IDA’s Partnership for Poverty Reduction · Pooling Donor Funds in Support of Health Sector Reform...

IDA’sPartnershipfor PovertyReduction

An IndependentEvaluation of Fiscal

Years 1994–2000

Catherine Gwin

2002The World Bank

Washington, D.C.

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

http://www.worldbank.org/oed

Copyright © 2002

The International Bank for Reconstruction and Development/THE WORLD BANK

1818 H Street, N.W.

Washington, D.C. 20433, USA

All rights reserved

Manufactured in the United States of America

First edition January 2002

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. The World Bank encourages dissemination of its work and will normally

grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use

of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is

paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers MA 01923, U.S.A., telephone 978-

750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to

reprint individual articles or chapters, please fax your request with information to the Republication Department,

Copyright Clearance Center, fax 978-750-4470.

All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank, at the address

above, or faxed to 202-522-2422.

Cover photo: World Bank photo library.

ISBN 0-8213-5052-8

Library of Congress Cataloging-in-Publication data has been applied for.

Printed on Recycled Paper

i i i

Contents

vii Acknowledgments

ix Foreword, Prefacio, Avant-Propos

xiii Executive Summary, Resumen, Résumé Analytique

xxi Abbreviations and Acronyms

xxiii Overview

1 Part I. The IDA Review and Its Context

3 1. Introduction3 IDA10–12 Replenishment Undertakings4 IDA’s Role in a Changing Development Environment10 Reviewing IDA’s Performance

13 Part II. The Program Dimension

15 2. Sharpening the Focus on Poverty Reduction15 Strengthening the Poverty Focus of Country Assistance Strategies16 Building in More Poverty Analysis and Monitoring18 Increasing Poverty-Targeted Interventions and Social Sector

Lending

23 3. Promoting Broad-Based Growth23 IDA’s Evolving Adjustment Operations26 Lags in Agriculture and Rural Poverty Reduction27 Strengthening Private Sector Development30 Looking Ahead

31 4. Integrating Gender, Environment, and Governance31 Addressing Gender35 Enhancing Environmental Sustainability40 Improving Governance43 Lessons

45 Part III. The Process Dimension

47 5. Improving Program and Country Selectivity47 Enhancing the Role of the Country Assistance Strategy50 Strengthening the System of Performance-Based Allocations

55 6. Building More Effective Partnerships55 Moving Beyond Project Participation57 Improving Aid Coordination59 Directions for the Future

61 Part IV. Getting to Results

63 7. Corporate Issues64 Accountability for Policy Compliance65 Instrument Choice67 Monitoring for Results68 Alignment Between Resources and Strategic Program Priorities

71 8. Implications for the Future71 Be More Selective72 Focus on Governance and Capacity Building72 Commit to Development Partnerships73 Clarify Objectives and Strengthen Accountabilities73 Better Align Resources with Priorities74 Develop a Long-Term Vision, Focused on Results74 Engage Developing Countries in Setting and Reviewing

Replenishment Commitments74 Define Commitments in Terms of Monitorable and Achievable

Objectives and Take Account of Their Realistic Costing

Annexes75 Annex A1: IDA’s Features77 Annex A2: Evolution of IDA’s Program79 Annex B: Poverty and Social Indicators80 Annex C1: IDA Commitments and Disbursements by Region82 Annex C2: IDA Commitments and Disbursements by Sector84 Annex D: IDA Portfolio Performance86 Annex E: Lending for Poverty-Targeted Interventions (PTIs)87 Annex F: Lending by Instrument89 Annex G: Methodological Note: Rating IDA’s Development

Performance94 Annex H: Chairman’s Summing Up—Meeting of the Executive

Directors (May 29, 2001)96 Annex I: Management Response

109 Endnotes

119 References

I D A’s P a r t n e r s h i p f o r P o v e r t y R e d u c t i o n

i v

Boxes4 Box 1.1. The IDA10–12 Replenishment Undertakings5 Box 1.2. Poverty, Growth, and Conflict6 Box 1.3. Trends in Official Development Assistance7 Box 1.4. Country Examples of IDA’s Assistance8 Box 1.5. Three Key Initiatives of the Late 1990s16 Box 2.1. IDA10–12 Commitments on Poverty17 Box 2.2. Stakeholders’ Views17 Box 2.3. IDA’s Contribution to Poverty Data Collection,

Analysis, and Dissemination18 Box 2.4. Where Did Increased Investments in the Social

Sectors Go?19 Box 2.5. Examples of IDA Support for Increased Social

Service Expenditures21 Box 2.6. Pooling Donor Funds in Support of Health Sector Reform

in Ghana24 Box 3.1. IDA10–12 Commitments on Improving the Poverty

Impact of Adjustment Lending24 Box 3.2. Poverty Impacts of Adjustment Lending in Uganda25 Box 3.3. Weak Broad-Based Growth, Despite Improved

Macroeconomic Stability, in Bolivia and Tanzania 26 Box 3.4. Programmatic Adjustment Lending to IDA Countries27 Box 3.5. Pro-Poor Rural Development Projects in Bangladesh

and India28 Box 3.6. IDA10–12 Undertakings on Private Sector Development29 Box 3.7. Bangladesh: Poverty Alleviation through Microfinance32 Box 4.1. IDA10–12 Commitments on Gender33 Box 4.2. Increasing Assistance for Girls’ Education34 Box 4.3. Maternal Health in Bangladesh35 Box 4.4. Characteristics of Good Practices in Integrating Gender35 Box 4.5. IDA10–12 Commitments on Environmental

Sustainability36 Box 4.6. Using the NEAP to Develop Strong Project Works38 Box 4.7. Eritrean Environmental Guidelines: Doing the Right

Thing Right40 Box 4.8. IDA10–12 Replenishment Commitments on Governance42 Box 4.9. Kenya: A Case Study of Withholding Lending to Foster

Governance Reforms48 Box 5.1. IDA10–12 Replenishment Commitments on Improving

Development Effectiveness through Selectivity49 Box 5.2. Joint IDA-IFC Country Assistance Strategies Highlight

Private Sector Development51 Box 5.3. The Performance-Based Allocation System51 Box 5.4. A New Approach to IDA Allocations to Post-Conflict

Borrowers52 Box 5.5. Suggested Changes for the Performance Rating Systems56 Box 6.1. IDA10–12 Replenishment Commitments on Improving

Development Effectiveness through Partnerships67 Box 7.1. Features of Alternative Lending Instruments

C o n t e n t s

v

Tables37 Table 4.1. Direct IDA Lending for the Environment and

Environmental Components, FY91–0066 Table 7.1. Breakdown of Operational Country Service Costs by

Activity (Direct Costs Only, in US$ millions)66 Table 7.2. Due Diligence Economic and Sector Work in IDA

Countries by Region (FY98–00)

Figure9 Figure 1.1. Improvements in Project Outcomes, Institutional

Development, and Sustainability in IDA/BlendCountries


v i

v i i

Acknowledgments

This report was prepared by an inde-pendent team, situated in the CorporateEvaluation and Methods Group of the

Operations Evaluation Department (OED) underthe overall oversight of, initially, Wendy Jarvieand, later, Nils Fostvedt. The team was led byCatherine Gwin (task manager and author of thereport) and included Stephen Eccles (back-ground study on IDA’s performance-based allo-cation system), John Eriksson (background studyon aid coordination), Alison Evans (backgroundstudy on poverty reduction), Pierre Landell-Mills (background study on governance), Mal-colm Rowat (background study on private sectordevelopment), John Shilling (background studyon environment), Olga Vybornaia (projectadministration and research assistance), andGeorgia Wallen (country consultation reportand research assistance).

