Capital adequacy, Asset quality,
Management efficiency, Earningsand Liquidity
CAR - Capital Asset RatioCAS - Country Assistance StrategyCBV - Central BankCBP - Central Bank of the PhilippinesCODE - Committee on Development
EffectivenessDBPC - Deposit Bank Credit to the Private
SectorDDSR - Debt and Debt Service ReductionDFC - Development Finance CorporationDFI - Development Finance InstitutionDOF - Department of FinanceDYB - Devlet Yatirim BankasiEAP - East Asia and PacificECA - Europe and Central AsiaEDI - Economic Development InstituteEIB - European Investment BankERL - Economic Reconstruction LoanESAF - Enhanced Structural Adjustment
FacilityESW - Economic Sector WorkFI - Financial InstitutionFIL - Financial Intermediary LoanFDI - Foreign Direct InvestmentFERIS - Foreign Exchange Risk Insurance
SchemeFFI - Financial Fragility IndexFML - Financial Markets LoanFOGADE - Deposit Insurance FundFSAC - Financial Sector Adjustment CreditFSAL - Financial Sector Adjustment LoanFSL - Financial Sector Adjustment
Related LoanFSRL - Financial Sector Restructuring
LoanFSU - Former Soviet Union RepublicsFTAL - Financial Sector Technical
Assistance LoanFY - Fiscal YearGDI - Gross Domestic InvestmentGDP - Gross Domestic ProductGDS - Gross Domestic SavingsIBRD - International Bank of
Reconstruction and DevelopmentICICI - Industrial Credit and Investment
Corporation of IndiaICR - Implementation Completion
ReportIDA - International Development
Association
IDBP - Industrial Development Bank ofPakistan
ID - Institutional DevelopmentIFC - International Finance CorporationIFS - International Financial StatisticsIIC - Industrial Investment CreditIMF - International Monetary FundIPRs - Initial Project ReviewsITPA - Industrial and Trade Policy
AdjustmentLAC - Latin America and the CaribbeanMAFI - Macro Fragility IndicatorMIFI - Micro-Institutional Fragility
IndicatorMIS - Management Information SystemMNA - Middle East and North AfricaNBFI - Non-Bank Financial IntermediaryNCB - National Commercial BankNDB - National Development BankNPAs - Non-Performing AssetsOD - Operational DirectiveOED - Operations Evaluation DepartmentOMS - Operational Manual StatementOP - Operational PolicyOPRIS - Operations Policy Research
Information SystemPAR - Performance Audit ReportPCR - Project Completion ReportPDIC - Procedures and the strengthening
of deposit insurancePFI - Participating Financial InstitutionPFP - Policy Framework PaperPICIC - Pakistan Industrial Credit and
Investment CorporationPIR - Performance Indicators RatingPR - President’s ReportPSDL - Private Sector Development LoanQAG - Quality Assistance GroupSAD - Sector Adjustment LoanSAL - Structural Adjustment LoanSAP - Structural Adjustment ProgramSAR - Staff Appraisal ReportSAS - South AsiaSECAL - Sector Adjustment LoanSIL - Specific Investment LoanSIM - Specific Investment and
Maintenance LoanSME - Small and Medium-size EnterprisesSOE - State-Owned EnterprisesSPO - State Planning OrganizationSSI - Small-scale IndustryTA - Technical AssistanceTAL - Technical Assistance LoanTKSB - Turkiye Sinai BankasiUEMOA - Union Economique et Monétaire
Ouest Africaine(since January 10, 1994)
UMOA - Union Monétaire Ouest Africaine(until January 10, 1994)
WDI - World Development Indicators
ABBREVIATIONS and acronyms
Financial SectorReform:
A Review of World Bank Assistance
W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T
1998
The World Bank
Washington, D.C.
Nicolas Mathieu
Copyright © 1998
The International Bank for Reconstruction
and Development/THE WORLD BANK
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.
All rights reserved
Manufactured in the United States of America
First printing June 1998
The opinions expressed in this report do not necessarily represent the views of the World Bank or
its member governments. The World Bank does not guarantee the accuracy of the data included in
this publication and accepts no responsibility whatsoever for any consequence of their use. The
boundaries, colors, denominations, and other information shown on any map in this volume do not
imply on the part of the World Bank Group any judgment on the legal status of any territory or the
endorsement or acceptance of such boundaries.
The material in this publication is copyrighted. Requests for permission to reproduce portions of it
should be sent to the Office of the Publisher at the address shown in the copyright notice above.
The World Bank encourages dissemination of its work and will normally give permission promptly
and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to
copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910,
222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A.
Design: The Magazine Group/Jeff Kibler
Cover photo: Curt Carnemark
ISSN 1011-0984
ISBN 0-8213-4291-6
Library of Congress Cataloging-in-Publication Data
Mathieu, N.
Financial sector reform: a review of World Bank assistance/Nicolas Mathieu
p. cm—(A World Bank operations evaluation study, ISSN 1011-0984)
Includes bibliographical references.
ISBN 0-8213-4291-6
1. Finance. 2. Financial institutions—Management. 3. World Bank. I. Title II. Series: World Bank
operations evaluation study.
HG173.M385 1998
332.1’532—dc21
98-36251
CIP
Printed on recycled paper.
iii
C o n t e n t s
vii Acknowledgmentsix Foreword, Prefacio, Préfacexiii Executive Summary, Resumen, Résumé Analytique
1 1. The Evaluation Framework2 The Conceptual Framework3 The Bank’s Approach to Financial Sector Operations3 The Bank’s Financial Lending Instruments4 The Bank Non-lending Instruments for Financial Sector Reform4 Sample Coverage and Methodology5 Data Limitations8 The Plan of the Study
9 2. Financial Sector Reform Operations9 Characteristics of Financial Systems in Sample Countries
10 The Situation in the Real Sector10 Program Design16 The Outcome as Measured by Performance Indicators17 Reasons for Mixed Outcomes21 Determinants of Performance23 The Real Sector24 Finance and Development
25 3. Institutional Development28 Key Factors Affecting Institutional Development29 The Sustainability of Institutional Components
33 4. Liberalization and Financial Crises35 Mexico38 Venezuela40 The Philippines41 Conclusions and Lessons
43 5. Bank Performance43 Economic and Sector Work45 Lending Operations46 Adjustment Strategies and Sequencing48 Partnerships with the IMF and IFC
53 6. Conclusions and Recommendations53 Main Findings54 Reasons for Mixed Outcomes54 Recommendations55 Program Design55 Bank Processes
59 Endnotes
61 Bibliography
iv
Annexes65 Table 1.1 List of 88 Financial Sector Adjustment-related Projects (FSLs):
FY85–FY9668 Table 1.2 Bank ESW Related to Financial Sector Reform: FY81–FY9571 Table 2.1 List of Sample Financial Sector Adjustment-related Operations (FSLs)
for 23 Countries: FY85–FY9672 Annex 2.2 Determinants of Project Performance: The Empirical Evidence74 Table 3.1 Sample Scorecard for Institutional Development75 Table 3.2 List of 24 Financial Sector Adjustment-related Projects (FSLs) for
Institutional Development Analysis76 Table 3.3 List of 58 Financial Intermediary Loans (FILs) for Institutional
Development Analysis78 Table 4.1 Macro-Fragility Indicators80 Table 4.2 Microeconomic/Institutional Fragility Indicators81 Table 4.3 Central Bank Fragility Indicators82 Annex 5 Financial Sector Reform: A Review of World Bank Assistance/
Management Response85 Annex 6 Report from CODE, Committee on Development Effectiveness
Boxes6 1.1 Methodology Used to Assess Financial Sector Reforms
11 2.1 Indonesia and the Sequencing of Reform12 2.2 Chile: A Success Story the Second Time Around22 2.3 Turkey: The “Ownership” Issue27 3.1 Sample and Methodology to Assess Institutional Development30 3.2 Pakistan: The Long, Difficult Road Toward Sustainability of Financial
Institutions34 4.1 Thailand’s 1997 Financial Crisis—What Did the Bank Do?36 4.2 Korea’s 1997 Financial Crisis—What Went Wrong With the Miracle
Economy?44 5.1 The Impact of Pakistan’s ESW on Lending Operations48 5.2 Sequencing Reforms in Senegal and the Union Monetaire Ouest Africaine
Figures4 1.1 Regional Distribution of Adjustment-related Operations in the Financial
Sector, FY85–9626 3.1 Number of FSLs and FILs Approved 1975–9737 4.1a FSAL and Financial Fragility in Mexico: Macroeconomic Level37 4.1b FSAL and Financial Fragility in Mexico: Microeconomic Institutional Level38 4.2a FSRL and Financial Fragility in Mexico: Macroeconomic Level38 4.2b FSRL and Financial Fragility in Mexico: Microeconomic Institutional Level39 4.3a FSAL and Financial Fragility in Venezuela: Macroeconomic Level39 4.3b FSAL and Financial Fragility in Venezuela: Microeconomic Institutional
Level40 4.4a FSAL and Financial Fragility in The Philippines: Macroeconomic Level40 4.4b FSAL and Financial Fragility in The Philippines: Microeconomic
Institutional Level41 4.4c Central Bank Fragility—The Philippines (1989–1996)
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
v
Tables7 1.1 Performance Indicators for the Financial Sector
10 2.1 Initial Financial Sector Conditions in Countries Studied14 2.2 Relevance of Bank-supported Policy Packages15 2.3 Hits, Misses, and Overkills16 2.4 Policy Implementation18 2.5 Outcomes: Financial Sector Performance Indicators20 2.6 A Detailed Comparison of Outcome Ratings for 23 Countries21 2.7 A Comparison of Outcome Ratings for Nineteen Countries21 2.8 Controlling for Initial Conditions in the Macroeconomy and the Financial
Sector23 2.9 Changes in the Real Economy Measured with Performance Indicators24 2.10 Comparison of Financial Sector Outcome with Real Economy Impact26 3.1 The Consolidated Results28 3.2 Results in Low- and Middle-income Countries28 3.3 Relative Success in Countries with High Loan Frequency (4 or More Loans,
FSLs Plus FILs) and Low Loan Frequency (1–3 Loans)45 5.1 Projects for Which Advance Economic and Sector Work was Done45 5.2 ESW and Financial Sector Program Design and Outcome (FY85–96)46 5.3 Bank Performance in Supporting Institutional Development
C o n t e n t s
vii
Acknowledgments
Nicolas Mathieu (task manager) acknowledges, with thanks,the contributions of: Salman Anees and Suparni Gunasekera(financial sector performance); Serdar Dalkir (liberaliza-tion policies); Sunit Gupta (institutional development); RajChhikara (financial crises); Derek White (Bank perfor-mance); Robert Cull (DECRG, econometric analysis);Deena R. Khatkhate (senior advisor); Robert M. Buckleyand Laurie Effron (OEDCR); Jasmine Mason-Anderson,Eneshi Irene K. Davis, Jacqueline Jackson and BarbaraYale who provided administrative assistance.
The report, part of a publication series, was producedby a team, led by Elizabeth Campbell-Pagé, consisting ofLeo Demesmaker, Kathy Strauss, Tsige Kagombe, RoshnaKapadia and Marie Daramy.
Director-General, Operations Evaluation Department: Robert Picciotto
Director, Operations Evaluation Department: Elizabeth McAllister
Manager, Sector and Thematic Evaluations: Roger Slade
Task Manager: Nicolas Mathieu
ix
F o r e w o r d
FOREWORDThis study analyzes the role ofthe Bank in helping client coun-tries to implement financial sec-tor reforms. It focuses on thecountry as the unit of analysis,rather than on individual loans,
and on performance indicators in thefinancial and real sectors. Even with-out full incorporation of lessons stillto be drawn from recent East Asiacrises, this study finds a satisfactoryoutcome in only 12 of the 23 coun-tries that it examines. Initial condi-tions appear to be significant indetermining the outcome of reforms,and prior sector work is importantfor ensuring that policy reforms inBank loans reflect a country’s initialconditions.
In general, adjustment loans aremore successful and more sustainablein promoting institutional develop-ment than are financial intermedia-tion loans; this may reflect the rela-tive ease of introducing prudentialregulations compared to strengthen-ing specific financial institutions. Thestudy develops financial fragilityindicators and applies them to threecountries that experienced financialcrises in the early 1990s (Mexico, thePhilippines and Venezuela), todemonstrate their use as well as theirlimitations in evaluating financialsectors. The conclusions reaffirm theidea that financial sector adjustmentis a complex, long-term process.
The main recommendations ofthe study are that the internal guide-lines on financial sector operations(OD 8.30) provide a valid frame-work for preparing operations insupport of financial sector reforms.The Bank should go beyond theseguidelines, however, by incorporatingbest practice on both substantiveissues and Bank processes, after suffi-cient time has elapsed to evaluate
PREFACIOEn este estudio se analiza elpapel del Banco Mundial paraayudar a los países miembros aaplicar reformas en el sectorfinanciero. El estudio se centraen los países como unidades de
análisis, no en cada uno de los prés-tamos, y en los indicadores deldesempeño en el sector financiero yel sector productivo de la economía.Si bien no se han incorporado plena-mente las enseñanzas que podránextraerse de las recientes crisis enAsia oriental, se concluye que sólo en12 de los 23 países examinados en elestudio se han obtenido resultadossatisfactorios. Al parecer, las condi-ciones iniciales son importantes paradeterminar el resultado de los refor-mas y los estudios sectoriales previosayudan a garantizar que las reformasde política incluidas en los préstamosdel Banco reflejen las condiciones ini-ciales del país.
En general, los préstamos parafines de ajuste son más eficaces parapromover el desarrollo institucionalque los préstamos de intermediaciónfinanciera, además de ser mássostenibles; esto podría deberse a quees relativamente más fácil incorporarnormas prudenciales que fortalecerdeterminadas institucionesfinancieras. En el estudio se propo-nen algunos indicadores de la fragili-dad financiera y se aplican a trespaíses que sufrieron crisis a comien-zos de la década de 1990 (México,Filipinas y Venezuela), a fin dedemostrar su utilidad y sus limita-ciones para la evaluación del sector.Las conclusiones del estudio reafir-man la idea de que el ajuste del sec-tor financiero es un proceso com-plejo y a largo plazo.
La conclusión más importantedel estudio es que las directricesinternas sobre operaciones en el sec-
PRÉFACELa présente étude analyse le rôlejoué par la Banque pour aiderses clients à entreprendre desréformes au niveau de leursecteur financier. Elle porte surles pays eux-mêmes, plutôt que
sur tel ou tel prêt, et sur les indica-teurs de performance au niveau dusecteur financier et du secteur réel del’économie. Même si les enseigne-ments restant à tirer des crises quiont récemment frappé l’Asie de l’Estn’y sont pas totalement pris encompte, l’étude fait état de résultatssatisfaisants dans 12 des 23 paysexaminés seulement. Il apparaît queles conditions de départ ont uneinfluence déterminante sur l’issue desréformes, et des études sectoriellespréalables contribuent fortement à ceque les réformes de politiquegénérale prévues dans les prêts de laBanque reflètent bien la situation ini-tiale d’un pays.
En général, les prêts à l’ajuste-ment contribuent de façon plus posi-tive et plus durable au développe-ment institutionnel des pays que lesprêts d’intermédiation financière ;cela témoigne peut-être du fait qu’ilest relativement plus facile d’intro-duire des réglementations pruden-tielles que de renforcer des établisse-ments financiers donnés. Desindicateurs de fragilité financièresont définis dans cette étude, et leurapplication au cas de trois pays quiont connu des crises financières audébut des années 90 (Mexique,Philippines et Venezuela) permet d’endémontrer à la fois l’utilité et les lim-ites comme outils d’évaluation dessecteurs financiers. Les conclusionsviennent réaffirmer l’idée selonlaquelle l’ajustement du secteurfinancier est un processus complexeet de longue haleine.
Selon les principales recomman-
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
x
and learn from the recent devel-opments in East Asia.
Substantive recommenda-tions include:
• better sequencing ofreforms (macrostability iscritical, interest ratederegulation and bankrecapitalization should bepreceded by a number ofconditions); some mea-sures to limit damage canbe taken while waitingfor those conditions to bein place;
• more attention to the implica-tions of systemic governancefactors and to the prudentialframework for financial insti-tutions, including developmentof adequate skills to ensurethey are effective; and to theindependence and solvency ofthe central bank;
• greater focus on promotingcompetition; and
• investing more resources in col-lecting information on riskexposure and governance offinancial sectors.
On processes:• realistic assessment of long-
term commitment to reform iscritical;
• use of a wide range of lendingand non-lending instrumentsshould be encouraged, includ-ing the new adaptable programloan;
• financial intermediary loansshould be used only in the con-text of an improved macro-and financial environment;
• good sector work is importantfor appropriate design of oper-ations;
• better coordination with theIMF and the IFC is necessary;and
tor financiero (OD 8.30) ofre-cen un enfoque válido para lapreparación de operaciones enrespaldo de las reformas del sec-tor. Sin embargo, la labor delBanco debería trascender estasdirectrices e incorporar prácti-
cas óptimas tanto en las áreas funda-mentales como en sus procedimien-tos, una vez que haya transcurrido eltiempo necesario para evaluar losrecientes acontecimientos de Asiaoriental y extraer las enseñanzas per-tinentes.
Las siguientes son algunas de lasrecomendaciones más importantesdel estudio:
• debe mejorarse la secuencia delas reformas (la estabilidadmacroeconómica es fundamen-tal; antes de ladesreglamentación de las tasasde interés y la recapitalizaciónde los bancos deben cumplirsevarias condiciones); algunas delas medidas destinadas a limi-tar los daños pueden adoptarsemientras se espera que se cum-plan dichas condiciones;
• debe prestarse mayor atención alas consecuencias de los factoressistémicos de la función degestión y a las normas de disci-plina y control de las institu-ciones financieras, incluido eldesarrollo de la capacidad nece-saria para garantizar su eficacia,así como a la independencia ysolvencia del banco central;
• debe hacerse un mayoresfuerzo por promover la com-petencia, y
• deben destinarse mayoresrecursos a la recolección deinformación sobre el riego y lagestión del sector financiero.
En lo que respecta a los proced-imientos, en el estudio se señala losiguiente:
dations de l’étude, les directivesinternes régissant les opérationsde la Banque dans le secteurfinancier (DO 8.30) constituentun cadre adéquat pour la prépa-ration de projets à l’appui desréformes dans ce secteur. Cepen-
dant, la Banque doit aller au-delà deces directives en incorporant les pra-tiques optimales pour ce qui con-cerne aussi bien les questions de fondque ses propres procédures, une foisqu’elle aura eu assez de temps pourévaluer les récents événements d’Asiede l’Est et en tirer les leçons.
Sur les questions de fond, lesrecommandations sont les suivantes :
• mieux échelonner les réformes(la stabilité macroéconomiqueest d’une importance capitale,et un certain nombre de condi-tions doivent précéder la déré-glementation des taux d’intérêtet la recapitalisation des ban-ques, certaines mesures pallia-tives pouvant être prises enattendant que ces conditionssoient réunies) ;
• prêter davantage attention auximplications des facteurs sys-témiques de gouvernance et aucadre prudentiel applicable auxétablissements financiers, ycompris à l’établissement decompétences adéquates pour engarantir l’efficacité, ainsi qu’àla question de l’indépendanceet de la solvabilité des banquescentrales ;
• mettre plus l’accent sur desmesures propres à encouragerla concurrence ; et
• investir davantage deressources dans la collecte d’in-formations sur l’exposition auxrisques et l’aspect gouvernancedes secteurs financiers.
Les procédures font l’objet desrecommandations suivantes :
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c eE
NG
LI
SH
ES
PA
NO
L
FR
AN
CA
IS
xi
F o r e w o r d
• the Bank has a key role toplay in helping to con-vince groups within bor-rowing countries of theimportance of financialreform.
• es fundamental hacer unaevaluación realista delcompromiso a largo plazocon la reforma;
• debería alentarse el uso deuna amplia gama deinstrumentos crediticios yno crediticios, incluidoslos nuevos préstamosadaptables para progra-mas;
• los préstamos a intermediariosfinancieros deberían otorgarsesolamente en el contexto de unclima macroeconómico yfinanciero más favorable;
• los estudios sectoriales de cali-dad son importantes para eldiseño adecuado de las opera-ciones;
• es necesario mejorar la coordi-nación con el FMI y la CFI, y
• al Banco le corresponde latarea fundamental de ayudar aconvencer a los diversos gruposen los países prestatarios sobrela importancia de la reformafinanciera.
• une évaluation réaliste del’adhésion à long termed’un pays au processus deréforme est d’une impor-tance capitale ;
• il faut encourager l’utilisa-tion d’un large éventaild’instruments de prêt ethors prêt, y compris lesnouveaux prêts à des pro-grammes évolutifs ;
• les prêts d’intermédiationfinancière doivent être utilisésuniquement dans le contexted’une situation macroé-conomique et financièreassainie ;
• des études sectorielles de qual-ité sont importantes pour laconception appropriée desopérations ;
• il convient d’assurer unemeilleure coordination avec leFMI et la SFI ; et
• la Banque a un rôle clé à jouerpour aider à convaincre lesgroupes concernés, dans sespays emprunteurs, de l’impor-tance des réformes financières.
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
Robert PicciottoDirector-General, Operations Evaluation Department
xiii
EXECUTIVE SUMMARYThe globalization of theworld’s financial markets hascreated a new world environ-ment, full of opportunities,but fraught with dangers andunexpected shocks, high-
lighted most recently in the coun-tries of East Asia. Resource flowsfrom developed to developing coun-tries have increased eight-fold in thelast decade, and growth in manydeveloping countries has reachedthe highest level in many years. Atthe same time, the promises of thisnew financial environment areclearly not without problems. Sinceglobalization began, more than 80countries—both developed anddeveloping—have had banking sec-tor crises, and in many countriesthe crises have been of unprece-dented scale.
The World Bank’s support forthe financial sector has exhibitedsimilarly contrasting patterns. Onthe one hand, many of the financialsector loans made in the late 1980sand 1990s have played a key role inrestoring growth with equity. At thesame time, however, the averageoutcome of Bank support for thefinancial sector has sharply deterio-rated from being one of the bestperforming sectors before globaliza-tion to currently being one of theweakest. In addition, the composi-tion and level of Bank support forthe financial sector has changed inbasic ways, focusing more onadjustment lending and less onfinancial intermediary lending, par-ticularly following the Bank’s 1992establishment of a new frameworkgoverning financial sector opera-tions—known as OperationalDirective (OD) 8.30. Finally, in thepast year the Bank has committedto a renewed emphasis on support-
RESUMENLa globalización de los merca-dos financieros mundiales hacreado un nuevo contextointernacional pleno de oportu-nidades que, no obstante, estáplagado de riesgos y crisis
inesperadas, como quedódemostrado recientemente en lospaíses de Asia oriental. En laúltima década, los flujos de recur-sos de los países desarrollados a lospaíses en desarrollo se han incre-mentado ocho veces, y muchospaíses en desarrollo han registradolos índices de crecimiento más altosen muchos años. Al mismo tiempo,este nuevo ambiente financiero, sibien prometedor, no está exento deproblemas. Desde el inicio del pro-ceso de globalización más de 80países tanto desarrollados como endesarrollo han sufrido crisis ban-carias, muchas de ellas de una mag-nitud sin precedentes.
El respaldo del Banco Mundialal sector financiero ha tenido con-trastes similares. Por un lado,muchos de los préstamos otorgadosa ese sector a fines de los añosochenta y en la década de 1990han sido cruciales para restablecerel crecimiento con equidad. Almismo tiempo, sin embargo, losresultados medios logrados con elapoyo del Banco al sectorfinanciero han empeorado acusada-mente; en efecto, antes de la global-ización dicho sector era uno de losque obtenía mejores resultados,pero en la actualidad su desempeñoes uno de los más deficientes.Además, la composición y el niveldel respaldo del Banco al sectorfinanciero han sufrido cambiosfundamentales, atribuyéndosemenos importancia a los préstamosa intermediarios financieros paraconcentrarse en préstamos con
RÉSUMÉ ANALYTIQUEL’intégration des marchésfinanciers à l’échelon mondiala créé un nouvel environ-nement porteur d’innom-brables opportunités maiségalement source de risques et
de chocs imprévus, comme entémoignent les récents événementsd’Asie de l’Est. Au cours de ladernière décennie, les apports deressources des pays développés auxpays en développement ont étémultipliés par huit et beaucoup depays en développement ont vu lacroissance atteindre des niveauxjamais observés depuis bienlongtemps. Mais en dépit de cespromesses, le nouveau contextefinancier n’est manifestement pasdénué de problèmes. Depuisl’amorce du processus de mondiali-sation, plus de 80 pays, développéset en développement, ont connudes crises bancaires qui ont été,pour beaucoup, d’une ampleur sansprécédent.
L’appui fourni par la Banquemondiale au secteur financierprésente un bilan tout aussi con-trasté. D’un côté, une bonne partiedes prêts accordés dans ce domaineà la fin des années 80 et dans lesannées 90 a fortement contribué àrétablir une croissance soucieused’équité. De l’autre, les résultatsobtenus en la matière se sont, enmoyenne, nettement dégradés, aupoint que cette partie du porte-feuille de prêts de la Banque, quiétait l’une des plus performantesavant la mondialisation, est aujour-d’hui parmi celles dont les résultatslaissent le plus à désirer. Il fautajouter à cela une modificationfondamentale du niveau et descomposantes de l’appui de laBanque, qui a fini par mettre l’ac-cent sur les opérations d’ajustement
E x e c u t i v e S u m m a r y
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
xiv
ing the financial sector, and in1997 it announced a newfinancial sector strategy.
This study examines theBank’s performance in thefinancial sector in this evolv-ing environment. It examines
the results on the ground of theBank’s performance over thisperiod of change and turbulence. Itanalyzes financial sector operationsthrough the prism of the 1992financial operational directive, andtraces the effects that these opera-tions had on country economic per-formance. These results, in turn,provide insights into whether coun-tries assisted by the Bank improveoverall economic performancewhen they follow the financial poli-cies recommended by the Bank. Ineffect, these results provide supportfor the metaphor, often used byPresident Wolfensohn, that thefinance sector is, in many respects,the lifeblood of the economy andfinancial sector reform is akin toheart surgery.
This study, for the first time,brings together an analysis of theoutcomes of financial sector adjust-ment lending and in particular, theBank’s treatment of financial crises.The question addressed is whetherthe Bank gave adequate attentionin these operations, to the fragilityof financial systems, and, indeed,whether financial sector fragility ismeasurable in ways that can helpavoid future problems? Finally, thestudy examines how firmlygrounded the Bank’s new financialsector strategy is in the Bank’sexperience—does the new strategybuild on the lessons learned by theBank? For example, does this expe-rience support the argument that alonger-term perspective on thefinancial sector is necessary—that
fines de ajuste, especialmentetras el establecimiento porparte del Banco, en 1992, deun nuevo marco normativopara las operaciones en el sec-tor financiero conocido comola directriz operacional 8.30
(OD 8.30). Finalmente, durante elúltimo año el Banco ha dado reno-vada importancia a su labor derespaldo al sector y en 1997 anun-ció la adopción de una nuevaestrategia al respecto.
En este estudio se examina eldesempeño del Banco en el sectorfinanciero en este contexto enevolución. Se analizan los resulta-dos en el terreno obtenidos por lainstitución durante este período decambios y turbulencias. También sepasa revista a las operaciones en elsector financiero a la luz de ladirectriz operacional de 1992 antesmencionada y se identifican losefectos de estas operaciones en eldesempeño económico de lospaíses. Esos resultados, a su vez,ayudan a determinar si los paísesque reciben asistencia del Bancologran mejorar el desempeño gen-eral de su economía cuando adop-tan las políticas financierasrecomendadas por la institución.En efecto, estos resultados dancredibilidad a la metáfora que elPresidente del Banco, JamesWolfensohn, usa frecuentemente alseñalar que el sector financiero es,en muchos aspectos, el órgano vitalde la economía y que la reforma endicho sector es una verdaderacirugía al corazón.
Este estudio constituye elprimer análisis de los resultados delos préstamos para ajuste del sectorfinanciero y, en particular, de lasmedidas adoptadas por el Bancopara hacer frente a las crisisfinancieras. Se procura determinar
plutôt que sur les prêts d’in-termédiation financière,notamment après l’adoptionen 1992 du nouveau cadrerégissant ses opérations dansle secteur financier (Directiveopérationnelle 8.30). Pour
finir, la Banque s’est engagée,durant l’année écoulée, à redonnerun nouvel élan à ses activités enfaveur du secteur financier, et aannoncé en 1997 une nouvellestratégie dans ce domaine.
C’est dans ce contexte mou-vant que la présente étude dresse lebilan de l’action de la Banque dansle secteur financier. Elle évalue lesrésultats obtenus sur le terraindurant cette période de change-ments et de turbulences. Elleanalyse l’action de la Banque auregard de sa directive opéra-tionnelle de 1992, et cerne les effetsde cette action sur les performanceséconomiques des pays. À partir delà, il est possible de déterminer siles pays bénéficiant de l’aide de laBanque obtiennent généralement demeilleurs résultats économiquesquand ils appliquent les politiquesrecommandées par celle-ci dans ledomaine financier. En fait, toutesces données viennent confirmerl’idée souvent exprimée par lePrésident Wolfensohn au moyen dela métaphore selon laquelle lesecteur financier est, à bien deségards, le coeur de l’économie, et laréforme de ce secteur tient de lachirurgie cardiaque.
Cette étude fait, pour la pre-mière fois, la synthèse des résultatsdes prêts à l’ajustement du secteurfinancier, en examinant en partic-ulier la réponse apportée aux crisesfinancières par la Banque. La ques-tion traitée est de savoir si celle-ci,dans les opérations en question, aprêté l’attention qu’il fallait à la
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
xv
financial reform is best seen asa process rather than a proj-ect? Similarly, has policyadvice and sector work hadan appreciable effect on finan-cial sector lending? In short,how well have financial sector
operations emphasized creatingincentives for prudent bankingpractices?
The study’s main finding is thatthe new financial sector strategy isindeed informed by the Bank’scomplex and turbulent experiencein the financial sector. Its new strat-egy is based on the lessons learnedduring one of the most volatilefinancial periods in history. But,the Bank’s learning process hasbeen a very slow one, and seriousconcerns remain. Indeed, these con-cerns have intensified. Financialsector issues are at the core of therecent crises in East Asia and theinternational machinery to foresee,prevent and manage such crises isunder stress. The discussion of apotential financial sector role forthe Bank, beyond its current coun-try focus, lies outside the scope ofthis report.
Setting aside these basic quali-fications, the main message is posi-tive. First, the study shows that theBank’s experience is consistent withthe growing conventional wisdomthat an effective financial sectorpolicy is at the heart of theprocesses which generate and nur-ture growth. As a result, the studyimplies that well-designed financialpolicies can be expected to con-tribute to greater growth. Second,the study indicates that the Bank’snew financial sector strategy, itsincreased emphasis on the need fora richer set of indicators to judgeperformance, and finally, its newlending instruments, such as
si en estas operaciones elBanco tomó debidamente encuenta la fragilidad de los sis-temas financieros y si, dehecho, es posible evaluardicha fragilidad de maneraque se puedan evitar proble-
mas en el futuro. Finalmente, seexamina hasta qué punto la nuevaestrategia del Banco para el sectorfinanciero se basa en la experienciaprevia de la institución: ¿seaprovecharon en la formulación deesta nueva estrategia las enseñanzasrecogidas? Por ejemplo, ¿confirmaesta experiencia el argumento deque para el sector financiero esnecesario adoptar una estrategia alargo plazo y de que la reformafinanciera debe considerarse másbien un proceso y no un proyecto?Igualmente, ¿han tenido el aseso-ramiento en materia de políticas ylos estudios sectoriales un efectoapreciable en las operaciones cred-iticias para el sector financiero? Enresumen, ¿en qué medida se haatribuido suficiente importancia enlas operaciones en el sectorfinanciero a la creación de incen-tivos para la adopción de prácticasbancarias prudentes?
La conclusión más importantedel estudio es que en la nuevaestrategia ha influido ciertamentela compleja y turbulenta experien-cia del Banco en el sectorfinanciero. La nueva estrategia estábasada en las enseñanzas recogidasdurante uno de los períodos demayor inestabilidad en la historia.Sin embargo, el proceso de apren-dizaje del Banco ha sido muy lentoy aún existen graves problemas. Dehecho, estos problemas se hanintensificado. La problemática delsector financiero es un aspecto fun-damental de las recientes crisis enAsia oriental y los mecanismos
fragilité des systèmesfinanciers, et si, d’ailleurs, lafragilité d’un secteur financierpeut se mesurer d’une manièrequi permette d’éviter desproblèmes ultérieurs. Enfin,l’étude examine à quel point
la nouvelle stratégie de la Banquepour le secteur financier est ancréedans sa propre expérience, et prendappui sur les enseignements qu’elleen a tirés. Il s’agit de savoir, parexemple, si cette expérience con-firme l’argument selon lequel uneperspective à plus long terme s’im-pose à l’égard du secteur financier,une réforme financière étant à con-sidérer de préférence comme unprocessus et non comme un simpleprojet, ou encore si les avis de poli-tique générale et les études secto-rielles ont eu un effet appréciablesur les prêts au secteur financier.Bref, dans quelle mesure les opéra-tions dans le secteur financier sesont-elles attachées à promouvoirdes règles de saine gestion dans ledomaine bancaire ?
La principale conclusion quiressort de l’étude est que la nou-velle stratégie de la Banque dans lesecteur financier est effectivementle fruit de l’expérience complexe etmouvementée qui a été la siennedans ce secteur, et qu’elle s’appuiesur les leçons tirées d’une des péri-odes les plus instables que l’histoireait connues au plan financier. Celadit, le processus d’apprentissage dela Banque a été très lent, et desérieux motifs de préoccupationsubsistent. Ils se sont même accrus.Les questions relatives au secteurfinancier sont en effet au coeurmême des crises récemment surv-enues en l’Asie de l’Est, et lesmécanismes internationaux permet-tant de prévoir, de prévenir etd’endiguer ces crises sont mis à
E x e c u t i v e S u m m a r y
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
xvi
Adaptable Lending, are firmlyrooted in the Bank’s experi-ence. Consequently, assumingthat the generic weaknesses ofthe global system areaddressed, the new strategyshould improve the Bank’s
impact. These favorable financialsector effects should, in turn, helpthe countries achieve higher, morestable rates of growth.
internacionales para pronos-ticar, prevenir y hacer frente aestas crisis están sujetos afuertes presiones. El análisisde una posible actuación delBanco Mundial en el sectorfinanciero, además de su
actual énfasis en los países, escapaal alcance de este informe.
Con excepción de estos proble-mas básicos, la evaluación funda-mental del informe es positiva.Primero, se indica que la experien-cia del Banco confirma la opinióngeneralmente aceptada y cada vezmás difundida de que para generary fomentar el crecimiento es indis-pensable una política financiera efi-caz. Por consiguiente, del informese desprende que las políticasfinancieras bien diseñadas deberíancontribuir a lograr un mayor crec-imiento. Segundo, en el estudio seseñala que la nueva estrategia delBanco para el sector financiero, lamayor insistencia de la instituciónen la necesidad de contar con unconjunto más amplio de indi-cadores para evaluar los resultadosy, finalmente, sus nuevos instru-mentos crediticios, como los prés-tamos adaptables, están sólida-mente cimentados en la experienciaprevia del Banco. Por lo tanto,suponiendo que se abordarán lasdeficiencias genéricas del sistemafinanciero mundial, la nuevaestrategia debería incrementar elimpacto de la labor del Banco. A suvez, este efecto favorable en el sec-tor financiero debería ayudar a lospaíses a aumentar y estabilizar sustasas de crecimiento.
mal. Mais la question du rôleque la Banque pourrait êtreamenée à jouer dans le secteurfinancier, au-delà de sonaction actuelle à l’échelon despays, déborde du cadre de cerapport.
Mis à part ces réserves fonda-mentales, le message essentiel estpositif. D’une part, l’étude montreque l’expérience de la Banque vadans le sens de l’idée de plus enplus communément admise selonlaquelle une politique rationnellepour le secteur financier est un élé-ment central des processus propresà engendrer et entretenir la crois-sance. L’implication de ce constatest que l’on peut s’attendre à ce quedes politiques financières bienconçues contribuent à une crois-sance plus soutenue. D’autre part,selon l’étude, la nouvelle stratégieque la Banque applique dans lesecteur financier, l’importanceaccrue qu’elle attache à la nécessitéd’une série plus fournie d’indica-teurs pour l’évaluation des perfor-mances et, en définitive, ses nou-veaux instruments de prêt, tels queles prêts évolutifs, sont tous des élé-ments fermement ancrés dans l’ex-périence de la Banque. De ce fait,dans l’hypothèse où des mesuressont prises pour remédier auxcarences génériques du système enplace à l’échelon mondial, la nou-velle stratégie devrait améliorerl’impact de la Banque, et ces effetsfavorables pour le secteur financierdevraient à leur tour aider les paysà atteindre des taux de croissanceplus élevés et plus stables.
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
EN
GL
IS
H
ES
PA
NO
L
FR
AN
CA
IS
1
The Evaluation Framework
11
In the past 15 years the world has experienced one of the most volatile periods in financial
history. An international debt crisis, high and variable inflation and real interest rates, and
banking crises created costly financial problems throughout the world and particularly in
developing countries. As dramatically illustrated in recent weeks, even the East Asian countries
with fiscal surpluses, high saving and investment rates, and some of the highest trade-to-GDP
ratios in the world, were not immune to financial turbulence (Claessens and Glaessner 1997).
The past 15 years saw the emergence of new financialinstruments and new types of institutions. To facilitate thedevelopment of financial institutions, restrictions onlenders were eliminated, and innovative ways of doingbusiness were adopted. Financial deregulation and liber-alization were attempted in parallel. Technological changemade the global financial environment a very differentplace. It helped the volume of private international capi-tal flows to grow at unprecedented rates.
Like other multilateral financial institutions, theWorld Bank has had to adapt its financial operations toadjust to the changes. It developed a comprehensiveapproach on financial sector issues in developing coun-tries and incorporated its findings in the Bank’s opera-tional manual. The financial sector Operational Direc-tive (OD) 8.30 (World Bank 1992), which is now beingrecast as Operational Policy (OP) 8.30, includes majorelements of sector development strategy and recommen-dations on the use of lending instruments for bothadjustment and financial intermediary lending. Recentdevelopments in East Asia confirm the wisdom of a pru-dent and comprehensive approach to financial sectorreform. However, the severity and breadth of the crisisand its complex antecedents call for a special review ofissues such as, economic governance, industrial policy,
interlocking ownership of enterprises and banks, foreigninvestment regulations and debt management. Thisstudy makes no such attempt.