Additional substantive input was provided byCharlotte Jones-Carroll (budget), Alice Galensonand John Nellis (private sector development), GitaGopal (gender), Daniel Morrow (Country Assis-tance Strategies), Sverrir Sigurdsson (social sec-tor development), and Warren van Wicklin(participation). Sohail Malik and Can Adamoglucontributed further statistical advice and researchsupport. Ted Rice, Sylvia Saborio, and Rita Trier-Somazzi assisted with in-country consultations.Annisa Cline-Thomas and Julia Ooro providedinvaluable administrative support. The team alsoacknowledges the help of Caroline McEuen in thedrafting of the overview, and Meta de Coquereau-mont and William Hurlbut in the editing of back-ground papers and this report.

The report benefited throughout from adviceand intellectual guidance offered by an advisorygroup of eminent experts: Nancy Birdsall, IngaBjork-Klevby, Kwesi Botchwey, Jalal AbdulLatif, Mohammed Syeduzzaman, and BernardWood. It also benefited from the advice, com-ments, and provision of information from manyBank colleagues, who are too numerous toacknowledge by name, but whose inputs wereessential to the background studies and over-all review.

Country consultations were held in Bangla-desh, Bolivia, Cambodia, Ghana, the KyrgyzRepublic, Mozambique, and Vietnam. In addi-tion, input was drawn from OED consultationscarried out in India and Uganda as part ofongoing Country Assistance Evaluations. Theseconsultations involved the participation of share-holder groups including government officials,nongovernmental organizations (NGOs), andprivate sector representatives, academics, anddonor agency representatives. The team is grate-ful to the IDA country directors and other col-leagues in the country offices, who helped makethe in-country consultations a success.

Workshops were held at the beginning andthe end of the review. The first, held in Bonnin December 1999 to discuss the evaluationdesign for the review, was hosted by the Cen-ter for Development Research and chaired by itsdirector, Joachim von Braun. The second washeld in November 2000 at World Bank head-quarters to discuss emerging findings. The teamwishes to thank the many Bank and externalexperts who participated in these meetings, in

particular C. Sanjivi Rajasingham, who representedIDA management. Financial assistance for theworkshops and country consultations was pro-vided by the Royal Ministry of Foreign Affairs,Department of Evaluation, Government of Norway.

The study was published in the Partnershipsand Knowledge Group (OEDPK) by the Out-reach and Dissemination Unit. The task teamincludes Elizabeth Campbell-Pagé (task team

leader), Caroline McEuen (editor), and JuicyQureishi-Huq (administrative assistant).

Director-General, Operations Evaluation: Robert Picciotto

Director, Operations Evaluation Department:

Gregory Ingram

Operations Adviser, Operations Evaluation Department:

Nils Fostvedt

Task Manager: Catherine Gwin


v i i i

FOREWORD

In the face of increasinglycomplex development

challenges, the past decade haswitnessed a broadening of theinternational development agendaand remarkable efforts bygovernments, voluntaryorganizations, and externalassistance agencies to learn fromexperience and improve theeffectiveness of developmentassistance. Against thisbackground, the IDA Deputiescommissioned this independentreview of the InternationalDevelopment Association (IDA)’sperformance in implementing theundertakings of the last threereplenishment agreements.

The seven-year period coveredby the review has been one ofconsiderable change—both inborrower countries and in thedevelopment system. IDA’sundertakings, while timely andrelevant, have continuously“raised the bar” on what wasexpected of both IDA and itsborrowers. It is thereforecommendable that the reviewrates IDA’s compliance with morethan 150 replenishmentundertakings as satisfactory, withqualifications.

The review establishes that IDAhas significantly improved itsportfolio performance, becomemore selective in its lendingallocations, recast its mission toaddress poverty reduction,strengthened the CountryAssistance Strategy (CAS) cycle,enhanced its responsiveness to

PREFACIO

En una coyuntura en que losdesafíos para el desarrollo

resultan cada vez más complejos, elpasado decenio se ha caracterizadopor la ampliación del programainternacional de desarrollo y larealización de esfuerzossignificativos por parte de losgobiernos, las organizacionesvoluntarias y los organismos deasistencia externa por aprender dela experiencia anterior y mejorar laeficacia de la asistencia para eldesarrollo. En esas circunstancias,los Suplentes de la AIF encargaroneste examen independiente deldesempeño de la AsociaciónInternacional de Fomento (AIF) en elcumplimiento de los compromisosde los acuerdos correspondientes alas tres últimas reposiciones.

En el período de siete años aque se refiere el estudio se hanproducido cambios considerables,tanto en los países prestatarioscomo en el sistema de desarrollo.Los compromisos de la AIF, sibien oportunos y pertinentes, hancontinuado “levantando el listón”de lo que se espera tanto de laAIF como de sus prestatarios. Porello, es encomiable que en elestudio se califique la observanciade la AIF con respecto a más de150 compromisos de reposicióncomo satisfactoria, con algunasreservas.

En dicho examen se estableceque la AIF ha mejoradosignificativamente el desempeñode su cartera, se ha mostrado másselectiva en las asignacionescrediticias, ha remodelado su

AVANT-PROPOS

Pendant la dernièredécennie, marquée par

des défis de plus en plus difficiles àrelever, le programme de travailde la communauté dudéveloppement n’a cesséde s’allonger et les gouvernements,les organisations bénévoles et lesbailleurs de fonds ont déployé desefforts remarquables pour tirer lesleçons de l’expérience et améliorerl’efficacité de l’aide audéveloppement. C’est dans cecontexte que les délégués à l’IDAont demandé que soit réalisée laprésente évaluation indépendantepour examiner les résultats obtenuspar l’Association internationale dedéveloppement sur la base desengagements pris dans le cadre desaccords au titre des trois dernièresreconstitutions des ressources.