Financial sector reform is defined as a set of actionsaimed at reducing or eliminating distortions in financialmarkets, deepening the financial sector, and strengthen-ing financial institutions. This study assesses the Bank’soperations relating to financial sector reform, the devel-opment of financial institutions, and the management offinancial crises in countries that approach it for financial,technical, or advisory support. Bank-supported opera-tions comprise two elements: financial sector liberaliza-tion (improving market structure and competition byremoving price and quantity distortions) and institu-tional infrastructure development (strengthening individ-ual financial institutions and prudential regulation andsupervision). The balance between these two has becomea major issue in light of the growing number of financialcrises in developing countries.
This study evaluates the result of a series of reformssupported by the Bank over the past decade (FY85–FY96). Since the Bank’s support of financial sectorreforms is relatively recent, many reform programs areongoing, and their outcomes are assessed for the mostrecent years of implementation. The analysis begins with
2
a country’s initial conditions, comparing stated reformobjectives with the conceptual framework. It then followsthe implementation of reforms, comparing their outcomeand impact with program objectives, using OD 8.30 cat-egories and related indicators as monitoring devices.
Another key objective of the Bank’s financial sectoroperations is to prevent major financial crises and tomitigate their impact when they do occur. The studyassesses the effectiveness of the Bank in this by examin-ing its contribution in preventing or attenuating therecent crises in Mexico, Venezuela, and the Philippines.Discussion of a potential financial sector role for theBank beyond its current country focus, however, liesoutside the scope of this study.
The Conceptual FrameworkThe framework of the Bank’s operations relating to thefinancial sector has been dictated by the ferment of ideasin the academic literature, as modified by experience inthe countries that undertook financial reforms over thepast two decades. The basic premise of financialreform—indeed, of all economic reform—is that gov-ernment intervention should not prevent market signalsfrom directing the allocation of resources. Voluntary,
market-based decisionmaking isfacilitated by an efficient financialsystem. Until the 1980s, however,the development and efficiency offinancial systems were severelyundermined by governmentefforts to promote economicdevelopment through interest ratecontrols, directed (and subsi-dized) credit programs, and heavyfiscal burdens. The result was
more distorted financial sectors, suboptimal savingrates, and misallocation of investment. Empiricalresearch shows a negative correlation between financialrepression and the real rate of growth. The financialreforms articulated by such intellectual pioneers asMcKinnon (1973) and Shaw (1973) included sweepingaway interest rate ceilings and directed credit programs,and making markets more competitive.
Increasing experience with financial reform, byitself, and as part of general economic reform, hasbrought profound changes in operational modes and inapproaches to financial liberalization. The new per-spective is affected by differences in the economic andinstitutional environment in which financial liberaliza-
tion has been initiated and implemented, the instru-mentality and interpretation of financial deregulation,and the relationship between fiscal deficits and financialreform. In the past, interest ceilings were dismantled inseveral developing countries without regard to the stateof the economy, the state of the banking system, the lia-bility structure of borrowing firms, and a host of otherconditions. Macroeconomic stability and reasonablysound banks are now viewed as prerequisites for finan-cial reform. Similarly, experts realize that financialliberalization may not improve financial institutions’allocation of resources if price distortions arising fromprotection and price controls persist. Policymakers nowknow that fiscal deficits must be reined-in before finan-cial liberalization can be fully implemented (Caprio andothers 1994; Cho and Khatkhate 1989; Khatkhate1996).
Recent academic forays into financial liberalizationhave highlighted problems in evaluating the effectivenessof financial reform. Until recently, the credit marketsthat prevailed before government intervention wereassumed to be perfect. But the market under reformshould be compared not with a perfect one, but with thedistorted one that existed before reform (Gertler andRose 1994). A rise in interest rates following financialliberalization causes borrowers’ net worth to fallbecause of the declining value of collateralizable assetsand the discounted value of future income streams. Anerosion of borrowers’ net worth raises the premium onexternal finance, reducing investment below the opti-mum. Since borrowers’ net worth links the financial sys-tem and the real sector, financial liberalization tends tobring about a shrinking of economic activity if properprecautions are not taken. Such insights affect not onlyhow to evaluate the effectiveness of financial reform buthow to sequence reforms (Caprio and others 1994).
The analytical underpinnings of the Bank’s evolvingnormative framework have been supported by operationalguidelines on prudential control and bank supervision,included in the core principles of the Basle Committee onBanking Supervision. Drawing on worldwide experience,25 principles were formulated, covering the prerequisitesfor effective banking supervision, licensing, and structure;prudential requirements; methods of bank supervision;information requirements; and cross-border banking.Since 1997 these have been a valuable reference, as is theIMF’s recent publication “Towards a Framework ForFinancial Stability,” which takes the blueprint for actionfurther.
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
Empiricalresearch shows anegative correlationbetween financialrepression and thereal rate of growth.
3
The Bank’s Approach to Financial Sector OperationsThe Bank’s approach has evolved, absorbing ideas fromthe academic literature and modifying them in the lightof experience in developing countries particularly andindustrial countries generally. The principle governingthe Bank’s policy has been that financial sector reformmust be viewed in the context of a country’s overall eco-nomic reform.
In the 1960s, when the Bank first became involvedin financial sector operations, the focal point for finan-
cial sector lending was often sta-ble interest rates in the credit anddevelopment institutions, (mostlygovernment-owned or govern-ment-dominated). Over time, theBank broadened its focus, recog-nizing that a more volatile worldfinancial environment createdpressures to liberalize the finan-cial sector, but that financialderegulation had to take intoaccount the country context. Sta-ble macroeconomic conditionsand strong regulations were pre-requisites for launching financialliberalization. Drawing fromexperience, the Bank supportedmarket-based lending undertaken
by competing financial institutions, the phasing-out ofsubsidized and targeted credit programs, and thestrengthening of policy, regulatory, and institutionalinfrastructure. This was predicated on providing anincentive framework through which the financial sectorcould gradually respond to market signals. In brief, theBank’s policy until the late 1980s was to support:
• Specific policy reforms to make the financial sys-tem more efficient.
• The creation or strengthening of financial institutions.• The promotion of finance to microenterprises
(World Bank 1989a).As experience with financial operations accumu-
lated, the Bank came to recognize the constraints of apurely market-based approach. Developing countriesoften lacked the staff, management skills, and legalframework needed for a viable banking system; theyneeded “properly designed administrative interventionto eliminate the existing barriers to market develop-ment” (World Bank 1989a). To deal with these con-straints, the Bank revised its policy in the early 1990s to
combine policy reform with support to specific institu-tions. It took a systemic approach to financial sectorinterventions. Explicit directions based on this approachare embodied in OD 8.30 (World Bank 1992) whichdefined “financial sector operations as structural andsectoral adjustment loans and technical assistance loansin support of financial sector reforms, as well as techni-cal assistance and all investment loans through financialintermediaries in the borrowing countries.” It “puttogether a policy perspective that concentrates on finan-cial liberalization and macro linkages, with an institu-tional perspective that concentrates on lending instru-ments and institutional development for the financialsector.” Thus OD 8.30 articulated directives to link thepolicy-based approach aimed at reducing financial sec-tor distortions with the institutional approach aimed atachieving specific objectives in the financial and real sec-tors. One result was to discourage directed credit—witha consequent dramatic decrease in lending aimed atsmall and micro-enterprises. The Bank has recentlyredrafted this directive to provide greater flexibility fortargeting intermediation lending (March 1997), clarifiedits approach to current financial sector issues (April1997), and strengthened its capacity to intervene rapidlyin case of financial crises (December 1997).
The Bank’s new view is that a well-functioningfinancial system is an ongoing process, not an event.Such a system requires a long-term view. In addition, thedesign of financial sector operations must be based onthorough economic and sector work (ESW), whichclearly articulates an economic environment in which afinancial sector can grow efficiently. The Bank now fullyacknowledges the importance of reducing financialfragility. The Bank’s new strategy emphasizes the need toestablish appropriate incentives in the client countriesfor prudent banking and to sequence adjustment policiestowards the ultimate success of the reform program.
The Bank’s Financial Lending InstrumentsThe Bank uses six types of lending instruments for finan-cial sector development. Five are aimed directly at thesector level: structural adjustment loans (SALs), sectoradjustment loans (SADs), technical assistance loans(TALs), and, occasionally, specific investment and main-tenance loans (SIMs) and specific investment loans(SILs). A sixth type, financial intermediary loans (FILs),supports reform through operations handled by individ-ual financial institutions. At times, FILs include a sectorreform program with a policy letter attached.
T h e E v a l u a t i o n F r a m e w o r k
Over time, theBank broadened itsfocus, recognizingthat a more volatileworld financialenvironmentcreated pressuresto liberalize thefinancial sector, but that financialderegulation had totake into accountthe countrycontext.
4
The Bank has made extensive use of financial sectoradjustment loan/credits (FSAL/C) to characterize loanswhose primary objective is to promote financial sectorreform.1 These are mainly—but not exclusively—SADs.FSALs, which acquired prominence in the 1980s, aredesigned to remove policy and institutional impedimentsto financial sector development. Their main objective isto bring about financial deepening and to make thefinancial sector more efficient and more responsive tomarket forces. They cover the full spectrum of financialsector issues, including bank restructuring, developmentof human resource skills, encouragement of competi-tion, and strengthening of financial infrastructure. Theyemphasize the importance of macroeconomic stability tosustain development of the financial system.
SALs have broader policy objectives. Even whenthey do not address financial sector issues directlythrough large financial sector policy components withfeatures similar to FSALs, they still have an indirecteffect on financial sector development, often includingconditionalities on two key relative prices—exchangerates and interest rates.
TALs have more limited but sharply-defined objec-tives: building the institutional capacity to manage a lib-eralized, market-based system and to sustain the positiveeffects of reform. Lending for technical assistance can beeither stand-alone (TAL) or part of a SAL, SAD, or FIL.
From FY85 to FY96 the Bank supported financialsector reform through 26 SALs for 21 countries, 33SADs (with strong financial components) for 28 coun-tries, 10 FILs for 10 countries, 8 TALs for 8 countries, 2SIMs for 2 countries, and 8 SILs for 8 countries.
As for regional distribution, the 88 adjustment-related operations show some concentration in LatinAmerica and the Caribbean (LAC) (more than 30percent of the world total), followed by Africa, andEurope and Central Asia (ECA) (figure 1.1). About 40percent of all adjustment-related operations are in SALsand 30 percent are in SADs, which include FSALs (seeAnnex Table 1.1). The adjustment operations are almostequally divided between ongoing and completedoperations.
The Bank’s Non-lending Instruments for Financial SectorReformThe Bank has prepared ESW reports analyzing manyissues bearing on the financial sector, taking intoaccount the macroeconomic settings. ESW reports mayhelp countries to develop a perspective on their prob-
lems with financial sector development and to designappropriate strategies to deal with those problems.Although ESW is supposed to precede financial sectorprojects, more often than not the reports follow projectpreparation and often even project implementation. TheBank issued 109 financial sector-related ESWs for 70countries between FY81 and FY95 (see Annex Table1.2).
Sample Coverage and MethodologyThe sample used for the study varies depending on theavailability and comparability of data and their quality.For evaluating the projects’ effects on the financial sys-tem and the real economy, and for evaluating the appro-priateness of project design, a sample of 23 countrieswas chosen (out of 58) with 43 financial sector adjust-ment-related loans (out of 88). These loans, called FSLsin this study, include mainly SADs, SALs, and TALs, andsome FILs with large financial sector policy compo-nents, as well as a few SIMs and SILs. The sampleincluded 29 completed and 14 ongoing projectsapproved between FY85 and FY96 (see Annex Table2.1).2 For assessing the Bank’s policies in the financial
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
FIGURE 1.1: REGIONAL DISTRIBUTION OFADJUSTMENT-RELATED OPERATIONS IN THEFINANCIAL SECTOR, FY85–96
AFR(22%)
EAP(8%)
ECA(21%)
LAC(34%)
MNA(8%)
SAS(7%)
AFR = AfricaEAP = East Asia and PacificECA = Europe and Central AsiaLAC = Latin America and the CaribbeanMNA = Middle East and North AfricaSAS = South Asia
Source: World Bank projects database
5
sector, this sample is considered sufficiently representa-tive of the universe of countries in which the Bank hadfinancial sector operations.3 The sample used for ana-lyzing institutional development includes 28 countrieswith 21 completed FSLs4 and 52 completed FILs.Admittedly, this sample is larger than the first; this isbecause the accent in this sample is on qualitative ratherthan quantitative indicators, and qualitative indicatorsare less demanding in time-series data. Varying the sam-ple size according to what is targeted for assessmentshould not introduce biases in conclusions, as a broadpreliminary examination of the Bank’s financial sectoroperations has shown that overall performance amongborrowing countries differs only at the margin, if not inspecific details.
We used the first sample to assess the effects of Bankassistance to reform the financial sector through adjust-ment lending. The methodology used (detailed in Box1.1 and in Table 1.1) traces Bank intervention at thecountry level, measuring stated objectives and instru-
ments employed against theimplementation record (outcomeand impact). Bank-supportedobjectives include improving mar-ket structure, enhancing competi-tion, and strengthening the finan-cial system. The instruments usedto achieve those objectives werepolicies, regulations, and institu-tional reform (see Annex Table2.2). “Outcome” means the directeffects of the policy instruments
used on the financial sector; “impact” means the finaleffects on the real economy. Both outcome and impactare measured by quantifiable indicators, such as thelevel of real interest rates, the spread between domesticlending and deposit rates, investment and saving ratios,the GDP growth rate, and the ratio of monetary assetsto GDP. These indicators are constructed on the basis ofBank research and experience from several financial sec-tor operations over the years, as reflected in OD 8.30.5
Indicators measure outcome and impact at the end ofthe reform period against of objectives set earlier. Therating scale adopted is binary, with “1” indicatingsatisfactory and “0” indicating unsatisfactory. The rat-ings for various indicators are aggregated and averagedout to assign an overall rating to a reform program.An overall rating equal to or below 0.5 is consideredunsatisfactory.
We used the second sample to evaluate FSLs andFILs for institutional development (ID), so the indicatorsare qualitative, not quantitative (see methodology inBox 3.1). Designed to assess performance in institu-tional development, they are based on findings in staffappraisal reports (SARs), project/implementation com-pletion reports (PCR/ICRs), and project performanceaudit reports (PARs). We devised a “score card” (seesample in Annex Table 3.1), which includes componentsor instruments, common to institutional development,giving equal weight to each instrument. Achievementunder each instrument is also scored using a binary sys-tem. The instruments include bank restructuring,improving financial infrastructure, and improving port-folio quality. One-by-one, 24 FSL (excluding 12 ongoingFSLs) and 58 FIL operations were scored. The scoreswere then aggregated and averaged.
To assess financial crises, we used the case studymethod for three countries(Mexico, the Philippines, andVenezuela.6 The assumption in using case studies wasthat experience in dealing with those crises was moreuseful than pure statistical indicators. We formulatedvarious financial indicators suited to the diagnosis andmonitoring of financial crises, and used these to con-struct a financial fragility index (FFI).
Data LimitationsThese were significant. Comparing countries was diffi-cult because statistical variables used were not compara-ble, so measures of outcome and impact should be takenmore as broad estimates of performance than as precisemeasures of how things went right or wrong.
Major limitations include:• Most adjustment programs in the sample are
relatively recent, so it is difficult to measure out-come or impact. Liberalization programs can beviewed as successful only when the reform issustained.
• The macroeconomic indicators emphasized in thestudy are generally useful for a long-term perspec-tive, but they are more reliable if supported by amicroeconomic analysis, such as an examinationof how the allocation of credit changes afterreform. The Bank has made some attempts in thisarea, but these were limited by the absence offirm-level data (Schrantarelli and others 1994).The Bank is carrying out research to develop andrefine the methodology for measuring the impactof reform on resource allocation.
T h e E v a l u a t i o n F r a m e w o r k
Bank-supportedobjectives includeimproving marketstructure,enhancingcompetition, andstrengthening thefinancial system.
6
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
T he methodologyin this study issimilar to the one
OED uses to evaluateprojects in general. Webegin by assessing initialconditions, then projectrelevance, outcome withrespect to relevant objec-tives, and the longer-termimpact. In this study,however, we put moreemphasis on the use ofperformance indicators,and we apply themethodology to bothcompleted and ongoingprojects. We focus moreon performance at thecountry level for a seriesof reforms supported bythe Bank rather than on asingle batch of reformssupported by one Bankoperation. Toward thisend, we use the word“program” instead of“project” to discuss ourfindings.• By initial conditions,
we understand thestate of macroeco-nomic and financialsector variables in thetime before approvalof a program’s firstproject. This time isusually up to a fewyears ahead of thefirst project approval.
• By program objec-tives, we understanda series of objectivesover time that gov-ernments establishedto reform their finan-cial sectors and forwhich they askedBank support.
• Program outcomerefers to changes thatoccur in outcomeindicators relative totheir values at the ini-tial conditions stage,as well as measure-ment of outcomeswith respect to theoriginal objectives.
• The real sectorimpact refers tochanges that occur inimpact indicators rel-ative to their valuesat the initial condi-tions stage. These arelonger-run effects ofthe financial sectoron the real sector.Changes in impactindicators areexpected to be posi-tively correlated withchanges in programoutcomes.The performance
indicators to evaluatefinancial sector reformcomponents are continu-ous time-series that fol-
low initial conditions,outcome, and impact of aparticular aspect of thereform. In this way finan-cial sector reform is mon-itored as an ongoingprocess and not as a one-time event (Sheng 1996).
In the rating of initialconditions, outcome, andimpact of a reform com-ponent, we use a binaryscale to interpret ourresults. For instance, takea reform program whoseobjective is to attain posi-tive real interest rates. Ifthe real deposit rate in acountry is negative andhas been so for a fewyears before approval ofthe program’s first proj-ect, we score the relevantindicator (real depositrates) as “0” or unsatis-factory, as an initial con-dition. However, after thefirst project is approvedand implementation is inprogress, if the realdeposit rate begins turn-ing positive (even if it isstill very low), we score itas a “1” or satisfactory,as an outcome of onecomponent of the reformprogram. Performanceindicators for all reformcomponents are scored inthis manner.
The reform compo-nents and the perfor-mance indicators moni-toring them are classifiedaccording to the generalfinancial sector objectivesprovided in OD 8.30: (1)macrostability policies,(2) financial sector poli-cies, (3) soundness of thebanking system, and (4)impact on the real sector. • Overall program out-
come ratings at thecountry level areobtained by calculat-ing a simple averageof performance indi-cators of macrostabil-ity policies, financialsector policies, andsoundness of thebanking system.
• Overall impact rat-ings are obtained bycalculating a simpleaverage of perfor-mance indicators inthe real sector cate-gory only.If the average score
for outcome indicators ismore than 0.5, we ratethe outcome as satisfac-tory. If the average is lessthan or equal to 0.5, werate the outcome asunsatisfactory. Impactratings are derived in thesame way.
BOX 1.1: METHODOLOGY USED TO ASSESS FINANCIAL SECTOR REFORMS
7
T h e E v a l u a t i o n F r a m e w o r k
TABLE 1.1: PERFORMANCE INDICATORS FOR THE FINANCIAL SECTORPERFORMANCE
OBJECTIVES OF POLICY PERFORMANCE INDICATORS OFADJUSTMENT POLICY IMPLEMENTATION INDICATORS OF IMPACT ON
PROGRAMS INSTRUMENTS RECORD OUTCOME REAL SECTOR
a GDS = gross domestic savings.b GDI = gross domestic investment.
Remove interestrate and creditdistortions
Removeimpediments tocompetition
Strengthenfinancial sectorinfrastructure
Strengtheninstitutions
Stablemacroeconomicenvironment
• Interest rateliberalization
• Move to indirectinstruments ofmonetary control
• Dismantling ofdirected credit
• Opening capitalaccount
• Privatization of banks
• Entry/exit laws
• Removal ofdifferential taxationof banks and otherfinancialintermediaries
• Foreign ownershiplaws
• Legal framework forcentral bank
• Legal framework forbanks
• Regulatoryframework for non-banking sector
• Identification ofrights and obligationsof financial agents
• Strengthening ofbanking supervision
• Bank restructuring• Bank institutional
reforms
• Number and categoriesof interest ratesliberalized
• Treasury bill rate,interest rate on reserves
• Directed credit as shareof total credit
• Level of foreignexchange controls
• Private assets as shareof total assets
• Number of new banks,concentration ratio
• New/old tax structureof banks
• Assets of foreign bankas share of total assets
• Central bank law
• Prudential regulations,banking company law
• Non-bank assets/totalfinancial sector assets,number of listedcompanies in stockexchange
• Liquidation andbankruptcy rules,payment systems,accounting and auditingstandards
• Capacity and authorityto supervise, licensingcriteria, supervisionsystems
• Capital replenishment,asset liquidation,independence andcapacity of board, MIS,human resourcedevelopment programs
Real positiveinterest rates,term structure ofinterest rates
M2/GDP
Banking systemfinancial ratios,lending/depositspread
Banking systemand central bankfinancial ratiosmarketcapitalization
Individual bankfinancial ratios
GDP growth
GDS/GDP,aGDI/GDPb
Level of directforeigninvestment
• Low inflation • Reduced current account deficit• Reduced fiscal deficit• Equilibrium of exchange rate
8
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
• Judgments of the health of banking systems aremade using indicators such as capital adequacy, butthese indicators do not throw light on what needsto be done to make banks sounder. For that, for-ward-looking indicators must be formulated. Thiscan be accomplished only through sophisticatedstatistical techniques of “stress testing” or“dynamic solvency testing” that can show howbanks withstand various shocks and how the risksthey face are correlated. The models based on thesetechniques focus attention on how banks makedecisions about portfolio assets and liabilities,avoiding the trap of universally prescribed ratiosfor different risk positions. So the evaluations ofbanking health are suggestive rather than definitive.
• The performance indicators used in the analysisare difficult to update even for inflation andM2/GDP (financial depth), variables for whichgood historical information is generally available.These performance indicators rely on both centralbank and national accounts data, which areupdated at different times. The banking sectorpresents the most problems. Data for indicatorslike asset quality (loans in arrears/total loans) andmanagement efficiency (expenses per unit ofprofit) are often not available.
The Plan of the StudyIn chapter 2 we evaluate the overall outcome and impactof financial sector reforms that the Bank supportedthrough FSLs. For this evaluation, which includes manyongoing loans, we use performance indicators derivedfrom the OD 8.30 conceptual framework (see Table1.1). Chapter 3 focuses on institutional development,with a larger sample including both FSL and FIL opera-tions, and a more detailed institutional “score card,”also derived from OD 8.30. Chapter 4 evaluates the out-come and impact of Bank intervention in financialcrises, with a modified set of performance indicators(fragility indicators) better adapted to identify crises andto monitor recovery. In chapter 5 we evaluate Bank per-formance in identifying and analyzing sector issuesthrough ESW; preparing and implementing lendingoperations, as well as overall sequencing issues in lend-ing strategies; and (3) coordinating the IBRD, IFC, andIMF in providing financial sector services to client coun-tries. In chapter 6 we conclude how the Bank fared whenwe use the operational directive OD 8.30 as a bench-mark, and what is needed beyond the directive to keepimproving support to financial sector reforms.
9
Financial SectorReform Operations
22
This chapter focuses on lending operations—mainly SALs, SADs, and TALs—that
support reform of the financial sector and sectorwide institutional changes. The
only FILs discussed in this section are those that included a sector reform program
with a policy letter.
After sketching the economies of the sample countries, we examine whether Bank
operations dealing with financial sector reform were appropriately or inadequately designed
(considering the initial conditions and the types of policiesrequired to improve efficiency), to see how well theyachieved their objectives over FY85–FY96. Then we ana-lyze the projects’ results over the same period.
Characteristics of Financial Systems in Sample CountriesOf the 23 countries in the sample, most had liberalizedinterest rates before undertaking financial reform, andthey had positive real deposit rates. In most countriespositive real interest rates were sustained over a periodup to and including the time when financial operationswere implemented, except in Mexico, Venezuela, andpossibly India. Some countries had introduced othertypes of liberalization—for example, the use of indirectmonetary instruments, a reduction in subsidized credits,and the easing of entry conditions for foreign or privatebanks (Indonesia, Malaysia, the Philippines). In manycountries (including Chile, Ghana, Indonesia, Mexico,the Philippines, and Turkey) previous adjustment pro-grams had included liberalization of the capital account.
Banking systems in many of the sample countrieswere highly concentrated. Banks were generally publicly-owned (in China, Ghana, Indonesia, the Republic ofKorea, Tanzania, and the South Asian countries), thoughthere were some private banks as well (in Chile, Mexico,
Turkey, and Venezuela). The oligopolistic and conglom-erate structure prevailing in many banking sectors pre-vented competition. A systematic review of the BankPresident’s Reports (PRs) and SARs for the sample coun-tries indicates that at the beginning of the adjustmentprogram only 17 percent of the countries had satisfactorycompetition in the financial sector (see Table 2.1).
Banks in several sample countries (such as India,Mexico, Pakistan, Poland, and Turkey) were initiallyclose to insolvency. Capital adequacy fell far short ofBasle standards. Banks were also inefficient and wereunable to meet the economy’s long-term resource needs.Publicly-owned banks, in particular, fell into this category.
The depth of the financial system, as measured bythe ratio of M2 to GDP, averaged close to 30 percent forsample countries on the eve of operations.1 Two-thirdsof this was accounted for by quasi-money and the restby narrow money (M1). The M2-to-GDP ratio washighest for India (50 percent), followed by Pakistan andChile (40 percent), Turkey and Tanzania (around 37percent), Venezuela (33 percent), Bangladesh and thePhilippines (30 percent), Indonesia (27 percent), andMexico and Ghana (15 percent). Deposit rates were sub-stantially negative in Mexico and Venezuela (below -50percent), and were also negative for Ghana (-22 percent)
10
and Tanzania (-9 percent). The domestic spreads inBangladesh, Indonesia and the Philippines were reason-able—4 to 5 percentage points—but they were high inChile, Ghana, India, and Tanzania (7 to 16 percentagepoints), and negative in China (-14 percentage points)and Venezuela (-6 percentage points).
The Situation in the Real SectorAnnual real GDP growth rates in all the sample countrieswere positive before reform, the average being 4.5 percent.Bangladesh, Tanzania, and Venezuela stood at the lowend, with growth rates declining to zero in Bangladesh andto -2 percent in Venezuela. At the high end, Turkey wasgrowing at 8 percent a year, followed by the Philippines atabout 6 percent. Gross domestic saving and investmentrates were mixed. Tanzania’s saving rate was almost zeroand those of Bangladesh and Ghana were barely morethan 2 percent and 5 percent of GDP, respectively. But thesaving rates in China and Indonesia stood at 30 percent,followed by 25 percent for Venezuela, and 22 percent forIndia. Chile, Mexico, the Philippines, and Turkey all had
saving rates between 15 and 20 percent at the beginning ofreform, but saving rates were falling in Mexico and thePhilippines, and rising in Chile and Turkey. In general,there was an imbalance between domestic savings andinvestment in most of the sample countries, with grossdomestic investment exceeding gross domestic savings.
Despite previous attempts at adjustment and stabi-lization, many countries launched their financial reformoperations in an unstable macroeconomic environment.Mexico, Turkey, and Venezuela had been suffering fromchronically high inflation. On the other hand, macro-economic stability prevailed in Ghana, Indonesia (Box2.1), and Chile (Box 2.2).
Program Design
Initial Conditions in the Financial SectorThe overall assessment of Bank programs begins byexamining initial conditions (see Box 1.1). To measurehow much initial conditions changed, or were modifiedby reform, we used performance indicators correspond-
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
TABLE 2.1: INITIAL FINANCIAL SECTOR CONDITIONS IN COUNTRIES STUDIEDCOUNTRY STATE OF(Fiscal year of FINANCIAL FINANCIALfirst financial DEGREE OF SECTOR STRENGTH OFsector operation) DISTORTIONS COMPETITION INFRASTRUCTURE INSTITUTIONS
Bangladesh (1990) 0 0 0 0Bolivia (1987) 0 0 0 0Chile (1986) 1 1 1 1China (1993) 0 0 1 0Côte d’Ivoire (1992) 0 0 0 0Egypt (1992) 0 0 0 0Ghana (1988) 1 0 0 0India (1995) 0 0 0 0Indonesia (1988) 0 0 0 0Kenya (1989) 0 0 1 0Korea (1985) 0 1 1 0Malawi (1991) 0 0 0 1Malaysia (1987) 1 1 0 0Mexico (1989) 0 0 0 0Morocco (1986) 0 0 1 0Pakistan (1989) 0 0 0 0Philippines (1989) 1 1 1 0Poland (1991) 0 0 0 0Senegal (1987) 0 0 0 0Tanzania (1992) 0 0 0 0Tunisia (1988) 0 0 1 0Turkey (1986) 0 0 0 1Venezuela (1990) 0 0 0 0
Percent satisfactory 17 17 30 13
Note: “1” is satisfactory on balance at time of appraisal, “0” is unsatisfactory.Source: PRs, SARs, and OED estimates based on time-series from World Development Indicators 1997 and International Finan-cial Statistics (IFS).
11
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
Initial conditions.Financial adjustmentoperations in Indone-
sia were undertakenwithin the framework ofBank support for abroader structural adjust-ment program. Followinga drop in oil prices in theearly 1980s and their col-lapse in 1986, Indonesiabegan a program thatincluded a major depreci-ation of the rupiah, areduction in publicspending, measures tostimulate exports andencourage foreign invest-ment, improvements toindustrial efficiency byopening up the economy,basic financial reform,and a long-term strategyto diversify the economyaway from an over-dependence on oil. Thisprogram entailed signifi-cant relaxation of the sys-tem of close regulationthat had characterized theIndonesian economy.
Bank initiatives. Lendingsupport for adjustmentwas provided after thefact, endorsing actionsthat were judged to con-form with the Bank’sassessment of Indonesia’seconomic priorities. Ini-tial support in the formof two trade policy loansin 1987–88 was followedby two private sectordevelopment loans
(PSDLs) in 1989–90, allwith financial sector com-ponents, and a financialsector development pro-ject in 1993. During thisperiod the Bank alsoapproved six FILs. TheBank’s operations havesupported the evolutionof Indonesia’s financialsystem from a state-domi-nated one, with fixedinterest rates and admin-istered credit, to a liberal-ized, competitive, andincreasingly privately-owned network, withmarket-determined inter-est rates. This has meantdeveloping an indirectmeans of monetary con-trol and strengthening thecentral bank’s supervisoryframework and enforce-ment capabilities.
Outcome. The heart of thegovernment’s adjustmenteffort in the second half ofthe 1980s and the early1990s was to restoreexternal equilibriumthrough real depreciationof the exchange rate,bringing public accountsinto better balance, open-ing the economy graduallyto increased competition.In the financial sector theliberalization of interestrates introduced under thetrade policy adjustmentloans, coupled withincreased public confi-dence in the banking sys-
tem, has led to substantialfinancial deepening. Themeasures introduced underthe PSDLs—especially per-mitting the entry of for-eign banks—increasedcompetition in the bankingsector to some extent. Theoverall structural adjust-ment program has beensuccessful in generatinghigh rates of saving andinvestment, and realizingrapid economic growth:but it did not succeed inreducing substantially thefragility of the bankingsystem, which in 1997became vulnerable to theeffects of the currencycrisis.
Assessment. After oilprices collapsed, Indone-sia’s immediate need wasto depreciate its currencyto restore external equi-librium and to cut publicspending because of thedecline in governmentrevenues and borrowingprospects. But to restorerapid growth from non-oil sources required fun-damental structuraladjustment. Initiatingstructural adjustmentafter implementing thebasic measures needed torestore macroeconomicstability was entirelyappropriate. The earlyinitiation of financial sec-tor reform was alsoappropriate. But liberaliz-
ing the financial sectorwithout first ensuringthat adequate legal, regu-latory, and supervisoryinfrastructure was inplace led to the prolifera-tion of weak banks andto a number of bank fail-ures, as clearly shownduring the currency crisisof 1997.
Lessons of experience.The audit report for thePDSLs draws a numberof important conclusionsabout the correctsequencing of reforms.First, the legal, regula-tory, and supervisoryenvironment of the finan-cial sector must bestrengthened before or atthe same time as entry isliberalized. Second, estab-lishing a strong legalinfrastructure for bankingactivities should have pri-ority from the start,rather than be one itemon the continuing agenda.This would have sent aclearer signal to the pri-vate sector that the gov-ernment was irrevocablycommitted to a funda-mental change in rela-tionships. Only inNovember 1997, did theBank approve a technicalassistance loan to directlystrengthen the structureand improve the sound-ness of the bankingsystem.
BOX 2.1: INDONESIA AND THE SEQUENCING OF REFORM
12
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
ing to OD 8.30’s five broad categories of financial sec-tor reforms: stability of the macroeconomic environ-ment, extent of distortions in the financial system (suchas negative interest rates and the existence of directedcredit), degree of competition, level of development offinancial infrastructure, and strength of the financialinstitutions. Initial conditions were assessed by studyingSARs of financial sector operations in our sample coun-tries (see Annex Table 2.1). Initial conditions were char-acterized as either satisfactory “1” or unsatisfactory “0”in all four categories (Table 2.1).
About 30 percent of the 23 sample countries hadfavorable initial conditions in development of theirfinancial sector infrastructure. About 17 percent wereconsidered to have acceptable levels of financial distor-tions. Only 17 percent of the countries began reformwith a high degree of competition, and only 13 percenthad financially-sound banks.
There were, of course, variations in initial condi-tions. Chile had satisfactory initial conditions in all fourareas; the Philippines had favorable initial conditions inthree areas. Korea and Malaysia had good initial condi-tions in two areas. Bangladesh, Bolivia,2 Egypt, India,Indonesia (Box 2.1), Mexico, Pakistan, Poland, Senegal,and Tanzania had the worst overall initial positions.
The Presence and Relevance of Bank PoliciesThe policies included in the country programs varied:
• Liberalization of interest rates was one of the for-mal loan conditions in: Ghana, Mexico, Tanzania,Turkey, Venezuela, and the South Asian countries.Ghana, Mexico, and Turkey complied with thiscondition at least partially, but neither Tanzanianor Venezuela complied at all. Compliance inBangladesh was admittedly formal. Although thedegree and quality of compliance in Pakistan had
Initial conditions.Chile was especiallyvulnerable to the
1982 debt crisis. Afterseveral years of stronggrowth in the late 1970s,its currency was highlyovervalued, its domesticsaving rate was low,supervision of the finan-cial system was lax, andforeign indebtedness hadrisen sharply. The rise ininterest rates and theabrupt termination of for-eign lending in 1982 ledto Chile’s worst economiccrisis since the GreatDepression. Economicgrowth then resumed,buoyed by the sustainedreal exchange rate devalu-ation, reduced imports,declining public spending,and rising private invest-
ment. After 1985, follow-ing the adoption of alonger-term program ofstructural adjustment,imports were further liber-alized, export earningsimproved, and the debt-GDP ratio declinedrapidly. Since 1988 Chile’seconomic growth has beenrapid, averaging morethan 7 percent a year.
Bank initiatives. TheBank strongly supportedChile’s growth-enhancingpost-1984 strategy withthree SALs (FY86–FY89),two industrial financeprojects (FY86–FY91), asmall- and medium-sizedindustry project (FY86–FY93), and a financialmarkets loan (FY91–FY93). The first two
SALs supported the gov-ernment’s structuraladjustment program(SAP) for 1985–87, theinitial goals of whichwere to accelerate anddiversify exports, correctChile’s savings/investmentdisequilibrium, andreduce the amount ofstress in the financial sys-tem. The second phase ofthe SAP, starting in late1986, focused on recapi-talizing banks and corpo-rations, and sought a sub-stantial increase in publicsavings, a reduced butmore efficient export-sup-portive public investmentprogram, strengthenedexport incentives, andimproved social security.The SAP thus incorpo-rated simultaneous and
mutually-reinforcingfinancial and real policyreforms.
The Bank’s FinancialMarkets Loan (FML) wasa hybrid loan with twobroad objectives: to sup-port policy changes aimedmainly at deepening thesecurities markets and toprovide term funds to theequipment-leasing sector.The policy component’sobjectives were to ensureprofitable investmentopportunities for theincreasing assets of thepension funds and the lifeinsurance companies, tostrengthen the banks, tosupport Chile’s balance ofpayments, to developCORFO (the state devel-opment bank) as a sec-ond-tier lender to the leas-
BOX 2.2: CHILE: A SUCCESS STORY THE SECOND TIME AROUND
13
not been explicitly reported, references to highinterest rates suggested interest rate liberalization.
• Credit reform was part of financial reform oper-ations in Bangladesh, Indonesia, Mexico, Pak-istan, the Philippines, and Turkey. Indonesia hadalready started phasing-out liquidity creditsfrom the central bank before reform, but thatrequirement was still included as part of condi-tionality. In the Philippines, directed creditschemes were transferred from the governmentto a development bank slated for privatization.Under Pakistan’s only FSAL and Turkey’s sec-ond, directed credit was curtailed to an extentsatisfactory to the Bank; this was not the casewith Mexico. In Bangladesh the directed pro-gram of lending to priority sectors through con-cessional refinancing from the central bank wasformally abolished.
• Indirect monetary instruments stressed by the IMFwere introduced as part of reform operations inBangladesh, Ghana, Indonesia, Tanzania, andTurkey. In all five of these countries, open marketoperations were initiated, but their effectiveness incontrolling the money supply was uncertain. InGhana, for example, the central bank had to beconvinced that it should use open market opera-tions as a monetary control instrument rather thanas a means of financing the fiscal deficit, thoughboth uses of open market operations are notmutually exclusive.
• Bank privatization was part of reform operationsin Chile (Box 2.2), India (1995), Indonesia, Mex-ico, Tanzania, and Venezuela.
• In Indonesia, measures to enhance competitionincluded reducing differentiation between publicand private banks by allowing state enterprises to
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
ing companies, and tostrengthen the govern-ment’s capacity to super-vise the financial sector.
Outcome. Chile’s struc-tural adjustment programwas remarkably successfuland in most respectsstands as a model forother countries facingsimilar initial problems.Chile’s growth perfor-mance over most of thepast decade has been thebest in Latin America. Itconsiderably exceededexpectations. Importsgrew more slowly in1985–88 than in 1982–84, exports grew morequickly, consumption andinvestment rose, savingsgrew rapidly, inflationdeclined, the fiscal deficit
fell sharply, foreign debtdropped, and unemploy-ment declined sharply.Because Chile has focusedon getting the economicfundamentals right,including dramaticallyraising the domestic sav-ing rate, its good eco-nomic performanceappears fully sustainable.