La période de sept anscouverte par cet examen a étéune période de mutationsprofondes — dans les paysemprunteurs et dans le systèmemis en place pour ledéveloppement. Les programmesentrepris par l’IDA, pour aussipertinents et opportuns qu’ilssoient, n’ont cessé d’accroître lespressions exercées sur l’IDA et surses bailleurs de fonds. Il faut doncse féliciter que le respect de plusde 150 des engagements luiincombant au titre desreconstitutions de ses ressourcesait été jugé satisfaisant, malgré uncertain nombre de réserves.

L’examen indique que l’IDA aconsidérablement amélioré laperformance de son portefeuille,

F o r e w o r d

i x

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

borrowers and partners’needs, augmented its fieldpresence, and diversifiedits lending instruments.With IDA’s help, a numberof IDA borrowers are in abetter position now than atthe beginning of the IDA10

period to achieve broad-basedgrowth and poverty reduction.

Yet poverty trends in most IDAcountries have been disappointing.This confirms that there is stillsome way to go for IDA to fulfillthe ambitions of its owners andborrowers. In particular, thelinkages between countryprograms and poverty outcomesneed to be better articulated.Private sector development,gender, environmental, social, andrural development strategies needto be clarified and mainstreamedinto country and sector assistancestrategies. More needs to be doneto enhance the quality ofgovernance and to buildinstitutional capacity in specificcountry contexts.

Of course, IDA’s performanceis only one of the factors thataffects the results observed on theground. Exogenous factors, thepace of borrowers’ reforms, andthe quality of non-IDA assistancealso intervene. On the whole, inrelation to the ambitiousobjectives of IDA country andsector programs, the review ratesthe development outcome ofIDA’s programs as partiallysatisfactory, with a trend towardimprovement. While the glass ismore than half full and recentperformance trends augurpositively for the future, there isno room for complacency.

In looking forward to IDA’snext replenishment, what is

misión para centrarse másen la reducción de lapobreza, ha reforzado elciclo de estrategias deasistencia a los países, hamostrado más flexibilidadante las necesidades de losprestatarios y asociados, ha

aumentado su presencia sobre elterreno y ha diversificado susinstrumentos de financiamiento.Con ayuda de la AIF, variosprestatarios de la Asociación estánahora en mejores condiciones queal comienzo de la décimareposición para conseguir uncrecimiento de amplia base y lareducción de la pobreza.

No obstante, las tendencias dela pobreza en la mayor parte delos países de la AIF han sidodecepcionantes. Ello confirma quees todavía mucho lo que quedapor recorrer para que la AIFresponda plenamente a lasambiciones de sus titulares yprestatarios. En particular, losvínculos entre los programas delos países y los resultados en lalucha contra la pobreza deberíanestar mejor articulados. Eldesarrollo del sector privado, lascuestiones de género y lasestrategias ambientales, sociales yde desarrollo rural debenclarificarse e incorporarse a lasestrategias de asistencia a lospaíses y los sectores. Es precisohacer todavía más por mejorar lacalidad de la gestión pública ydesarrollar la capacidadinstitucional en el contexto de lospaíses concretos.

Naturalmente, el desempeñode la AIF es sólo uno de losfactores que influyen en losresultados observados en lapráctica. Intervienen tambiénfactores exógenos, el ritmo de la

a fait preuve de plus desélectivité dans le choix del’appui aux pays, recentréson action sur la luttecontre la pauvreté, renforcéle cycle de la Stratégied’aide-pays, accru sacapacité d’adaptation aux

besoins de ses emprunteurs et deses partenaires, renforcé saprésence sur le terrain etdiversifié ses instruments de prêt.Grâce à l’aide de l’IDA, plusieursemprunteurs de l’Association sontaujourd’hui mieux à même qu’audébut de la période d’IDA 10 demettre l’accent sur la promotiond’une croissance largementrépartie et d’une réduction del’incidence de la pauvreté.

Pourtant, les progrès accomplissur le front de la pauvreté sontdécevants. L’IDA doit doncpoursuivre sa transformation afinde pouvoir répondre pleinementaux attentes de ses actionnaires etde ses emprunteurs. Enparticulier, des liens mieux définisdoivent être établis entre lesprogrammes destinés aux pays etles résultats à obtenir sur le frontde la pauvreté. Un plan plus clairde promotion du développementdu secteur privé, de la parité dessexes, de l’action sur le planenvironnemental et social et dudéveloppement rural doit fairepartie intégrante des stratégiesd’aide aux pays et aux secteurs.Par ailleurs, les efforts doivent sepoursuivre pour améliorer laqualité de la gouvernance etrenforcer les capacitésinstitutionnelles en fonction de lasituation propre aux différentspays.

De toute évidence, laperformance de l’IDA n’est quel’un des facteurs qui déterminent


x

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

needed most is toconsolidate the IDAmandates and to focus onfurther improvements inthe effectiveimplementation of CASsand programs in thecontext of Poverty

Reduction Strategy Papersprepared by IDA borrowers. Thiswill call for even greater countryand program selectivity, far moreeffective donor coordination andharmonization, as well assystematic monitoring andevaluation, focusing on resultsand the Millennium DevelopmentGoals.

reforma en los paísesprestatarios y la calidad dela asistencia no procedentede la AIF. En conjunto, enrelación con los ambiciososobjetivos de los programasde la Asociación en lospaíses y en los sectores, el

examen califica sus resultados entérminos de desarrollo comoparcialmente satisfactorios, contendencia a mejorar. Si bien másde la mitad de la botella está llenay las tendencias de los resultadosrecientes son prometedoras parael futuro, no hay lugar para lacomplacencia.

Mirando al futuro, hacia lapróxima reposición de losrecursos de la AIF, lo que más senecesita es consolidar losmandatos de la Asociación yconcentrarse en la introducciónde nuevas mejoras en laaplicación eficaz de las estrategiasde asistencia los países y de losprogramas en el contexto de losdocumentos de estrategia delucha contra la pobreza (DELP)preparados por los prestatariosdel AIF. Ello requerirá unaselectividad todavía mayor encuanto a los países y programas,una coordinación y armonizaciónmucho más eficaz de losdonantes, así como unseguimiento y evaluaciónsistemáticos, centrados en losresultados y en los objetivosinternacionales de desarrollo.

les résultats sur le terrain,ceux-ci tenant aussi à desfacteurs exogènes ainsiqu’au rythme des réformesmenées par lesemprunteurs et à la qualitéde l’aide des partenaires del’institution. De manière

générale, l’impact desprogrammes de l’IDA au plan dudéveloppement est jugépartiellement satisfaisant, et enprogrès par rapport aux résultatsantérieurs. Bien que le verre soitplus qu’à moitié rempli et quel’évolution récente de la situationlaisse bien augurer de l’avenir, ilne faudrait pas pêcher par excèsd’optimisme.