The FML was suc-cessful in promoting sub-stantial development ofChile’s capital markets.The loan facilitated a con-siderable increase in thevolume of domestic shareissues, high pension fundand life insurance com-pany returns, and a largeincrease in stock prices. Inaddition to the develop-ment of the capital mar-kets, securities market
infrastructure wasimproved, with the intro-duction of competitivecomputerized trading,improvements in riskassessment services, andthe establishment of acentralized securitiesdepository. The leasingcomponent of the FMLwas an unqualified suc-cess. Another positive fea-ture of the loan was thatthe regulatory changes itsupported have increas-ingly corrected the marketfailure that justified theBank’s directed lending toleasing companies, thusrendering future lendingof this type unnecessary.Finally, technical assis-tance provided under theloan was unusually well-managed and successful.
The institutional objec-tives of the loan weremet, and CORFO hasemerged as a successfulsecond-tier lending institution.
Lessons of Experience. Ahighly complex financialsector project can be suc-cessful provided: (1) themacroeconomic, financialand industrial frameworkis sound; (2) the govern-ment is fully committedto the project; and (3)the Bank and the govern-ment are able to drawupon a sound base ofsectoral and institutionalknowledge.
BOX 2.2: CHILE (CONTINUED)
14
make deposits with either, and between banks andnon-banks by reducing reserve requirements (Box2.1). A measure similar in intent was the ceiling ontaxes on deposits in the Philippines and opening ofthe banking system to foreign banks in Tanzania.
The important question is whether policies incorpo-rated in project designs were relevant to initial condi-tions. Bangladesh, Indonesia, Mexico, Pakistan, andPoland were the only five countries in which Bank pro-grams had provided policy instruments to improve ini-tial conditions in all four reform areas (Table 2.2). InIndia, Morocco, Senegal, and Tanzania, programs hadpolicy instruments to alleviate problems in three areas.In Malawi and Venezuela, only one area incorporatedappropriate policy instruments. For Egypt andMalaysia, none of the required policies was incorpo-rated in program design.
We identified matches or mismatches between initialconditions and the Bank’s policy package. If that pack-
age was well-designed for improving initial conditions,it was called a “hit.” If the program policies wereunsuited to improving poor initial conditions, it wasdefined a “miss.” And if the policies prescribed wereunnecessary because initial conditions were relativelygood or did not need major improvement, they wereconsidered to be “overkill” (Table 2.3).3
Hits averaged 64 percent, misses 25 percent, andoverkills 11 percent. Using our threshold of 50 percentfor satisfactory performance, the Bank’s diagnosisappears marginally satisfactory. Removing distortionsand strengthening institutions showed the most hitswith 78 percent. Bangladesh, Indonesia, Korea, Mex-ico, Pakistan, and Poland fared best, with hits in allfour areas. Egypt, however, scored misses in all fourareas.
Competition policies had fewest hits.4 In somecountries the Bank took a perfunctory approach to com-petition within the banking system and sidestepped
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
Note: “X” indicates the presence of a policy instrument; (u) indicates unsatisfactory initial conditions in a particular reform area.Source: PRs and SARs.
TABLE 2.2: RELEVANCE OF BANK-SUPPORTED POLICY PACKAGESCOUNTRY DEVELOPMENT(Fiscal year of OF FINANCIAL STRENGTHENINGfirst financial REMOVAL OF INCREASE IN SECTOR OFsector operation) DISTORTIONS COMPETITION INFRASTRUCTURE INSTITUTIONS
Bangladesh (1990) X (u) X (u) X (u) X (u)Bolivia (1987) X (u) (u) X (u) X (u)Chile (1986) X XChina (1993) (u) (u) X X (u)Côte d’Ivoire (1992) (u) (u) X (u) X (u)Egypt (1992) (u) (u) (u) (u)Ghana (1988) X (u) X (u) X (u)India (1995) X (u) X (u) (u) X (u)Indonesia (1988) X (u) X (u) X (u) X (u)Kenya (1989) X (u) (u) X X (u)Korea (1985) X (u) X (u)Malawi (1991) (u) X (u) (u)Malaysia (1987) (u) (u)Mexico (1989) X (u) X (u) X (u) X (u)Morocco (1986) X (u) X (u) X X (u)Pakistan (1989) X (u) X (u) X (u) X (u)Philippines (1989) X X X (u)Poland (1991) X (u) X (u) X (u) X (u)Senegal (1987) X (u) (u) X (u) X (u)Tanzania (1992) X (u) (u) X (u) X (u)Tunisia (1988) X (u) (u) X X (u)Turkey (1986) X (u) (u) X (u) XVenezuela (1990) X (u) (u) (u) (u)
15
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
competition between banks and non-bank financialintermediaries (NBFIs) altogether. In some countries thegovernment used NBFIs to introduce more outside com-petition, but neither the Bank nor the government wasprepared to take substantive action regarding competi-tion within the banking system. Generally, competitionpolicies seem to be either absent from or superficiallydesigned in most Bank projects. Future projects need tofocus more on promoting competition and strengtheninginstitutions.
Policy implementation under financial sector opera-tions also varied (Table 2.4).
Admittedly, implementation performance can bejudged only if the policies are included in a project. Ofthe 16 countries whose programs included policy pre-
scriptions to remove distortions, 10 countries imple-mented reforms. Only three of the nine countries whoseprograms contained policy prescriptions to increasecompetition carried out the expected reforms. Of the 17countries for which development of financial sectorinfrastructure was an objective, 12 were able to imple-ment the required policies. And of the 19 countrieswhose programs contained policy prescriptions for insti-tutional strengthening, only 8 implemented them satis-factorily according to the objectives of the reformprogram.
In only four countries—Chile, Côte d’Ivoire (post-1994), Mexico, and Senegal—was implementation suc-cessful in all areas covered by the policy reform pack-ages. On the other hand, Bangladesh, India, Korea,
Note: Figures in parenthesis indicate average “hits,” “overkills” or “misses” in percentages.Source: PRs, SARs, ICRs, PCRs and PARs.
TABLE 2.3: HITS, MISSES, AND OVERKILLSCOUNTRY DEVELOPING(Fiscal year of FINANCIALfirst financial REMOVING PROMOTING SECTOR STRENGTHENINGsector operation) DISTORTIONS COMPETITION INFRASTRUCTURE INSTITUTIONS
Bangladesh (1990) Hit Hit Hit HitBolivia (1987) Hit Miss Hit HitChile (1986) Hit Hit Overkill OverkillChina (1993) Miss Miss Overkill HitCôte d’Ivoire (1992) Miss Miss Hit HitEgypt (1992) Miss Miss Miss MissGhana (1988) Overkill Miss Hit HitIndia (1995) Hit Hit Miss HitIndonesia (1988) Hit Hit Hit HitKenya (1989) Hit Miss Overkill HitKorea (1985) Hit Hit Hit HitMalawi (1991) Miss Hit Miss HitMalaysia (1987) Hit Hit Miss MissMexico (1989) Hit Hit Hit HitMorocco (1986) Hit Hit Overkill HitPakistan (1989) Hit Hit Hit HitPhilippines (1989) Hit Overkill Overkill HitPoland (1991) Hit Hit Hit HitSenegal (1987) Hit Miss Hit HitTanzania (1992) Hit Miss Hit HitTunisia (1988) Hit Miss Overkill HitTurkey (1986) Hit Miss Hit OverkillVenezuela (1990) Hit Miss Miss MissPercentage of:
Hits (64) 78 48 52 78Overkills (11) 4 4 26 9Misses (25) 17 48 22 13
16
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
Morocco, Turkey and Venezuela had poor implementa-tion records.5
The Outcome as Measured by Performance IndicatorsHow did the Bank’s adjustment operations do in oursample of 23 countries? To assess program outcomes,we use 16 performance indicators derived from the OD8.30 conceptual framework (these are outlined in Table1.1). Unfortunately the availability of data on these indi-cators varies from country to country (Table 2.5). Asassessed by our PIR methodology (see Box 1.1) appliedto 29 completed and 13 ongoing projects, financial sec-tor reform supported by the Bank yielded satisfactoryoutcomes in 12 countries—52 percent of the total (Table2.5), for the period of adjustment programs examined(FY85–96). The macroeconomic dimension was wellunderstood by the Borrower, and results are generallygood, with a satisfactory rating of 74 percent. Still,countries remain vulnerable to exogenous shocks. Thesector dimension, however, achieved a satisfactory rat-
ing of only 52 percent. The Bank actively—and at timessuccessfully—supported policies to remove distortionsand to develop financial infrastructure. Competitionpolicies, however, were either missing or not well-imple-mented. The poor results are in soundness of bankingsystems, with an average satisfactory rating of 35 per-cent—indicating that in most cases reforms did notreach the core of the financial system. But data are rela-tively scarce in this area, and results may thus differwhen more information becomes available.
Program outcomes vary considerably across ele-ments of the financial system in all countries. More thantwo-thirds of the countries were rated as having a satis-factory outcome in M2-GDP ratio, whereas only one-third was rated satisfactory in bank profits. Althoughthe capitalization of commercial banks improved intwo-thirds of the countries, only in about one-fourth didcentral banks have satisfactory capital adequacy ratios.Financial repression generally declined as a result offinancial sector operations, but bank restructuring
TABLE 2.4: POLICY IMPLEMENTATIONCOUNTRY DEVELOPMENT(Fiscal year of OF FINANCIAL STRENGTHENINGfirst financial REMOVAL OF INCREASE IN SECTOR OFsector operation) DISTORTIONS COMPETITION INFRASTRUCTURE INSTITUTIONS
Bangladesh (1990) XBolivia (1987) X XChile (1986) X XChina (1993)Côte d’Ivoire (1992) X XEgypt (1992)Ghana (1988) X XIndia (1995)Indonesia (1988) X X XKenya (1989) X XKorea (1985)Malawi (1991)Malaysia (1987)Mexico (1989) X X X XMorocco (1986) XPakistan (1989) X X XPhilippines (1989) X XPoland (1991) X XSenegal (1987) X X XTanzania (1992) XTunisia (1988) X XTurkey (1986) XVenezuela (1990)
Note: ‘X’ indicates the satisfactory implementation of a policy instrument in a particular reform area.Source: PCRs, ICRs, and PARs.
17
remained a problem in many countries. These outcomes,however, do not take account of the recent effects of the1997 crisis in East Asia, e.g. Korea, Indonesia6 and thePhilippines. The recent financial crises in Korea andIndonesia clearly indicate that strong macroeconomicperformance can temporarily cover the weak underlyingcondition of the financial system. Actually, the weaknessof the banking system was already identified for bothcountries with our rating methodology applied to pre-crisis information (Table 2.5).
It is interesting to compare outcome assessmentsbased on PIRs with those in PCRs and ICRs, PARs, theAnnual Review of Portfolio Performance (ARPP) forprojects under implementation, and the report of theQuality Assistance Group (QAG) on FILs.7 One canmake such a comparison only for outcomes (effect onthe financial system), since PCRs, PARs, and the ARPPdo not rate projects for impact (effect on the real sector).And even a comparison of outcomes is difficult becausethe project samples are not the same. The ARPP coversthe Bank’s portfolio of projects that are being imple-mented, while PCR/PAR ratings are normally for com-pleted projects only. Furthermore, PCR/PAR ratings areavailable for only 19 of our sample countries.
Using PCR/PAR methodology on the full sample of23 countries, we find that 58 percent show satisfactoryperformance; using ARPP methodology, 78 percent aresatisfactory. But only 52 percent of the countries showsatisfactory performance using the PIR methodology(Table 2.6). Eleven countries had higher satisfactory rat-ings under the PCR/PAR system than under the PIR sys-tem. The same 11 countries also had higher satisfactoryratings in the ARPP system. Eighteen countries had satis-factory ratings under the ARPP, compared with only 12in PIR. One plausible reason for fewer satisfactory rat-ings in this study may be that the PIR indicators givegreater weight to conditions in the banking system. InChina, India, Pakistan, the Philippines, Senegal, andTunisia the banking systems have either stagnated ordeteriorated according to the assessment of performanceindicators.
The discrepancy in the satisfactory ratings betweenthe PIR and the other Bank rating systems could reflectdisparate sample sizes. To remove this sample bias, wecompared performance results for a subset of 19 coun-tries out of the 23, retaining only those for which all therating methodologies have been systematically applied.8
Even in the adjusted sample, the PIRs show that projectoutcomes are less satisfactory (Table 2.7).
Reasons for Mixed OutcomesFrom the results obtained so far (Table 2.5), the out-comes of Bank-supported reforms, though not uniformlyunsatisfactory, have been below program objectives. Wefocus here on possible reasons for those outcomes, draw-ing on OD 8.30 and academic studies on the role of thereal economy, the fiscal factor, and, more generally, theresponsibility of governments.
Government PoliciesFinancial sector operations should be viewed as part ofthe structural transformation of an economy. If real sec-tors lag in reform, they hinder the growth of the finan-cial sector. And without enterprise restructuring, therewill be few good opportunities for banks to lend. Ifbanks are restructured but enterprises are not, bankrestructuring will only have to be repeated, at a muchhigher cost each time, as experience in Africa and LatinAmerica suggests. In Central Africa one bank wasrestructured three times. In Mauritania recapitalizedbanks generated losses for the state amounting to 50 to60 percent of GDP. The Chilean experiment underscoresthe lesson that highly-leveraged, undercapitalized, andweak enterprises tend to unravel a banking system. So,financial sector operations should fit neatly with therestructuring of the enterprise sector (Khatkhate 1993;World Bank 1995a; Caprio and Klingebiel 1996).
Fiscal policy is important because there is a closelink between the fiscal deficit and financial saving andthe banking systems. A larger fiscal deficit makes it dif-ficult to fully liberalize interest rates because of raisingthe cost of government debt. It therefore hinders finan-cial savings. But the impact of the fiscal deficit on finan-cial intermediaries cuts both ways. While a larger fiscaldeficit represses financial intermediaries in that it doesnot permit the government to abandon interest rate reg-ulation or directed credit, as in India, it helps the finan-cial sector to appropriate a substantial part of inflationtax resulting from the fiscal deficit, as in Chile duringthe first half of the 1980s. The result depends onwhether the benefit of inflation exceeds or falls short ofthe costs imposed by the fiscal deficit.
Financial intermediaries focus on their own andtheir borrowers’ net worth, which makes the timing offinancial liberalization crucial. If, for example, the realeconomy is in a cyclical downturn because of an exter-nal shock, such as adverse terms of trade or high inter-est rates transmitted from abroad, borrowers’ net worthdeclines and hence the cost of the external finance they
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
18
need rises. As a result, real investment falls, leading to adrop in the net worth of the financial institutions. Thislink between financial reform and the net worth of bor-rowers and lenders means that governments shouldmove aggressively on financial reform in good times andmore slowly in bad times (Caprio and others 1994).Latin American countries ignored the critical impor-tance of timing financial reform, and their experimentswere unsuccessful. Korea had the opposite experience.The Korean Government initiated financial liberaliza-tion measures only after a significant deceleration of
inflation and abatement of recessionary tendencies.Doing so ensured a realistic set of relative prices and aprofitable corporate sector (Cho and Khatkhate 1989).
The Bank’s operations may be seen as less success-ful than expected partly because the Bank expectsresults in two to three years. But meaningful resultsfrom interventions in the financial sector may take along time to emerge (Sheng 1996)—as long as 8 to 10years for Bank-supported structural reforms. Because ofthe Bank’s high expectations, it tends to urge govern-ments to take quick action, such as completely deregu-
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
MACROSTABILITY FINANCIAL STRUCTURE
IFC’S
CURRENT REAL MONEY GLOBAL
FISCAL ACCOUNT EX- DEPTH REAL DO- MAR- CAPITAL
INDI- INFLA- DEFICIT/ BALANCE/ CHANGE M2/ AVER- DEPOSIT MESTIC FOREIGN KET MARKET AVER-
COUNTRY CATORS TION GDP GDP RATES GDP AGE RATES SPREAD SPREAD RATE INDEX AGE
Bangladesh 14 1 1 1 1 1.0 1 0 1 0 0.5Bolivia 13 1 1 0 1 1 0.8 1 0 0 0.3Chile 15 1 1 1 1 1 1.0 1 1 1 1 1.0China 18 0 1 1 1 1 0.8 0 1 0 0 0 0.2Côte d’Ivoire 10 0 0 1 1 0.5 1 1.0Egypt 15 0 1 1 0 1 0.6 0 1 0 0 1 0.4Ghana 13 1 1 0 1 0 0.6 1 1 1 1.0India 10 1 1 1.0 1 1 0 1 0.8Indonesia 15 1 1 0 1 1 0.8 1 1 1 1 0 0.8Kenya 12 1 0 1 0 1 0.6 1 0 0 1 0.5Korea 15 1 1 1 0 1 0.8 1 1 1 1 1.0Malawi 13 0 1 1 0 0.5 0 1 0 0 0.3Malaysia 16 0 1 0 1 1 0.6 1 1 1 1 0 0.8Mexico 13 1 1 0 1 1 0.8 1 1 1 1.0Morocco 14 0 1 1 0 1 0.6 1 0 1 1 0.8Pakistan 13 0 0 0 1 1 0.4 0 0 0.0Philippines 18 1 0 0 1 1 0.6 1 0 1 1 0 0.6Poland 18 1 1 1 0 1 0.8 0 0 0 1 0 0.2Senegal 12 1 0 0 0 0.3 1 0 0 1 0.5Tanzania 13 1 0 0 1 1 0.6 0 0 1 0.3Tunisia 12 0 0 0 0 0 0.0 1 1.0Turkey 17 0 0 1 1 0 0.4 0 0 1 0 0.3Venezuela 17 0 1 1 1 0 0.6 1 1 1 1 0 0.8Percent satisfactory 74 52
Note:1. “1” means satisfactory, “0” means unsatisfactory; an average rating greater than 0.5 is “satisfactory,” an average rating of
0.5 or less is “unsatisfactory.2. This table includes macroeconomic variables even when Bank programs are often narrowly based because the reform of the
sector depends critically on the macroeconomic environment.If the latter is not conducive, the sector concerned would not show improvement despite a well-conceived program and itssatisfactory implementation.
3. “S” means satisfactory, “U” means unsatisfactory.Source: International Financial Statistics (IFS), World Development Indicators 1997.
TABLE 2.5: OUTCOMES: FINANCIAL SECTOR PERFORMANCE INDICATORS
19
lating interest rates, and privatizing and recapitalizingbanks, though such measures are not so well thought-out. These recommendations create more problemsthan solutions in the long-term.
The Political Economy of ReformOED’s definition of ownership (Johnson 1993) is partic-ularly apt in financial sector work. It calls for joint iden-tification of program goals, consensus within the gov-ernment leadership, upfront actions to demonstrateintellectual conviction and broad outreach regarding
reform goals within the body politic. It is especially dif-ficult to secure genuine commitment to financial sectorreform, because these sectors are often characterized byconcentrated ownership, where the owners and thosewith political power are closely connected. The questionis often raised: is financial sector reform more likely tobe successful if borrowers take more ownership ofreform? Clearly, any reform that a country does notaccept stands very little chance of success. The urge forreform must come from within the country. All that anoutside agency, like the Bank, can do is to nudge the
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
BANKING SYSTEM OVERALL RATING
COM- CENTRAL
BANKS’ BANKS’ BANKS’ MERCIAL BANK
CAPITAL ASSET MANAGE- BANK’S BANKS’ CAPITAL
ADEQUACY QUALITY MENT EARNINGS LIQUIDITY ADEQUACY AVERAGE AVERAGE RATING3
Bangladesh 0 0 0 0 0.0 0.50 UBolivia 1 0 1 0.7 0.64 SChile 1 0 1 1 0.8 0.92 SChina 0 0 0 0 1 0 0.2 0.38 UCôte d’Ivoire 1 1 1 1.0 0.75 SEgypt 1 0 0 0.3 0.46 UGhana 1 0 0 0.3 0.64 SIndia 0 0 0.0 0.63 SIndonesia 1 0 0 0.3 0.69 SKenya 0 0.0 0.50 UKorea 1 1 0 0 0.5 0.77 SMalawi 1 1 0 0.7 0.45 UMalaysia 0 0 0 0 0.0 0.50 UMexico 0 0 0 0.0 0.64 SMorocco 1 0 1 0.7 0.67 SPakistan 0 0 1 0 0.3 0.27 UPhilippines 1 1 1 1 1 1 1.0 0.75 SPoland 1 0 0 0 1 0 0.3 0.44 USenegal 1 0 0.5 0.40 UTanzania 1 1 0 0.7 0.55 STunisia 1 1 0 0 0.5 0.30 UTurkey 0 1 1 1 1 1 0.8 0.53 SVenezuela 0 0 0 0 0 0.0 0.47 UPercent satisfactory 35 52
TABLE 2.5: OUTCOMES (CONTINUED)
20
country through offers of financial assistance and tech-nical advice (see Box 2.3). The question is, when is acountry willing to embark on financial sector and otherreforms? Experiences over the past decade show thatcountries embrace reform when the following three con-ditions prevail (Williamson 1994):
• The political party changes and the party in powerdevelops a new program. Or, the party in powermay change its stance on reforms when the coun-try finds itself in a serious crisis.
• Either the bureaucracy through which a rulingparty implements its program changes when thepolitical party changes, or it is persuaded to acceptnew policies.
• Citizens change their perception of reform.Clearly, none of these three conditions is easily met
unless reforms are seen to be unavoidable and thereforedesirable, or unless the fruits of reform are seen quicklyand are credible. After the debt and oil crises, a wave of
reforms in several developing countries in Latin Americaand Asia accelerated. In Chile, ownership of reform camewith a change of regime. Reform also took place whencentrally-planned regimes collapsed. India initiatedreform only when it faced a dire economic crisis in 1991.
Accepting ownership of reform at a political level,however, does not always ensure smooth sailing. It hasbeen a universal experience in transition and developingcountries that the bureaucracy stalls reform even whenit is announced by the party in power. At the same timethe bureaucracy is impelled to change its attitude if theleadership persists with reform and results become visi-ble. Reform means upsetting vested interests. But theirresistance can be overcome if the ultimate beneficiariesof reforms—investors, producers, and wage earners—gain from reform, or lose less than they expected. Whenreform succeeds, both bureaucrats and politicians feelpressure from below and find it difficult to retreat(Williamson 1994). Ownership of reform can also be
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
TABLE 2.6: A DETAILED COMPARISON OF OUTCOME RATINGS FOR 23 COUNTRIESANNUAL
COUNTRY PROJECT DIFFERENCES REVIEW OF(Fiscal year of PERFORMANCE COMPLETION/ BETWEEN PIR PORTFOLIOfirst financial INDICATORS AUDIT REPORTS AND PCR/PAR PERFORMANCE PIR-ARPPsector operation) (PIRS) (PCRS/PARS) RATINGS (ARPP) DISCONNECT
Bangladesh (1990) 0.50 0.00 0.50 1.00 -0.50Bolivia (1987) 0.64 1.00 -0.36 1.00 -0.36Chile (1986) 0.92 1.00 -0.08 1.00 -0.08China (1993) 0.38 1.00 -0.62Côte d’Ivoire (1992) 0.75 1.00 -0.25 1.00 -0.25Egypt (1992) 0.46 0.00 0.46Ghana (1988) 0.64 1.00 -0.36 1.00 -0.36India (1995) 0.63 1.00 -0.37Indonesia (1988) 0.69 1.00 -0.31 0.00 -0.31Kenya (1989) 0.50 0.00 0.50 1.00 0.50Korea (1985) 0.77 0.00 0.77 1.00 -0.23Malawi (1991) 0.45 1.00 -0.55Malaysia (1987) 0.50 0.00 0.50 1.00 -0.50Mexico (1989) 0.64 1.00 -0.36 1.00 -0.36Morocco (1986) 0.67 1.00 -0.33 1.00 -0.34Pakistan (1989) 0.27 1.00 -0.73 1.00 -0.73Philippines (1989) 0.75 0.00 0.25 1.00 -0.26Poland (1991) 0.44 1.00 -0.81 1.00 -0.81Senegal (1987) 0.40 1.00 -0.60 1.00 -0.60Tanzania (1992) 0.55 0.00 0.55 0.00 0.66Tunisia (1988) 0.30 1.00 -0.70 1.00 -0.67Turkey (1986) 0.53 0.00 0.53 0.00 0.53Venezuela (1990) 0.47 0.00 0.47 0.00 0.47Number of countries 23 19 19 23 23Satisfactory 12 11 18Unsatisfactory 11 8 5Success rate (%) 52 58 -6 78 -26
Source: PCRs, ICRs, PARs, ARPP, and OED estimates based on time-series from Central Bank Annual Reports, IFS, and WDI.
21
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
assured if it is accelerated so that it becomes irreversible.This has happened both in transition economies, such asPoland, and in developing countries, such as India.
Against this background some generalizations canbe made. Most distortions in the financial sector are per-petuated by the government’s close association with thebanking system. If an arm’s-length relationship can bebuilt between the banking system and the government,reform is likely to take on greater urgency. This maymean fiscal reform, which will prevent governmentsfrom taxing financial systems; privatization of publicly-owned banks, which will neutralize the influence of anentrenched bureaucracy; establishment of independentcentral banks, which will take a more detached view ofhow the financial system functions; and creation of anincentive-compatible system under which all involvedhave stakes in maintaining an efficient financial sector.
Program DesignThe design of adjustment operations also greatly influ-ences outcomes, as the percentage of “hits” in Table 2.4suggests. By and large, adjustment operations yielded
good outcomes when projects were well-designed andpolicy instruments well-coordinated. But good projectdesign alone is not enough. Six of the sample coun-tries—Bangladesh, Indonesia, Korea, Mexico, Pakistan,and Poland—had hits in all four policy areas, suggestingsatisfactory project designs, but only Indonesia, Koreaand Mexico showed satisfactory outcome rating for theperiod under review (up to FY96). On the other hand,of ten countries with more than two hits but fewer thanfour, only eight—Chile, Côte d’Ivoire, Ghana, India,Morocco, the Philippines, Tanzania, and Turkey—hadsatisfactory outcomes. Project failure in countries withwell-designed projects—or project success in countrieswith moderately relevant designs—may be attributed toother determinants of performance, such as the externalenvironment, other policies, or real sector performance.It may also be explained by the fact that outcome andimpact depend on how much the authorities are willingto implement broad-based programs rather than nar-rower or more specifically targeted programs.
Determinants of PerformanceSince financial sector operations are relatively new andmany of them are not completed, performance data arerelatively limited. Nevertheless, we attempt to identifythe determinants of project performance through statis-tical and econometric analysis.9
To explore why financial sector loans perform dif-ferently in different countries, we compare changes inkey outcome variables three years after FSLs areapproved. A country’s outcomes are separated into threegroups (Table 2.8) according to two categories of initialconditions—one macroeconomic, the degree of infla-tion, and the other microeconomic, deposit bank creditto private enterprises (DBPC) as a share of GDP. Theseinitial conditions are then linked to changes in the ratios
TABLE 2.7: A COMPARISON OF OUTCOME RATINGSFOR NINETEEN COUNTRIES
OEDSTUDY PCR/
(PIR) PAR ARPP QAGa
Satisfactoryrating (%) 58 58 79 60
a QAG ratings are not strictly comparable, as the QAG rat-ing reflects the concept of “project at risk,” rather than “sat-isfactory outcome,” and is applied only to the current portfo-lio of FILs. For the purpose of comparison, however, we callthe portion of the QAG portfolio that is not “at risk” (60percent of FILs) “satisfactory.”Source: IFS, WDI, PCRs, PARs, OPRIS.
TABLE 2.8: CONTROLLING FOR INITIAL CONDITIONS IN THE MACROECONOMY AND THE FINANCIAL SECTOR INITIAL CONDITIONS
I II IIIINFLATION<25%; INFLATION>25%; INFLATION<25%;
DEPOSIT BANK PRIVATE DEPOSIT BANK PRIVATE DEPOSIT BANK PRIVATECREDITS <20% CREDITS <20% CREDITS >20%
INDICATOR OF GDP OF GDP OF GDP
Change 3 years after initial (percentage points)conditions:
M2/GDP +8.22 +4.64 -0.23DBPC/GDP +5.69 +2.33 -0.70
Source: IFS database
22
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
Initial conditions. Thecentral problem withTurkey’s financial sys-
tem in the mid-1980s wasthe banks’ poor financialcondition, reflecting thedifficulties of their clients,especially the publicenterprises. These diffi-culties were exacerbatedby macroeconomic insta-bility and very high realinterest rates resultingfrom excessive publicborrowing. Clearly, thesuccess of the financialreform program was con-tingent on bringing publicspending under much bet-ter control. Other finan-cial policy problems hadto be dealt with as well assuch institutional prob-lems as inadequateaccounting proceduresand unsatisfactory pru-dential supervision offinancial institutions.
FSAL objectives. Theobjectives of the firstFSAL, approved in 1986,were to reform financialpolicies, strengthen thebanking system, and
develop capital markets.FSAL II, a follow-up pro-ject approved in 1988,focused on filling-in thegaps under the sameheadings. In addition,more explicit emphasiswas placed on improvingmacroeconomic condi-tions than under FSAL I.FSAL I and FSAL II bothprovided technical assis-tance oriented towardinstitution-building, train-ing the staff of bankinginstitutions, preparingrelated studies, andautomating informationsystems.
FSAL outcome. Theenvisaged reforms wereonly partially completed.On the positive side, abanking crisis wasaverted and prudentialregulation, external audit-ing, and off-site monitor-ing of banks and capitalmarket institutions werebrought up to interna-tional standards. Thecapital base of the bank-ing sector was strength-ened, and nonperforming
loans were appropriatelyclassified and providedfor. The size of the publicbanking sector wasreduced both relativelyand absolutely. The rela-tive size of preferentialcredits declined, and sub-sidized credit programswere curtailed. However,government borrowingcontinued to drive-up realinterest rates, and crowd-out private sector bor-rowing, and public over-spending helped generatemacroeconomic instabil-ity, reversing or jeopar-dizing some sectoralreforms.
Assessment. The PAR,noting that public over-spending was partly theresult of public enterprisedeficits and that the prob-lems of the financial sec-tor were also closelylinked to questionableloans to the same publicenterprises, asked why acommon solution tomacroeconomic andfinancial sector problemswas not sought directly in
public enterprise reform.This suggests that theBank failed to come fullyto grips in its adjustmentlending strategy with oneof the most critical eco-nomic issues in Turkey.This unwillingness toinsist on the fulfillment ofkey macroeconomic con-ditions undermined theeffectiveness of the firstFSAL. The Bank imposedtougher conditions onFSAL II and showed muchgreater willingness toinsist on these being met.
Lessons of experience.Experience from the twoFSALs indicates incom-plete government owner-ship of the reform pro-gram, partly because oflack of a political consen-sus, a reflection of contin-ued political oppositionto some of its measuresby entrenched groups.The Bank was alsoresponsible for not hav-ing asked for ownershipof the most critical com-ponents of the reform.
BOX 2.3: TURKEY: THE “OWNERSHIP” ISSUE
23
F i n a n c i a l S e c t o r R e f o r m O p e r a t i o n s
of M2 and DBPC to GDP three years after the latestloan approval. Data were not available for several coun-tries, so a smaller sample of 18 countries was used.10
The statistical results still remain significant.The comparison shows that changes in financial
depth (measured as M2/GDP) are highly sensitive to ini-tial conditions. Improvements under FSLs were greatest incountries with low inflation and low bank credit to theprivate sector.11 In countries starting with inflation below25 percent and DBPCs of less than 20 percent of GDP, forexample, M2/GDP improved by an average of 8.22 per-centage points, and DBPC ratios improved by an averageof 5.69 percentage points. On the other hand, in countrieswith inflation lower than 25 percent and a DBPC higherthan 20 percent of GDP, the reform’s effect on both ratioswas disappointing. Thus both macroeconomic and micro-economic initial conditions appear to be key determinantsof a reform project’s impact on the financial system.
Other factors that affect outcome include policieson prudential regulation. Econometric analysis (Cull
1997) shows that attempts to improve prudential con-trols have an important bearing on the outcome of theBank’s interventions in the financial sector. As in the pre-vious analysis of initial conditions, firmer conclusionscould be derived from larger samples. (see Annex 2.2).
The Real SectorWe also use the performance indicators to assess theprobable effect of financial sector reform on the econ-omy. The indicators are macroeconomic structural vari-ables and the level of foreign direct investment. Gener-ally, the impact of the Bank’s interventions on the realeconomy has been satisfactory in more than half of our23 sample countries (see Table 2.9). Bank interventionshave had a satisfactory impact on net direct foreigninvestment in 16 countries, on the gross domestic invest-ment ratio in 17 countries, and on the GDP growth ratein 18 countries. In general, the impact in each countrydiffers, depending on the scores for each component ofthe macroeconomy.
TABLE 2.9: CHANGES IN THE REAL ECONOMY MEASURED WITH PERFORMANCE INDICATORSCOUNTRY(Fiscal year of first NET FOREIGN OVERALL RATINGfinancial sector GDP DIRECT (Satisfactory/operation) GROWTH GDSa/GDP GDIb/GDP INVESTMENT AVERAGE unsatisfactory)
Bangladesh (1990) 1 0 1 0 0.50 UnsatisfactoryBolivia (1987) 1 0 0 1 0.50 UnsatisfactoryChile (1986) 1 1 1 1 1.00 SatisfactoryChina (1993) 1 1 1 1 1.00 SatisfactoryCôte d’Ivoire (1992) 1 1 0 1 0.75 SatisfactoryEgypt (1992) 0 0 0 1 0.25 UnsatisfactoryGhana (1988) 1 0 1 1 0.75 SatisfactoryIndia (1995) n.a. n.a.Indonesia (1988) 1 1 1 1 1.00 SatisfactoryKenya (1989) 1 0 1 0 0.50 UnsatisfactoryKorea (1985) 1 1 1 0 0.75 SatisfactoryMalawi (1991) 0 0 1 0 0.25 UnsatisfactoryMalaysia (1987) 1 1 1 1 1.00 SatisfactoryMexico (1989) 1 0 1 1 0.75 SatisfactoryMorocco (1986) 1 0 1 1 0.75 SatisfactoryPakistan (1989) 0 1 1 1 0.75 SatisfactoryPhilippines (1989) 1 0 1 1 0.75 SatisfactoryPoland (1991) 1 0 1 1 0.75 SatisfactorySenegal (1987) 1 0 1 0 0.50 UnsatisfactoryTanzania (1992) 1 0 1 1 0.75 SatisfactoryTunisia (1988) 1 0 0 0 0.25 UnsatisfactoryTurkey (1986) 1 1 1 1 1.00 SatisfactoryVenezuela (1990) 0 0 0 1 0.25 UnsatisfactoryPercent satisfactory 64
Note: “1” indicates satisfactory; “0” indicates unsatisfactory.a GDS: gross domestic savings.b GDI: gross domestic investment.Source: OED estimates based on time-series from World Development Indicators 1997.
24
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
TABLE 2.10: COMPARISON OF FINANCIAL SECTOR OUTCOME WITH REAL ECONOMY IMPACTCOUNTRY(Fiscal year of SATISFACTORY UNSATISFACTORYfirst financial SATISFACTORY UNSATISFACTORY REAL SECTOR REAL SECTORsector operation) OUTCOME OUTCOME PERFORMANCE PERFORMANCE
Bangladesh (1990) X XBolivia (1987) X XChile (1986) X XChina (1993) X XCôte d’Ivoire (1992) X XEgypt (1992) X XIndia (1995) X n.a.Indonesia (1988) X XKenya (1989) X XKorea (1985) X XMalawi (1991) X XMalaysia (1987) X XMexico (1989) X XMorocco (1986) X XPakistan (1989) X XPhilippines (1989) X XPoland (1991) X XSenegal (1987) X XTanzania (1992) X XTunisia (1988) X XTurkey (1986) X XVenezuela (1990) X XPercent satisfactory 52 48 64 36
Note: “Outcome” refers to effect on the financial system; “impact” refers to effect on the real economy.Source: OED estimates based on time-series from International Financial Statistics and World Development Indicators 1997.
Finance and DevelopmentAmong the 23 sample countries the proportion with a sat-isfactory real sector performance was higher than thatwith satisfactory financial sector outcomes—64 percentcompared with 52 percent (see Table 2.10). Note that in10 of the 23 countries programs were satisfactory in bothareas. And in 7 countries performance was unsatisfactoryin both the real and financial sectors. This suggests a closelink between reform of the financial sector and develop-ment of the real economy, as found in recent academicand Bank studies (Levine 1997). Nevertheless, a strongmacroeconomic performance can temporarily cover theweak underlying condition of the financial system.
As presented in detail by Levine (1997), the natureof the link between finance and development is ambigu-ous at both theoretical and empirical levels. In this study,for example, the financial sector outcome was unsatis-factory for four countries, but the real sector perfor-mance was satisfactory. Why? These countries mighthave benefited from other reforms, such as trade liberal-ization, the removal of price controls, and deregulationof the industrial sector. They may have also benefitedfrom favorable exogenous events, such as changes ininternational trade and prices.12
25
Institutional Development
33
The two major aspects of institutional development are: strengthening financial
sector infrastructure and strengthening individual institutions. Strengthening financial
infrastructure usually involves developing a legal framework for the central bank,
the banking system, and non-bank financial institutions, as well as laws and regulations
defining the rights and obligations of financial agents. Strengthening individual financial
institutions requires better bank supervision, the restructuring and privatization of banks, and
organizational and managerial changes within banks.Institutional development is critical to financial reformbecause without high-quality technical capabilities andgeneral institutional efficiency, resource mobilizationand efficient allocation remain weak.1
To support institutional development in the financialsector, the Bank relies on SADs, other loans related to sec-toral adjustment (financial sector components in SALsand TALs), and FILs. The difference between sector-related loans (SADs, SALs, and TALs) and FILs in institu-tional development is large. Financial sector loans are pol-icy-based lending instruments. They support changes in awhole range of financial infrastructure, including laws,regulations, and administrative practices to strengthencapital markets and to improve the soundness of bankingsystems. In FILs institutional development has a narrowerfocus (it refers to the viability of the institution itself, asreflected in its governance, management, technical capa-bility, resources, portfolio, and profitability).
From the late 1960s to the late 1980s, FILs were afavored lending device. But in 1979, when the second oilshock translated into rising inflation and falling invest-ment, FILs began to decline. Except for a temporaryincrease in 1986–88—especially in LAC countries fol-lowing an investment boom in the Western Hemi-
sphere—FILs have shown a steady decline, offset some-what by a corresponding rise in FSLs (Figure 3.1).
This section reviews institutional development inFSLs2 and FILs made to 24 countries.3 Institutionaldevelopment components in FSLs (Annex Table 3.2)include measures to strengthen financial infrastructureand institutions (such as the restructuring and privatiza-tion of banks). Twenty-four completed FSLs and fifty-eight completed FILs (Annex Table 3.3) were reviewed(approved between FY82 and FY92).4 The selected insti-tutional components were portfolio quality, resourcemobilization, financial performance, and technical capa-bility. In assessing institutional development (seemethodology in Box 3.1), we characterize initial condi-tions of financial institutions and the regulatory frame-work, and then observe progress under Bank-supportedprograms. We also evaluate Bank performance.