Dans la perspective de laprochaine reconstitution desressources de l’IDA, il faudrasurtout renforcer les missions del’Association et s’attacher à encoreaméliorer l’exécution desstratégies et des programmesd’aide aux pays dans le contextedes Documents de stratégie deréduction de la pauvreté préparéspar les pays emprunteurs. Cedernier point passe notammentpar une plus grande sélectivité auniveau des pays et desprogrammes, des mesuresrésolues en faveur d’unecoordination et d’uneharmonisation de l’aide, et untravail systématique de suivi etd’évaluation, axé sur les résultatset sur les objectifs internationauxde développement.

F o r e w o r d

x i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

Robert PicciottoDirector-General, Operations Evaluation

EXECUTIVESUMMARY

Over the seven-year periodcovered by this review, IDA

significantly enhanced its relevanceand improved its portfolioperformance. It recast its mission,repositioned its country assistanceprograms, increased itsresponsiveness to borrowers,strengthened its field presence, anddiversified its lending instruments.Recently, it has intensified its aidcoordination efforts in the context ofthe CDF and PRSP initiatives.

Hence, IDA in FY00 is verydifferent from IDA in FY94. Still,disappointing progress in povertyreduction, especially in Africa,points to the need to sustain thetransformation of IDA in order torespond fully to new views ofcountry ownership, conditionality,and country-based programmingof development assistance. Focuson governance, capacity building,and development results; a movefrom ad hoc to structured donorcoordination; and an explicit rolefor IDA at the global level remaincritical to maintaining the positivemomentum of IDA’s developmenteffectiveness.

With IDA’s help, a number ofits borrowers are in a betterposition now than at thebeginning of the IDA10 period tofocus on broad-based growth andpoverty reduction goals. Still, thelinkages between countryprograms and poverty outcomesneed to be better articulated;clearer private sector development,gender, and environmental/socialsustainability strategies need to be

RESUMEN

Durante ese período de sieteaños, la AIF aumentó

significativamente su relevancia ymejoró los resultados de su cartera.Remodeló su misión, revisó susprogramas de asistencia a lospaíses, aumentó su capacidad derespuesta a los prestatarios, reforzósu presencia sobre el terreno ydiversificó sus instrumentos definanciamiento. Recientemente, haintensificado sus esfuerzos decoordinación de la ayuda en elcontexto de las iniciativas delMarco Integral de Desarrollo (MID)y de los documentos de estrategiade lucha contra la pobreza (DELP).

Por ello, en el ejercicio de 2000la AIF es muy diferente de la AIFde 1994. No obstante, losdecepcionantes progresos en lareducción de la pobreza, sobretodo en África, revelan la necesidadde continuar la transformación dela AIF con el fin de responderplenamente a las nuevas ideassobre la titularidad de los países, lacondicionalidad y la programaciónde la asistencia al desarrollo basadaen los países. La atención al sistemade gobierno, el desarrollo de lacapacidad y los resultados deldesarrollo; el paso de coordinaciónimprovisada de los donantes haciauna coordinación más estructurada,y una intervención explícita de laAIF en el plano mundial continúansiendo elementos críticos paramantener el impulso positivo de laeficacia de la AIF en términos dedesarrollo.

Con ayuda de la AIF, varios desus prestatarios se encuentran en

RÉSUMÉ ANALYTIQUE

Au cours des sept annéescouvertes par cette période,

l’IDA a considérablement renforcéla pertinence de ses interventions eta amélioré la performance de sonportefeuille. Elle a recentré sonaction, réorienté ses programmesd’aide aux pays, accru sa capacitéd’adaptation aux besoins desdemandeurs, renforcé sa présencesur le terrain et diversifié sesinstruments de prêt. Dans le cadredu CDI et des DSRP récemmentlancés, elle a également redoubléd’efforts pour permettre unemeilleure coordination de l’aide.

Aussi l’IDA s’est-elleprofondément transformée entreles exercices 94 et 00. Et pourtant,le manque de progrès sur le frontde la pauvreté, surtout en Afrique,montre qu’il faut poursuivre cettetransformation afin que l’aidefournie réponde pleinement auxnouveaux critères d’adhésion, deconditionnalité et de pilotage parles pays. L’importance à accorder àla bonne gestion des affairespubliques, au renforcement descapacités et aux résultats en termesde développement, l’abandon desmesures ponctuelles au profit d’unmécanisme structuré decoordination de l’aide et le rôlebien établi que l’IDA doit jouer auniveau mondial restent en outre depuissants moteurs de l’améliorationde l’efficacité du développement.

Grâce à l’aide fournie par l’IDApendant cette période, plusieursemprunteurs de l’Association sontaujourd’hui mieux à même demettre l’accent sur la promotion

E x e c u t i v e S u m m a r y

x i i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

well integrated into CountryAssistance Strategies (CASs);and more needs to be doneto integrate broad-basedpoverty reduction strategies,governance, andinstitutional developmentissues into macroeconomic

and sector strategies andinterventions within specificcountry contexts.

With management’s attention toaccelerating organizationalchanges over this period, IDA’scountry focus and responsivenessare stronger, with some 24 countrydirectors now in the field.Innovations in lending instrumentshave increased flexibility andinstitutional development impact.Still, the translation of key IDAobjectives into monitorableprograms and verifiable results islargely an unfinished business. Inparticular, corporate issues ofaccountability for policycompliance, investment inanalytical work and capacitybuilding, tracking progress,evaluating results, and aligningresources with program prioritiesall need closer attention andfurther action, as recentmanagement reviews andproposals indicate.

IDA in ContextIDA is only one among manyexternal assistance agencies,accounting for 14–18 percent ofofficial development assistance inthe 1990s. The performance ofborrowers and other developmentpartners, as well as changes in theglobal economic environment, allplay a role. IDA can and shouldbe held accountable, however, forhow well it makes hard choicesand necessary tradeoffs in

mejor situación ahora que alcomienzo del período de ladécima reposición paraconcentrarse en las metasde reducción de la pobrezay de un crecimiento deamplia base. No obstante,deben articularse mejor los

vínculos entre los programas de lospaíses y los resultados en la luchacontra la pobreza. Se necesitanestrategias sobre el desarrollo delsector privado, el género y lasostenibilidad ambiental/social quesean más claras y estén mejorintegradas en las estrategias deasistencia a los países. Hay quehacer todavía más para integrar lasestrategias de amplia base para lareducción de la pobreza y lascuestiones relacionadas con elsistema de gobierno y el desarrolloinstitucional en las estrategiasmacroeconómicas y sectoriales yen las intervenciones en paísesconcretos.

Gracias a los cambiosintroducidos por la administracióndurante ese período, la atenciónde la AIF a los países y sucapacidad de respuesta a losmismos son más sólidas, y ahorase encuentran sobre el terrenounos 24 directores a cargo depaíses. Las innovaciones de losinstrumentos crediticios hanaumentado la flexibilidad y losefectos del desarrolloinstitucional. No obstante, latraducción de los objetivos clavede la AIF en programas que sepuedan supervisar y en resultadoscomprobables es todavía unatarea pendiente. En particular, lascuestiones institucionales de larendición de cuentas sobre laobservancia de las políticas, lainversión en estudios analíticos ydesarrollo de la capacidad, el

d’une croissance largementrépartie et d’une réductionde l’incidence de lapauvreté. Il n’en reste pasmoins que des liens mieuxdéfinis doivent être établisentre les programmesdestinés aux pays et les

résultats à obtenir sur le front dela pauvreté ; qu’un régime plusclair de promotion dudéveloppement du secteur privé,de la parité des sexes et de laviabilité sur le plan social etenvironnemental doit faire partieintégrante des stratégies d’aideaux pays ; et que les effortsdoivent se poursuivre pour queles stratégies de lutte contre lapauvreté et les questions liées à lagouvernance et audéveloppement institutionnelsoient prises en compte dans lesstratégies et les actionsmacroéconomiques et sectoriellesmenées en fonction de lasituation propre aux différentspays.

La direction s’étant attachée àaccélérer la réorganisation internependant la période étudiée,l’intervention de l’IDA estaujourd’hui mieux adaptée àchaque emprunteur et sa capacitéd’ajustement est plus grande,grâce notamment à l’affectationsur le terrain de 24 des directeursd’opérations pour les pays. Lelancement de nouveauxinstruments de prêt a donné uneplus grande souplessed’intervention et a accru l’impactsur le développementinstitutionnel. La mise en place deprogrammes aux indicateursmesurables et aux résultatsvérifiables demeure toutefois unobjectif clé qui est loin d’êtreatteint. Les responsabilités


x i v

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

deploying its concessionalresources in fulfillment ofcorporate commitments,including both its lendingand nonlending advisoryservices, and howeffectively it shapes itsassistance to the priorities

and circumstances of borrowers.It is largely in these terms—ofIDA’s relevance (that is, whetherIDA “did the right things,” takinginto account the actions of others)and its efficacy and efficiency(that is, did it “do things right”)—that this review evaluates IDA.

A Strengthened PovertyOrientation. IDA has sharpenedthe poverty focus of its analyticalwork, policy dialogue, andlending. Broad-based growth,human resource development,and protection of vulnerablegroups have become commonobjectives of IDA CASs. IDA hasmade a major contribution topoverty analysis and data. It hasestablished a strong presence inthe social sectors, introducingnew lending instruments andapproaches to better deal withcomplex institutionaldevelopment issues. It has alsoimplemented its commitment toimprove the poverty and socialdimensions of structuraladjustment operations. Still, thelinkages between countryassistance programs and povertyoutcomes need to be betterarticulated.

The Challenge of Broad-BasedGrowth. IDA has helped manyborrowers move into position todirect greater efforts towardpoverty reduction goals than atthe beginning of the review

progreso del seguimiento,la evaluación de losresultados y laacomodación de losrecursos a las prioridadesprogramáticas son aspectosque deben ser objeto deatención más estrecha y de

nuevas medidas, como indican losrecientes exámenes y propuestasde la administración.

La AIF en contextoLa AIF es sólo una de las muchasorganizaciones de asistenciaexterna. En el decenio de 1990,representó entre el 14% y el 18%de la asistencia oficial para eldesarrollo. El desempeño de losprestatarios y otros asociados en eldesarrollo, así como los cambiosen el entorno económico mundial,son factores influyentes. La AIFpuede y debe dar cuentas decómo toma decisiones difíciles yllega a las necesarias soluciones decompromiso cuando despliega susrecursos concesionarios paracumplir los compromisosinstitucionales, incluidos loscréditos y los servicios de asesoría,y de cómo adapta eficazmente suasistencia a las prioridades ycircunstancias de los prestatarios.Es fundamentalmente desde estaperspectiva —relevancia de la AIFpara los objetivos básicos dedesarrollo, y su eficiencia oeficacia en función de los costospara conseguir los objetivosdeclarados— como se evalúa a laAIF en el presente examen.

Una orientación más firmecontra la pobreza. La AIF haorientado de forma más específicala lucha contra la pobreza en susestudios analíticos, el diálogo sobrepolíticas y los créditos. El

internes pour le respectdes politiques, lesinvestissements dans letravail d’analyse et lerenforcement des capacités,le suivi des progrès,l’évaluation des résultats etle couplage des ressources

aux actions prioritaires sont tousdes aspects qui demandent uneattention plus soutenue et desmesures supplémentaires, commela direction le reconnaît elle-même dans des propositions etexamens récents.

L’aide de l’IDA dans soncontexteL’IDA, qui a contribué pour 14 à18 % au total de l’aide publiqueau développement dans lesannées 90, n’est aussi qu’unorganisme d’aide extérieure parmide nombreux autres. Laperformance des emprunteurs etdes autres partenaires pour ledéveloppement ainsi quel’évolution de la conjonctureéconomique mondiale jouent unrôle important dans l’obtentiondes résultats. L’IDA peut et doitêtre tenue responsable, toutefois,du degré d’efficacité avec lequel,au prix de choix difficiles etd’arbitrages indispensables, elledéploie ses ressourcesconcessionnelles pour remplir sesengagements institutionnels (tantdans le cadre de ses services deprêt que de ses services de conseilhors-prêt), comme de l’efficacitéavec laquelle elle adapte sonassistance aux priorités et à lasituation de ses emprunteurs. C’estessentiellement en ces termes — àsavoir en termes de pertinence del’IDA, face aux grands objectifs dudéveloppement, et de l’efficacitéet l’efficience dont elle a fait


x v

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

period. In countriescommitted to reform, IDA’ssupport has contributed toincreased economicstability, fewer distortions,and improvedinfrastructure development.But acceleration of broad-

based, job-creating growthremains a major challenge.

The Critical Factor ofGovernance. Good governanceaffects all other areas of programemphasis. IDA was slow tocomply with its IDA10governance undertakings, butover the past four years it hasgiven priority to public sectorreform in country assistanceprograms. Lending for publicsector reforms has increased, andinstitutional development issuesare addressed in analytical workand lending. Public sectorcapacity building andaccountability remain major tasks,and greater coordination amongexternal agencies is needed toconsistently cover the broadgovernance agenda.

Gender and EnvironmentalMainstreaming. IDA has madelimited progress in integratinggender and environment, in partbecause of a lack of countryinterest. But IDA has also lackedclear accountability for gender andenvironmental mainstreaming.Where IDA’s assistance hascontributed to strong results—notably in girls’ education—therehas been significant countryownership, effective partnerships,and rigorous analytic workunderpinning lending. In similarcircumstances, environmentalassistance has helped to increase

crecimiento de amplia base,el desarrollo de los recursoshumanos y la protección delos grupos vulnerables sehan convertido en objetivoscomunes de las estrategiasde asistencia a los países dela AIF. La Asociación ha

realizado una importantecontribución en forma de análisis ypresentación de datos sobre lapobreza. Ha establecido una firmepresencia en los sectores sociales,mediante la introducción denuevos instrumentos y conceptosde financiamiento para hacer mejorfrente a las complejas cuestionesdel desarrollo institucional. Hahecho también realidad sucompromiso de mejorar lasdimensiones del ajuste estructuralrelacionadas con la pobreza y losaspectos sociales. No obstante, espreciso articular todavía mejor losvínculos entre los programas deasistencia a los países y losresultados frente a la pobreza.