Initial conditions (prevailing before FSLs wereapproved) were weak, but institutional performanceimproved significantly, especially after FSLs supportedmeasures that strengthened financial infrastructure(Table 3.1). Initial conditions for FILs were better thanthose for FSLs, but progress under FIL programs wasmarginal after measures were taken to improve the insti-tutional capacity of individual banks (Table 3.1). At the
26
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
FIGURE 3.1: NUMBER OF FSLs AND FILs APPROVED 1975–97
0
5
10
15
20
25
30
35
40
45
50
Num
ber
FILs
FSLs
1975 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97
FSLs include SADs, SALs, and TALs with financial sector components.Source: Operations Information System, World Bank
TABLE 3.1: THE CONSOLIDATED RESULTSINITIAL IMPLEMENTATION
CONDITIONS RECORD OUTCOME SUSTAINABILITYa
FSLsStrengthening financial infrastructure 0.21 0.93 0.80 0.71Strengthening institutions, of which 0.17 0.71 0.62 0.75Bank restructuring 0.12 0.67 0.50 0.67
FSLsb
Portfolio quality 0.52 0.42 0.40 0.30Resource mobilization 0.28 0.37 0.31 0.33Financial performance 0.63 0.42 0.41 0.47Technical capability 0.58 0.65 0.59 0.63
Overall rating 0.38 0.44 0.42 0.42
Notes: Methodology is given in Box 3.1.(a) Sustainability is defined here as the probability of maintaining the project achievements in the institutional area. (b) FIL scores relate strictly to the financial institutions’ performance and do not cover the enterprise sector.Source: PRs, SARs, and OED estimates
27
end of FIL operations, performance scores for both out-come and sustainability remained more or less at thesame level as they had been initially. Thus, FSLs seem to
have been a more effective lend-ing instrument than FILs formeeting objectives.
Notable progress was madein strengthening financial infra-structure under FSLs, but out-comes by bank restructuring andinstitutional-strengthening weregenerally weak. This was not sur-prising, as improving the legalsystem is relatively easy, but
restructuring the banking system takes a long time. If theperiod covered were longer, the scores for restructuringmight improve. The narrow gap between initial condi-tions and outcomes for the components of institutional
development in FILs could be attributed to the time con-straint. Institutional development may be greater overthe long-haul than in the short-run, because it requiresskill development, which is a long, drawn-out process.
Institutional development programs were rated bycountry groups, broken down by level of income (Table3.2). Initial conditions were better in middle-incomecountries than in low-income countries for both types ofinstruments. Outcomes were also better in middle-income countries under both FSLs and FILs. Butimprovements with respect to initial conditions in low-income countries were substantial for FSLs. Although itis difficult in general to achieve good results in low-income countries, FSLs appear to be a relatively efficientinstrument for institutional development.
Initial conditions were better for countries withgreater loan frequency (four or more loans), and out-comes were better too (Table 3.3). This may be because
I n s t i t u t i o n a l D e v e l o p m e n t
T he assessment isbased on find-ings in SARs,
PCR/ICRs, and PARs. Weuse a sample of 24 FSLsand 58 FILs for 24countries.
The methodology toassess institutional devel-opment follows the sameapproach as that forBank’s interventions infinancial sector liberaliza-tion (Chapter 2). We eval-uate initial conditions,implementation, and out-come with respect to pro-gram objectives. Then welook at the impact or sus-tainability of the reformprogram. Because of thespecificity of institutionaldevelopment, there are afew differences. Themethodology for institu-tional development usesloan instrument categories
(FSLs compared withFILs) as the focus ofassessment, unlike themethodology for liberal-ization, which supports acountry focus. Anotherdifference lies in the typesof indicators used. Toassess financial liberaliza-tion, we used quantitativeindicators; for institu-tional development, weused qualitativeindicators.
In practice wedevised a “score card”(Annex Table 3.1) listingthe two main objectivesof institutional develop-ment and distinguishingthem according towhether they are morelikely to be supported bya FSL or by a FIL. Thetwo main institutionaldevelopment objectivesare to (1) strengthen
financial sector infra-structure; and (2)strengthen individualinstitutions.
These objectives areachieved by using regula-tory and institutionalreform measures at thelevel of individual banks.Financial sector infra-structure measures mainlysupported by FSLsinclude, but are not lim-ited to, changes in centralbank law, capital andmoney market regulationsand supervision practicesof central banks. Institu-tional reform measuresfor individual banksunder FILs include, butare not limited to,improving governance,management, and portfo-lio quality of the banks.
Using a binary sys-tem, performance indica-
tors were constructed tomonitor performanceunder each reform mea-sure for initial conditions,the policy implementationrecord, outcomes, andsustainability. Satisfactoryscores were rated as “1”and unsatisfactory oneswere rated as “0”. Eachindicator was given thesame weight in the scor-ing system.
The scores undereach subcategory of ini-tial conditions, implemen-tation progress, outcome,and sustainability wereaggregated and averagedfor FSLs and FILs sepa-rately. In addition, wederived aggregated resultsby regions, level of percapita GDP, and intensityof Bank lending.
BOX 3.1: SAMPLE AND METHODOLOGY TO ASSESS INSTITUTIONAL DEVELOPMENT
Improving thelegal system isrelatively easy, butrestructuring thebanking systemtakes a long time.
28
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
countries seeking Bank loans more often made a greatereffort to improve performance so that they could con-tinue receiving loans. It may also be because of theBank’s more effective monitoring of loan implementa-tion (or those who execute a country’s program maydevelop expertise as time goes by and more loans comein). Certainly, portfolio quality and resource mobiliza-tion show better outcomes in countries with high loanfrequency. We find at the institutional level an echo ofwhat was identified in the conceptual framework, and
observed more generally in adjustment operations, thatreform in the financial sector is a process, not an event.
Key Factors Affecting Institutional DevelopmentDifferent components of institutional development areinfluenced by different factors. Whether financial infra-structure is improved, for example, depends on howdetermined governments are to put through and imple-ment the necessary legislation. The same is true forrestructuring banks (the whole system as well as individ-
TABLE 3.2: RESULTS IN LOW- AND MIDDLE-INCOME COUNTRIESINITIAL IMPLEMENTATION
CONDITIONS RECORD OUTCOME SUSTAINABILITY
LOW MIDDLE LOW MIDDLE LOW MIDDLE LOW MIDDLE
FSLsStrengthening financial
infrastructure 0.13 0.30 0.94 0.95 0.82 0.82 0.66 0.83Strengthening institutions: 0.08 0.22 0.67 0.73 0.47 0.68 0.65 0.80
a. Central bank supervisionpractices 0.17 0.25 0.67 0.79 0.60 0.79 0.80 0.86
b. Restructuring of banks 0.00 0.18 0.67 0.67 0.33 0.58 0.50 0.75
FSLsPortfolio quality 0.31 0.60 0.36 0.45 0.36 0.42 0.21 0.39Technical capability 0.46 0.63 0.58 0.68 0.50 0.64 0.58 0.68Resource mobilization 0.18 0.32 0.20 0.44 0.20 0.36 0.10 0.48Financial performance 0.55 0.65 0.57 0.35 0.50 0.38 0.43 0.48
Source: PRs, SARs, and OED estimates
TABLE 3.3: RELATIVE SUCCESS IN COUNTRIES WITH HIGH LOAN FREQUENCY (4 OR MORE LOANS, FSLs PLUSFILs) AND LOW LOAN FREQUENCY (1–3 LOANS)
INITIAL IMPLEMENTATIONCONDITIONS RECORD OUTCOME SUSTAINABILITYFREQUENCY FREQUENCY FREQUENCY FREQUENCY
HIGH LOW HIGH LOW HIGH LOW HIGH LOW
FSLsStrengthening financial
infrastructure 0.29 0.17 0.92 0.89 0.94 0.70 0.83 0.61Strengthening institutions
including 0.34 0.08 0.79 0.63 0.78 0.51 0.68 0.77Bank restructuring 0.40 0.00 0.80 0.62 0.80 0.38 0.60 0.69
FSLsPortfolio quality 0.52 0.47 0.47 0.21 0.43 0.21 0.40 0.15Technical capability 0.53 0.62 0.67 0.55 0.62 0.45 0.71 0.45Resource mobilization 0.32 0.21 0.46 0.11 0.38 0.11 0.50 0.11Financial performance 0.66 0.50 0.45 0.36 0.50 0.29 0.52 0.36
Source: PRs, SARs, and OED estimates
29
ual banks) and for central banks’ exercising supervision.The quality of the macroeconomic environment appearsto have little or no impact on outcomes in those areas.But other components of institutional development, suchas portfolio quality, resource mobilization, and financialperformance, depend heavily on macroeconomic stabilityand the creditworthiness of bank clientele—two factorsbank managements cannot control.
In this context it is not surprising that outcomes ininstitutional development under FSLs have been uni-formly better than those under FILs(with the exceptionof improving technical ability (whose outcome is betterthan that of portfolio quality, resource mobilization,and financial performance). Macroeconomic failuressuch as inflation militated against banks’ efforts toimprove their institutional profile. This was exacer-bated by the weak financial position of state enter-prises, core borrowers who invariably reneged on theirloans. Many countries also had to contend with over-bearing government interventions, government micro-management of bank operations, and inadequatemargin spreads. Development banks in particular, themain participants under FILs, are generally prohibitedfrom accepting domestic deposits, making themfinancially dependent on the government or on multi-lateral institutions.
Two conjectures arise from the preceding observa-tions. First, macroeconomic stability should be ensuredwhen institutional development programs are initiated.FSLs may include instruments to achieve that state; FILsdo not. That is the principal reason why FSLs have grownat the expense of FILs in recent years. Second, since nei-ther FSLs nor FILs are designed to directly address enter-prise restructuring, there is no automatic linkage betweenfinancial intermediary reforms and financial reform in theproductive sector.5 Without such a link, the “evergreen-ing” of bad bank loans may continue.
More generally, the Bank has been closely involved inrestructuring and privatizing state enterprises, but may nothave coordinated its separate programs directed at the realand financial sectors. A recent Bank study (Pohl and oth-ers 1997) shows that in countries that privatized rapidly,the privatized firms improved their profitability relativelyquickly, making government intervention in banks (andBank support) unnecessary. By contrast, in countries withslow privatization, loan portfolios of commercial banks’were even worse than expected. Rapid privatization in the
industrial sector is therefore an important precondition forBank intervention in the financial sector. But we must adda caveat here. Privatization, though necessary, does notsucceed unless an appropriate, transparent regulatorymechanism is in place or the ground work is assiduouslyprepared to find the buyers of equity in privatized banks.Mexican bank privatization failed to prevent collapse ofbanks, because care was not taken to choose buyers on thebasis of their records and because bank supervision wasinadequate to prevent deterioration of banks’ loan quality.Monetary authorities were lax in controlling expansion ofbank credit. In Ghana, potential buyers were scarcebecause of the obstacles placed in the way of allowing for-eigners to own bank stocks.
The Sustainability of Institutional ComponentsTo gauge the success of adjustment-related operationssuch as FSLs and FILs, it is not enough to look at out-comes. It is also important to know whether programbenefits are sustainable for both the borrowing countriesand the Bank. Sustainability is vital if countries’ overalldevelopment programs are to be maintained. Withoutsustainability, they must start projects all over again,using scarce human resources. The Bank’s task managersmust scrutinize programs with a view to removing loop-holes in their designs or implementation so that Bankinterventions can be more efficient in the long-run.
It is striking that institutional development is farmore sustainable under FSLs than under FILs (Table 3.1and Box 3.2). The record on sustainability has also beenuniformly higher in middle-income countries under bothFSLs and FILs (Table 3.2). The same holds true, but notas strongly, for countries with higher loan frequencies(Table 3.3).
FSLs and FILs differ in sustainability, partly becauseof structural factors that directly affect institutionaldevelopment (Box 3.2). For FSLs inadequate attentionwas given to the payments system, accounting andauditing standards, collateral laws and regulations, andeducation programs in accounting and banking that arecritical to sustainability. For FILs performance waslower because of uneven professionalism in banking,limited on-site training of staff, and inadequateimprovements in the quality of banks’ portfolios.
I n s t i t u t i o n a l D e v e l o p m e n t
30
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
T he Bank’sinvolvement inlending through
financial intermediaries tosupport private industrialdevelopment and tostrengthen lending institu-tions in Pakistan startedin the late 1950s. Banklending focused for a longtime on individual devel-opment finance institu-tions and national com-mercial banks.
The Pakistan Indus-trial Credit and Invest-ment Corporation(PICIC) project, approvedin 1980, was the Bank’seleventh loan to thatinstitution. The IndustrialDevelopment Bank ofPakistan (IDBP) project,approved in FY82,focused on lending to
small and medium-sizeenterprises (SMEs). Withthe Industrial InvestmentCredit (IIC) project(FY84), the Bank adopteda multi-institutionalapproach, lending toPICIC, IDBP, and thenational commercialbanks (NCBs). Two moreIIC projects followed(FY86 and FY89). Finally,the Bank was involved inlending to small-scaleindustry (SSI) with threeSSI projects (FY81, FY84,and FY87).
Outcome. Most of theBank’s credit operations,although accomplishingtheir basic task of trans-ferring Bank resources tofinance investment, werenot successful at financing
viable projects throughefficient, financially-soundinstitutions. And imple-mentation of their techni-cal assistance componentswas decidedly mixed.Despite the Bank’s longinvolvement with PICIC,the recurrence of funda-mental problems suggeststhat trying to transforman inefficient, nominallyprivate but partly state-owned, government-con-trolled institution into avigorous and financially-sound intermediary was atbest a herculean task andat worst a forlorn hope.IDBP continued to sufferfrom inadequate recover-ies, low profitability, fail-ure to mobilize domesticresources, and failure tokeep abreast of change. It
made some progress indiversifying its portfolioand areas of business, butwas unsuccessful in itsarrears recovery program.The poor results of the(first) IIC are ascribed to anumber of external factorsand deficiencies inappraisal and projectselection. The financialperformance of PICIC andIDBP declined after 1987,as portfolios and loan col-lection rates worsened.Their financial positionswere adversely affected bythe government’s repres-sion of the banking sys-tem. Under the second IICproject, 92 percent of thesubloan commitmentswere made to the textilesector. This lack of diver-sification contributed to
BOX 3.2: PAKISTAN: THE LONG, DIFFICULT ROAD TOWARD SUSTAINABILITY OF FINANCIAL INSTITUTIONS
31
I n s t i t u t i o n a l D e v e l o p m e n t
deterioration in theparticipating financialinstitutions’ (PFIs’) bal-ance sheets. Most of thePFIs experienced collec-tion and portfolio prob-lems, there was consider-able managementturnover, and the govern-ment’s interference inmanagement adverselyaffected PFI performance.
In the first SSI proj-ect, lending through thecommercial banks provedsatisfactory, the banksbuilt up their subprojectlending capabilities, andsound projects werefinanced, although techni-cal assistance was notquite as successful. Thesecond SSI projectincluded a line of creditthrough IDBP, acting as
an apex institution, tofinance subprojectsthrough the commercialbanks and a technicalassistance component.Unfortunately, IDBPturned out to be an ineffi-cient apex agency and didnot supervise the nationalcommercial banks (NCBs)properly. The NCBs didnot adequately appraiseand monitor the subloansand as a result had majorrecovery problems. Infact, during implementa-tion of the third SSI proj-ect, the Bank was obligedto declare four out of fiveNCBs ineligible to furtherparticipate in the project.This was intended toforce them to take reme-dial actions. Three ofthem failed to participate
for the remaining years ofthe project life.
Lessons of experience. Animportant lesson fromPakistan is the need tomaintain the presence,analysis, and dialogue onthe sector and to follow-up with regular opera-tions, preferably adjust-ment-type, to reinforcewhatever progress ismade under one opera-tion. The Bank has had astop-and-go approach tothe financial sector inPakistan; essentially theBank backed-off when itshould have continued.The Bank made onefinancial sector adjust-ment loan in FY89 andnine years later approvedanother, the Banking Sec-
tor Adjustment Loan.Pakistan would have ben-efited from much earlierintroduction of financialsector reform, particu-larly reform aimed atestablishing uniform andtransparent accounting,developing legal, regula-tory, and enforcementinfrastructure; andstrengthening the bankingsystem. Institution-build-ing through FILs is moretime-consuming and vul-nerable to political inter-vention than the policy-based components ofSADs, which provide formarket-based determina-tion of interest rates andimprovements in institu-tional infrastructure.
BOX 3.2: PAKISTAN (CONTINUED)
33
Liberalization andFinancial Crises
44
Policies for financial sector reform must take into account the likelihood of financial
crises, because banking systems are often in financial distress on the eve of reform.1
Indeed, financial crises emerge from reform policies or are magnified by them, depend-
ing on how the reform policies are conceived, adapted to initial conditions, and sequenced. The
distress of banks, especially of publicly-owned banks, is caused both by interventionist govern-
ment policies and by weak internal bank management which results in poor credit evaluation
procedures, undiversified loan portfolios, and depar-tures from standard commercial banking practices. Theregulatory apparatus that oversees bank operations isusually inadequate because of regulatory authorities’limited capacities. Governments undertake financial lib-eralization to eliminate or reduce quantitative controlson banks’ assets and liabilities (Goldstein and Turner1996; Khatkhate 1993; Sunderarajan and Balino 1991).
A delayed policy response to macroeconomic imbal-ances that emerge in more financially integratedeconomies precipitates financial crises, because the mar-gin for policy error is small. This was well-illustrated bythe 1997 Thailand crisis. Despite evidence that the bahtwas appreciating in real terms because of the drop inexports, the large current account deficit, and the appre-ciation of the U.S. dollar relative to the yen, the mone-tary authority maintained the fixed baht-dollar parity.To defend it, domestic interest rates were raised. As aresult banks’ nonperforming assets increased sharply,leading to the bankruptcy of some financial institutions(Box 4.1). The 1997 Korea crisis is also a case of macro-economic imbalances revealing deep structural weak-nesses in the financial sector (Box 4.2).
When a financial system is fragile, it is vulnerable toeconomic shocks; fragility is a measure of susceptibility
to crises. Policy measures that reduce a financial system’sfragility tend to increase its ability to withstand shocks.In this chapter we develop asimple framework for quantify-ing this measure—a financialfragility index (FFI)—based ona set of fragility indicators.
The financial reform strat-egy embodied in the Bank’sadjustment operations has threedimensions: (1) reducing gov-ernment interventions that dis-tort the banking system, (2)restructuring and revitalizing banks in distress becauseof bad loans carried over from before the reforms set inimproving supervision and prudential control of banks,privatizing state-owned banks, and (3) establishingmodern accounting and prudential norms. This strategyminimizes the likelihood of financial crises.
Financial crises are common in financial systems allover the world but have been especially severe in devel-oping countries. In about a dozen countries, includingfour studied here, estimated losses from resolving finan-cial crises exceeded 10 percent of GDP or more (Caprioand Klingebiel 1996; Lindgren and Saal 1996). Some fac-
Policy measuresthat reduce a
financial system’sfragility tend to
increase its abilityto withstand
shocks.
34
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
tors causing banking crises are macroeconomic andexogenous to the financial system: others are endogenous.Exogenous factors include fluctuations in terms of trade,relative prices, international interest rates, exchange rates,and inflation. But the factor most relevant to adjustmentoperations and associated policies is endogenous: inade-quate preparation for financial liberalization. Financialreforms often generate new risks for banks that, withoutproper precautions, can precipitate a financial crisis.Interest rate liberalization deprives banks of the protec-tion they previously enjoyed. Paradoxically, creditexpands quickly, despite higher real interest rates. Bankmanagers accustomed to controlled interest rates are nei-ther skilled nor experienced enough to evaluate newsources of risk, including credit, interest rate, exchangerate, and other market risks. Lacking the skills to evaluateloans and faced with new competition, both foreign anddomestic banks may make poor credit decisions.
Bank adjustment operations can help preventendogenous financial crises if they are attuned to specific
country circumstances and balance the components ofreform, such as removing distortions, improvingaccounting standards, instituting prudential controlsand bank supervision, and correctly sequencing reformpolicies (Goldstein and Turner 1996; Khatkhate 1996;Sunderarajan and Balino 1991). The reforms also createstronger financial systems that can better weather exoge-nous shocks. In this section we examine how effectivelythe Bank’s financial sector operations and interventionsdeal with financial crises. We rely on case studies, whichallow us to identify the precise nature of interventions inspecific types of financial crises. Case studies also allowus to identify common patterns among crises by exam-ining different factors.
The case studies focus on financial crises in Mexico,the Philippines, and Venezuela—each presents a differentperspective on Bank intervention. Mexico’s first financialcrisis occurred in 1987, before the approval of a financialsector adjustment program. The second occurred afterBank operations began. The financial crisis in the Philip-
The genesis of acrisis. Recentevents indicate
that the 1997 crisis wasprecipitated by a delay inpolicy response to emerg-ing macroeconomicimbalances. Exportgrowth, vibrant and sus-tained until 1996, slowedunder the pressure ofheavy competition fromChina and Vietnam inlabor-intensive products.Given Thailand’s relianceon foreign capital tofinance domestic invest-ment, the current accountdeficit rose to around 8percent in 1996. Despitethe evidence that the bahtwas appreciating in realterms, the fixed exchange
rate was maintained.When the macroeconomicimbalance reached criticallevels, foreign capitalinflows shriveled, andforeign banks started towithdraw their loansfrom businesses anddomestic banks, despitetheir commitment torolling them over. Eventu-ally, the authorities wereleft no alternative than tofloat the baht.
The ensuing devalua-tion revealed the fragilityof the Thai banking sys-tem. Banks were alreadysaddled with large non-performing assets,amounting to 9 to 10 per-cent of total bank loans.Their financial position
worsened further whenthe burden of banks’ andtheir customers’ repay-ment of foreign currencyliabilities in domestic cur-rency ballooned. TheMinistry of Finance andthe Bank of Thailandextended liquidity tobanks to tide them overthe pressure on theirresources. But the loansto corporate customers,groaning under the repay-ment burden, becamenon-performing, furtheraggravating the banks’financial position.
The currency crisiswould have affected thebanking system less if thesystem had been suffi-ciently resilient to adverse
macroeconomic fluctua-tions and asset price gyra-tions. Commercial banksand non-bank financialinstitutions did not havethe capability to efficientlymanage their portfoliorisks, and the authoritiesdid not have in place anadequate framework ofprudential control andsupervision. Loans werenot made on strict businessstandards, because politicalinfluences interfered withbank management. Aboveall, the readiness of theauthorities to bail-outfinancial institutions in dis-tress, either because oftheir own management dif-ficulties or because theconsequences of exchange
BOX 4.1: THAILAND’S 1997 FINANCIAL CRISIS—WHAT DID THE BANK DO?
35
L i b e r a l i z a t i o n a n d F i n a n c i a l C r i s e s
pines took place after approval of the Bank’s financial sec-tor operations; the crisis in Venezuela occurred afterapproval.
A low-fragility system is sound, stable, and well-functioning. Here, we use well-known indicators todevelop an aggregate view of system fragility. From anassessment of their contribution to financial fragility,two sets of indicators, macroeconomic and micro-economic, are scored on a scale of 1 to 3, with 1 denot-ing low fragility and 3 high fragility. A simple averageof these scores for each set generates a macroeconomicand microeconomic FFIs.2 The macroeconomic fragil-ity index is an aggregate of 14 macroeconomic indica-tors covering internal and external balances, real inter-est rate levels, real sector growth, Brady bond spreads,stock market prices, capital flows, debt, credit growth,and terms of trade exposure (see Annex Table 4.1). Themicroeconomic index is an aggregate of 18 indicators,including market structure of the banking sector, fi-nancial soundness of the banking sector, and central
bank, and progress in enterprise restructuring (AnnexTable 4.2).
MexicoThe Bank had two financial sector operations in Mexicoin the period under study: one in 1989 (L3085), whichwas completed in 1993 and one in 1995 (L3911), sched-uled for completion in 1998. During the first financial sec-tor operation (1989–93) the macroeconomic FFI declinedslightly, from 1.7 to 1.6, reflecting a negligible improve-ment in Mexico’s overall financial position. Moreover, themicroeconomic FFI increased from 1.8 to 2.1, revealingthe banking system’s growing fragility. Between the twoprograms the country’s overall financial position wors-ened. Both FFIs increased sharply between 1993 and1994, the first year of the second Bank-supported pro-gram, indicating that even the limited effects of the firstprogram could not be sustained. Then between 1994 andthe first quarter of 1995, both FFIs increased again, sug-gesting the possibility of trouble (Figures 4.1a and 4.1b).
rate changes created seriousmoral hazard problems forthe banking system.
World Bank’s Involve-ment. Although the WorldBank’s borrowing pro-gram has been modestduring the decade begin-ning in 1987, the Bankhas been a close observerof the Thai financial sys-tem through various stud-ies it carried out. A majorwork was undertaken toanalyze Thai macroeco-nomic policy in the 1970sand the 1980s as a part ofa global study, but itbarely touched on theThai financial sector. Thiswork was followed by acomprehensive report in
1990 on the Thai financialsector carried out with theconcurrence of the Thaiauthorities. The report,while recognizing the vig-orous growth of the Thaieconomy, and its otherfavorable aspects, didspotlight, albeit cautiously,some of the features of thefinancial sector thatcaused concern. Thereport, however, over-played the strengths ofbank supervisors, suggest-ing that they were efficientand that the skills of indi-vidual examiners werehigh. The Bank used thereport to understand theThai financial system butdid not use policy dia-logue to bring the portents
of future trouble to theattention of the authori-ties. Another Bank study,prepared in the Bank’sCountry EconomicsDepartment in 1990,expressed the same appre-hension about the Thaifinancial sector, but againwith a lack of urgency.
The Country Assis-tance Strategy prepared in1994 had generallyeuphoric overtones. Insome places, however, thereport mentioned the diffi-culties in maintainingexternal balance in theface of large saving andinvestment imbalances. Inproposing new ESW pro-grams, it suggested thatmore attention should be
focused on financial sectorreform and domestic sav-ings mobilization. Butnone of those suggestionswere implemented. At thefirst sign of macroeco-nomic imbalances emerg-ing and export growthslowing, the financial crisisgathered speed and inten-sity, prompting in Septem-ber 1997 the approval ofa Bank TA loan tostrengthen financial infra-structure. Another Bankloan to support therestructuring of financecompanies was approvedin December 1997 for theamount of US$ 350 mil-lion. Both loans were pre-pared without the benefitof recent ESW.
BOX 4.1: THAILAND (CONTINUED)
36
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
Korea’s strikingand sustainedeconomic growth
since the 1970s wasbrought about by thegovernment’s interven-tionist policies, whichconformed to a marketsystem because ofKorea’s peculiar institu-tional characteristics. TheKorean economic systemcomprised two subsys-tems: a public systemcommitted to develop-mental goals and a pri-vate system pursuingprofit motives. Under thisdouble system the gov-ernment, the financialsector, and private pro-ducers worked togetherefficiently for a longperiod. But in the pastfive years the Koreaneconomy reached nearfull employment, andwage pressure mounted.Korean industries facedgrowing competition inlabor-intensive exportsfrom China and Vietnam.Once the discipliningforce of exports lostmomentum, Korea’s GDPgrowth started to decline.And in such an environ-ment the weaknesses ofboth the corporate sectorand the financial system,
which had been coveredby the country’s highgrowth rate, came to thesurface. Corporate bank-ruptcies increased, and,consequently, the bank-ing system’s non-per-forming assets, whichwere already high andhad been masked byweak accounting stan-dards, ballooned. TheKorean authorities con-tinued to direct the finan-cial system to lend to theloss-making corporatesector. The governmentthen compensated thefinancial system throughcentral bank assistance.When the financial crisiserupted in Thailand inJuly 1997, it echoed inneighboring East Asiancountries, includingKorea, leading to a lossof foreign investor confi-dence. Even when therewere clear signals ofimpending trouble, theKorean authorities triedto avert the crisis byusing the same interven-tionist policies. This time,however, these effortscould not be sustained,and the financial sectoras well as the real econ-omy fell into a majoreconomic crisis.
The World Bank’sInvolvement. The Bankwas aware of seriousweaknesses in Korea’sfinancial sector, as evi-denced in its reports. Four1993 OED audit reportson Korea’s industrialfinance and financialintermediary projectsspecifically indicated thatcompetition betweenbanks had not been intro-duced, interest rate liber-alization was limited, andthe supervision and pru-dential control of finan-cial intermediaries hadnot been changed. A 1993Bank financial sectorstudy gave a similarreport, emphasizing thegrowing amount of non-performing assets and theweakness of accountingstandards. The studyargued that governmentcontrols had led to finan-cial market segmentationresulting in inefficientlending policies and non-performing assets. Thelatter posed a seriousproblem for the financialsector, and a lack oftransparency in account-ing standards underesti-mated the size of theseassets. The report sug-gested total abolition of
directed credit. As a fol-low-up to the financialsector study, the Bankcooperated with Koreanauthorities to design ablueprint for a phasedprogram of completefinancial deregulationbeginning in 1994. Butimplementation was half-hearted, and whateverwas achieved was drivenmore by the government’sattempt to qualify forOECD status (and as aresult of pressure fromthe US government) thanit was by Korean convic-tion. Though the Bankcould diagnose the ills ofthe Korean financial sys-tem, it could not persuadethe country to undertakesubstantial reforms,because of the limitedleverage it had withKorea. The 1997 crisisprompted the Koreanauthorities to adopt anew attitude towards thestructural problems of thefinancial sector. InDecember 1997 the Bankapproved a US$ 3 billioneconomic reconstructionloan upon strong declara-tions from the new gov-ernment that Korea waswilling to embark on seri-ous structural reforms.
BOX 4.2: KOREA’S 1997 FINANCIAL CRISIS: WHAT WENT WRONG WITH THE MIRACLE ECONOMY?
37
L i b e r a l i z a t i o n a n d F i n a n c i a l C r i s e s
The message from the FFIs can be compared withthe conclusions reached in the PCR (No. 11638) and thePAR (No. 12077). Both reports describe the outcomes ofthe adjustment program as “highly satisfactory” and
express few doubts about their sustainability. Althoughboth reports were issued in 1993 (the PCR in Februaryand the PAR in June), the picture they provide is mis-leading. By early 1993, and certainly by the middle ofthat year, there were clear signs of a significant increasein financial fragility and a reversal of many programoutcomes. Given how rapidly the Mexican economywas changing in early 1993, the PAR could have reliedon more current data to better evaluate project perfor-mance and the prospects for sustainability.
In formulating the FSAL, the Bank committed errorsof both commission and omission. The FSAL programwas put together quickly in an attempt to respond to theimmediate needs of the Mexican financial sector. Thus,the quality of project preparation and design may havesuffered. Without undertaking economic and sectorwork, the Bank had little direct knowledge of bank reg-ulation and supervision, and of development banks. Asa result, there was too much pressure on local institu-tional capacity to properly implement reforms. Further-more, the project relied on action plans for institutionaldevelopment, rather than on concrete measures andappropriate incentives, and there was little Bank follow-up to implement those plans. The action plans on bankregulation and supervision were either not implementedor poorly implemented, mainly because of limitations inlocal capacity.
A major flaw of the FSAL, which might have con-tributed to increased microeconomic fragility, was theabsence of any conditionality for the loan portfolios ofcommercial banks. A recent Bank study concluded thata heavy concentration of consumer loans in the bankingsystem’s portfolio may have been a major cause—morethan the devaluation of the peso itself—of the bankingcrisis in 1994–95 (Wilson, Caprio, and Sanders, 1997).Further, dominance by large financial groups of thebanks (as well as their misleading accounting and audit-ing standards) did not disclose the actual deteriorationof asset quality. This deterioration went undetectedbecause of regulatory forbearance that allowed bankowners to recoup their losses.
The sustainability of the FSAL was underminedlargely by a combination of government policies andexternal and domestic shocks which led to the crisis inlate December 1994. The financial sector reforms dic-tated under the FSAL were neither comprehensive nordeep enough to strengthen the financial system to with-stand such shocks. Regardless of the optimistic assess-ments of the PCR and PAR, the Mexican financial sys-
FIGURE 4.1a: FSAL AND FINANCIAL FRAGILITY INMEXICO: MACROECONOMIC LEVEL
1989Initial
conditions
1993
Outcome
1994 1995-Q1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mac
roec
onom
ic F
FI
1.7 1.6
2.8 2.9
Impact
FIGURE 4.1b: FSAL AND FINANCIAL FRAGILITY INMEXICO: MICROECONOMIC INSTITUTIONAL LEVEL
1989Initial
conditions
1993
Outcome
1994 1995-Q1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mic
roec
onom
ic F
FI
1.8
2.1
2.83.0
Impact
Source: OED estimates
38
tem was extremely fragile at the end of 1993. It col-lapsed under the shocks of 1994. A major cause of thatcollapse was the weak system of prudential oversightand supervision instituted for newly-privatized banks.
The Mexican macroeconomy and banking systemwere very weak when the Bank provided a financial sec-tor reconstruction loan (FSRL) in early 1995 (Figures4.2a and 4.2b). One objective of that loan was to helpthe Mexican government contain the crisis. A longer-term goal was to help Mexico restructure its financialsystem to prevent the recurrence of systemic crises. Adistinguishing feature of the FSRL was that it directlyand effectively addressed the fragility of the financialsystem at the microeconomic and macroeconomic levels.As a result both measures of financial fragility declinedsharply. The macroeconomic FFI declined from 2.9 inthe first quarter of 1995 to 1.7 in the first quarter of1996, reflecting mainly lower inflation, fiscal and cur-rent account deficits, real interest rates and domesticcredit growth. During the same period the microeco-nomic FFI declined from 3 to 2.7, indicating someimprovement of the banks’ financial position and aslightly more competitive environment.
The positive effects of the FRSL reforms persisteduntil early 1997 (Figures 4.2a and 4.2b). The macroeco-nomic fragility index decreased slightly from 1996-Q1to 1997-Q1, mainly as a result of lower inflation. Themicroeconomic index also declined, as the publicbecame more confident in keeping deposits in commer-cial banks, and the central bank’s financial positionstrengthened. The quality of commercial bank supervi-sion improved as a result of more intensive technicalassistance programs. The level of the microeconomicFFI, however, remained above 2 in the first quarter of1997, indicating that the financial sector was still fragileand vulnerable to crises. Effective technical assistancemay help improve supervision, but policies to provideadequate staff and financing and to ensure autonomy ofthe supervisors are more important.
VenezuelaVenezuela enjoyed the highest per capita income in LatinAmerica during the early 1970s. Following the 1973 oilboom, however, the economy was increasingly misman-aged, governed by a set of highly complex regulations,subsidies, direct state interventions, inadequate socialsector policies, and excessive reliance on petroleum rev-enues. Eventually, the economic and social problems ofthe country reached crisis by 1988. A new administration
took office in 1989 and almost immediately began asweeping economic stabilization and reform. As a conse-quence Venezuela became eligible for a Brady Plan pro-gram of debt and debt service reduction (DDSR). In July1989 the government proposed a program of new bor-
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
FIGURE 4.2a: FSRL AND FINANCIAL FRAGILITY INMEXICO: MACROECONOMIC LEVEL
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mac
roec
onom
ic F
FI
2.8 2.9
1.7
1.4
1994 1995-Q1 1996-Q1 1997-Q1
Initial conditions Current outcome
FIGURE 4.2b: FSRL AND FINANCIAL FRAGILITY INMEXICO: MICROECONOMIC INSTITUTIONAL LEVEL
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mic
roec
onom
ic F
FI
2.83.0
2.7
2.4
1994 1995-Q1 1996-Q1 1997-Q1
Initial conditions Current outcome
Source: OED estimates
39
rowing and DDSR to its external creditors, consistentwith the Brady Plan announced earlier that year. TheFSAL operation (Loan 3224-VE) approved by the Bankin 1990 and later evaluated by OED should be viewedagainst this background (see World Bank 1995g).
The FSAL supported a program of further interest rateliberalization and bank privatization. It also aimed toenhance competition and strengthen both the regulatoryframework and individual financial institutions. Bankstudies completed in 1988–89 showed the necessity of cor-recting insolvency problems as early as possible. At thattime the financial system was oligopolistic in structure,and there were close links between major borrowers andbanks. Venezuela was particularly vulnerable to financialcrises. The FSAL proposals were not commensurate withthe magnitude of Venezuela’s financial problems.
Indeed, the macroeconomic fragility index forVenezuela, relatively low in 1990, increased substan-tially during program implementation (Figure 4.3a). Thederegulation of interest rates, began as early as 1989,became dangerously destabilizing because the marketstructure remained oligopolistic and poorly-regulated.In addition, the macroeconomic environment becameincreasingly unstable. Two military coup attempts inFebruary and November 1992 challenged implementa-tion of the stabilization-cum-adjustment program initi-ated in 1989. Then oil prices declined substantially in1993. Macroeconomic management deteriorated sub-stantially in 1993 and 1994—the fiscal deficit grew from3.9 percent of GDP to 14.5 percent, the exchange ratebecame overvalued, and domestic credit rose sharply.
The microeconomic fragility indicator which washigh in 1990 (Figure 4.3b), declined only slightly andremained high during and after implementation. Thisindex was high to begin with because the banking sys-tem was highly oligopolistic and insolvent. Fragilityremained high throughout because measures to restruc-ture commercial banks, strengthen the supervisory func-tion of the superintendency of Banks and Financial Insti-tutions, and reduce bank fraud were not taken. Thegovernment did not approve the new banking lawbefore 1993, even though it had already liberalizedinterest rates. After the law was approved, authorities’actions to deal with the financial crisis were not consis-tent with the procedures set by the law regarding inter-vention and liquidation of financial institutions.
The outcome of the FSAL was unsatisfactory mainlybecause the program it supported did not deal with sol-vency problems as quickly as was required and because
the government liberalized interest rates without imple-menting a regulatory and supervisory framework. Fur-ther, the Bank program did not include appropriate policyinstruments to increase competition, develop financialinfrastructure, and strengthen financial institutions.
L i b e r a l i z a t i o n a n d F i n a n c i a l C r i s e s
FIGURE 4.3a: FSAL AND FINANCIAL FRAGILITY INVENEZUELA: MACROECONOMIC LEVEL
1990Initial conditions
1993–94 Outcome
1996–97 Impact
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mac
roec
onom
ic F
FI
1.5
2.6
1.8
FIGURE 4.3b: FSAL AND FINANCIAL FRAGILITY INVENEZUELA: MICROECONOMIC INSTITUTIONAL LEVEL
1990Initial conditions
1993–94 Outcome
1996–97 Impact
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mic
roec
onom
ic F
FI
2.9
2.62.4
Source: OED estimates
40
The PhilippinesIn 1987 the Bank carried out a study of the financial sec-tor in the Philippines. It identified five problem areas:bank supervision, the protection of depositors, the costof intermediation, poor savings mobilization, and thedelivery of long-term credit. These problem areasbecame the starting point of a dialogue with the govern-ment for the preparation of a FSAL (Loan 3049-PH) inlate 1988. At that time the financial soundness of thecentral bank was not considered to be an issue.