El desafío del crecimiento deamplia base. La AIF ha ayudadoa muchos prestatarios a colocarseen situación de orientar más susesfuerzos a la reducción de lapobreza que al comienzo delperíodo del examen. En los paísescomprometidos con la reforma, elapoyo de la AIF ha contribuido aaumentar la estabilidadeconómica, reducir lasdistorsiones y mejorar eldesarrollo de la infraestructura.Pero la aceleración de uncrecimiento de amplia base ygenerador de empleo continúasiendo todavía un gran desafío.

El factor crítico de la gestiónpública. Un buen sistema degobierno es un factor crítico que

preuve à cet égard — quele présent examen évaluel’IDA.

Une focalisation accruesur la pauvreté. L’IDA arecentré son travaild’analyse, son dialogue sur

l’action à mener et ses opérationsde financement sur le combat dela pauvreté. Croissance largementrépartie, valorisation desressources humaines et protectiondes plus démunis font désormaispartie des objectifs traditionnelsde ses SAP. L’IDA a réalisé unimportant travail de collecte et dediffusion de données et d’analysessur la pauvreté. Elle est désormaistrès présente dans les secteurssociaux et a mis au point denouveaux instruments de prêt etde nouvelles démarches pourmieux faire face à des problèmescomplexes de développementinstitutionnel. L’IDA a égalementpris en compte la nécessitéd’améliorer la dimension socialeet les aspects liés à la pauvretédans les opérations d’ajustementstructurel. Il n’en reste pas moinsque des liens mieux définisdoivent être établis entre lesprogrammes d’aide aux pays etles résultats à obtenir sur le frontde la pauvreté.

La difficulté de promouvoirune croissance largementrépartie. Grâce à l’appui del’IDA, nombre de ses emprunteurssont en mesure de déployer desefforts plus importants pouratteindre les objectifs de réductionde la pauvreté qu’au début de lapériode couverte par le présentexamen. Dans les pays déterminésà entreprendre des réformes, lesopérations d’ajustement de l’IDA


x v i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

public and governmentinterest, reversed landdegradation, improvedincomes from arid lands,and strengthenedenvironmental capacity.

More EffectivePartnerships. To further increaseIDA’s development effectiveness,IDA10–12 strengthened the linkbetween building more effectivepartnerships through participationand aid coordination. Progresswas moderate during IDA10–11,but the CDF and PRSP initiativesintroduced after the IDA12negotiations have given significantimpetus to this aid reform agenda.Greater aid coordination at thecountry level and increased policyand procedural harmonization atthe agency level are necessary.

ConclusionsIn looking forward to IDA’s nextreplenishment, what is neededmost is to consolidate the IDAmandates, within the context ofCDF and PRSP implementation,and to focus on furtherimprovements in theimplementation of CountryAssistance Strategies andprograms. Improvedimplementation needs to giveparticular attention to country andprogram selectivity; a determinedmove to donor coordination andharmonization to reduce theburden of high aid transactioncosts on borrowers; as well assystematic monitoring andevaluation, focusing on resultsand the Millennium DevelopmentGoals. Full, multi-year funding ofCAS lending and nonlendingservices as well as newcommitments (including new IDA

influye en todas las demásesferas contempladas en losprogramas. La AIF tardó encumplir sus compromisosen materia de gestiónpública formulados en ladécima reposición derecursos, pero en los cuatro

últimos años ha dado prioridad ala reforma del sector público en losprogramas de asistencia a lospaíses. Los créditos para la reformade este sector han aumentado, ylas cuestiones del desarrolloinstitucional se consideran en losestudios analíticos y en lasactividades de financiamiento. Eldesarrollo de la capacidad y larendición de cuentas del sectorpúblico continúan siendo tareasimportantes, y se necesita unamayor coordinación entre losorganismos externos para abarcarel amplio programa pendiente eneste terreno.

Integración de las cuestionesrelacionadas con el género y elmedio ambiente. La AIF harealizado pocos progresos en laintegración de las cuestionesrelacionadas con el género y elmedio ambiente, en parte porfalta de interés de los países.Cuando la asistencia de la AIF hacontribuido a tener resultadosfavorables —sobre todo en laeducación de las niñas—, hahabido una importanteidentificación de los países conlos programas, asociacioneseficaces y rigurosos estudiosanalíticos como base delfinanciamiento. En circunstanciassemejantes, la asistencia ambientalha ayudado a aumentar el interéspúblico y gubernamental, hainvertido la tendencia a ladegradación de la tierra, ha

ont contribué à promouvoirla stabilitémacroéconomique, àréduire les distorsionséconomiques et à améliorerl’infrastructure dudéveloppement.L’accélération d’une

croissance largement répartie etcréatrice d’emplois restenéanmoins une tâche redoutable.

La gouvernance, une dimensionclé. La bonne gestion des affairespubliques est un facteur influantde façon déterminante sur tous lesautres domaines prioritaires desprogrammes. L’Association a tardéà honorer les engagements prispour IDA 10 sur le plan de lagouvernance mais au cours desquatre dernières années, elle adonné la priorité aux réformes dusecteur public dans lesprogrammes d’aide aux pays. Ellea accru le volume de ses crédits autitre de réformes du secteur publicet elle s’attaque aux problèmes derenforcement institutionnel dans lecadre de ses études analytiques etde ses opérations de financement.Le renforcement des capacités dusecteur public et la nécessité dejustifier de l’emploi des fondspublics dans les pays emprunteurscontinuent d’exiger des effortsconsidérables ; il faudra parailleurs améliorer la coordinationdes interventions des organismesde financement extérieur pours’assurer de la fourniture d’unappui systématique au vasteprogramme de l’amélioration de lagouvernance.

Prise en compte de la paritédes sexes et del’environnement. Les progrèsaccomplis par l’IDA au plan de


x v i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

undertakings) is essentialto align resources withprogram priorities.