Overall, the FSAL was successful in reducing thefragility of the financial sector (see World Bank 1996b).The macroeconomic fragility index was 1.8 to start withand declined while the stabilization policies were effec-tive during the period 1994–96. The index, however,substantially increased in 1997 (Figure 4.4a), reflectingthe effects of the turmoil in East Asia capital marketsand a relatively large trade deficit. The microeconomicfragility index fell after 1989 as a result of a continuouseffort to strengthen the central bank’s supervisory func-tions, but increased in 1997, although not as fast as themacro index (Figure 4.4b).
A serious central bank crisis, however, occurred dur-ing project implementation. The financial fragility of thecentral bank had grown over the years to the point thatit became bankrupt in 1993. Central bank deficits cal-culated on a cash basis appeared in the mid-1980s andincreased substantially in the early 1990s. Cumulativelosses from 1983 until 1993 (the year the central bankwas recapitalized) amounted to 180 billion pesos (12.2percent of GDP).
Since the crisis was confined to the central bankand did not spread throughout the financial system, theindicators, which are systemic in nature do not tracewhat happened to the central bank. Therefore, weadded a central bank fragility index to follow ex postthe intensity of the crisis. This fragility index is an aver-age of five indicators related to profitability and sol-vency ratios as well as to quality of governance andshort-term treasury bill rates (Annex table 4.3). Becausethe crisis was built up over several years, the fragilityindex reflects high losses and insolvency in 1992 andearlier (Figure 4.4c).
In 1993 the central bank was completely restruc-tured—and thus strengthened, as shown by the substan-tial decrease of the fragility indicator. The process wassuccessful but costly, partly because the interventioncame so late and partly because the entire institutionwas fully restructured. Earlier diagnosis of the central
bank’s financial difficulties and earlier application ofremedies might have averted the need for drastic restruc-turing and reduced the cost of adjustment.
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
FIGURE 4.4a: FSAL AND FINANCIAL FRAGILITY INTHE PHILIPPINES: MACROECONOMIC LEVEL
1989Initial
conditions
1993 Outcome
1994–96 Impact
1997 Projected
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mac
roec
onom
ic F
FI
1.8
1.5 1.6
2.0
FIGURE 4.4b: FSAL AND FINANCIAL FRAGILITY INTHE PHILIPPINES: MICROECONOMIC INSTITUTIONALLEVEL
1989Initial
conditions
1993 Outcome
1994–96 Impact
1997 Projected
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mic
roec
onom
ic F
FI 2.4
1.81.6 1.7
Source: OED estimates
41
Conclusions and LessonsAll three countries began with a high microeconomicfragility index. It was reduced in the Philippines, and(after a second loan) in Mexico. It did not decline inVenezuela. The main difference in performance is attrib-utable to the rapidity and effectiveness of interventionsthat improved the financial position of the central bankand the soundness of the banking system. The macro-economic indexes show better results in the short-term,as expected, since it is relatively easier to stabilize aggre-gates than to change institutions.
The point of this fragility indicator analysis is not tosuggest that the FFIs can predict crises with any degreeof precision. It is to show that far too little data onfinancial sector performance has been collected,processed, and analyzed in a systematic way and thatmore rigorous surveillance of financial sectors is neces-sary. As the recent crisis in East Asia makes dramaticallyclear, lack of regular and reliable monitoring informa-tion on financial system risk exposures can have seriousconsequences.
Some broad conclusions may be drawn from theseexperiences:
• The standard indicators (those derived from OD8.30 guidelines) used to assess outcomes in thefinancial sector under Bank-supported programs
are not sufficient for assessing the fragility of afinancial system.
• The costs of financial crises can be enormous: theadditional costs of assessing fragility should beseen in this light. A set of reasonably simple indi-cators can be developed to provide insights intofinancial fragility. In addition, the Bank shouldpursue its recent initiative to systematically ana-lyze the determinants of banking crises in view ofestimating the probability of occurrence (seeDemirgüç-Kunt and Detragiache 1997). Moreattention should be given to this kind of analysisand to the development of evaluative measures atthe onset of a project.
• The Bank should continue to monitor financialsystems long after financial sector operations areconcluded (Mexico). It might be useful for theBank to seek feedback from, and to coordinateactivities with, the IMF in its efforts to identifypotential crises, since the IMF has key surveillanceresponsibilities in regard to the financial sector(see Chapter 6).
• Financial sector reform is not sustainable unless itis comprehensive and penetrates down to the insti-tutional level (Mexico). This is possible only ifsolid ESW is completed before operations andonly if technical assistance components are part ofthe package.
• Effective prudential regulation and supervision arevitally important to the soundness of the financialsystem (Mexico, Venezuela).
• Deregulating interest rates can be dangerouslydestabilizing if done when macroeconomic condi-tions are unstable, financial markets are oligopo-listic, many borrowers show serious financial dis-tress, connected lending is common, banks are ina fragile financial condition, banking supervisionis incompetent, and banking legislation and regu-lations are weak (Venezuela) (see World Bank1995d).
• In preparing financial sector studies, the Bankshould pay more attention to the financial per-formance of the central bank and, as much aspossible, review off-balance sheet items, especiallyguarantees and derivatives operations (thePhilippines).
L i b e r a l i z a t i o n a n d F i n a n c i a l C r i s e s
FIGURE 4.4c: CENTRAL BANK FRAGILITY—THEPHILIPPINES (1989–1996)
1989–91 1992 1993 Outcome
1994–95 Impact
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Mic
ro-F
ragi
lity
Inde
x (M
IFI) 3.0 3.0
1.8
1.4
Initial conditions
Source: OED estimates
43
Bank Performance
55
For an adjustment project to succeed, the Bank must take a comprehensive view, drawing
on its own and outside knowledge and resources. Bank staff must have a complete picture
of the economy. Even a well-designed project may be affected by adverse macroeconomic
conditions, such as high and rising inflation or overvalued exchange rates, or by microeconomic
distortions, such as price controls, subsidies, lack of competition, or restricted trade regimes.
ESW should precede the preparation of an adjustment operation, placing the Bank’s project
work securely in the context of broad macroeconomicpolicy and strategy imperatives. This work should thenbe supplemented by consultation with the IMF, whichhas a different (but not fundamentally divergent) focuson issues, and with the IFC, which has expertise in pri-vate sector issues. The Bank’s performance can bejudged by how well it designs projects to suit a particu-lar country situation. In this chapter we assess Bank per-formance with institutional development, sequencing ofadjustment strategies, and cooperation and coordinationwith the IMF and the IFC.
Economic and Sector Work Much ESW was, in fact, prepared before OD 8.30(World Bank 1992) and did not always cover the itemsspecified there. Reports in the 1980s generally dealt withthe macroeconomic setting and structure of the financialsystem. Now, as specified in OD 8.30, other factors,especially resource mobilization, interest rate liberaliza-tion, and directed credit are also examined. Some atten-tion was paid to the strengthening of financial institu-tions and the importance of the informal financial sectorin the 1980s. Reports in the early 1990s cover these andother OD issues more systematically, placing moreemphasis on regulatory reform.
The extent to which ESW was attached to the proj-ect cycle and the impact on project outcome give a mea-sure of the complementarity between ESW and projectlending. A 1995 Bank study (Schneider 1995) found thatless than half of Bank projects had been preceded byESW in the three years before project approval. Whenconsidering the entire set of the 88 FSLs in this study,and extending the period of presence (or absence) ofESW to five years preceding project approval, we foundthat no ESW was done for an even higher percentage ofprojects (55 percent). This has certainly deprived thepotential borrower of the benefit of in-depth back-ground information and a well-argued long-term strat-egy to develop the sector (Box 5.1). But ESW was com-pleted for the financial sector in countries (Botswana,Burundi, Colombia, Lesotho, Mauritius, Nigeria,Rwanda, Thailand, and former Yugoslavia) for whichthere was no Bank lending for financial sector adjust-ment. This might suggest that ESW was considered as asubstitute for lending.
Turning to our sample of 23 countries we alsoobserve a large proportion of projects not preceded byESW within five years (47%). When using the PIRmethodology of Chapter 2 to measure the performanceof Bank-supported reforms at country level, we find that
44
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
Bank Operations.Pakistan’s eco-nomic reform
program began in 1977when the governmentembarked on a newinvestment strategy,emphasizing a leadingrole for the private sector.The results were stronglypositive: real privateinvestment in large- andmedium-scale industrialprojects grew at 13 per-cent a year in 1977–82,and real GDP growthaveraged 6.7 percent over1977–83. The Bank sup-ported the strategy withadjustment lending. ASAL (FY82) addressedstructural imbalances andan Export DevelopmentLoan (FY86) supportedthe introduction of amore open trade regime.An FSAL to strengthenmonetary management,the prudential framework,and the banking sectorwas approved in FY89.Financial sector reformwas initiated, and Banksupport came, althoughfairly late in the reformprocess. Bank assistanceto the real sector and thestrengthening of financialinstitutions through FILs,however, dates back tothe late 1950s (Box 3.2).
ESW. The first compre-hensive financial sectorreview was carried out in1987 and identified sev-eral structural weak-nesses, including over-reg-ulation of the monetarysystem, inadequate super-vision of financial institu-tions, an unstable strat-egy for governmentborrowing , and ineffi-cient institutional creditmarkets. In response, thegovernment drew up afinancial sector adjust-ment program. The pro-gram included measuresto remedy the weaknessesidentified in the Bankreview, but did notaddress directly reformsof development financeinstitutions (DFIs), insur-ance, or the stock market,nor did it address thebasic question of thesequencing of reforms.The FSAL supported pro-gram implementation.Although progress wasmade on the institutionalfront, some objectives offinancial reform(increasedresource mobilization,greater investment, moreefficient allocation, andhigher output growth—were not reached. In ret-rospect it is clear that theFSAL should have gone
further, placing moreemphasis on fiscalreform, greater indepen-dence for the centralbank, and more trans-parency in the centralbank’s operations.
Another Bank study,Restructuring the Finan-cial System: Building onFinancial Reform in Pak-istan, distributed in April1993, four years afterapproval of the FSAL,again drew attention topersistent deficiencies inthe financial system.These include weak andinefficient financial insti-tutions, with weak port-folios, poor debt recov-ery, excessive creditconcentration, and inade-quate capitalization; inad-equate prudential regula-tion and lax enforcement:inadequate central banksupervisory capacity; apoor framework for loanrecovery; the need forfurther bank privatiza-tion; and the need torestructure the DFIs.There were also manyunresolved policy issues.Evidently, financial deci-sionmaking must be freedfrom government interfer-ence, and financial insti-tutions like NCBs mustbe prepared for and sub-
jected to stronger compe-tition. Selected issues, butnot the most fundamen-tal, were supposed to beaddressed by the Finan-cial Sector Deepening andIntermediation Project,but the conditionalityremains unfulfilled, fundsundisbursed, and theBank is planning to can-cel the loan.
Lessons of Experience. Itis very difficult to mountsuccessful loans to bechanneled by DFIs,through unreformed, oronly partly-reformed statebanks in a heavily dis-torted financial sector. Itis also very difficult toachieve the objectives offinancial sector reform ina macroeconomic envi-ronment characterized bypublic overspending,inflation, and high realinterest rates. ESW cer-tainly deepened theknowledge of the sectorand identified majorstructural issues. ButESW alone is not enoughto develop a timely andclear financial sectorstrategy with an appro-priate sequencing of pol-icy measures.
BOX 5.1: THE IMPACT OF PAKISTAN’S ESW ON LENDING OPERATIONS
45
B a n k P e r f o r m a n c e
only one-third of the countries with no previous ESWhave a satisfactory outcome, while two-thirds of thecountries with at least one ESW before project approvalhave a satisfactory outcome (Table 5.2). In addition,when using the “hits and misses” methodology of Chap-ter 2 as a proxy for good project design, we find a posi-tive influence of ESW: in those countries where the Bankcarried out ESW on the financial sector no more thanfive years before at least one of the adjustment-relatedloans in the financial sector, there were fewer “misses”,while in countries where the Bank had not carried outsuch ESW, there were more likely to be more “misses”(Table 5.2).
These results suggest a link between ESW and devel-oping successful Bank financial sector operations fromdesign to outcome for the period examined (FY85–96).1
The design of the adjustment loans for Bolivia, Indone-sia, and Morocco were built on knowledge acquiredthrough in-depth Bank studies on the financial sector.Chile’s successful financial market operation benefitedfrom a preceding Bank report on industrial finance.ESW, however, is not a guarantee of successful lending.Substantial ESW was prepared for Bangladesh andKenya and delivered before project preparation, but theprogram outcomes were unsatisfactory, because reformswere not properly sequenced.
Lending OperationsTo assess the Bank’s performance in lending, we adopttwo different perspectives: (1) lending operationsemphasizing financial sector liberalization; and (2) oper-ations emphasizing institutional development.
It is necessary to point out that our evaluationmethod is constrained. A more complete evaluationwould rely on enterprise-level data in the borrowingcountries, which would enable us to see how reformchanges the financing patterns and the allocation ofresources at the enterprise level, and thus to assesswhether the efficiency gains from reform outweighlosses from post-reform crises.2 The implication is that
TABLE 5.1: PROJECTS FOR WHICH ADVANCEECONOMIC AND SECTOR WORK WAS DONENUMBER OFESW REPORTSPRECEDINGA PROJECT 0 1 2
Project status TotalActive 19 14 5 38Completed 29 19 2 50Total number
of projects 48 33 7 88Percentage 55 37 8 100
Note: This table tallies the number of ESW reports done inthe 5-year period preceding each project.Source: Operations Information System, World Bank.
TABLE 5.2: ESW AND FINANCIAL SECTOR PROGRAM DESIGN AND OUTCOME (FY85–96)PROGRAM DESIGN PROGRAM OUTCOME
≥ 2 MISSES 0–1 MISS UNSATISFACTORY SATISFACTORY
No ESW Côte d’Ivoire Poland Malawi Côte d’IvoireMalawi Senegal Malaysia Tanzania
Malaysia Tanzania PolandVenezuela Senegal
Venezuela
At least one ESW China Bangladesh Bangladesh BoliviaEgypt Bolivia China Chile
Chile Egypt GhanaGhana Kenya IndiaIndia Pakistan Indonesia
Indonesia Tunisia KoreaKenya MexicoKorea Morocco
Morocco The PhilippinesMexico TurkeyPakistan
The PhilippinesTunisiaTurkey
Source: Tables 2.3 and 2.5 and Annex Tables 1.1 and 1.2.
46
OED, QAG, and DEC should insist that the Bank col-lects firm-level data before project preparation.
Operations Focusing on Market ReformsFor the 17 countries in our 23-country sample for whichBank performance data on adjustment-related opera-tions are available, there is a strong positive relationshipbetween Bank performance and project performance. Incases in which Bank performance was strong,3 88 per-cent of the project outcomes were satisfactory. Eighty-one percent of the project outcomes were probably sus-tainable, and, 69 percent led to substantial institutionaldevelopment.
Lending for Institutional DevelopmentIn general, Bank adjustment operations and financialintermediary loans tend to succeed when the borrower iscommitted to reform, and Bank involvement and prob-lem-solving are intense and continuous. Bank perfor-mance is better under FSLs than under FILs (Table 5.3),partly for reasons outside the Bank’s control. Programimplementation is difficult under FILs when a country’smacroeconomic conditions deteriorate, because thoseconditions deeply affect portfolio quality and bank prof-itability. In five of nine countries FILs failed partlybecause of exchange rate fluctuations or an upsurge ininflation, and partly because of government interferenceand inadequate appraisal of borrowers’ capabilities.OED’s Annual Review provides supporting evidencethat Bank performance in the financial sector for FILsclosed between 1988 and 1995 was poor (the Bank wastoo optimistic about portfolio management). When theBank stays involved, combining close monitoring withassistance, institutional development is boosted. Butwhen Bank support lessens, institutional developmenttends to slump.
Weak Bank performance in institutional develop-ment under FILs was inevitable since FILs concentrated
largely on DFIs which have often been required to makehigh-risk loans at low interest rates to projects with higheconomic rates of return but not necessarily with highfinancial rates of return. This often translated into heavylosses. DFIs’ difficulties were exacerbated by the foreignexchange risks involved with Bank lending. Thus, it hasbeen a Herculean task to make DFIs viable, competitive,and sustainable—unless they were allowed to operate ina liberalized financial and economic environment whereloans were priced on the basis of risk.4
Adjustment Strategies and SequencingIn analyzing the effectiveness of the Bank’s adjustmentstrategy, the question is often asked whether the out-come of financial sector operations is vitiated by thewrong sequencing of financial reforms. This issue arosein Bank experiences in Chile, Indonesia, and Turkey, allof which are in our sample. In the early 1980s the finan-cial sector did not perform well if deregulation was ini-tiated in the midst of macroeconomic instability in coun-tries with little or no prudential control and supervision.Toward the end of the 1980s, when the same countrieshad developed sound macroeconomies and well-designed regulatory mechanisms, financial reforms suc-ceeded. In Indonesia, which had fairly stable macroeco-nomic conditions but weak regulations, financial reformfared well, but only up to a point. Reform would havebeen more successful if strong regulations had been inplace. The same was true of Chile in the late 1980s.Turkey’s financial liberalization was launched in a goodmacroeconomic setting in the mid-1980s, but began tounravel as the macroeconomy deteriorated.
Sequencing is extremely important and should begiven prominence in Bank adjustment operations. Giventhe enormous heterogeneity of country circumstancesand institutions, it may be difficult, if not impossible, tooutline a cast-iron formula for sequencing. Still, countrystudies and other Bank resources suggest some broad,empirically tested guidelines for a practical approach.
Sequencing of financial reform is multidimensional: • First, it is dictated by the state of the economy—
whether it is in a cyclical upswing or down-swing—which in turn determines the borrower’snet worth. This net worth rises in good times anddeclines in bad times, so that lending institutionsreduce the premium for debt finance in good timesand raise it in bad times. Financial reform shouldthus proceed faster when the borrowers’ net worthis rising and slower (World Bank experience
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
TABLE 5.3: BANK PERFORMANCE IN SUPPORTINGINSTITUTIONAL DEVELOPMENT
QUALITY AT PROJECTENTRY SUPERVISION
(percent satisfactory)FSLsa 74 90FILs 53 49
Note: a The FSLs include SADs, SALs, and TALs focusing onfinancial reform. Quality at entry includes identification/preparation and appraisal processes.Source: PRs, SARs, ICRs, and PCRs and OED PARs.
47
1996a). This has been well supported by experi-ence in the sample countries. For instance, Caprioand Klingebiel (1996) point out that, “Malaysianauthorities followed this strategy, re-controllinginterest rates when the economy was experiencingfinancial distress in the mid-1980s, and resumingdecontrol as net worth improved. Korean andIndonesian authorities also profited from earlypositive shocks.”
• Second, the pace of reform must be adjustedaccording to initial conditions. “If banks have lim-ited skills, do not have reliable financial informa-tion on which to base credit decisions, are notmotivated to make prudent lending and risk-tak-ing decisions, reforms can go awry” (World Bank1996a). Likewise, if directed credit is predominantbefore reforms, its removal or phasing-outdepends on the quality of supervision, which canensure prudent risk-taking. If supervision is poor,the abrupt ending of directed credit can lead tofinancing of real estate, generating an asset pricebubble (World Bank 1996a).
• Third, macroeconomic stability is important—sometimes vital—to the success of a structuraladjustment program. The most crucial element isprobably an appropriate exchange rate. Senegaland Indonesia pose contrasting examples. InSenegal financial sector reform stalled because ofan overvalued exchange rate. But reform gatheredmomentum after a depreciation. Indonesia, onthe other hand, depreciated its exchange rate atthe beginning of financial reform, reduced infla-tion, and curbed public expenditures (Johnson1997; Khatkhate 1993; Villanueva and Mirakhor1990).
• Fourth, financial sector reform has close linkswith real sector reform. The real sector must func-tion properly. Both public and private enterprisesare often inefficient and unprofitable. As long asthese inefficiencies continue, the financial sectorwill be less effective, as we have seen in Africa andLatin America, where restructured banks becameimmediately unviable because their borrowerswere unprofitable (mostly public) enterprises. Thesame happened in Eastern Europe (see Pohl andothers 1997; Johnson 1997; Khatkhate 1993; Vil-lanueva and Mirakhor 1990). However, financialsector reforms should not necessarily be post-poned until the real sector is restructured.
It should be noted that an optimal sequence ofreforms, though theoretically appealing, is difficult todesign: one can proceed only by trial and error.
• Policymakers must consider macroeconomic sta-bility and the building of financial infrastructurebasic to reform. The mainpillar of macroeconomicstability is the country’sfiscal situation. Whilefinancial repression isdetrimental to financialsystem development, itssudden easing wouldenlarge the fiscal deficitunless the governmentcan develop other sourcesof revenue. Thus the gov-ernment should be cau-tious in eliminating at a stroke its reliance on thefinancial institutions for resources.
• Financial infrastructure should be built at the startof reform. It should include developing skills,information, and appropriate incentive systems;improving accounting and auditing systems;establishing a regulatory system with properly-trained regulators; and devising modern financialsector legislation. This is a very time-consumingtask, but once started governments should notretreat, even when conditions may warrant slow-ing other reforms. For example, Malaysianauthorities re-controlled interest rates for sometime (G. Caprio and Klingebiel 1996).
• Introducing bank recapitalization and vigorouscompetition should come after a sustained efforthas been made to improve the incentive system forbanks. Incentives refer to the effort that banks andtheir regulators devote to ensuring that their insti-tutions are sound; they include high capitalrequirements, liability on bank owners, and ade-quate debt-collecting procedures (World Bank1995a). Once the right incentives are in place,recapitalization will be rewarded. There will be lit-tle risk of banks falling back into a situationwhere their net worth is negative because they willemploy newly-infused capital more efficiently.And the banks will be able to face competitionfrom new entrants more confidently.
The sequencing strategy outlined here should not beinterpreted rigidly except in regard to what is central to
B a n k P e r f o r m a n c e
Policymakersmust consider
macroeconomicstability and the
building of financialinfrastructure basic
to reform.
48
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
reforms. There should be enough room to shift andshape some elements depending on the overall positionof the economy.
Partnerships with the IMF and IFC
Coordination with the IMFCountry studies provided evidence that collaborationamong the Bank, the IMF, and bilateral donors inmounting reform operations was usually close andjointly beneficial. For example, in Chile the two institu-tions supported mutually reinforcing policy measures,and cooperated closely in securing private sector financ-ing for Chile’s reform program and facilitated commer-cial bank and Paris Club debt restructuring. In Turkeythe IMF, although not directly involved in the country atthat time, collaborated with the Bank in preparing ashadow IMF program in conjunction with FSAL 2. TheBank also played an important role in mobilizing Japan-ese cofinancing. In Pakistan, the Bank and the IMF
jointly designed the medium-term economic programand worked closely on the preparation of the FSAL andthe Policy Framework Paper (PFP).
Bank-IMF coordination in carrying out financial sec-tor operations improved after formal arrangementsbetween them were strengthened and clarified in 1989.Still, there is potential for improving collaboration. IMFstaff noted that in many countries most of the easy reformshave already been undertaken under ESAFs and SALs. Butfinancial sectors often remain fragile, sometimes danger-ously so. The IMF increasingly sees the strengthening offinancial systems as pivotal to the continued success ofadjustment and reform efforts. In dealing with fragilebanking systems, the IMF recognized the need to draw onBank resources. IMF staff interviewed acknowledged thatthe Bank’s extensive involvement with country commercialbanking systems has given its staff a distinct comparativeadvantage in dealing with such institutional problems.IMF staff would welcome far more Bank support in thisarea, particularly in countries where the Bank is not
Initial Conditions.Like many Africancountries, Senegal has
suffered chronically froma severely limited physi-cal and human resourcebase, a high populationgrowth rate that hinderseconomic advancement,frequent droughts,increasing desertification,and major swings interms of trade. In theearly post-Independenceyears these economicproblems were exacer-bated by, among otherthings, excessive indus-trial protection, govern-ment overspending, inef-ficient public enterprises,excessive market regula-tion, rigidities in the
labor market, and aninadequately-regulatedand inefficient bankingsystem. By the late1970s, Senegal’s eco-nomic stagnation anddeclining per capitaincome impelled the gov-ernment to seek IMF andBank assistance.
Bank and IMF initiatives.The IMF responded withan Extended Fund Facilityin 1980. In addition,beginning in 1980, theBank initiated a series ofSALs aimed at alleviatingsome of the problems.SAL III included a finan-cial sector reform compo-nent. Later, in conjunc-tion with SAL IV, the
Bank mounted a separateFSAL. The FSAL was pre-ceded by regional finan-cial reforms announcedby the Union MonetaireOuest Africaine (UMOA)in August 1989. Thesereforms included elimi-nating the sectoral alloca-tion of credit, partiallyliberalizing interest rates,eliminating preferentialrates, implementing a newsystem of crop-financing,introducing a new systemfor determining borrowercreditworthiness, andrevising bank-by-bankcredit ceilings.
Financial Sector Adjust-ment. The FSAL, intendedto complement and
extend the UMOAreforms, was a compre-hensive project coveringall four areas typicallypart of the Bank’s finan-cial sector reform: remov-ing distortions, strength-ening the financialsystem, strengtheningindividual financial insti-tutions, and enhancingcompetition. Specifically,the FSAL embraced mon-etary and credit reform,bank regulation andsupervision reform, bankrestructuring and liquida-tion, government divesti-ture of bank shares, pro-visions for the recovery ofbad debts, provisions forgovernment payment ofannual liabilities arising
BOX 5.2: SEQUENCING REFORMS IN SENEGAL AND THE UNION MONETAIRE OUEST AFRICAINE
49
B a n k P e r f o r m a n c e
directly engaged in financial sector operations. It was sug-gested that annual Bank-IMF consultations, possibly alongthe lines of present annual public expenditure review con-sultations, might be useful for identifying priority coun-tries and areas for the Bank’s institutional analysis.
There are many examples of close cooperationbetween the two institutions in designing successfulfinancial reform programs, providing technical assis-tance, and providing complementary loans. Inevitably,at a working level, staff of the two institutions do notalways agree on specific diagnoses and recommenda-tions. There is room for improving the sharing of infor-mation and analysis, reducing the scope for duplicationof efforts and avoiding the rapid escalation of disagree-ments to the senior management of both institutions. Apaper on “Bank-Fund Collaboration in StrengtheningFinancial Sectors,” was discussed by both Boards inmid-1997. This paper sets out a broad division ofresponsibilities, as well as procedures for collaboration.The Fund will be mainly concerned with bilateral and
multilateral surveillance of banking and the broaderfinancial sector. It will emphasize, to a greater extentthan in the past, identifying vulnerabilities in financialsystems that could have major macroeconomic implica-tions. It will also suggest corrective policies. The Bankwill continue to help develop member’s countries finan-cial sectors, in the context of its ESW and lending pro-grams. There are overlaps in the responsibilities betweenthe two institutions, notably banking supervision andregulation and legislation. Detailed consultation proce-dures have been mapped-out. More needs to be done,however, to ensure full and timely exchange of informa-tion and early coordination between the Bank and theIMF for immediate response to financial crises.
Bank-IFC CoordinationFrom the outset the World Bank Group recognized thatthe IFC was to be the focal point for development workin financial markets. Until fairly recently the Bank hadlittle involvement in securities market development. In
from bank restructuring,and a program to estab-lish “grass-roots” banks.The Bank also approvedthree loans and an IDAcredit to create and sup-port a state developmentbank, SOFISEDIT, to pro-mote and support indus-trial development.
Outcome. Although con-siderable progress wasmade, the overall result ofthe SALs was disappoint-ing, partly because ofexternal factors, such asovervaluation of the CFAfranc during much of theperiod, and partly becauseof internal factors, includ-ing entrenched politicalopposition to change;
incomplete governmentownership of, and com-mitment to, the reformprocess; and weak supplyresponses to the improvedincentives. In the financialsector, the Bank underesti-mated the severity of theproblems. By 1989, afterthree SAL operations, thebanking system was still insevere crisis. The FSALfinally succeeded inreforming the financialsector and restructuringand recapitalizing thebanking system, oncemore fundamentalreforms to the role of theregional central bank(BCEAO), initiated by theFund and agreed to byUMOA, were in place.
Assessment. Senegal’sexperience clearly demon-strates that propersequencing of elements inan economic adjustmentand structural reform pro-gram is vital. With regardto the sequencing offinancial versus realreforms, the PAR on theFSAL argues that finan-cial reform paved the wayfor a successful devalua-tion, the typically adverseeffects of which were sur-mounted with relativeease by the strengthenedbanking system. The ben-eficial economic impact ofthe 1994 exchange ratedepreciation, reinforcedby structural adjustment,further strengthened the
financial system. As far asthe sequencing of finan-cial sector reforms is con-cerned, the basic UMOAreforms clearly had to bein place before the Sene-galese financial systemcould be reformed.
Lessons of Experience. Alesson to be derived fromthe Bank’s overall experi-ence in Senegal is that theBank needs to place consid-erably more emphasis on“getting it right” in the firstplace, be it with regard tothe basics of the country’seconomic situation, its sec-toral reform requirements,or the need for individualinstitutions or particularforms of lending.
BOX 5.2: SEQUENCING (CONTINUED)
50
the past decade, however, the number of IFC technicalassistance projects undertaken in coordination with theBank has escalated sharply. There were only six jointprojects in 1971–79, and only four in 1980–85. Between1986 and 1990 the number rose to 79, and in 1991–95it climbed to 90. In 1971–95 overall, 24 percent of IFCtechnical assistance projects were undertaken in con-junction with the Bank. These joint projects tended to bemore successful in persuading governments to undertakeneeded reforms. In the financial sector, (mostly thebanking sector) Bank-IFC collaboration has been evencloser: the 82 joint projects undertaken in the financialand banking sectors represented 47 percent of IFC’stechnical assistance operations in those sectors.
IFC officials interviewed for this study confirmedthat there is a complete exchange of documents betweenthe two institutions, ensuring that each is kept formallyaware of the other’s operations. As with the IMF, a senseof joint mission has often been the cement for close coop-eration. Reporting on Bank-IFC coordination in devel-oping Hungary’s capital markets, an IFC study notedthat, “a sense of higher mission and the excitement ofworking on the first official stirrings of capitalism in theCommunist bloc overcame any natural feelings of terri-toriality.” The report also noted mutual respect for eachinstitution’s areas of skill. Still, there is a risk of possibleconflict of interest between the IFC, as a shareholder infinancial institutions, and the Bank and Fund, given theirroles in restructuring.
The potential for differences between the two insti-tutions is inherent in the IFC’s instinctive search for pri-vate sector solutions to problems with financial sectordevelopment (for example, setting-up a market forbank bonds and notes where none existed before) asopposed to the Bank’s working with governments andrequiring sovereign guarantees. The Bank’s orientationhas sometimes led to proposals for government involve-ment in what are, in most industrial countries, mainlyprivate arrangements. Although the machinery cur-rently used for exchanging information and views isgenerally satisfactory for resolving such issues, someproblems remain.
First, the two institutions differ greatly in size. Thescope and complexity of Bank operations means thatcoordinating with the Bank imposes a very heavy loadon IFC staff. And because the IFC is smaller, some Bankstaff are not fully aware of the technical assistance it canprovide and do not feel compelled to coordinate with it.Informal contacts and good will resolve most of these
problems, but IFC personnel are sometimes frustrated.And while the Bank and the IFC exchange preliminaryproject documents, such as summaries of investmentenquiries (SIQs) and initial project reviews (IPRs) theIFC staff would welcome even earlier notification aboutimpending projects, so that they have a better opportu-nity to influence their design.
According to IFC staff, a continuing difference ofviews had surfaced in a number of countries about theBank’s channeling of IDA funds at low-interest ratesthrough apex institutions to private financial intermedi-aries. In some countries (such as Zambia) the Bank wascontemplating making IDA funds available to financialintermediaries at interest rates below LIBOR—substan-tially less than rates that the IFC would charge. Thisappeared to conflict with the Bank’s own stated princi-ple of discouraging subsidized lending operations. IFCofficials noted that competition from the EuropeanInvestment Bank (EIB) and African Development Bank(AfDB), which also provided subsidized funding, pres-sured the Bank to do likewise. Although compromiseson IDA lending were reached in particular cases, theneed for a generic mechanism to resolve issues of pricingwithin the World Bank Group was noted both by theIFC and QAG.
Conclusions on IMF and IFC Coordination Some broad conclusions may be drawn about the Bank’sstrategy with regard to IMF and IFC partnerships. Thefirst priority is to recognize thatfinancial reform is a long,drawn-out process. Instead ofthinking in terms of two-orthree-year lending operations,the Bank must develop long-term strategies, which must beincluded in a CAS. But thisrequirement should not be anexcuse for lethargy in pushingthrough financial sector reform.That strategy should incorpo-rate all instruments, especiallynon-lending services, to helpeffect the desired changes. Tenyears or more may be required to complete financialreform in a country, and far more emphasis must beplaced on financial infrastructure. Still, there may beoccasions to hasten financial sector reforms to ensuremacroeconomic stabilization.
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
Ten years ormore may be
required tocomplete financial
reform in acountry, and farmore emphasis
must be placed onfinancial
infrastructure.
51
B a n k P e r f o r m a n c e
This long-term strategy should be shared with allinterested parties, both inside and outside the BankGroup. The Bank would support the development of theframework within which private financial intermediariescan function effectively, while the IFC would help estab-lish or privatize financial intermediaries, which can thenserve as models. The strategy should also include inter-action and agreement with the IMF on monetary poli-
cies, banking supervision, taxation of the financial sys-tem, and the like. Collaboration at that level was for-mally discussed by the Board members of the Bank inAugust 1997, in which they further clarified the divisionof responsibilities between the Bank and the Fund(5.19). It is important for the regional multilateral devel-opment banks, bilateral donors, and other parties tosign on to a common financial development strategy.
53
Conclusions andRecommendations
66
The performance of Bank assistance in financial sector reform has not been strong. It
was not until the late 1980s that the Bank started to emphasize the need for financial
sector adjustment lending; it was in 1992 that the Bank’s financial sector operations
began to take a systematic perspective of macroeconomic and financial sector issues with the
Operational Directive (OD) 8.30. It was only in 1997 that the Bank articulated the importance
of a long-term perspective and the importance of economic sector work in policy advice. The
East Asia crisis has provided fresh evidence of the Bank’stendency to underrate the seriousness of financial sectordysfunctions.
Main FindingsAccording to OD 8.30, the Bank’s policy instruments forsector liberalization and institutional developmentinclude: (1) deregulation of interest rates and elimina-tion of directed credit; (2) privatization of banks toenhance competition; (3) improvement of regulatoryregimes; and (4) strengthening of financial institutions.The major findings of this study are:
• In only 4 of 23 sample countries did Bank pro-grams provide policy instruments to improve con-ditions in all four areas of reform. In some areasthe Bank policy package was deemed inappropri-ate given the country’s economic situation. A largeproportion (78 percent) of policy packagesmatched initial conditions in the categories ofinstitution-building, and elimination of distor-tions. Reform aimed at encouraging competitionscored the fewest matches.
• Project implementation was weakest in enhancingcompetition. Strengthening individual institutionsalso remained weak, with the implementation rate
less than 50 percent. Policies relating to financialinfrastructure improvements had the best imple-mentation record: 71 percent.
• In 10 of 23 countries both program outcome(the effect on the financial sector) and impact(the effect on the real sector) were satisfactory.In 7 countries outcome and impact were unsat-isfactory. This suggests close links betweenfinancial sector reform and the growth of thereal economy.
• The ratings used in the study are on average simi-lar to PAR/PCR ratings, and are lower than ARPPratings. The discrepancies between the evaluationshere and the ARPP evaluation arise in countrieswhere the performance of the banking system hasstagnated or deteriorated. This aspect of financialsector soundness, well-referenced in OD 8.30, hasnot always received the attention needed toachieve satisfactory results in Bank operations.
• Initial conditions greatly influence outcomes.These were best in countries with low inflationand low levels of bank credit to the private sector.Outcomes were also greatly affected by non-bankfinancial regulation and prudential control, aswell as by the design of adjustment operations.
54
• Both FSLs and FILs support institutional develop-ment, but their specific objectives are not thesame. FSL objectives are related to financial infra-structure, while FILs seek to strengthen financialinstitutions more directly. Initial conditions underFSLs were very weak, but improved greatly afterthe loans were made. By contrast, initial condi-tions were better under FILs, but hardly changedas a result of FIL operations.
• Outcomes in low and middle-income countriesshowed improvement over initial conditions underFSLs and FILs. But the performance of bothgroups of countries has been consistently betterunder FSLs than under FILs.
• Institutional development is much more sustain-able under FSLs than under FILs. It has been muchmore consistently so in middle-income countriesthan in low-income countries. Critical to sustain-ability are good payments systems and higheraccounting and auditing standards.
Case studies of financial crises in Mexico, the Philip-pines, and Venezuela suggest that:
• The set of standard indicators used to assess finan-cial sector outcomes under Bank-supported pro-grams is not adequate for assessing the fragility ofa financial system.
• Financial sector reform is not sustainable unless itis comprehensive and penetrates to the institu-tional level. Operations must include ESW andtechnical assistance components. Effective pru-dential regulation and supervision are vital to thesoundness of a financial system.
• Deregulating interest rates can be destabilizing ifmacroeconomic conditions are unstable, financialmarkets are oligopolistic, borrowers are in seriousfinancial distress, connected lending is common,the financial condition of banks is fragile, bankingsupervision is ineffective, and regulatory bankinglegislation is weak.
• In preparing financial sector studies, the Bankshould pay more attention to the financial perfor-mance of the central bank and, as much as possi-ble, should review off-balance sheet items, espe-cially guarantees and derivatives operations.
Reasons for Mixed OutcomesIt is clear from this study that the outcomes of the Bank’sadjustment operations, though not uniformly unsatisfac-tory, were below what was expected. We focus here on
possible reasons for those poor outcomes, drawing onthe Bank’s OD 8.30 and academic studies on the role ofthe real economy, the fiscal factor, and, more generally,the responsibility of governments:
• Financial sector operations should be viewed aspart of the structural transformation of an econ-omy. Without restructuring enterprises, there willbe few good opportunities for banks to lend, andeven if banks are restructured, that restructuringwill only have to be repeated, at a much highercost. Fiscal policy is important because there is aclose link between fiscal deficits and the bankingsystem.
• Financial intermediaries focus on their own andtheir borrowers’ net worth, which makes the tim-ing of financial liberalization crucial. This linkimplies that governments should move aggres-sively on financial reform in good times and moreslowly when borrowers’ net worth has beenreduced by negative shocks, such as recessions orlosses from terms of trade (Caprio and others1994).
• In borrowing countries political ownership ofreform does not always ensure smooth sailing(Williamson 1994). The bureaucracy must changeits ways and attitudes,and implement reformsquickly so that the resultsbecome visible. Anyreform requires upsettingvested interests, but theirresistance can be over-come if the ultimate bene-ficiaries—investors, pro-ducers, and wage earners—gain.1
• The Bank’s operationsmay be seen as less successful than expected, partlybecause the Bank expects results in two to threeyears. But meaningful results from interventions inthe financial sector may take eight to ten years.Thus, outcomes should be evaluated over a longerperspective.
RecommendationsThe OD 8.30 guidelines and their revised (draft) versionin OP 8.30 are comprehensive, consistent with pastBank experience, and relevant to current financial sectorissues. They reflect recent academic findings on financial
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
The bureaucracymust change its
ways and attitudes,and implement
reforms quickly sothat the resultsbecome visible.
55
liberalization and institution-building. By definition,however, both documents have limitations. They aregeneral guidelines, not intended to recommend a partic-ular strategy. For that reason they do not automaticallygenerate good practice. To improve Bank support offinancial reforms in future, the first recommendation isto go back to the guidelines as a general reference forcomprehensive reform. Then the recommendations is tofocus on the five main aspects of financial sector strat-egy where weaknesses were identified.
Recommendation 1: Returning to the OperationalPolicies DirectiveThe Bank should aim at a more integrated approach tofinancial sector development. When the Bank incorporatesfinancial sector development strategies into CAS docu-ments it should also ensure that sooner or later (this is amatter of sequencing) all major relevant aspects of thefinancial sector, as described in OD 8.30, especially compe-tition, are dealt with in a systematic way. In addition, theBank should occasionally initiate and support regionalreforms (as in West African Central Bank reforms).
Recommendation 2: Going Beyond the OperationalPolicy and Toward Good Practice OD 8.30 is now being revised. The directive clarifiedand formalized the Bank’s position on the financial sec-tor, which it slowly built over the years from conceptualwork and lending experience. OD 8.30 has anchoredmany good practices and has been recently updated tofurther clarify current issues, such as directed credit andrestructuring of financial intermediaries. Beyond the rec-ommendation to use the directive as a fundamental ref-erence for establishing complete sector strategies andsound lending, this study points out additional areasthat still need attention.
Program Design
Recommendation 2.1: Proper Sequencing of ReformsThe general strategy should be to first initiate reformonly if the macroeconomy is stable, ensuring specificallythat the fiscal situation is good. Second, a financialinfrastructure, which includes the formation of skills,development of information, setting of an appropriateincentive system for banks, improvement of accountingand auditing standards, and establishment of regulatorymarkets and legislation, should be in place. Both partsshould be taken as central to financial reform and
should not be stalled, even when other reforms, such asinterest rate liberalization or phasing-out of directedcredit, must be slowed or held in abeyance because ofunexpected events. If the proper incentives are offered, itis possible to advance reforms such as, easing entry ofnew banks and liberalizing the capital account. The paceof reforms will vary depending on the institutionalstrengths and ownership of the major players. The EastAsia crisis will be rich in additional lessons.
Recommendation 2.2: Emphasizing Competition andFinancial InfrastructureFinancial liberalization is incomplete without well-con-ceived measures for introducing competition. Reformsshould be designed to have long-term effects alwayskeeping in mind the ultimate objective of establishing acompetitive environment. Conventional instruments ofenhancing competition within banking systems—such asbank privatization, legal changes in company laws,banking laws, foreign ownership laws, bankruptcy laws,and regulatory changes to remove different treatment ofdifferent banks—should constitute the basis of a policyto enhance competition. But these conditions are neces-sary, not sufficient. Authorities must create an environ-ment, through regulation, to prevent collusive behavioramong banks and conglomerate relationships betweenbanks and nonfinancial groups (Denizer 1997).
A major lesson from the financial crises in Mexico,Venezuela, and the Philippines is that financial sectorreform is not sustainable unless it is comprehensive andpenetrates to the institutional level. Strengthening finan-cial infrastructure requires, first, analysis of the centralbank’s governance, independence, skill level, capacityand incentives to supervise. Second, attention must begiven to the payments system, accounting and auditingstandards, collateral laws, and regulations in client coun-tries. Third, professionalism in banking and accountingis often neglected. Training programs in these areasshould firmly anchor the reform process. All FSLs shouldaddress these important developmental issues.
Bank Processes
Recommendation 2.3: Timely ESW and ContinuousMonitoringOD 8.30 states clearly that lending should be based onESW. But many financial sector operations were carriedout without ESW. Analysis suggests that previous ESWstrengthens the design of financial sector operations and
C o n c l u s i o n s a n d R e c o m m e n d a t i o n s
56
contributes to their success. In-depth, policy-orientedeconomic and institutional analysis should invariablyprecede project initiation to ensure that the criticalissues are clearly defined from the outset and that amacroeconomic environment conducive to reform ismaintained.
The Bank should monitor and evaluate more con-tinuously and more systematically achievements ofcountries’ financial sector adjustment programs, withperformance indicators focusing on structural changesand relevant to the objectives of these programs.Increased coordination with the IMF, who regularlyreviews and appraises macro and financial sector poli-cies (Article IV Consultation) and IFC, who monitorscapital markets, would reduce duplication of effort andsharpen the diagnosis. In the context of financial sectoradjustment programs supported by the Bank, perfor-mance indicators should be carefully selected and con-tinuously monitored by the Bank and the governmentauthorities. The indicators should be relevant to the spe-cific objectives of the adjustment program and shouldfocus on structural changes in the financial system.Changes in these indicators should be used to evaluatethe relevance and efficacy of the adjustment program.
The Bank’s relevant policy documentation should beupdated to ensure that specific financial sector indica-tors are included in reform operations. Financial reformoperations must use appropriate and generally accept-able banking indicators, which are essential for ascer-taining the weaknesses in a system and for monitoringprogress or deterioration.
Recommendation 2.4: Judicious Use of Classical andNew Lending InstrumentsThe Bank already has a considerable number of instru-ments to support reforms in the financial sector (SALs,SADs, FILs, SIMs, TALs) and has used them all. Eachlending instrument, however has its own relative advan-tage: SALs for macroeconomic stability and structuraladjustments, SADs (also called FSALs) for financial sec-tor liberalization and development of financial infra-structure, and TALs for strengthening financial systems,and individual financial institutions.
The optimal sequence of instruments should reflectthe optimal sequence of reforms. In general, SALsshould pave the way for SADs, and FILs should be usedin a reform framework supported by SADs. A TAL canbe used at all times, but its content will differ at variousstages of the reform program, emphasizing system-wide
weaknesses at the beginning and individual financialinstitutions later. Since institution-building is a slowprocess, the new instrument, adaptable program loans(APLs) which disburse according to the pace of institu-tional progress, could become essential in supportingfinancial reforms. But the prospect of a financial crisis ina country could generate a loss of confidence among for-eign creditors and require up-front disbursements oflarge amounts from the Bank and other multilateralorganizations to compensate the short-term capital out-flows. Provided the country has good long-term repay-ment capabilities and is willing to embark on a majorlong-term reform program, the Bank could decide tomake such a loan (ERL), as evidenced in Korea recently.
Financial intermediation projects, and the health offinancial institutions they help maintain, are vulnerableto the effects of recessions and abrupt changes in gov-ernment policy. In general, Bank lending through FILsfor institution-strengthening has not been successful.The Bank should focus first on establishing an environ-ment in which FIs can become profitable and competi-tive. Lending to FIs for institution-strengthening shouldbe undertaken within a phased financial sector reformprogram, in which the government has credibly estab-lished that it will maintain good policies. The proposedevolution of FIs within a financial sector programshould be made explicit. If the Bank is to increase itssupport of private sector development, this aspect ofBank performance must be improved on.
While the Bank intends to develop technical assis-tance programs and loans, there are very few completed“stand-alone” projects in the financial sector on whichto base conclusions for future loans. Case studies oftechnical assistance incorporated in Bank loans showthat the Bank should commit itself to providing techni-cal assistance only under the following conditions: (1)the need is clearly established (as opposed to being onlyperceived); (2) the Bank has the resources to design andsupervise assistance effectively; and (3) the governmentis committed to specific actions that will ensure theeffectiveness of assistance.
Recommendation 2.5: More Effective Partnershipswith the IMF, IFC, and EDIThe first priority is to recognize that financial reform isa long, drawn-out process. This long-term view shouldbe shared with all interested parties, inside and outsidethe Bank Group. The strategy should include interactionand agreement with the IMF on monetary policies,
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f W o r l d B a n k A s s i s t a n c e
57
C o n c l u s i o n s a n d R e c o m m e n d a t i o n s
banking supervision, taxation of the financial system,and the like. In August 1997 the Board of the Bank, for-mally discussed further clarifying the division of respon-sibilities between the Bank and the Fund. It is importantthat regional multilateral development banks, bilateraldonors, and other parties sign on to a common develop-ment strategy.
The scope and complexity of Bank operationsmeans that coordinating with the Bank imposes a heavyload on IFC staff. While the Bank and the IFC nowexchange preliminary project documents, such as sum-maries of investment inquires (SIQs) and initial projectreviews (IPRs), Bank staff should give IFC staff even ear-lier notification about impending projects, so that theyhave more opportunity to influence project design.
The Economic Development Institute (EDI) of theWorld Bank trains decisionmakers to help promotefinancial sector development through its banking andfinance program. This has led to successful outcomes incountries of the Former Soviet Union during the last fiveyears. EDI should be encouraged and given appropriate
resources to further develop its banking and financetraining program in sound bank management, bankingpolicy, regulatory issues, and bank failures.
Recommendation 2.6: Better Partnerships withInstitutions in Borrowing CountriesThe Bank should support a consensus among partiesinvolved in project design and implementation in clientcountries. This is part of promoting borrower “owner-ship,” a prerequisite for project success. Country casestudies suggest that Bank support of financial sectoradjustment operations is more successful when the keyparties are convinced of the need for reform than whenit is trying to persuade reluctant counterparts with dif-ferent priorities. To ensure continuity of reform, it iscritical for the Bank to maintain a close liaison with bor-rower governments, even in adverse political circum-stances. As an outside, objective analyst of the financialsector, the Bank should bring lessons of experience fromother countries and disseminate these to focus earlyattention on reform needs.
59
Chapter 11. OD 8.30 (World Bank 1992: 12) defines FSALs as instru-
ments “for simultaneously supporting a broad range of policy andinstitutional reforms to improve the functioning of financial mar-kets and to strengthen the ability of financial institutions to mobi-lize and allocate financial resources.”
2. Relatively few projects were undertaken in the financialsector in the mid-1980s. The number of projects increased only atthe end of the decade and in the early 1990s, with a growing num-ber of SADs, also called FSALs, focusing on the financial sector,which, the Bank believed, should complement, through their tar-geted sector reform, the overall reform in SALs. Another reasonfor increasing recourse to FSALs was the Bank’s realization of thelimits of FILs as the main tool to promote development of thefinancial sector.
3. Among economies in transition, China and Poland wereselected because the Bank has had a longer experience of support-ing financial sector reforms in these countries.
4. FSLs include SALs, SADs, and TALs with financial sectorcomponents.
5. See “Financial Systems and Development” in World Bank(1989b); “Financial Sector Operations” in World Bank (1992);World Bank 1993b; World Bank 1995e; and World Bank 1995f.
6. Among countries that had financial crises, Mexico, thePhilippines, and Venezuela were chosen because OED had done themost PARs of Bank financial sector operations for these countries.
Chapter 21. M2 stands for broad money. It is measured as the sum of
money (IFS line 34) and quasi-money (IFS line 35).
2. Although Bolivia eliminated many interest rate distortionsin August 1985, our performance indicators show that interestrates were erratic and mostly negative before the approval of thePublic Financial Management Operations I (C1809-BO). Interestrate spreads were also high at that time, which is consistent witha lack of competition within the banking system (World BankReport No. 6765-BO).
3. Overkill may be too strong a word in the sense thatevery country needs continuous improvement in all these areas,as well as others not specifically targeted. But given the limitedfinancial and human resources in each country, policy packagesmust be selective and tailored to initial conditions. A redundantpolicy prescription may hamper proper implementation of theprogram.
4. In Venezuela, although reform policies included measuresto promote competition, the audit report observed that competi-tion policies were included as a mere formality and stipulated rel-atively weak conditionality.
5. Some explanation about Korea and India is necessarybecause although their implementation records were poor, theiroutcomes were satisfactory (see Table 2.5). In Korea imple-mentation of banks’ interest rate liberalization was poor, but itwas offset by the rapid growth of non-bank financial interme-diaries, with freedom to set interest rates, which competed with
banks. In India the reason for the discrepancy between poorprogram implementation and a good outcome seems to lie inthe extreme financial repression present on the eve of the pro-gram, so that even a poorly-implemented program could yieldpositive results.
6. Performance indicators for Indonesia showed on balancesatisfactory outcomes up to 1996 for all the financial sector com-ponents, except the banking system, which even then was unsatis-factory. In 1997, a year after our sample, almost all the other per-formance indicators deteriorated sharply, to the point that theoverall outcome for Indonesia would have been unsatisfactory.The 1997 crisis was generated by an illiquid banking system whichbecame rapidly insolvent, together with a pegged exchange ratepolicy. The weakness of the banking system is clearly revealed bymicro fragility indicators (CAMEL). Our analysis of the Bankresponse to financial crisis (see Chapter 4) gives more emphasis tomicro fragility indicators, and the role the lack of of financial sec-tor soundness can play in financial crises.
7. In 1996 the QAG commissioned a review of the Bank’sportfolio financial intermediary loans as part of its portfolioinvestment program. The FIL portfolio was chosen as it was foundto have a higher ratio of projects at risk than the portfolios ofother instruments of lending. As of June 1996 the Bank’s portfo-lio included 52 ongoing FILs. Of these, 15 FILs are formally ratedas having a problem (that is, their rating is “unsatisfactory” fordevelopment objectives or implementation progress or both). Inaddition, six projects are classified as potential problem projects.These 21 FILs are classified “at risk” and make up 40 percent ofall FILs in the portfolio. See World Bank (1997c).
8. The 19 countries are: Bangladesh, Bolivia, Chile, Côted’Ivoire, Ghana, Indonesia, Kenya, Korea, Malaysia, Mexico,Morocco, Pakistan, the Philippines, Poland, Senegal, Tanzania,Tunisia, Turkey, and Venezuela.
9. This section draws on Cull (1997). Here, we consider onlythe last Bank financial sector operation in each country.
10. The five countries taken out of the original sample are:China, India, Indonesia, Poland, and Tanzania.
11. The rationale for a negative relationship between infla-tion and DBPC and M2/GDP is as follows: Gertler and Rose(1994) find an empirically-strong positive relationship betweenhigher per capita income and depth of the financial system as mea-sured by DBPC/GDP. On the basis of this, it is presumed that ifDBPC/GDP ratio is initially low, that is, the financial system isunderdeveloped, then M2/GDP should rise if a FSL program isimplemented successfully. There will be a negative relationshipbetween the initial condition of DBPC/GDP and M2/GDP afterthree years of FSL. The negative relationship between low infla-tion as an initial condition and high M2/GDP after a successfulFSL is self-explanatory. For details, see R. Cull (1997).
12. In one country (Bolivia) the outcome of the reform pro-gram was satisfactory, while the performance of the real sectorremained unsatisfactory. The gains from financial sector reformmight have been transmitted to the real economy with a time-laglonger than the period under review because of institutional iner-tia or financial institutions’ failure to respond rapidly to marketsignals.
ENDNOTES
60
Chapter 31. This theme is reflected in the Bank’s Operational Manual
Statement (OMS 3.73), which applied to FILs approved beforeFebruary 1992, when OD 8.30 was issued. The SADs follow OD8.30, even those approved before its issue.
2. FSLs include SADs, as well as large financial sector com-ponents in SALs and TALs.
3. In carrying out this analysis, we referred to SARs and sub-sequent audit documents (PCRs, ICRs, PARs) and to the relevantESW. Financial sector issues are also occasionally raised in struc-tural adjustment and sector adjustment loan reports.
4. One FIL began a few months before 1982 and nine FSLsafter 1992. The nine FSLs, as ongoing cases, are excluded fromthis analysis.
5. SALs, however, may include both public enterprise andfinancial sector reforms.
Chapter 41. For the purposes of this chapter a financial crisis is
defined as a situation in which a significant proportion offinancial institutions have their liabilities exceeding the marketvalue of their assets, leading to runs, the collapse of somefinancial firms, and government intervention. A financial cri-sis, then, is a situation in which an increase in the share of non-performing loans, mounting losses (because of foreignexchange exposure, interest rate mismatch, contingent liabili-ties), and a decline in the value of assets cause systemic sol-vency problems in a financial system and lead to liquidations,mergers, or restructuring.
2. A caveat: the aggregated index is an unweighted sumof indicators, and several of the indicators are not indepen-dent of each other, so the FFI has limitations. Demirgüç-Kuntand Detragiache (1997) developed a more selective approach,using an econometric model to estimate the probability of abanking crisis in which real interest rates and inflation arehighly significant in all specifications. But the exchange ratedoes not have an independent effect once inflation and termsof trade are controlled for. The fiscal surplus is also not sig-nificant. Although we fully acknowledge the merits of thisapproach for predicting financial crises, we prefer to relyupon a more extensive set of macroeconomic and microeco-nomic indicators for assessing ex post the impact of Bankintervention.
Chapter 51. More general results on the positive impact of ESW on
the quality of Bank projects can be found in a recent study byKlaus Deininger, Lyn Squire, and Swati Basu (1997).
2. The work carried out in the Bank by Caprio, Hansonand Associates (1994) showed that in general financial reformhas positive efficiency effects on firms.
3. Strong Bank performance is defined here as satisfactoryperformance at project identification, appraisal, and implementa-tion. Criteria to assess Bank performance are (1) at identifica-tion/preparation, the degree of involvement of the governmentand beneficiaries, the project’s consistency with the Bank’s coun-try strategy, the extent to which the project is grounded in eco-nomic and sector work; (2) at appraisal, the quality of technical,financial, and institutional capacities analysis; the incorporationof lessons learned; the readiness of implementation; and the suit-ability of the lending instrument; and (3) for supervision, the qual-ity of reporting on the progress of project implementation, thecapacity to assess implementation problems, the quality of advicegiven to the implementing agency, the degree of enforcement ofloan covenants, and the flexibility in suggesting modifications.
4. Although the cases of ICICI in India and NDB in SriLanka suggest that it can be done.
Chapter 61. World Bank (1995b) adds three more dimensions to the
political economy: political acceptability, feasibility, and cred-ibility of reforms.
Notes1. This section draws on Cull (1997). In this section only the
latest Bank financial sector operation in each country is consideredfor the statistical and econometric analysis. For reasons of dataavailability, five countries have been taken out of the original sam-ple (China, India, Indonesia, Poland, and Tanzania).
2. Robust growth results, the focus of Levine and Renelt (1992),are also found in the institutional work of Knack and Keefer (1995).
3. Other policy variables achieved significance with the pre-dicted sign (recapitalization, strengthening of bank supervisionpractices) when entered in the regressions one-by-one (Cull 1997).
4. Additional conclusions from regressions and data analysisnot included here are found in Cull (1997).
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
61
Anderson, Robert E, Simeon Djankov, Gerhard Pohl,and Stijn Claessens 1997. “Privatization and Restruc-turing in Central and Eastern Europe.” ViewpointNote 123, World Bank Group Finance, Private Sectorand Infrastructure Network. Washington, D.C.
Aryeetey, Ernest 1994. “Supply and Demand forFinance of Small Enterprises in Ghana.” World BankDiscussion Paper 251. Washington, D.C.
Baer, Herbert L., and Gerard Caprio Jr. 1995. “BankRecapitalization: If and When.” DEC Policy ReviewNote 2. World Bank, Development Economic Depart-ment, Washington, D.C.
Bekaert, Geert 1995. “Market Integration and Invest-ment Barriers in Emerging Stock Markets.” TheWorld Bank Economic Review.
Bennett, Lynn, Mike Goldberg, and J. D. von Pischke1994. “Basing Access on Performance to Create Sus-tainable Financial Services for the Poor in Nepal.”Paper prepared for a Conference on Financial Ser-vices and the Poor, The Brookings Institution, Wash-ington, D.C. (September).
Blejer, Mario I. and Silvia B. Sagari 1988. “Sequencingthe Liberalization of Financial Markets.” Finance &Development (March).
Boratav, Korkut 1993 “External Financial Liberaliza-tion in Developing Countries: Some Adverse Conse-quences.” METU Studies in Development (1–2).
Borensztein, Eduardo 1994. “Structural Policies inDeveloping Countries.” IMF Paper on Policy Analy-sis and Assessment PPA/94/19. International Mone-tary Fund, Washington, D.C.
Buckberg, Elaine 1995. “Emerging Stock Markets andInternational Asset Pricing.” The World Bank Eco-nomic Review 9 (1).
Caprio, Gerard, Jr. 1997. “Safe and Sound Banking inDeveloping Countries; We’re Not in Kansas Any-more.” Policy Research Working Paper 1739, WorldBank, Policy Research Department, Finance and Pri-vate Sector Development Division, Washington, D.C.
Caprio Gerard, Jr., and Daniela Klingebiel 1996. “BankInsolvencies: Cross-country Experience.” WorldBank, Policy Research Department, Washington,D.C.
____ 1997. “Bank Insolvency: Bad Luck, Bad Policy, orBad Banking?” In Michael Bruno and BorisPleskovic, eds., Proceedings of the Annual WorldBank Conference on Development Economics, 1996,Washington, D.C.: World Bank.
Caprio, Gerard, Jr., Izak Atiyas, James A. Hanson, andAssociates 1994. Financial Reform: Theory and Prac-tice. New York: Cambridge University Press.
Cho, Y. J. and D. Khatkhate 1989. Lessons of FinancialLiberalization in Asia: A Comparative Study. WorldBank Discussion Paper 50. Washington, D.C.
Cull, Robert 1997. “Financial Sector Adjustment Lend-ing: A Mid-course Analysis.” Policy Research Work-ing Paper 1804. World Bank, Policy Research Depart-ment, Washington, D.C.
Christen, Robert Peck, Elizabeth Rhyne, and Robert C.Vogel 1994. “Maximizing the Outreach of Microen-terprise Finance: the Emerging Lessons of SuccessfulPrograms.” IMMC Consulting Assistance for Eco-nomic Reform Draft Paper, September.
Claessens, Stijn 1995. “The Emergence of Equity Invest-ment in Developing Countries: Overview.” TheWorld Bank Economic Review 9 (1).
Claessens, Stijn, and Thomas Glaessner 1997. “AreFinancial Sector Weaknesses Undermining the EastAsian Miracle?” World Bank, Washington, D.C.
Claessens, Stijn, Susmita Dasgupta, and Jack Glen 1995.“Return Behavior in Emerging Stock Markets.” TheWorld Bank Economic Review. 9 (1).
Claessens, Stijn, Michael P. Dooley, and Andrew Warner1995. “Portfolio Capital Flows: Hot or Cold?” TheWorld Bank Economic Review 9 (1).
Corden, Max W. 1993. “Exchange Rate Policies forDeveloping Countries.” The Economic Journal 103.
Demirgüç-Kunt, Asli, and Enrica Detragiache 1997.“The Determinants of Banking Crises: Evidence fromIndustrial and Developing Countries.” PolicyResearch Working Paper 1828, World Bank, PolicyResearch Department, Washington, D.C.
Deininger, Klaus, Lyn Squire, and Swati Basu 1997,“Does Economic Analysis Improve the Quality ofForeign Assistance?” Draft Internal (World Bank)Paper.
Denizer, Cevdet 1997. “The Effects of Financial Liberal-ization and New Bank Entry on Market Structure andCompetition in Turkey.” Policy Research WorkingPaper 1839, World Bank, Policy Research Depart-ment, Washington, D.C.
Faruqi, Shakil, ed. 1994. “Financial Sector Reforms,Economic Growth, and Stability: Experiences inSelected Asian and Latin American Countries.”World Bank Economic Development Institute Semi-nar Series.
BIBLIOGRAPHY
62
Galbis, Vicente 1993. “High Real Interest Rates UnderFinancial Liberalization: Is There a Problem?” IMFWorking Paper WP/93/7, International MonetaryFund, Washington, D.C.
____ 1994. “Sequencing of Financial Sector Reforms: AReview.” IMF Working Paper WP/94/101, Interna-tional Monetary Fund, Washington, D.C.
____ 1995. “Financial Sector Reforms in Eight Coun-tries: Issues and Results.” IMF Working PaperWP/95/141, International Monetary Fund, Washing-ton, D.C.
Gertler, M., and A. Rose 1994. “Finance, Public Policyand Growth.” In Financial Reform: Theory and Expe-rience. Caprio Gerard, Jr., Izak Atiyas, and JamesHanson, eds. New York: Cambridge University Press.
Gohlke, Gene, and Kenneth J. Berman 1994. “Regula-tion of Money Market Mutual Funds and Lessons forDeveloping Countries.” LATAD Seminar Paper,World Bank. LAC Technical Department AdvisoryGroup. Washington, D.C.
Goldstein M., and P. Turner 1996. “Banking Crisis inEmerging Economics: Origins and Policy Options,Bank for International Settlement (BIS).” Basle.
de Gregorio, Jose, and Pablo E. Guidotti 1995. “Finan-cial Development and Economic Growth.” WorldDevelopment 23 (3): 433–48.
Hanna, Donald P. 1994. Indonesian Experience withFinancial Sector Reform. World Bank DiscussionPaper 237.
Harvey, Campbell R.1995. “The Risk Exposure ofEmerging Equity Markets.” The World Bank Eco-nomic Review 9 (1).
Hellman, Thomas, Kevin Murdock, and JosephStiglitz 1995. “Financial Restraint: Towards A NewParadigm.”
International Finance Corporation, 1996a. “IFC CapitalMarkets Review.” IFC Board of Directors, Washing-ton D.C.
____ 1996b. “IFC’s Experience in Promoting EmergingMarket Investment Funds, 1977–1995.” Washington,D.C.
____ 1996c. “IFC’s Experience with Promoting Leasingin Developing Countries, 1977–1995.” Washington,D.C.
International Monetary Fund 1997a. “Towards aFramework for Financial Stability.” Washington,D.C.
____ 1997b. “Systemic Bank Restructuring and Macro-economic Policy.” Washington D.C.
Jaramillo, Fidel, Fabio Schiantarelli, and Andrew Weiss1993. “Capital Market Imperfections Before andAfter Financial Liberalization.” Policy ResearchWorking Paper 1091. World Bank, Policy ResearchDepartment, Washington, D.C.
Johnson, Barry R. 1997. “Speed of Financial SectorReform: Risks and Strategy Included in SequencingFinancial Strategies in Developing Countries.” A.Harwood and B. Smith, eds., Washington, D.C.: TheBrookings Institution.
Johnson, John H., Sulaiman S. Wasty 1993. “BorrowerOwnership of Adjustment Programs and the PoliticalEconomy of Reform.” World Bank Discussion Paper199.
Kessides, Christine, et al. eds. 1989. “Financial Reformin Socialist Economies.” EDI Seminar Series, WorldBank, Economic Development Institute, Washington,D.C.
Khandker, Shahidur, Baqui Khalily, and Zahed Khan1994. “Is Grameen Bank Sustainable?” HumanResources Development and Operations PolicyWorking Paper 23. World Bank, Human ResourcesGroup, Washington, D.C.
Khatkhate, Deena R. 1988a. “Assessing the Impact ofInterest Rates in Less Developed Countries.” WorldDevelopment 16 (5).
____ 1988b. “Assessing the Level of Interest Rates inLess Developed Countries.” Economic and PoliticalWeekly.
____ 1992 “Financial Liberalization and FinancialCrises.” Economic and Political Weekly November27, 1992
____ 1993. “Indian Banking System Restitution andReform Sequencing.” Washington, D.C.
____ 1996. “Financial Liberalization: A RevisionistView.” The Indian Journal of Economics 304.
Khatkhate, Deena R., and Yoon-Je Cho 1989. Lessonsof Financial Liberalization in Asia: A ComparativeStudy. World Bank Discussion Paper 50. Washington,D.C.
Khatkhate, Deena R., and Ismail Dalla 1995. RegulatedDeregulation of the Financial System in Korea. WorldBank Discussion Paper 292. Washington, D.C.
Khatkhate, Deena R., and Klaus-Walter Riechel 1980.“Multipurpose Banking: Its Nature, Scope, and Rele-vance for Less Developed Countries.” InternationalMonetary Fund Staff Paper 27. Washington, D.C.
Khatkhate, Deena R., and Delano P. Villanueva 1978.“Operation of Selective Credit Policies in Less Devel-
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
63
oped Countries: Certain Critical Issues.” WorldDevelopment (6). Washington, D.C.
Klaus Deininger, Lyn Squire, and Swati Basu 1997.“Does Economic Analysis Improve the Quality ofForeign Assistance?” World Bank, Washington, D.C.
Lamdany, Ruben 1993. “Russia: The Banking SystemDuring Transition.” A World Bank Country Study.Washington, D.C.
Levine, Ross 1997. “Financial Development and Eco-nomic Growth: Views and Agenda.” Journal of Eco-nomic Literature XXXV 2: 688–726.
Lindgren, C., G. Garcia, and M. Saal 1996. “BankSoundness and Macroeconomic Policy.” Interna-tional Monetary Fund, Washington, D.C.
Long, Millard, and Izabela Rutkowska 1995. “The Roleof Commercial Banks in Enterprise Restructuring inCentral and Eastern Europe.” Policy Research Work-ing Paper 1423. World Bank, Washington, D.C.
McKinnon R. I. 1973. Money and Capital in EconomicDevelopment. Washington, D.C.: The BrookingsInstitution.
____ 1994. “Gradual versus Rapid Liberalization inSocialist Economies: Financial Policies and Macro-economic Stability in China and Russia Compared.”In Michael Bruno and Boris Pleskovic, eds., Proceed-ings of the World Bank Annual Conference on Devel-opment Economics 1993. World Bank, Washington,D.C.
____ 1995. “Notes on the Overborrowing Syndrome.”Paper presented at the EDI Conference on ManagingEconomic Reform In An Uncertain World, WorldBank, Washington, D.C. May 30–June 2.
Mishkin, Frederic S. 1997. “Understanding FinancialCrises: A Developing Country Perspective.” InMichael Bruno and Boris Pleskovic, eds., Proceedingsof the Annual World Bank Conference on Develop-ment Economics. 1996. World Bank, Washington,D.C.
Page, John 1994. “The East Asian Miracle: Building aBasis for Growth.” Finance & Development (March).
von Pischke, J. D. 1994. “Measuring the Performance ofSmall Enterprise Lenders.” Paper prepared for a con-ference on Financial Services and the Poor, TheBrookings Institution, Washington, D.C., September.
Pohl, Gerhard, Robert E. Anderson, Stijn Claessens, andSimeon Djankov 1997. “Privatization and Restruc-turing in Central and Eastern Europe: Evidence andPolicy Options.” World Bank Technical Paper 368.Washington, D.C.
Popiel, Paul A. 1994. Financial Systems in Sub-SaharanAfrica: A Comparative Study. World Bank DiscussionPaper 260. Washington, D.C.
Ray, Anandarup 1995 “Directed Credit.” DEC PolicyReview Note 1. World Bank, Development Econom-ics Department, Washington, D.C.
Schandler, Susan, Adam Bennett 1995a. “IMF Condi-tionality: Experience Under Stand-By and ExtendedArrangements–Part I: Key Issues and Findings.”Occasional Paper 128. International Monetary Fund.Washington, D.C.
____ 1995b. “IMF Conditionality: Experience UnderStand-By and Extended Arrangements–Part II: Back-ground Papers.” Occasional Paper 129. InternationalMonetary Fund. Washington, D.C.
Shaw Edward, S. 1973. Financial Deepening in EconomicDevelopment. New York: Oxford University Press.
Schneider, Robert 1995. “Economic Sector Work andResults on the Ground.” World Bank, Washington,D.C.
Schrantarelli, F., Izak Atiyas, Gerard Caprio, James Har-ris, and A. Weiss 1994. “A Credit Where It Is Due? AReview of the Macro and Micro Evidence on the RealEffects of Financial Reform.” Financial Reform: The-ory and Experience. G. Caprio, Izak Atiyas, and J.Hanson, eds., New York: Cambridge University Press.
Sheng, Andrew, ed. 1996. “Bank Restructuring: Lessonsfrom the 1980s.” World Bank, Washington, D.C.
Shilling, John D., and Yan Wang 1995. “Managing Cap-ital Flows in East Asia.” EAP Regional Study (Draft).World Bank, East Asia and the Pacific Region, Wash-ington, D.C.
Steel, William, and Ernest Aryeetey 1994. “InformalSavings Collectors in Ghana: Can They Intermedi-ate?” Finance & Development (March).
Stiglitz, Joseph E. 1994. “The Role of the State in Finan-cial Markets.” In Michael Bruno and Boris Pleskovic,eds., Proceedings of the World Bank Annual Confer-ence on Development Economics. Washington, D.C.:World Bank
Stiglitz, Joseph E., and A. Weiss 1981. “CreditRationing in Markets with Imperfect Information.”American Economic Review 71.
Sundararajan, V., and Tomas Balino, eds., 1991. Bank-ing Crises: Cases and Issues. International MonetaryFund, Washington, D.C.
Sundararajan V., and Lazaros Molho 1988. “FinancialReform in Indonesia.” Finance & Development(December).
B i b l i o g r a p h y
64
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
Swinburne, Mark 1995. “Financial Liberalization inTransition Economies: Problems and Issues.” Interna-tional Monetary Fund, Monetary and ExchangeAffairs Dept., Washington, D.C.
Tanzi, Vito 1996. “Fiscal Federalism and Decentraliza-tion: A Review of Some Efficiency and Macroeco-nomic Aspects.” In Michael Bruno and BorisPleskovic, eds., Annual World Bank Conference onDevelopment Economics 1995. Washington, D.C.:World Bank
Tesar, Linda, and Ingrid M. Werner 1995. “U.S. EquityInvestment in Emerging Stock Markets.” The WorldBank Economic Review 9 (1).
Villanueva, Delano, and Abbas Mirakhor 1990. “Strate-gies for Financial Reforms.” IMF Staff Papers 37 (3).
Vittas, Dimitri, and Yoon Je Cho 1994. “The Role ofCredit Policies in Japan and Korea.” Finance &Development (March).
Williamson, John, ed. 1994. Political Economy of Pol-icy Reforms. Washington, D.C.: Institute for Interna-tional Economics.
Wilson, B., Gerard Caprio, and A Saunders 1997.“Mexico Banking Crisis: Devaluation and Asset Con-centration Effect.” World Bank, Washington, D.C.
World Bank 1989a. “Report of the Task Force on Finan-cial Sector Operations.” Washington, D.C.
____ 1989b. World Development Report 1989. NewYork: Oxford University Press.
____ 1992. “Operational Manual Operational DirectiveOD 8.30.” Washington, D.C.
____ 1993a. “Adjustment in Africa: Reforms, Resultsand the Road Ahead” Policy Research Report 2, Pol-icy Research Department. Washington, D.C.
____ 1993b. “Handbook on Financial Sector Opera-tions.” Operations Policy Department, Washington,D.C.
____ 1995a. “Bank Recapitalization: If and When.”DEC Policy Review Note 2. Development EconomicsDepartment, Washington, D.C.
____ 1995b. Bureaucrats in Business: The Economicsand Politics of Public Ownership. New York: OxfordUniversity Press. Washington, D.C.
____ 1995c. Directed Credit. DEC Policy Review Note1. Development Economics Department, Washing-ton, D.C.
____ 1995d. Interest Rate Deregulation. DEC PolicyReview Note 3. Development Economics Depart-ment, Washington, D.C.
____ 1995e. “Performance Indicators for AdjustmentPrograms: A First Edition Note.” Development Eco-nomics Department, Office of the Vice President,Washington, D.C.
____ 1995f. “The Use of Sectoral and Project Perfor-mance Indicators in Bank-Financed Financial SectorOperations.” Financial Sector Development Unit,Washington, D.C.
____ 1995g. “Venezuela: Financial Sector AdjustmentLoan.” Report 14814, Operations EvaluationDepartment, Washington, D.C.
____ 1996a. Guidelines for Reforming Finance. DECPolicy Review Note 4. Development EconomicsDepartment, Washington, D.C.
____ 1996b. “Philippines: Financial Sector AdjustmentLoan and Cottage Enterprise Finance Project.”Report 15834, Operations Evaluation Department,Washington, D.C.
____ 1997a. “Adaptable Lending—New InvestmentInstruments.” Memorandum to the Executive Direc-tors, August 15. Washington, D.C.
____ 1997b. “Bank-Fund Collaboration on Strengthen-ing Financial Sectors.” Prepared jointly by the Staff ofthe World Bank and the International MonetaryFund, Washington, D.C.
____ 1997c. “Portfolio Improvement Program: Reviewof Financial Intermediary Lending Portfolio.” WorldBank, Financial Sector Department, Washington,D.C.
____ 1997d. “The Road to Financial Integration, PrivateCapital Flows to Developing Countries.” Interna-tional Economics Department. Washington, D.C.
____ 1997e. “The World Bank’s Role in the FinancialSector.” OED report presented to the Committee onDevelopment Effectiveness (CODE 97-19), Opera-tions Evaluation Department. Washington, D.C.
65
ANNEXESAN
NEX
TABL
E 1.