Finally, the reportsuggests fine-tuning of thereplenishment process.Although consistent withthe evolving development

paradigm, replenishmentundertakings have been perceivedto be both overdetermined andoverloaded. Greater realism aboutwhat IDA and its borrowers couldreasonably accomplish in a three-year period is desirable. Equally,the replenishment process needsstronger connections with alldevelopment partners. IDA, inconsultation with its borrowers,should develop a longer-termvision focused on results. Itshould engage developingcountries in both settingreplenishment priorities andmonitoring IDA performance, andit should define commitments interms of monitorable andachievable objectives.

aumentado los ingresosprocedentes de las tierrasáridas y reforzado lacapacidad ambiental. Perola AIF carece también deun sistema claro derendición de cuentas sobrela integración de las

cuestiones relacionadas con elgénero y el medio ambiente.

Asociaciones más eficaces. Paraconseguir una mayor eficacia de laAIF en términos de desarrollo,entre la décima y duodécimareposiciones se intensificó elvínculo entre el establecimientode asociaciones más eficacesmediante la participación y lacoordinación de la ayuda. Esteprogreso fue moderado durante ladécima y undécima reposiciones,pero las iniciativas del MID y losDELP introducidas después de lasnegociaciones de la duodécimareposición han dado considerableimpulso a este programa dereforma de la ayuda. Se necesitatambién una mayor coordinaciónde la asistencia en los países yuna mayor armonización depolíticas y procedimientos en elplano institucional.

ConclusiónDe cara a la próxima reposiciónde los recursos de la AIF, lo quemás se necesita es consolidar losmandatos de la Asociación, en elcontexto del MID y los DELP, yconcentrarse en la introducción denuevas mejoras en la aplicaciónde los programas de asistencia alos países. Las mejoras en laaplicación deben prestar especialatención a la selectividad depaíses y programas, a lacoordinación y armonización delos donantes con el fin de reducir

l’intégration de cesquestions à sesprogrammes ont été limitésen partie par le manqued’intérêt manifesté par lespays. L’appui del’Association a néanmoinscontribué à l’obtention de

bons résultats, notamment dans ledomaine de l’éducation des filles,du fait de la large adhésion despays à l’action menée, del’efficacité des partenariats avecles autres intervenants, et desbases solides fournies par untravail d’analyse rigoureux. Lescirconstances sont similaires pourl’environnement, dans lequell’appui de l’IDA a aidé àsensibiliser le public et lesautorités aux questionsenvironnementales et a contribué,par exemple, à un renversementde la tendance à la dégradationdes sols, à un accroissement desrevenus tirés des terres arides et àun renforcement des capacités.Mais l’IDA n’est pas non plusclairement comptable de l’actionqu’elle doit mener pour intégrerces deux dimensions à sesactivités.

Bâtir des partenariats plusefficaces. Pour accroître encoreplus l’efficacité de l’aide de l’IDA,IDA 10-12 ont mis l’accent sur lerenforcement du lien entre laconstitution de partenariats plusefficaces basés sur la participationet la coordination de l’aide. Sil’amélioration a été faible pendantIDA 10-11, les initiatives du CDIet des DSRP lancées après lesnégociations d’IDA 12 ont donnéune forte impulsion à ces effortsde réforme. Il est nécessaire demieux coordonner l’aide auniveau des pays et de mieux


x v i i i

EN

GL

ISH

FR

AN

ÇA

IS

ES

PA

NO

L

la carga de los elevadoscostos de transacción de laayuda que recae sobre losprestatarios, así como a unmayor fortalecimiento de lacapacidad y unseguimiento y evaluaciónsistemáticos, centrados en

los resultados y en los objetivosinternacionales de desarrollo. Paraarmonizar los recursos con lasprioridades de los programas esnecesario un financiamientocompleto y plurianual de losservicios crediticios y nocrediticios de las estrategias deasistencia los países, así como ellogro de nuevos compromisos(también de la AIF).

Finalmente, en el informe seaconseja la introducción depequeños ajustes en el procesode reposición de recursos. Si bienen consonancia con el paradigmade desarrollo en evolución, loscompromisos de reposiciónparecen estar excesivamentedeterminados y sobrecargados.Sería aconsejable un mayorrealismo acerca de lo que la AIF ysus prestatarios podrían conseguirrazonablemente en un período detres años. De la misma manera, elproceso de reposición necesitavinculaciones más estrechas conlos asociados en el desarrollo. LaAIF, en consulta con susprestatarios, debería estableceruna visión a más largo plazocentrada en los resultados.Debería conseguir que los paísesen desarrollo fijaran prioridadesde reposición y supervisaran eldesempeño de la AIF, y deberíadefinir los compromisos enfunción de objetivos que puedanser objeto de seguimiento yasequibles.

harmoniser les grandesorientations et lesprocédures au niveau desorganismes bailleurs defonds.

ConclusionDans la perspective de la

prochaine reconstitution desressources de l’IDA, il faudrasurtout promouvoir les missionsde l’Association, dans le cadre dela mise en oeuvre du CDI et desDSRP, et s’attacher à encoreaméliorer l’exécution desstratégies et des programmesd’aide aux pays. Ce dernier pointpasse notamment par une plusgrande sélectivité au niveau despays et des programmes, desmesures résolues en faveur d’unecoordination et d’uneharmonisation de l’aide afin deréduire la charge de coûts detransaction qui pèsent lourdementsur les emprunteurs, et un travailsystématique de suivi etd’évaluation, axé sur les résultatset sur les objectifs internationauxde développement. Lefinancement intégral et surplusieurs années des activités deprêt et des services hors prêt desSAP ainsi que des nouveauxengagements (y compris ceux prisdans le cadre de la reconstitutiondes ressources) est indispensablesi l’on veut que des moyensadéquats soient affectés auxactions prioritaires.

Enfin, le rapport d’évaluationpropose un ajustement duprocessus de reconstitution desressources. En effet, on a lesentiment que les engagementspris en cette occasion sontsurdéterminés et trop lourds,même s’ils vont bien dans le sensde l’évolution du modèle de


x i x

FR

AN

ÇA

IS

ES

PA

NO

L

développement. Il seraitsouhaitable de faire preuved’un plus grand réalismelorsqu’on envisage ce quel’IDA et ses emprunteurspeuvent raisonnablementaccomplir en trois ans. Ilfaut aussi que ce processus

se déroule en relation étroite avectous les acteurs dudéveloppement. En consultationavec ses emprunteurs, l’IDA doitdonc définir une vision à longterme s’articulant autour desrésultats souhaités. Elle doitassocier les pays endéveloppement à la définition despriorités et au suivi de saperformance, et les engagementsqu’elle prend doiventcorrespondre à des objectifsréalisables, accompagnésd’indicateurs de suivi.