1: L
IST
OF
88 F
INAN
CIAL
SEC
TOR
AD
JUST
MEN
T-R
ELAT
ED P
RO
JECT
S (F
SLs)
: FY
85–F
Y96
MA
JOR
LO
AN
/L
EN
DIN
GL
OA
NC
RE
DIT
PR
OJE
CT
FIS
CA
L
PR
OJE
CT
IN
STR
U-
AM
OU
NT
NU
MB
ER
IDC
OU
NT
RY
RE
GIO
NP
RO
JEC
T N
AM
EY
EA
RST
AT
US
ME
NT
(Mil
lion
)
C22
830
110
Ben
inA
FRSt
ruct
ural
Adj
ustm
ent
Loa
n II
1991
Com
plet
edSA
L55
C24
720
289
Bur
kina
Fas
oA
FRPr
ivat
e Se
ctor
Ass
ista
nce
1993
Act
ive
SIL
7C
2303
011
99C
ôte
d’Iv
oire
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Prog
ram
1992
Com
plet
edSA
D20
0C
1911
089
2G
hana
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
I19
88C
ompl
eted
SAD
100
C23
180
911
Gha
naA
FRFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t II
1992
Act
ive
SAD
100
C21
480
1050
Gui
nea
AFR
Priv
ate
Sect
or P
rom
otio
n19
90C
ompl
eted
SAD
50C
2653
010
78G
uine
aA
FRFi
nanc
ial S
ecto
r O
pera
tion
1995
Act
ive
SAL
23C
1798
098
3G
uine
a B
issa
u A
FRSt
ruct
ural
Adj
ustm
ent
Loa
n 19
87C
ompl
eted
SAL
15C
2049
013
10K
enya
AFR
Fina
ncia
l Sec
tor
Ope
rati
on19
89C
ompl
eted
SAD
120
C18
340
1511
Mad
agas
car
AFR
Indu
stry
and
Tra
de P
olic
y A
djus
tmen
t C
redi
t19
87C
ompl
eted
SAD
83C
2104
015
40M
adag
asca
rA
FRFi
nanc
ial S
ecto
r an
d Pr
ivat
e E
nter
pris
e D
evel
opm
ent
Proj
.19
90A
ctiv
eSI
M48
C22
210
1655
Mal
awi
AFR
Fina
ncia
l Sec
tor
and
Ent
erpr
ise
Dev
elop
men
t Pr
ojec
t19
91A
ctiv
eFI
L32
C27
260
1866
Mau
rita
nia
AFR
Priv
ate
Sect
or D
evel
opm
ent
Prog
ram
1995
Act
ive
SAD
30C
2607
018
11M
ozam
biqu
eA
FRFi
nanc
ial S
ecto
r C
apac
ity-
build
ing
Proj
ect
1994
Act
ive
TA
L9
C18
020
2340
Sene
gal
AFR
Stru
ctur
al A
djus
tmen
t L
oan
III
1987
Com
plet
edSA
L93
C20
770
2355
Sene
gal
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Prog
ram
1990
Com
plet
edSA
D45
C23
080
2809
Tanz
ania
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
1992
Com
plet
edSA
D20
0C
2771
035
620
Tanz
ania
AFR
Fina
ncia
l Ins
titu
tion
s D
evel
opm
ent
Proj
ect
1996
Act
ive
SIL
10.9
C24
960
2962
Uga
nda
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
1993
Act
ive
SAD
100
C24
230
3623
Chi
naE
AP
Fina
ncia
l Sec
tor
Tech
nica
l Ass
ista
nce
Proj
ect
1993
Act
ive
TA
L60
L29
370
3962
Indo
nesi
aE
AP
Seco
nd T
rade
Pol
icy
Adj
ustm
ent
Loa
n19
88C
ompl
eted
SAD
300
L30
800
3974
Indo
nesi
aE
AP
Priv
ate
Sect
or D
evel
opm
ent
Loa
n19
89C
ompl
eted
SAL
350
L32
670
3994
Indo
nesi
aE
AP
Priv
ate
Sect
or D
evel
opm
ent
Loa
n II
1991
Com
plet
edSA
L25
0L
3526
039
70In
done
sia
EA
PFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1993
Act
ive
FIL
307
L25
7141
20K
orea
EA
PIn
dust
rial
Fin
ance
Pro
ject
II
1985
Com
plet
edSA
D22
2C
2037
041
99L
ao, P
DR
EA
PSt
ruct
ural
Adj
ustm
ent
Cre
dit
1989
Com
plet
edSA
L40
L27
70/1
4281
Mal
aysi
aE
AP
Dev
elop
men
t Fi
nanc
e Pr
ojec
t19
87C
ompl
eted
SIL
65L
3049
045
17Ph
ilipp
ines
EA
PFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1989
Com
plet
edSA
D30
0
L35
850
8278
Arm
enia
EC
AIn
stit
utio
n-bu
ildin
g L
oan
1993
Act
ive
TA
L12
L33
970
8308
Bul
gari
aE
CA
Stru
ctur
al A
djus
tmen
t L
oan
I19
92C
ompl
eted
SAL
250
L38
030
8401
Est
onia
EC
AFi
nanc
ial I
nsti
tuti
ons
Dev
elop
men
t Pr
ojec
t19
95A
ctiv
eFI
L10
L31
910
8478
Hun
gary
EC
AFi
nanc
ial S
yste
ms
Mod
erni
zati
on P
roje
ct19
90A
ctiv
eSI
M66
L33
470
8492
Hun
gary
EC
ASt
ruct
ural
Adj
ustm
ent
Loa
n II
1991
Com
plet
edSA
L25
0L
3867
085
08K
azak
stan
EC
AFi
nanc
e an
d E
nter
pris
e D
evel
opm
ent
Proj
ect
1995
Act
ive
SIL
62L
3795
085
27L
atvi
aE
CA
Ent
erpr
ise
and
Fina
ncia
l Sec
tor
Res
truc
turi
ng P
roje
ct19
94A
ctiv
eSI
L25
66
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNE
X TA
BLE
1.1:
(CO
NTI
NU
ED)
MA
JOR
LO
AN
/L
EN
DIN
GL
OA
NC
RE
DIT
PR
OJE
CT
FIS
CA
L
PR
OJE
CT
IN
STR
U-
AM
OU
NT
NU
MB
ER
IDC
OU
NT
RY
RE
GIO
NP
RO
JEC
T N
AM
EY
EA
RST
AT
US
ME
NT
(Mil
lion
)
L36
9585
29L
atvi
aE
CA
Agr
icul
tura
l Dev
elop
men
t Pr
ojec
t19
95A
ctiv
eFI
L35
L38
660
8536
Lit
huan
iaE
CA
Ent
erpr
ise
and
Fina
ncia
l Sec
tor
Ass
ista
nce
Proj
ect
1995
Act
ive
FIL
25C
2721
084
09M
aced
onia
EC
AFi
nanc
ial a
nd E
nter
pris
e Se
ctor
Adj
ustm
ent
Cre
dit
1995
Act
ive
SAD
85L
3247
085
85Po
land
EC
AFi
nanc
ial I
nsti
tuti
ons
Dev
elop
men
t Pr
ojec
t19
91A
ctiv
eSA
D20
0L
3341
085
88Po
land
EC
ASt
ruct
ural
Adj
ustm
ent
Loa
n I
1991
Com
plet
edSA
L30
0L
3599
085
89Po
land
EC
AE
nter
pris
e an
d Fi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1993
Act
ive
SAL
450
L37
340
8828
Rus
sia
EC
AFi
nanc
ial I
nsti
tuti
ons
Dev
elop
men
t Pr
ojec
t19
94C
ompl
eted
SIL
200
L36
660
8848
Slov
ak
EC
AE
cono
mic
Rec
over
y L
oan
1994
Com
plet
edSA
L80
Rep
ublic
L36
360
8854
Slov
enia
EC
AE
nter
pris
e an
d Fi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1994
Act
ive
SAD
80L
2714
089
62Tu
rkey
EC
AFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1986
Com
plet
edSA
D30
0L
2964
089
87Tu
rkey
EC
AFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
II19
88C
ompl
eted
SAD
400
L35
580
6047
Arg
enti
naL
AC
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n19
93C
ompl
eted
SAL
400
L37
090
5988
Arg
enti
naL
AC
Cap
ital
Mar
ket
Dev
elop
men
t Pr
ojec
t19
94A
ctiv
eFI
L50
0L
3710
060
62A
rgen
tina
LA
CC
apit
al M
arke
t Te
chni
cal A
ssis
tanc
e Pr
ojec
t19
94A
ctiv
eT
AL
8.5
L39
2640
904
Arg
enti
naL
AC
Ban
k R
efor
m L
oan
1996
Act
ive
SAD
500
C18
090
6160
Bol
ivia
LA
CPu
blic
Fin
anci
al M
anag
emen
t O
pera
tion
I19
87C
ompl
eted
TA
L11
.5C
1925
061
59B
oliv
iaL
AC
Fina
ncia
l Sec
tor
Adj
ustm
ent
Proj
ect
1988
Com
plet
edSA
D70
C22
980
6184
Bol
ivia
LA
CSt
ruct
ural
Adj
ustm
ent
Prog
ram
1992
Com
plet
edSA
L40
L26
250
6627
Chi
leL
AC
Stru
ctur
al A
djus
tmen
t L
oan
1986
Com
plet
edSA
L25
0L
2767
066
28C
hile
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n II
1987
Com
plet
edSA
L25
0L
2892
066
33C
hile
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n II
I19
88C
ompl
eted
SAL
250
L31
430
6634
Chi
leL
AC
Fina
ncia
l Mar
kets
Ope
rati
on19
90C
ompl
eted
FIL
130
L25
180
6923
Cos
ta R
ica
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n I
1985
Com
plet
edSA
L80
L28
970
7103
Ecu
ador
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t O
pera
tion
1988
Com
plet
edSA
D10
0L
3609
070
98E
cuad
orL
AC
Priv
ate
Sect
or D
evel
opm
ent
Proj
ect
1993
Com
plet
edFI
L75
L35
330
7210
Gua
tem
ala
LA
CE
cono
mic
Mod
erni
zati
on L
oan
1993
Com
plet
edSA
L12
0C
2669
072
59G
uyan
aL
AC
Fina
ncia
l Sec
tor
and
Bus
ines
s E
nvir
onm
ent
Proj
ect
1995
Act
ive
SAD
15.5
C27
460
3460
5G
uyan
aL
AC
Priv
ate
Sect
or D
evel
opm
ent
Adj
ustm
ent
Cre
dit
1995
Act
ive
TA
L3.
5L
2990
073
83H
ondu
ras
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n I
1989
Com
plet
edSA
L50
L28
480
7445
Jam
aica
LA
CT
rade
and
Fin
anci
al S
ecto
r A
djus
tmen
t O
pera
tion
I19
87C
ompl
eted
SAD
40L
3303
074
69Ja
mai
caL
AC
Tra
de a
nd F
inan
cial
Sec
tor
Adj
ustm
ent
Ope
rati
on I
I19
91C
ompl
eted
SAD
30L
3085
076
91M
exic
oL
AC
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n19
89C
ompl
eted
SAD
500
L38
380
4049
7M
exic
oL
AC
Fina
ncia
l Sec
tor
Tech
nica
l Ass
ista
nce
Proj
ect
1995
Act
ive
SAD
1,00
0L
3911
034
161
Mex
ico
LA
CFi
nanc
ial S
ecto
r R
estr
uctu
ring
Adj
ustm
ent
Prog
ram
1995
Act
ive
TA
L23
.6C
2302
077
81N
icar
agua
LA
CE
cono
mic
Rec
over
y C
redi
t19
92C
ompl
eted
SAL
110
67
A n n e x e s
ANNE
X TA
BLE
1.1:
(CO
NTI
NU
ED)
L37
740
7917
Para
guay
LA
CPr
ivat
e Se
ctor
Dev
elop
men
t Pr
ojec
t19
95A
ctiv
eSI
L50
L34
890
8050
Peru
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1992
Act
ive
SAD
400
L30
810
8145
Uru
guay
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n II
1989
Com
plet
edSA
L14
0L
3224
082
08V
enez
uela
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1990
Com
plet
edSA
D30
0
L33
520
4963
Alg
eria
MN
AE
nter
pris
e an
d Fi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1991
Com
plet
edSA
D35
0L
3834
035
704
Alg
eria
MN
AE
cono
mic
Reh
abili
tati
on S
uppo
rt L
oan
1995
Act
ive
SAL
150
C24
020
5167
Egy
ptM
NA
Tech
nica
l Ass
ista
nce
Proj
ect
for
Priv
atiz
atio
n an
d
1992
Act
ive
TA
L9
Ent
erpr
ise
and
Ban
king
Sec
tor
Ref
orm
sL
2604
054
14M
oroc
coM
NA
Indu
stri
al T
rade
Pol
icy
Adj
ustm
ent
Loa
n19
86C
ompl
eted
SAD
200
L33
650
5495
Mor
occo
MN
AFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1991
Act
ive
SAD
235
L39
280
5522
Mor
occo
MN
AFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1996
Act
ive
SAD
250
L29
620
5718
Tuni
sia
MN
ASt
ruct
ural
Adj
ustm
ent
Loa
n I
1988
Com
plet
edSA
L15
0L
3424
057
42Tu
nisi
aM
NA
Eco
nom
ic a
nd F
inan
cial
Ref
orm
s Su
ppor
t L
oan
1992
Com
plet
edSA
L25
0
C21
520
9528
Ban
glad
esh
SAS
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
1990
Com
plet
edSA
D17
5L
3856
010
563
Indi
aSA
SFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1995
Act
ive
FIL
700
C20
460
1033
4N
epal
SAS
Stru
ctur
al A
djus
tmen
t C
redi
t II
1989
Com
plet
edSA
L60
L30
290
1032
7Pa
kist
anSA
SFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1989
Com
plet
edSA
D15
0L
3808
010
470
Paki
stan
SAS
Fina
ncia
l Sec
tor
Dee
peni
ng a
nd I
nter
med
iati
on P
roje
ct19
95A
ctiv
eFI
L21
6C
2484
010
419
Sri L
anka
SAS
Priv
ate
Fina
nce
Dev
elop
men
t Pr
ojec
t19
93A
ctiv
eSI
L60
68
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNE
X TA
BLE
1.2:
BAN
K ES
W R
ELAT
ED T
O F
INAN
CIAL
SEC
TOR
REF
OR
M:
FY81
–FY9
5
RE
PO
RT
NO
.R
EG
ION
CO
UN
TR
YT
ITL
ET
YP
ED
AT
E
7690
AFR
Bot
swan
aFi
nanc
ial P
olic
ies
for
Div
ersi
fied
Gro
wth
SR8/
1/89
1097
8A
FRB
urun
diA
Fin
anci
al S
ecto
r R
evie
wSR
7/1/
9240
66A
FRB
urun
diFi
nanc
ial S
ecto
r R
epor
tSR
7/1/
8260
28A
FRC
amer
oon
Fina
ncia
l Sec
tor
Rep
ort
SR6/
1/86
3113
AFR
Côt
e D
’Ivo
ire
Fina
nce
in t
he D
evel
opm
ent
of t
he I
vory
Coa
stE
R12
/1/8
194
27A
FRE
thop
iaFi
nanc
ial S
ecto
r R
evie
wSR
6/1/
9147
66A
FRG
ambi
aFi
nanc
ial S
ecto
r R
evie
wSR
10/1
/83
1342
3A
FRG
hana
Fina
ncia
l Sec
tor
Rev
iew
: Bri
ngin
g Sa
vers
and
In
vest
ors
Toge
ther
SR12
/29/
9489
11A
FRG
hana
Tow
ards
a D
ynam
ic I
nves
tmen
t R
epon
seSR
10/1
/90
9809
AFR
Gui
nea
Bis
sau
Priv
ate
Sect
or D
evel
opm
ent
Stud
ySR
12/1
/91
1143
8A
FRK
enya
Tapp
ing
Ken
ya’s
Pot
enti
al: A
PSD
Str
ateg
ySR
12/1
/92
5619
AFR
Ken
yaA
gric
ultu
ral C
redi
t Po
licy
Rev
iew
SR8/
1/85
5246
AFR
Ken
yaIn
dust
rial
Fin
ance
SR4/
1/85
8021
AFR
Les
otho
Fina
ncia
l Sec
tor
Rev
iew
SR4/
1/90
9817
AFR
Mad
agas
car
Fina
ncia
l Pol
icie
s fo
r D
iver
sifi
ed G
row
thSR
3/1/
9210
057
AFR
Mad
agas
car
Rur
al F
inan
ce S
ecto
r R
evie
wSR
2/1/
9290
09A
FRM
alaw
iFi
nanc
ial P
olic
ies
for
Sust
aina
ble
Gro
wth
SR2/
19/9
210
443
AFR
Mau
riti
usFi
nanc
ial S
ecto
r R
evie
wSR
6/1/
9210
269
AFR
Moz
ambi
que
Fina
ncia
l Sec
tor
Stud
ySR
9/1/
9213
911
AFR
Nig
eria
The
Nig
eria
n R
ural
Fin
anci
al S
yste
m: A
sses
smen
tSR
1/26
/95
and
Rec
omm
enda
tion
s12
781
AFR
Nig
eria
The
Nig
eria
n B
anki
ng S
yste
m: F
inan
cial
Ass
essm
ent
and
SR3/
16/9
4K
ey I
ssue
s fo
r D
istr
ess
Res
olut
ion
9864
AFR
Nig
eria
The
Fin
anci
al S
ecto
r: I
ssue
s an
d O
ptio
nsSR
10/1
/91
4051
AFR
Nig
eria
Fina
ncia
l Int
erm
edia
tion
SR2/
1/83
8934
AFR
Rw
anda
Fina
ncia
l Sec
tor
Rev
iew
SR5/
1/91
4457
AFR
Sier
ra L
eone
Fina
ncia
l Sec
tor
Stud
ySR
2/1/
8490
99A
FRU
gand
aFi
nanc
ial S
ecto
r R
evie
wSR
5/7/
9184
98A
FRZ
aire
Priv
ate
Sect
or I
ncen
tive
s: C
urre
nt P
olic
ies
and
Prop
osal
s E
R3/
1/90
for
Ref
orm
1238
7A
FRZ
ambi
aFi
nanc
ial S
ecto
r D
evel
opm
ent
SR10
/1/9
3
1349
2E
AP
Chi
naFi
nanc
ial S
ecto
r R
efor
ms:
Cur
rent
Sta
tus
and
Issu
esSR
9/6/
9484
15E
AP
Chi
naFi
nanc
ial S
ecto
r R
evie
w: F
inan
cial
Pol
icie
s an
d E
R6/
29/9
0In
stit
utio
nal D
evel
opm
ent
1311
2E
AP
Indo
nesi
aN
on-b
ank
Fina
ncia
l Sec
tor
Stud
ySR
6/7/
9494
98E
AP
Indo
nesi
aD
evel
opin
g Pr
ivat
e E
nter
pris
eSR
5/9/
9181
59E
AP
Indo
nesi
aFi
nanc
ial S
ecto
r R
epor
tSR
5/1/
9055
01E
AP
Indo
nesi
aPo
licie
s an
d Pr
ospe
cts
for
Lon
g-te
rm F
inan
cial
Dev
elop
men
tE
R7/
1/85
69
A n n e x e s
ANNE
X TA
BLE
1.2:
(CO
NTI
NU
ED)
4566
EA
PIn
done
sia
Rur
al C
redi
t St
udy
SR6/
1/83
1137
3E
AP
Kor
eaFi
nanc
ial S
ecto
r St
udy
SR7/
1/93
3288
EA
PK
orea
Fina
ncia
l Sec
tor
Rep
ort
SR11
/1/8
065
30E
AP
Mal
aysi
aA
Stu
dy o
f D
evel
opm
ent
Fina
nce
Inst
itut
ions
SR1/
1/87
1005
3E
AP
Phili
ppin
esC
apit
al M
arke
t St
udy
ER
2/1/
9271
77E
AP
Phili
ppin
esFi
nanc
ial S
ecto
r St
udy
SR8/
1/88
8403
EA
PT
haila
ndFi
nanc
ial S
ecto
r St
udy
SR5/
25/9
040
85E
AP
Tha
iland
Pers
pect
ives
for
Fin
anci
al R
efor
mSR
7/1/
8313
153
EA
PV
ietn
amFi
nanc
ial S
ecto
r R
evie
w: A
n A
gend
a fo
r Fi
nanc
ial S
ecto
r SR
3/1/
95D
evel
opm
ent
9223
EA
PV
ietn
amT
rans
form
ing
A S
tate
-ow
ned
Fina
ncia
l Sys
tem
: SR
4/15
/91
A F
inan
cial
Sec
tor
Stud
y
6941
EC
AH
unga
ryD
evel
opm
ent
and
Ref
orm
of
Fina
ncia
l Mar
kets
SR6/
1/89
1181
8E
CA
Rus
sia
The
Ban
king
Sys
tem
in t
he T
rans
itio
nSR
9/1/
9344
59E
CA
Turk
eySp
ecia
l Eco
nom
ic R
epor
t: P
olic
ies
for
the
Fina
ncia
l Sec
tor
ER
9/1/
8378
69E
CA
Yug
osla
via
Fina
ncia
l Sec
tor
Res
truc
turi
ng: P
olic
ies
and
Prio
riti
esSR
11/1
/89
1296
3L
AC
Arg
enti
naC
apit
al M
arke
t St
udy
SR12
/21/
9411
673
LA
CA
rgen
tina
Fina
ncia
l Sec
tor
Rev
iew
SR11
/1/8
971
76L
AC
Arg
enti
naSe
curi
ties
Mar
kets
and
Mai
n N
on-b
ank
Fina
ncia
l Ins
titu
tion
sSR
3/1/
8864
18L
AC
Arg
enti
naB
anki
ng S
ecto
r: T
he N
eed
for
Ref
orm
SR12
/1/8
613
873
LA
CB
oliv
iaH
ow L
egal
Res
tric
tion
s on
Col
late
ral L
imit
Acc
ess
to
SR12
/1/9
4C
redi
t in
Bol
ivia
1107
5L
AC
Bol
ivia
Res
truc
turi
ng f
or G
row
th: T
he R
emai
ning
Age
nda
for
ER
9/1/
92PE
Ref
orm
and
PSD
6765
LA
CB
oliv
iaB
anki
ng S
ecto
r St
udy
SR11
/1/8
811
581
LA
CB
razi
lT
he D
evel
opm
ent
of t
he B
razi
lian
Cap
ital
Mar
ket
SR10
/7/9
494
58L
AC
Bra
zil
Cap
ital
Mar
kets
Iss
ues
SR3/
1/91
7725
LA
CB
razi
lSe
lect
ed I
ssue
s of
the
Fin
anci
al S
ecto
rSR
3/1/
9082
47L
AC
Bra
zil
The
Dile
mm
a of
Bra
zil’s
Sta
te B
anki
ng S
yste
mSR
1/1/
9027
90L
AC
Bra
zil
Fina
ncia
l Sys
tem
s R
evie
wSR
11/1
/80
7737
LA
CC
hile
Indu
stri
al F
inan
ce: S
ecto
r R
epor
tSR
1/1/
8911
724
LA
CC
olom
bia
Fina
ncia
l Sec
tor
Ref
orm
SR6/
1/93
8276
LA
CC
olom
bia
Fina
ncia
l Sec
tor
Stra
tegy
Pap
erSR
12/1
/89
6337
LA
CC
olom
bia
Com
mer
cial
and
Dev
elop
men
t B
anks
SR3/
1/86
4274
LA
CC
olom
bia
The
Col
ombi
an I
nves
tmen
t B
anki
ng S
ecto
r an
d R
elat
ed
SR8/
1/83
Fina
ncia
l Sec
tor
Issu
es68
21L
AC
Cos
ta R
ica
Sele
cted
Fin
anci
al S
ecto
r Is
sues
SR3/
1/88
8601
LA
CD
omin
ican
Rep
ublic
Fina
ncia
l Sec
tor
Stra
tegy
ER
4/1/
9052
70L
AC
Ecu
ador
Bri
ef R
evie
w o
f th
e Fi
nanc
ial S
ecto
rSR
4/1/
8578
19L
AC
Gua
tem
ala
Fina
ncia
l Sec
tor
Rep
ort
ER
7/1/
90
70
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNE
X TA
BLE
1.2:
(CO
NTI
NU
ED)
RE
PO
RT
NO
.R
EG
ION
CO
UN
TR
YT
ITL
ET
YP
ED
AT
E
1170
5L
AC
Guy
ana
Priv
ate
Sect
or D
evel
opm
ent
SR6/
1/93
1030
7L
AC
Guy
ana
From
Eco
nom
ic R
ecov
ery
to S
usta
ined
Gro
wth
ER
4/1/
9273
63L
AC
Hon
dura
sFi
nanc
ial S
ecto
r Su
rvey
ER
7/1/
8811
823
LA
CM
exic
oC
ount
ry E
cono
mic
Mem
oran
dum
ER
5/16
/94
9198
LA
CM
exic
oD
evel
opm
ent
Ban
ks I
ssue
s: A
Fra
mew
ork
of A
naly
sis
and
SR12
/1/9
0Su
gges
ted
Ban
k St
rate
gy43
16L
AC
Peru
Bri
ef R
evie
w o
f th
e Fi
nanc
ial S
ecto
rSR
2/1/
8393
73L
AC
Tri
nida
d an
d Fi
nanc
ial S
ecto
r St
udy
SR5/
1/91
Toba
go88
99L
AC
Uru
guay
Fina
ncia
l Sec
tor
Dev
elop
men
t an
d R
estr
uctu
ring
Iss
ues
ER
6/1/
9010
393
LA
CV
enez
uela
Term
Fin
anci
ng a
nd C
apit
al M
arke
ts D
evel
opm
ent
SR2/
1/92
1194
0M
NA
Alg
eria
Cap
ital
Mar
kets
Stu
dyE
R6/
1/93
7241
MN
AA
lger
iaT
he A
lger
ian
Fina
ncia
l Sys
tem
SR5/
1/88
2981
MN
AA
lger
iaL
e Fi
nanc
emen
t D
es E
nter
pris
es E
t L
e Sy
stem
e B
anca
ire
SR4/
10/8
010
790
MN
AE
gypt
Fina
ncia
l Pol
icy
for
Adj
ustm
ent
and
Gro
wth
SR9/
1/93
1033
7M
NA
Egy
ptR
efor
m a
nd D
evel
opm
ent
of t
he S
ecur
itie
s M
arke
t:
SR2/
1/92
Rec
omm
enda
tion
to
the
Gov
ernm
ent
1315
3M
NA
Iran
Cap
ital
Mar
kets
and
Fin
anci
al I
nsti
tuti
ons
SR9/
28/9
413
183
MN
AL
eban
onFi
nanc
ial P
olic
y fo
r St
abili
zati
on, R
econ
stru
ctio
n,
ER
7/27
/94
and
Dev
elop
men
t11
557
MN
AM
oroc
coD
evel
opin
g Pr
ivat
e In
dust
ry in
Mor
occo
SR7/
1/93
4957
MN
AM
oroc
coFi
nanc
ial S
ecto
r St
udy
SR12
/1/8
452
63M
NA
Tuni
sia
Fina
ncia
l Sec
tor
Rep
ort
SR12
/1/8
5
6901
SAS
Ban
glad
esh
A P
rogr
am f
or F
inan
cial
Sec
tor
Ref
orm
SR12
/1/8
740
98SA
SB
angl
ades
hFi
nanc
ial S
ecto
r R
evie
wSR
9/1/
8210
489
SAS
Indi
aSt
abili
zing
and
Ref
orm
ing
the
Eco
nom
yE
R5/
1/92
8264
SAS
Indi
aFi
nanc
ial S
ecto
r R
epor
t: C
onso
lidat
ion
of t
heSR
6/1/
90Fi
nanc
ial S
yste
m66
61SA
SIn
dia
Cre
dit
and
Cap
ital
Mar
kets
Stu
dySR
2/1/
8711
637
SAS
Paki
stan
Res
truc
turi
ng t
he F
inan
cial
Sys
tem
: Bui
ldin
g on
E
R4/
1/93
Fina
ncia
l Ref
orm
7049
SAS
Paki
stan
Fina
ncia
l Sec
tor
Rev
iew
SR12
/1/8
793
39SA
SSr
i Lan
kaFi
nanc
ial I
nsti
tuti
ons
Stud
ySR
2/1/
91
71
A n n e x e s
ANNE
X TA
BLE
2.1:
LIST
OF
SAM
PLE
FIN
ANCI
AL S
ECTO
R A
DJU
STM
ENT-
REL
ATED
OPE
RAT
ION
S (F
SLs)
FO
R 2
3 CO
UN
TRIE
S: F
Y85–
FY96
MA
JOR
LO
AN
/L
EN
DIN
GL
OA
NC
RE
DIT
PR
OJE
CT
FIS
CA
L
PR
OJE
CT
IN
STR
U-
AM
OU
NT
NU
MB
ER
IDC
OU
NT
RY
RE
GIO
NP
RO
JEC
T N
AM
EY
EA
RST
AT
US
ME
NT
(Mil
lion
)
C23
030
1199
Côt
e d’
Ivoi
reA
FRFi
nanc
ial S
ecto
r A
djus
tmen
t Pr
ogra
m19
92C
ompl
eted
SAD
200
C19
110
892
Gha
naA
FRFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t I
1988
Com
plet
edSA
D10
0C
2318
091
1G
hana
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
II19
92A
ctiv
eSA
D10
0C
2049
013
10K
enya
AFR
Fina
ncia
l Sec
tor
Ope
rati
on19
89C
ompl
eted
SAD
120
C22
210
1655
Mal
awi
AFR
Fina
ncia
l Sec
tor
and
Ent
erpr
ise
Dev
elop
men
t Pr
ojec
t19
91A
ctiv
eFI
L32
C18
020
2340
Sene
gal
AFR
Stru
ctur
al A
djus
tmen
t L
oan
III
1987
Com
plet
edSA
L93
C20
770
2355
Sene
gal
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Prog
ram
1990
Com
plet
edSA
D45
C23
080
2809
Tanz
ania
AFR
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
1992
Com
plet
edSA
D20
0C
2771
035
620
Tanz
ania
AFR
Fina
ncia
l Ins
titu
tion
s D
evel
opm
ent
Proj
ect
1996
Act
ive
SIL
10.9
C24
230
3623
Chi
naE
AP
Fina
ncia
l Sec
tor
Tech
nica
l Ass
ista
nce
Proj
ect
1993
Act
ive
TA
L60
L29
370
3962
Indo
nesi
aE
AP
Seco
nd T
rade
Pol
icy
Adj
ustm
ent
Loa
n19
88C
ompl
eted
SAD
300
L30
800
3974
Indo
nesi
aE
AP
Priv
ate
Sect
or D
evel
opm
ent
Loa
n I
1989
Com
plet
edSA
L35
0L
3267
039
94In
done
sia
EA
PPr
ivat
e Se
ctor
Dev
elop
men
t L
oan
II19
91C
ompl
eted
SAL
250
L35
260
3970
Indo
nesi
aE
AP
Fina
ncia
l Sec
tor
Dev
elop
men
t Pr
ojec
t19
93A
ctiv
eFI
L30
7L
2571
4120
Kor
eaE
AP
Indu
stri
al F
inan
ce P
roje
ct I
I19
85C
ompl
eted
SAD
222
L27
70/1
4281
Mal
aysi
aE
AP
Dev
elop
men
t Fi
nanc
e Pr
ojec
t19
87C
ompl
eted
SIL
65L
3049
045
17Ph
ilipp
ines
EA
PFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1989
Com
plet
edSA
D30
0L
3247
085
85Po
land
EC
AFi
nanc
ial I
nsti
tuti
ons
Dev
elop
men
t Pr
ojec
t19
91A
ctiv
eSA
D20
0L
3341
085
88Po
land
EC
ASt
ruct
ural
Adj
ustm
ent
Loa
n I
1991
Com
plet
edSA
L30
0L
3599
085
89Po
land
EC
AE
nter
pris
e an
d Fi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1993
Act
ive
SAL
450
L27
140
8962
Turk
eyE
CA
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n19
86C
ompl
eted
SAD
300
L29
640
8987
Turk
eyE
CA
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n II
1988
Com
plet
edSA
D40
0C
1809
061
60B
oliv
iaL
AC
Publ
ic F
inan
cial
Man
agem
ent
Ope
rati
on I
1987
Com
plet
edT
AL
11.5
C19
250
6159
Bol
ivia
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t Pr
ojec
t19
88C
ompl
eted
SAD
70C
2298
061
84B
oliv
iaL
AC
Stru
ctur
al A
djus
tmen
t Pr
ogra
m19
92C
ompl
eted
SAL
40L
2625
066
27C
hile
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n I
1986
Com
plet
edSA
L25
0L
2767
066
28C
hile
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n II
1987
Com
plet
edSA
L25
0L
2892
066
33C
hile
LA
CSt
ruct
ural
Adj
ustm
ent
Loa
n II
I19
88C
ompl
eted
SAL
250
L31
430
6634
Chi
leL
AC
Fina
ncia
l Mar
kets
Ope
rati
on19
90C
ompl
eted
FIL
130
L30
850
7691
Mex
ico
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1989
Com
plet
edSA
D50
0L
3838
040
497
Mex
ico
LA
CFi
nanc
ial S
ecto
r Te
chni
cal A
ssis
tanc
e Pr
ojec
t19
95A
ctiv
eSA
D10
00L
3911
034
161
Mex
ico
LA
CFi
nanc
ial S
ecto
r R
estr
uctu
ring
Adj
ustm
ent
Prog
ram
1995
Act
ive
TA
L23
.6L
3224
082
08V
enez
uela
LA
CFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
1990
Com
plet
edSA
D30
0C
2402
051
67E
gypt
MN
ATe
chni
cal A
ssis
tanc
e Pr
ojec
t fo
r Pr
ivat
izat
ion
and
1992
Act
ive
TA
L9
Ent
erpr
ise
and
Ban
king
Sec
tor
Ref
orm
sL
2604
054
14M
oroc
coM
NA
Indu
stri
al T
rade
Pol
icy
Adj
ustm
ent
Loa
n19
86C
ompl
eted
SAD
200
L33
650
5495
Mor
occo
MN
AFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1991
Act
ive
SAD
235
L39
280
5522
Mor
occo
MN
AFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
1996
Act
ive
SAD
250
L29
620
5718
Tuni
sia
MN
ASt
ruct
ural
Adj
ustm
ent
Loa
n I
1988
Com
plet
edSA
L15
0L
3424
057
42Tu
nisi
aM
NA
Eco
nom
ic a
nd F
inan
cial
Ref
orm
s Su
ppor
t L
oan
1992
Com
plet
edSA
L25
0C
2152
095
28B
angl
ades
hSA
SFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t19
90C
ompl
eted
SAD
175
L38
560
1056
3In
dia
SAS
Fina
ncia
l Sec
tor
Dev
elop
men
t Pr
ojec
t19
95A
ctiv
eFI
L70
0L
3029
010
327
Paki
stan
SAS
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n19
89C
ompl
eted
SAD
150
L38
080
1047
0Pa
kist
anSA
SFi
nanc
ial S
ecto
r D
eepe
ning
and
Int
erm
edia
tion
Pro
ject
1995
Act
ive
FIL
216
72
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNEX 2.2: DETERMINANTS OF PROJECT PERFORMANCE: THE EMPIRICAL EVIDENCE
An econometric analysis was carried out to identify themain factors affecting outcome. For reasons of dataavailability, the analysis was limited to a sample of 18countries. The analysis was applied to a limited numberof specific factors, and to only a few outcomes. In spiteof these limitations, the analysis provides useful firstindications of the relative magnitude of the effects, andpaves the way for further quantitative analysis in thiscomplex area.1
Regression Analysis
HypothesisThe purpose of this regression analysis is to controlsimultaneously for sectoral and macroeconomic initialconditions, and to help identify salient factors affectingoutcome. The explanatory variables selected to reflectinitial conditions are inflation, to indicate the level ofmacrostability, and deposit bank credit to the privatesector (DBPC) as a percentage of GDP, to reflect thelevel of financial sector development. The growth liter-ature provides another important candidate for inclu-sion: international trade, here measured as the ratio ofimports to GDP (IMP/GDP).
The economic literature points to initial real percapita GDP as another explanatory variable. But Gertlerand Rose (1994) found that countries with high percapita GDP at the time of intervention also had rela-tively well-developed financial systems, as measured byDBPC/GDP. In the specifications that follow, therefore,we use DBPC/GDP in place of per capita income.
With respect to initial conditions, explanatory vari-ables will include the inflation rate, IMP/GDP, andDBPC/GDP, all measured in the year of the first policychange supported by the FSL. The remaining explana-tory variables are also institutional and capture featuresspecific to Bank-supported policies, according to theclassifications of the “score card” (Annex Table 3.1).These latter variables (interest rate distortion, non-bankfinancial regulation and prudential regulation) areentered in the regressions one-by-one to identify theirspecific contribution to outcome.
ResultsCaveat. Given the data limitations and the multiple fac-tors that affect financial sector performance in the yearsfollowing policy changes, the results are less robust than
recent ones from the growth literature. While most ofthose studies use samples of 50-100 countries, the sam-ple here is much smaller.2 In the other studies, growthrates and explanatory variables are averaged over longperiods. In this analysis many of the policy reforms sup-ported by the Bank were undertaken in 1993-96. Com-plete sets of IFS data are available through 1995 only,although the benefits of policy changes are expected tooperate with at least a two or three-year lag. The poten-tial sample is accordingly cut substantially. With so fewdegrees of freedom, the results should be interpretedwith caution.
As the previous growth analysis indicated, initialfinancial sector and macroeconomic conditions hadimplications for the change in outcome (Table 2.8). Theregression results essentially restate that countries withrelatively underdeveloped financial systems and lowinflation experienced the largest postinterventionimprovements, as measured by M2/GDP.
At least two major policy variables had significantcoefficients in the regression exercise; interest rate dis-tortions and bank prudential regulations, with the latterhaving the largest effect on outcome.3 The relativelylarge positive coefficient on the prudential regulationsdummy variable and its statistical significance suggestthat development of the financial sector has been sub-stantially linked with World Bank support to thestrengthening of the regulatory framework.
Overall, the regression analysis indicates that finan-cial deepening, measured as M2/GDP, was positivelyassociated with low inflation and an initially underde-veloped financial sector. Controlling for initial macro-economic and financial sector conditions, certainreforms, especially strengthening of prudential regula-tions, were associated with relatively large increases inM2/GDP.4
73
A n n e x e s
OUTCOME CHANGE IN FINANCIAL DEEPENING (M2/GDP)-REGRESSION RESULTS
MACRO + FINANCIAL + MACRO + FINANCIAL + MACRO + FINANCIAL + EXPLANATORY TRADE + INTEREST RATE TRADE + NONBANK TRADE + PRUDENTIALVARIABLE DISTORTION FINANCIAL REGULATION REGULATIONS
Constant 0.92 3.27 -1.19(0.20) (0.64) (0.24)
Group I 9.85 7.87 7.46(2.90) (1.95) (2.36)
Group II 1.37 3.52 3.14(0.38) (0.94) (0.96)
IMP/GDP -0.16 -0.14 -0.8(1.06) (0.83) (0.55)
Removal of interest 5.13Rate distortions (1.67)Non-bank financial regulation 0.40
(0.12)Prudential regulations 5.26
(1.93)Adjusted R-squared 0.293 0.141 0.331Number of observations 18.000 18.000 18.000
Note: t-statistics in parentheses; group dummies are defined below:The intercept term for Group III is reflected by the value of the constant term in the regression.Group I - Inflation < 25 %; Deposit Bank Private Credits < 20 % of GDPGroup II- Inflation >25 %; Deposit Bank Private Credits < 20 % of GDPGroup III- Inflation < 25 %; Deposit Bank Private Credits >20 % of GDP
74
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNEX TABLE 3.1: SAMPLE SCORECARD FOR INSTITUTIONAL DEVELOPMENT
COUNTRY: REGION: INCOME LEVEL:
Bank Program:Approval year:Closing year:Objectives:Instruments:
INITIAL IMPLEMENTATIONINSTITUTIONAL PERFORMANCE CONDITION RECORD OUTCOME SUSTAINABILITY
I. Financial System Infrastructure1. Changes in central bank law2. Changes in banking law3. Changes in laws governing financial
(non-bank) institutions 4. Capital and money market
regulations5. Accounting and audit standards6. Collateral laws and regulations7. Payments system
II. Strengthen Institutions1. Supervision practices of central bank2. Restructuring, privatization, or
strengthening of banks or both
III. Sustainability of Banks1. Overall rating on sustainability
a. Governanceb. Managementc. Technical capabilityd. Resource mobilizatione. WB loan disbursementf. Portfolio qualityg. Financial performanceh. Risk avoidancei. Subsidization
IV. Technical Assistance
Total scoreRating
Borrower PerformanceBank Performance
75
A n n e x e s
ANNE
X TA
BLE
3.2:
LIST
OF
24 F
INAN
CIAL
SEC
TOR
AD
JUST
MEN
T-R
ELAT
ED P
RO
JECT
S (F
SLs)
FO
R I
NST
ITU
TIO
NAL
DEV
ELO
PMEN
T AN
ALYS
IS
LO
AN
/CR
ED
ITN
UM
BE
RR
EG
ION
CO
UN
TR
YT
ITL
ED
AT
E
L35
58L
AC
Arg
enti
naFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
2/16
/93
C21
52SA
SB
angl
ades
hFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t6/
5/90
C18
09L
AC
Bol
ivia
Publ
ic F
inan
cial
Man
agem
ent
5/28
/87
C19
25L
AC
Bol
ivia
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
6/16
/88
C22
98L
AC
Bol
ivia
Stru
ctur
al A
djus
tmen
t L
oan
II9/
17/9
1L
2625
LA
CC
hile
Stru
ctur
al A
djus
tmen
t L
oan
10/2
2/85
L27
67L
AC
Chi
leSe
cond
Str
uctu
ral A
djus
tmen
t L
oan
11/2
0/86
L28
92L
AC
Chi
leT
hird
Str
uctu
ral A
djus
tmen
t L
oan
12/1
5/87
L31
43L
AC
Chi
leFi
nanc
ial M
arke
ts P
roje
ct12
/14/
89C
2303
AFR
Côt
e d’
Ivoi
reFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t10
/1/9
1L
2897
LA
CE
cuad
orFi
nanc
ial S
ecto
r A
djus
tmen
t L
oan
12/2
2/87
C19
11A
FRG
hana
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
5/31
/88
C20
49A
FRK
enya
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
6/27
/89
L30
85L
AC
Mex
ico
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n6/
13/8
9L
3365
MN
AM
oroc
coFi
nanc
ial S
ecto
r D
evel
opm
ent
Proj
ect
6/3/
91L
3029
SAS
Paki
stan
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n3/
28/8
9L
3049
EA
PPh
ilipp
ines
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n5/
4/89
C18
02A
FRSe
nega
lT
hird
Str
uctu
ral A
djus
tmen
t Pr
ogra
m5/
21/8
7C
2077
AFR
Sene
gal
Fina
ncia
l Sec
tor
Adj
ustm
ent
Cre
dit
12/1
8/89
C23
08A
FRTa
nzan
iaFi
nanc
ial S
ecto
r A
djus
tmen
t C
redi
t10
/17/
91L
3424
MN
ATu
nisi
aE
cono
mic
and
Fin
anci
al R
efor
ms
12/1
2/91
L27
14E
CA
Turk
eyFi
rst
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n6/
10/8
6L
2964
EC
ATu
rkey
Seco
nd F
inan
cial
Sec
tor
Adj
ustm
ent
Loa
n6/
21/8
8L
3224
LA
CV
enez
uela
Fina
ncia
l Sec
tor
Adj
ustm
ent
Loa
n6/
12/9
0
76
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNE
X TA
BLE
3.3:
LIST
OF
58 F
INAN
CIAL
IN
TER
MED
IAR
Y LO
ANS
(FIL
s) F
OR
IN
STIT
UTI
ON
AL D
EVEL
OPM
ENT
ANAL
YSIS
LO
AN
/CR
ED
ITN
UM
BE
RR
EG
ION
CO
UN
TR
YT
ITL
ED
AT
E
L27
93L
AC
Arg
enti
naSm
all a
nd M
ediu
m-s
cale
Ind
ustr
ial C
redi
t11
-Jan
-95
L20
63L
AC
Arg
enti
naSe
cond
Ind
ustr
ial C
redi
t Pr
ojec
t01
-Dec
-81
C11
17SA
SB
angl
ades
hSe
cond
Shi
lpa
Ban
k Pr
ojec
t03
-Dec
-90
L24
34/C
1491
EA
PC
hina
Chi
na I
nves
tmen
t B
ank
2 Pr
ojec
t03
-Oct
-91
L22
26/C
1313
EA
PC
hina
Chi
na I
nves
tmen
t B
ank
1 Pr
ojec
t03
-Oct
-91
L26
59/C
1663
EA
PC
hina
Chi
na I
nves
tmen
t B
ank
3 Pr
ojec
t03
-Oct
-91
C17
63/L
2783
EA
PC
hina
Indu
stri
al C
redi
t 4
Proj
ect
03-M
ar-8
7 L
3075
EA
PC
hina
Indu
stri
al C
redi
t 5
Proj
ect
30-M
ay-8
9
L26
73L
AC
Ecu
ador
Smal
l-sc
ale
Ent
erpr
ise
Cre
dit
III
11-M
ar-9
4 L
2672
LA
CE
cuad
orIn
dust
rial
Fin
ance
Pro
ject
14-M
ar-9
2 L
2096
LA
CE
cuad
orFi
fth
Dev
elop
men
t B
anki
ng P
roje
ct17
-Dec
-90
L22
21L
AC
Ecu
ador
Smal
l-sc
ale
Ent
erpr
ise
Cre
dit
I29
-Jun
-89
L18
79L
AC
Ecu
ador
Smal
l-sc
ale
Ent
erpr
ise
Cre
dit
II29
-Jun
-89
L20
74M
NA
Egy
ptFi
fth
Dev
elop
men
t In
dust
rial
Ban
k03
-Oct
-91
L18
42M
NA
Egy
ptM
isr
Iran
Dev
elop
men
t B
ank
Proj
ect
04-M
ay-9
0 L
1804
MN
AE
gypt
Dev
elop
men
t Fi
nanc
e C
ompa
ny27
-Jun
-88
L20
51SA
SIn
dia
Thi
rtee
nth
and
Four
teen
th I
ndus
tria
l Cre
dit
and
Inve
stm
ent
Cor
pora
tion
Pro
ject
s26
-Jun
-90
L18
43SA
SIn
dia
Thi
rtee
nth
and
Four
teen
th I
ndus
tria
l Cre
dit
and
Inve
stm
ent
Cor
pora
tion
Pro
ject
s26
-Jun
-90
L28
00E
AP
Indo
nesi
aB
RI/
KU
PED
ES
Smal
l Cre
dit
Proj
ect
18-A
pr-9
4 L
2277
EA
PIn
done
sia
Fift
h B
ank
Pem
bang
unan
Ind
ones
ia (
BA
PIN
DO
) Pr
ojec
t16
-Aug
-91
L24
30E
AP
Indo
nesi
aT
hird
Sm
all-
scal
e E
nter
pris
e D
evel
opm
ent
Proj
ect
16-A
pr-9
0 L
2011
EA
PIn
done
sia
Smal
l Ent
erpr
ise
Dev
elop
men
t Pr
ojec
t12
-Jun
-89
L18
17A
FRK
enya
Four
th I
ndus
tria
l Dev
elop
men
t Pr
ojec
t31
-Dec
-92
C17
38A
FRK
enya
Seco
nd K
enya
Ind
ustr
ial E
stat
es (
KIE
) Sm
all-
scal
e In
dust
ry C
redi
t Pr
ojec
t16
-Dec
-92
L23
09E
AP
Kor
eaSe
cond
Ind
ustr
ial F
inan
ce P
roje
ct30
-Jun
-93
L21
44E
AP
Kor
eaSe
cond
Cit
izen
s N
atio
nal B
ank
and
SME
Pro
ject
25-J
un-9
3 L
1933
EA
PK
orea
Dev
elop
men
t B
ank
Proj
ect
26-J
un-9
0 L
2004
EA
PK
orea
Four
th S
mal
l and
Med
ium
Ind
ustr
y B
ank
Proj
ect
04-M
ay-9
0 L
1829
EA
PK
orea
Cit
izen
s N
atio
nal B
ank
Proj
ect
05-D
ec-8
6 L
2515
EA
PK
orea
Smal
l and
Med
ium
Ind
ustr
y Pr
ojec
t16
-Apr
-85
L25
71E
AP
Kor
eaSe
cond
Ind
ustr
ial F
inan
ce P
roje
ct06
-Jun
-85
C32
21A
FRM
alaw
iFi
nanc
ial S
ecto
r an
d E
nter
pris
e D
evel
opm
ent
Proj
ect
25-F
eb-9
1
L27
70E
AP
Mal
aysi
aD
evel
opm
ent
Fina
nce
Proj
ect
27-S
ep-9
3 L
2471
EA
PM
alay
sia
Smal
l-sc
ale
Ent
erpr
ise
Proj
ect
06-M
ay-9
3
77
A n n e x e s
ANNE
X TA
BLE
3.3:
(CO
NTIN
UED)
L23
25L
AC
Mex
ico
Seco
nd S
mal
l and
Med
ium
Ent
erpr
ise
Ind
ustr
y D
evel
opm
ent
Proj
ect
30-J
un-9
4 L
1881
LA
CM
exic
oT
hird
Sm
all a
nd M
ediu
m E
nter
pris
e I
ndus
try
Dev
elop
men
t Pr
ojec
t30
-Jun
-94
L27
47L
AC
Mex
ico
Indu
stri
al T
echn
olog
y D
evel
opm
ent
Proj
ect
30-J
un-9
4 L
2546
LA
CM
exic
oSe
cond
Sm
all a
nd M
ediu
m M
inin
g D
evel
opm
ent
Proj
ect
30-J
un-9
4 L
2142
LA
CM
exic
oC
apit
al G
oods
Ind
ustr
ies
Dev
elop
men
t Pr
ojec
t25
-Sep
-90
L18
20L
AC
Mex
ico
Smal
l and
Med
ium
Min
ing
Dev
elop
men
t Pr
ojec
t17
-May
-90
L24
87M
NA
Mor
occo
Ele
ctri
cal a
nd M
echa
nica
l Ind
ustr
ies
Proj
ect
29-J
un-9
4 L
2038
MN
AM
oroc
coSe
cond
Sm
all-
scal
e In
dust
ry P
roje
ct23
-Apr
-90
L20
37M
NA
Mor
occo
Nin
th B
anqu
e N
atio
nale
Pou
r L
e D
evel
oppe
men
t E
cono
miq
ue (
BN
DE
) Pr
ojec
t26
-Oct
-89
L33
65M
NA
Mor
occo
Fina
ncia
l Sec
tor
Dev
elop
men
t Pr
ojec
t03
-Jun
-91
L26
48/C
1646
SAS
Paki
stan
Seco
nd I
ndus
tria
l Inv
estm
ent
Proj
ect
10-J
un-9
4 L
2380
SAS
Paki
stan
Indu
stri
al I
nves
tmen
t C
redi
t16
-Jun
-93
C11
86SA
SPa
kist
anSe
cond
Ind
ustr
ial D
evel
opm
ent
Ban
k Pr
ojec
t17
-May
-89
C10
19SA
SPa
kist
anE
leve
nth
Indu
stri
al C
redi
t an
d In
vest
men
t C
orpo
rati
on P
roje
ct01
-Dec
-86
L30
38E
AP
Phili
ppin
esFo
urth
Sm
all a
nd M
ediu
m I
ndus
trie
s D
evel
opm
ent
Proj
ect
15-M
ar-9
5 L
3123
EA
PPh
ilipp
ines
Indu
stri
al I
nves
tmen
t C
redi
t Pr
ojec
t30
-Jun
-93
L19
84E
AP
Phili
ppin
esD
evel
opm
ent
Cor
pora
tion
s an
d Sm
all a
nd M
ediu
m I
ndus
trie
s I
21-J
un-9
0 L
2169
EA
PPh
ilipp
ines
Dev
elop
men
t C
orpo
rati
ons
and
Smal
l and
Med
ium
Ind
ustr
ies
II21
-Jun
-90
C11
36A
FRSe
nega
lIn
vest
men
t Pr
omot
ion
(Thi
rd S
OFI
SED
IT)
Proj
ect
04-F
eb-9
1
L25
22M
NA
Tuni
sia
Exp
ort
Indu
stri
es P
roje
ct06
-Jan
-94
L26
47E
CA
Turk
eySm
all a
nd M
ediu
m-s
cale
Ind
ustr
y Pr
ojec
t01
-Dec
-93
L19
98E
CA
Turk
eyIn
dust
rial
Cre
dit
Proj
ect
04-M
ay-9
0 L
2093
EC
ATu
rkey
Indu
stri
al D
evel
opm
ent
Cre
dit
I29
-Jun
-89
L19
52E
CA
Turk
eyIn
dust
rial
Dev
elop
men
t C
redi
t II
29-J
un-8
9
78
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNEX TABLE 4.1: MACRO-FRAGILITY INDICATORS
INDICATORS LINKS TO FINANCIAL FRAGILITY
1. Inflation rate (annual percentchange in CPI)
2. Fiscal balance (percent ofGDP)
3. Current account balance(percent of GDP)
4. Exchange rate position
5. Real interest rates: a. Depositsb. Loans
6. Gross domestic savings(GDS) (percent of GDP)a. privateb. public
7. Real GDP growth a. levelb. volatility
8. Brady bond spreads (spreadover US T-bonds)
9. Stock market prices
A high and variable inflation environment generally adds to the fragility of the financialsystem by causing disintermediation and preventing the emergence of long-term debt con-tracts. A reasonable degree of price stability is essential for a strong and stable financialsystem. The fragility threshold level of inflation will probably vary from country-to-coun-try and can only be determined by examining country experiences. However, casual obser-vation suggests that inflation rates in excess of 20 percent are damaging to financial stabil-ity in most countries.
Large and persistent fiscal deficits create strong incentives for governments to repress thefinancial system, and often lead to excessively high real interest rates and high levels ofnonperforming loans, thus adding to financial fragility. Fiscal deficits are particularly dam-aging when private saving rates are low.
Large current account deficits tend to damage the financial system, because they often leadto reserve losses, capital outflows, and excessive domestic and foreign borrowings. Recentevidence suggests that CADs in excess of 8 percent of GDP tend to be highly destabilizingfor the economy and the financial systems.
Over-valued real exchange rates distort price incentives, put upward pressure on real inter-est rates, and lead to loan defaults. Dual exchange rate regimes. Flexible versus pegged orcurrency board regimes.
Excessively high real interest rates over extended periods greatly increase the fragility ofthe financial system and often lead to banking crises. High real deposit rates (in excess of10 percent) may reflect greater and perhaps imprudent risk-taking by bankers, while highlending rates (in excess of 30 percent) are often caused by “distress borrowing” and mayattract destabilizing portfolio capital inflows. On the other hand, negative or very low realinterest rates lead to disintermediation and credit mis-allocation, thus weakening the finan-cial system. In addition, very high spreads may indicate serious and damaging incentiveproblems on the part of bankers (Sheng 1996; Galbis 1993).
Low domestic saving rates (especially private) are found closely correlated with fragilityand weakness of financial systems. East Asian economies with very high saving rates gener-ally have stable financial systems. By contrast, LAC countries with very low savings rates(less than 15 percent) have fragile financial systems. The exception is Chile, with a currentsaving rate around 24 percent (up from 5.4 percent in 1980) and a sound and stable finan-cial system. The Mexican private savings rate, which dropped from 18.8 percent to a lowof 9.1 percent in 1994, is widely believed to have aggravated the 1994–95 financial crisis.A GDS rate of 20 percent can perhaps be taken as a reasonable fragility threshold value forthis indicator. (The Economist, December 9, 1995).
A stable, growing economy tends to promote sound and strong financial systems, whiledeclines and volatility in economic growth are often associated with increased fragility ofthe banking sector and proneness to crises (Sheng 1996).
The fragility threshold for most developing countries: sustained 4–5 percent growth.
The secondary market yields on Brady bonds, commonly measured by spreads over compa-rable U.S. T-bonds, are a reliable barometer of investor confidence. Sustained high spreads(in excess of 300 basis points) probably indicate a seriously fragile financial situation. Forexample, Mexican Brady bonds had spreads well in excess of 300 basis points for monthspreceding the 1994–95 crisis. So did those of Argentina and Venezuela.
79
A n n e x e s
ANNEX TABLE 4.1: (CONTINUED)
INDICATORS LINKS TO FINANCIAL FRAGILITY
10. Capital flowsa. FDIb. Portfolio investment
11. External debta. Levelb. Maturity structure
12. Monetization/FinancialDepth (M2 as percent ofGDP)
13. Domestic credit growth(in real terms)a. Private sectorb. Public sector
14. Terms-of-trade exposure(trend and variance ofTerms-of-Trade)
Large and unexpected declines in stock market prices are associated with increased finan-cial fragility. Similarly, rapid growth in the stock market, often fueled by portfolio capitalor speculative activity financed by bank credit (stock market bubbles), produces a highlyfragile financial system.
If not managed properly, large and volatile capital flows (especially portfolio investmentsor “hot capital”) can greatly add to financial fragility, making the system vulnerable toshocks and crises. Large capital inflows tend to overvalue the exchange rate, depressexports, and worsen capital account deficits. Likewise, large and unexpected capital out-flows (capital flight, portfolio investments) cause interest rates to rise, leading to increasesin loan defaults and banking distress.
A steady increase in long-term FDI is usually a stabilizing factor for the financial system. Sustained annual portfolio capital inflows exceeding 10 percent of GDP are probably areasonable fragility threshold for most developing countries.
Excessive external debt, especially when concentrated at the short end of the maturitystructure, tends to increase financial fragility.
The fragility threshold will vary with the size and structure of a country’s economy.
A relatively deep financial system (higher M2 ) tends to be more resilient and better able toabsorb shocks. Financial fragility threshold: 30 percent.
Excessive growth in real credit is widely found to increase financial fragility and lead tobanking crises (Sundararajan and Balino 1991; Gavin and Hausmann, 1995; Kaminskyand Reinhart 1996). Empirical analysis by Caprio and Klingebiel (1996) finds close associ-ation between real credit growth rates in excess of twice real GDP growth rates, and bank-ing crises in many countries. The reason is that rapid credit growth in the banking systemoften leads to a deterioration of loan quality by hampering effective credit assessment bybankers and supervision, in part because of non-receipt of timely information. Rapid creditgrowth may also reflect a decline in loan quality as distressed banks resort to extension orrollover of bad loans to cover the extent of the damage of non-performing loans (ever-greening). The latter serves as an example of the problems in the real sector spilling intoand weakening the financial sector.
Bank lending over-concentrated (more than 30 percent ) in sectors subject to frequent andsubstantial changes in terms of trade (TOT Exposure) can greatly increase banking fragilityand lead to banking crises. (Sheng 1996). TOT changes may arise from exogenous pricedeclines or as a result of trade or exchange rate policy reforms, often as part of SAL pro-grams. Examples include the Chilean financial crisis (copper price declines, 1974–87), theMalaysian banking crisis (tin, rubber, and oil price declines, 1983–86), the farm financialcrisis in the US Mid-West (farm exports decline brought on by the over-valued U.S. dollar),and Texas bank failures (energy price declines in the mid-1980s). Moreover, volatility interms of trade appears to be particularly detrimental to the banking sector in the presenceof shallow financial markets (Caprio 1996).
Note: Indicators 1–4 define the macroeconomic environment in which a banking or a financial system operates. Establishing astrong and stable macroeconomic environment is essential to reducing financial fragility and promoting a strong financial/bank-ing system.
80
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNEX TABLE 4.2: MICROECONOMIC/INSTITUTIONAL FRAGILITY INDICATORS
INDICATORS LINKS TO FINANCIAL FRAGILITY
1. Banking market structure
2. Capital adequacy (Banks’capital/total assets)
3. Deposit insurance schemes
4. Asset quality (NPLs/totalloans)
5. Loan-to-deposit ratio
6. Maturity profile of thedeposit base in the bankingsystem
7. Currency/deposit ratio
8. Loan portfolio concentration
9. Enterprise/firm indebtedness(debt-to-asset ratio)
10. Connected lending (percent-age of banks owned byindustrial or financialconglomerates)
11. Banks’ exposure to realestate (percent of total loansto real estate)
12. Market risk exposure (non-interest income/total income)a. Interest rate risksb. Foreign exchange risksc. Other derivatives/
guarantees
Competitive market structure probably provides the optimal mix of low fragility and highefficiency; oligopolistic or cartelized market structures can be less fragile but are also lessefficient and tend to undercut the financial liberalization and development process. Forexample, even though Mexico shared the same debt crisis in the early 1980s as other heav-ily-indebted countries, it avoided a banking crisis because of nationalization, although at ahigh cost in the form of inefficient credit allocation by state-owned banks.
Inadequate capital to cushion loan losses is a key determinant of fragility in the bankingsector. The minimum risk-weighted capital-asset ratio of 8 percent required by the BasleCommittee is probably too low for most developing countries. Fragility threshold: 8percent.
A well-designed and fully-funded deposit insurance sector can promote financial stability,especially when accompanied by an effective supervision system. On the other hand,deposit insurance without effective supervision invariably leads to a highly fragile bankingsystem because of moral hazard problems.
A key determinant of financial fragility. As a rule of thumb, non-performing loans inexcess of about 10 percent greatly weaken the banking system and often lead to bankingcrises. (Sheng 1996; Caprio 1996).
Higher values (in excess of 100 percent) suggest a banking system that may be over-reach-ing its resources; may reflect excessive reliance on interbank money markets or foreign bor-rowings, both leading to increased fragility. Very low values (less than 30 percent) indicatea banking system with a weak franchise, implying greater credit risks and fragility (Corri-gan 1991; Rojas-Suarez and Weisbrod 1995).
A very short maturity structure of bank deposits reflect a general loss of confidence andhence increased fragility in the banking system. During the period leading up to the Argen-tine bank crisis in 1989, most of the deposit base had only seven days’ maturity.
An excessively high currency/deposit ratio may often reflect a lack of public confidence inthe banking system, leading to dis-intermediation and increased financial fragility.
Sectoral or geographical concentration of loan portfolios permits insufficient diversificationof risks, leaving the banking system vulnerable to shocks and crises.
The presence of a large number of highly-indebted (especially with a large proportion ofshort-term debt) or insolvent firms and enterprises often leads to “distress borrowing” anda fragile banking system. Fragility threshold: debt-to-asset ratios exceeding 0.5 for morethan one-third of banks.
Excessive lending to related firms under an interlocking system of bank and firm owner-ship or a conglomerate structure often leads to a highly fragile financial system. Fragility threshold: 25 percent (Sheng 1996).
Over-exposure to real estate lending tends to be a significant source of bank fragilitybecause of property booms and speculative bubbles, often leading to banking and financialcrises. Fragility threshold: 30 percent (Sheng 1996).
A high share of non-interest income can be taken as indicative of weakness in core businessand an over-reliance on risky, often off-balance-sheet, activities involving market risks.
Fragility threshold: 30 percent.
81
A n n e x e s
ANNEX TABLE 4.2: (CONTINUED)
INDICATORS LINKS TO FINANCIAL FRAGILITY
13. Previous bank restructuring
14. Previous enterpriserestructuring
15. DFI solvency/reform
16. Central Bank:a. Capital adequacyb. Foreign exchange reservesc. Off-balance-sheet
operations
17. Domestic accounting andauditing standards
18. Prudential bank regulationand supervision systems andcapacity
The presence of weak banks that have not been properly restructured poses a significantthreat to the stability of the financial system.
Public or private enterprises carrying a significant portion of nonperforming bank loansmust be restructured (with banks) to minimize financial fragility. An improved legal andregulatory framework is essential to facilitate corporate workouts and debt restructuring,and thus to minimize their negative impact on financial stability.
Highly-indebted or insolvent development banks and other Development Finance Institu-tions (DFIs) often pose a serious threat to the stability of the financial system.
Adequate levels of capital and foreign exchange reserves are essential for a central bank toeffectively discharge its various functions in support of a strong and stable financial sys-tem. Using the central bank to finance government budget deficits and undertake other fis-cal activities undermines its effectiveness and leads to increased fragility in the financialsystem. Off-balance-sheet operations by a central bank can greatly increase risks and causefinancial instability.
Poor-quality accounting and auditing systems greatly exacerbate the problem of asymmet-ric information, thus undermining public confidence in the integrity of the financial systemand greatly adding to its fragility.
Effective prudential regulation and supervision is a key determinant of financial stability. Intheir absence, financial systems remain dangerously fragile and prone to financial crises, asshown, for example, by the Chilean banking crisis of 1981.
ANNEX TABLE 4.3: CENTRAL BANK FRAGILITY INDICATORS
INDICATORS MEASUREMENT
1. Central bank profitability
2. Short-term interest rates
3. Foreign liability coverageratio
4. Solvency
5. Quality of governance
Deficit/surplus from the income statement.
91-day Treasury Bill rate.
Ratio of foreign loans payable to international reserves; figures taken from central bankbalance sheets.
Total liablities/total assets; figures taken from central bank balance sheets.
Includes central bank independence, effectiveness of management, and adequacy of its reg-ulatory powers.
82
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
ANNEX 5: FINANCIAL SECTOR REFORM: A REVIEW OF WORLD BANK ASSISTANCE/MANAGEMENT RESPONSE
Management Response
Agreed, subject to the advice of the Board that theOP should be limited to financial intermediary lend-ing, rather than to the broader financial sector opera-tions covered by the OD. The financial intermediarylending policy content of OD 8.30 (with some refine-ments and clarifications) is being replaced by OP8.30, Financial Intermediary Lending. The draft pol-icy, which will be submitted shortly to the ExecutiveDirectors for approval, includes a general frameworkfor financial intermediary lending. Such lending: (i) isbased in a CAS that discusses the critical developmen-tal importance of the financial sector and the justifica-tion and sequencing of Bank interventions; (ii) is con-sidered in a satisfactory macroeconomic and policyenvironment; (iii) supports appropriate financialinfrastructure, including prudential regulations andsupervision; (iv) establishes minimum eligibility crite-ria for participating financial institutions (FIs); and (v)requires appraisal of each operation to cover, interalia, the strategic sectoral significance of the opera-tion/project, economic justification, and consistencywith financial sector development objectives.
Efforts are underway to step up financial sectordiagnostic work and ensure that CASs systematicallydraw on such diagnoses and cover financial sectorissues, including competition, where they are identi-fied as important.
Agreed. The basic principle that relevant ESW shouldlay a foundation for financial sector operations iswell recognized. In January 1998, the Boardapproved a program to reinforce the Bank’s financialsector work, and a special budget for this program.A detailed strategy to implement this program isunder preparation. It includes a proposal to conductfinancial sector assessments initially for all countriesvulnerable to crises and sequentially to expandthem to other countries. These assessments will
Recommendations
Financial Sector Strategy1. OD 8.30 and its associated OP define a frame-work for financial sector operations. CASs docu-ments should ensure that all major relevant compo-nents of the directive, and especially competition,are dealt with in a systematic way.
ESW and Project Preparation2. The value of in-depth policy-oriented economicand institutional analysis of the financial sector isclear. Much greater emphasis should be given toESW prior to project initiation to ensure that thecritical issues are clearly identified from the outset.
83
A n n e x e s
review, inter alia, the status and performance of thereal/enterprise sector in the context of their impact on the performance of financial institutions andvice versa. In addition, a common diagnostic frame-work and methodology is being developed whichwill help improve the quality of ESW.
Agreed. As noted in para. 2, the Board authorized aspecial budget to reinforce the Bank’s financial sectorwork. An important component of the program willbe the development of a monitoring and surveillancemechanism based on key performance indicators, atboth the country and project/operation levels.
Agreed. The Bank approach under the reinforcementprogram should lead to more sustained attention tofinancial sector development, particularly in vulnera-ble countries. APLs allow for flexibility in the designand implementation of diversified Bank interven-tions in the financial sector. A few such projects arealready under preparation. However, additionalwork is needed to develop new approaches (e.g.,more effective ways to develop capital markets anddeliver technical assistance) to cater to the needs ofthe financial sector of client countries.
Agreed. In addition to financial sector assessmentsand detailed diagnoses under the reinforcement pro-gram, we will have special TA operations, financedby trust funds and Bank loans. This TA and associ-ated funding will be provided if the country is will-ing to commit to the agreed programs of reform.Various triggers will be used to gauge this commit-ment. Also, additional resources are being commit-ted to strengthen supervision, and we are developing new products to better integrate TA with lending.
EDI’s current work program—its own as wellas the programs it has undertaken in partnership
Performance Indicators.3. The Bank should allocate more resources tomonitor and evaluate countries’ financial sectorprograms, with performance indicators focusing onstructural changes and system resiliency.
Lending Instruments and Program Monitoring4. Financial sector development is a process not anevent. Hence, more attention should be given tosustaining this process. In particular, the new lend-ing instrument, adaptable program loans (APLs),which disburse according to the pace of institu-tional progress, should become important in sup-porting financial reforms.
5. Technical assistance is of the utmost importanceto financial sector development. However, TALshave not performed effectively. As a result, the Bankshould commit itself to providing technical assis-tance only under the following conditions: (1) longterm commitment is clearly established; (2) the Bankinvests sufficient resources to design and supervisethe assistance effectively; (3) the government is com-mitted to specific actions that will ensure the effec-tiveness of assistance; and (4) EDI should beencouraged to further develop its banking andfinance training program in the areas of sound
84
F i n a n c i a l S e c t o r R e f o r m : A R e v i e w o f Wo r l d B a n k A s s i s t a n c e
with the FPSI Network and other operational units-already encompasses current financial sector issues(including sound bank management, systemic bankcrisis management, capital market development,microfinance, etc.). EDI programs continue toevolve to respond to changing client needs.
Agreed. A framework for “Bank-Fund Collabora-tion on Strengthening Financial Sectors.” preparedjointly by the two institutions, was discussed andendorsed in September 1997 by the Boards of bothinstitutions. Among other things, this frameworkprovides for an exchange of information. Sincethen, Bank and Fund staff have been workingclosely together on crisis countries. At present, IFCand the Bank routinely consult each other withrespect to their financial sector operations; this col-laboration takes place at the project as well as atthe sector strategy formulation level. In many coun-tries the Bank and IFC are preparing jointly CASsthat include financial sector concerns. The princi-ples of collaboration between the Bank and IFC arealso included in the draft OP 8.30.
bank management, banking policy, regulatoryissues, and bank failures.
IMF6. The Bank’s interaction with the IMF and the IFCis of increased importance. The Bank’s countrystrategies should include interaction and agreementwith the IMF on full and timely exchange of infor-mation and early coordination. Similarly, while theBank and the IFC exchange preliminary project doc-uments, Bank staff should give IFC staff even earliernotification about impending projects, so that theyhave more opportunity to influence their design.
85
On May 13, 1998 the committee on Development Effec-tiveness (CODE) reviewed a report Financial SectorReform: A review of World Bank Assistance (SecM98-221) prepared by the Operations Evaluation Depart-ment (OED), together with a draft Managementresponse (CODE98-26). The report analyzes the Bank’srole in helping client countries to implement financialsector reforms prior to the recent East Asia crisis. Itfocuses on the country as the unit of analysis, ratherthan on individual loans, and on performance indica-tions in the financial and real sectors to evaluate out-comes. Thus, it examines results rather than inputs.Even without fully incorporating the lessons still to bedrawn from the East Asia crisis, the study finds satisfac-tory outcome in only 12 (52 percent) of the countriesthat were examined. In promoting institutional develop-ment, adjustment loans were found to be more success-ful and more sustainable than financial intermediationloans. The report suggests that the East Asia crisis pro-vides new evidence of the Bank’s tendency to underratethe seriousness of financial sector dysfunctions.
The report finds that the Bank’s new financial sectorstrategy is firmly grounded in the Bank’s experience inthe sector and is based on the lessons learned from thisexperience, although the learning process was slow andserious concerns remain. The report suggests that theinternal guidelines on financial sector operations(OD8.30) provide a valid framework for preparingoperations in support of financial sector reforms. How-ever, the Bank should go beyond these guidelines andincorporate best practice on substantive issues and Bankprocess, after sufficient time has elapsed to evaluate andlearn from the recent developments in East Asia.
The committee welcomed the opportunity to focus onthis timely, relevant, and comprehensive report on thecritical issue of financial sector reform, and commendedboth the OED report and the forward-looking Manage-ment response. Committee members appreciated in par-ticular the innovate methodology used in the OEDreport. They noted that the report will be published witha summary translated into French and Spanish. OEDwill work closely with the Sector Board to disseminatethe evaluation findings.
Committee members noted the report’s findings thatresults were better in those countries where the Bankcarried out prior economic and sector work (ESW). Thisconfirmed prior evaluation findings about ESW’s posi-tive contribution to the design of reform programs. Itwas also suggested that short, focused and diagnosticsector notes would be useful. The importance of ade-quate and timely technical assistance and supervisionwas also highlighted. Members also commented on theimportance of ensuring that CAS’s properly addressfinancial sector issues and their links to macro issues andreal sectors: realistically assessing ex ante long-termcommitment to reform; monitoring and evaluation ofprogress; using the full range of lending and non-lendinginstruments, including adaptable program loans (APLs);involving the local institutions; having in place a trans-parent regulatory mechanism to ensure success of priva-tization; improving synchronization and sequencing ofreforms; and ensuring a long-term, sustained approachto financial reform and viewing it as a long-termprocess, rather than an event.
The Committee was generally supportive and wasencouraged by the Management’s constructive and for-ward-looking response outlining the Bank’s efforts tostrengthen its ability to help clients address financialconcerns. However, members raised a number of issues,including the need to do more analysis and draw lessonsfrom the experience in the sector including in Mexicoand East Asia, and assess progress made since the estab-lishment in January of the Special Financial OperationsUnit and allocation of incremental authority for workon the financial sector program. Committee memberswere particularly interested in progress on he recruit-ment, internal organization, cooperation/coordinationwith the IMF and IFC, and in whether the Bank has acritical mass including staff skills, knowledge manage-ment and instruments to move forward in the very chal-lenging area.
Next Steps: The Committee looks forward to discus-sions of these important issues in future Board AndCommittee meetings, and to the forthcoming FinancialSector Strategy Paper.
A n n e x e s
ANNEX 6: REPORT FROM CODE/COMMITTEE ON DEVELOPMENT EFFECTIVENESS
OED Study Series
Evaluation and Development: The Institutional Dimension (1998)
1997 Annual Review of Development Effectiveness (1998)
India: The Dairy Revolution (1998)
The World Bank’s Experience with Post-Conflict Reconstruction (1998)
Financial Sector Reform: A Review of World Bank Assistance (1998)
Rebuilding the Mozambique Economy: Assessment of a Development Partnership (1998)
Agricultural Extension and Research: Achievements and Problems in National Systems (1997)
Fiscal Management in Adjustment Lending (1997)
Reforming Agriculture: The World Bank Goes to Market (1997)
Paddy Irrigation and Water Management in Southeast Asia (1997)
1995 Evaluation Results (1997)
Zambia Country Assistance Review: Turning an Economy Around (1997)
The Aga Khan Rural Support Program: A Third Evaluation (1996)
Lending for Electric Power in Sub-Saharan Africa (1996)
Industrial Restructuring: World Bank Experience, Future Challenges (1996)
Social Dimensions of Adjustment: World Bank Experience, 1980–93 (1996)
1994 Evaluation Results (1996)
Ghana Country Assistance Review: A Study in Development Effectiveness (1995)
Evaluation and Development: Proceedings of the 1994 World Bank Conference (1995)
Developing Industrial Technology: Lessons for Policy and Practice (1995)
The World Bank and Irrigation (1995)
1993 Evaluation Results (1995)
Structural and Sectoral Adjustment: World Bank Experience, 1980–92 (1995)
Gender Issues in World Bank Lending (1995)
The World Bank’s Role in Human Resource Development in Sub-Saharan Africa: Education, Training, and Technical Assistance (1994)
1992 Evaluation Results (1994)
New Lessons from Old Projects: The Workings of Rural Development in Northeast Brazil (1993; contains summaries in French, Portuguese and Spanish)
World Bank Approaches to the Environment in Brazil (1993; contains summaries in French, Portuguese, and Spanish)
Trade Policy Reforms under Adjustment Programs (1992)
World Bank Support for Industrialization in Korea, India, and Indonesia (1992)
Population and the World Bank: Implications from Eight Case Studies (1992)
The Aga Khan Rural Support Program in Pakistan: Second Interim Evaluation (1990)
The Operations Evaluation Department (OED), an inde-pendent evaluation unit reporting to the World Bank’sexecutive directors, rates the development impact andperformance of all the Bank’s completed lending opera-tions. Results and recommendations are reported to theexecutive directors and fed back into the design andimplementation of new policies and projects. In additionto the individual operations and country assistance pro-grams, OED evaluates the Bank’s policies and processes.
Operations evaluation studies, World Bank discus-sion papers, and all other documents are available fromthe World Bank InfoShop.
Summaries of studies and the full text of the Précisand Lessons & Practices can be read on the Internet athttp://www.worldbank.org/html/oed/index.htm
How To Order OED PublicationsDocuments listed with a stock number and price codemay be obtained through the World Bank’s mail orderservice or from its InfoShop in downtown Washington,DC. For information on all other documents contact theWorld Bank InfoShop.
Ordering World Bank PublicationsCustomers in the United States and in territories notserved by any of the Bank’s publication distributors maysend publication orders to:
The World BankP.O. Box 960Herndon, VA 20172-0960Fax: (703) 661-1501Telephone: (703) 661-1580The address for the World Bank publication database onthe Internet is: http://www.worldbank.org (click onpublications).
Internet: http://www.worldbank.org//. From the WorldBank homepage, select publications. E-mail: [email protected] number: (202) 522-1500Telephone number: (202) 458-5454
The World Bank InfoShop serves walk-in customersonly. The InfoShop is located at:
701 18th Street, NWWashington, DC 20433, USA
All other customers must place their orders throughtheir local distributors.
Ordering via E-MailIf you have an established account with the World Bank,you may transmit your order via electronic mail on theInternet to: [email protected]. Please include youraccount number, billing and shipping addresses, the titleand order number, quantity, and unit price for each item.
The OED Working Paper Series is produced byOEDPK (Partnerships and Knowledge). Copies orinquiries about these or other publications in our seriescan be obtained by calling 202-473-5365 or email:[email protected].
OPERATIONS EVALUATION DEPARTMENT PUBLICATIONS