x x

FR

AN

ÇA

IS

x x i

AfDB African Development BankAFR Africa RegionAPL Adaptable Program LoanARDE Annual Review of Development EffectivenessCAE Country Assistance EvaluationCAS Country Assistance StrategyCDF Comprehensive Development FrameworkCEM Country Economic MemorandumCFAA Country Financial Accountability AssessmentCG Consultative GroupCGAP Consultative Group to Assist the PoorestCPAR Country Procurement Assessment ReviewCPFA Country Profile of Financial AccountabilityCPIA Country Policy and Institutional AssessmentDfID Department for International DevelopmentEAP East Asia and Pacific RegionECA Europe and Central Asia RegionESSD Environmentally and Socially Sustainable Development NetworkESW Economic and sector workGEF Global Environment FacilityGDP Gross domestic productGNP Gross national productHIPC Heavily Indebted Poor CountryHNP Health, Nutrition, and Population SectorIBRD International Bank for Reconstruction and Development (World Bank)IDA International Development AssociationIFC International Finance CorporationLAC Latin America and Caribbean RegionLIL Learning and Innovation LoanM&E Monitoring and evaluationMIGA Multilateral Investment Guarantee AgencyMNA Middle East and North Africa RegionNEAP National Environmental Action PlanNGO Nongovernmental OrganizationOD Operational DirectiveODA Official Development AssistanceOECD Organisation for Economic Co-operation and DevelopmentOED Operations Evaluation DepartmentPA Poverty AssessmentPACT Partnership for Capacity BuildingPBA Performance-based allocationPER Public Expenditure ReviewPPP Purchasing power parityPREM Poverty Reduction and Economic Management NetworkPRSP Poverty Reduction Strategy PaperPSD Private sector development

ABBREVIATIONS AND ACRONYMS


x x i i

QAG Quality Assurance GroupSA South Asia RegionSPA Special Program with AfricaSSP Sector and Thematic Strategy PaperSSR Social and Structural ReviewSWAps Sector-wide approachesWBI World Bank InstituteWDR World Development Report

x x i i i

Overview

The mission of IDA is to support efficient and effective programs to reducepoverty and improve the quality of life in its poorest member countries … thefocus must be on sustainability—to achieve enduring development impactwith an environmentally sustainable framework—and equity—to removebarriers and open up opportunities for the disadvantaged.

—IDA12 Replenishment Report

The International Development Association (IDA)is a unique instrument of development cooper-ation, focused on assisting the world’s poorestcountries. The establishment of IDA in 1960—asa separately funded but integral component of theWorld Bank Group—represented a bold innova-tion and an unprecedented commitment by theinternational community to improving global wel-fare. During the past 40 years, IDA has made sub-stantial and distinctive contributions to growth andpoverty reduction in low-income countriesthrough a combination of concessional finance,analytical work, and aid coordination services.

During the seven-year period covered by thisreview (FY94–00), IDA committed just over $42billion to some 77 low-income borrower coun-tries. These resources have supported countryefforts to accelerate sustainable economic growthand poverty reduction through improved eco-nomic policies and investments; expanded accessto basic education, health, and other social ser-vices; and strengthened public sector manage-

ment and governance. IDA has also assistedwith the unprecedented challenges faced bycountries in transition and the special needs ofcountries emerging from conflict. While persistentpoverty, increasing inequality, and conflict char-acterize the period, a number of countries whereIDA programs are being implemented haverecorded declines in the incidence of poverty,strengthened social development, and begunto tackle governance and public sector reforms.

At the request of IDA’s donor governmentsand its executive directors, this report reviewsthe IDA10–11 program and, on an interim basis,IDA12. It covers IDA’s performance in imple-menting the undertakings set out in the Replen-ishment Reports for each of these periods. Themethods used for this review are described inAnnex G. The final report was presented to theBoard on May 29, 2001. A summary of the dis-cussion at that meeting is attached as Annex H.The Bank management response to the findingsof the review is attached as Annex I.

Summary of FindingsThe IDA10–12 replenishment undertakings (sum-marized in box 1) have been highly relevant andtimely. They encouraged IDA to adapt to a newdevelopment paradigm, embodying a compre-hensive approach to poverty reduction thatreflects the accumulated lessons of develop-ment experience. Individually, each area ofoperational emphasis makes sense. But togetherthey have proved extraordinarily demanding—for both IDA and its borrowers. Even in coun-tries where the commitment to poverty reductionand sustainable development is strong, buildingconsensus for change, reforming policies, andstrengthening institutional capacities are formi-dable tasks, which, in countries with limitedhuman and financial resources, require toughchoices and tradeoffs.

In evaluating IDA’s performance, it is impor-tant to place its assistance in context. IDA pro-vides a small share of the resources that countriesuse to pursue their development priorities—thelargest share comes from the countries them-selves. IDA is also only one among many exter-nal assistance agencies, accounting for 14–18percent of official development assistance inthe 1990s. It cannot determine the choices thatcountries make, although it can and does sup-port and influence them. Nor can IDA alone be

held accountable for country development out-comes. The performance of borrowers and otherdevelopment partners, as well as changes inthe global economic environment, all play arole in determining outcomes. IDA can andshould be held accountable, however, for howwell it makes hard choices and necessary trade-offs in deploying its concessional resources tofulfill corporate commitments, including both itslending and its nonlending advisory services, andhow effectively it shapes its assistance to the pri-orities and circumstances of borrowers. It islargely in these terms—that is, of IDA’s relevance(whether IDA “did the right things,” taking intoaccount the actions of others) and its efficacy andefficiency (did it “do things right”)—that thisreview evaluates IDA. (For a concise summaryof the report, see box 2.)

Compliance Overall, IDA’s compliance with the formidablelist of replenishment undertakings has been sat-isfactory, albeit with important qualifications.The implementation of IDA undertakings hasrepositioned country programs to better respondto development priorities. IDA has sharpened thepoverty focus of its Country Assistance Strategies(CASs), analytical work, and lending. It hasmade a substantial contribution to poverty datacollection and analysis; established a strongpresence in the social sectors; and improved thepoverty and social dimensions of its adjustmentoperations. More recently, IDA has markedlyexpanded its work in support of good gover-


x x i v

In broad terms, the three IDA replenishment agreements—reports that set forth IDA’s goals for each three-year funding cycle—endorse poverty reduction as IDA’s overarching objective and instruct IDA to:• Sharpen the poverty focus of its support for country development.• Direct its assistance, in support of that objective, to expanding access to basic social services, fostering broad-based growth,

promoting good governance, and integrating gender and environmental considerations into development efforts.• Increase its development effectiveness through more selective, more participatory, and better-coordinated country assistance

programs.The specific undertakings related to these instructions center on six program priorities (poverty, social sectors, private sec-

tor development, gender, environment, and governance) and four processes (Country Assistance Strategies, performance-based allocation system, participation, and aid coordination).

I D A 1 0 – 1 2 R e p l e n i s h m e n t U n d e r t a k i n g sB o x 1

IDA10–12 replenishment undertakingshave been highly relevant and timely.

nance as a key factor in country developmentefforts.

To strengthen the effectiveness of its coun-try programs, IDA has also enhanced the role ofits CASs as a tool for comprehensive diagnosisand program planning. It has strengthened thelink between country performance and lending;increased stakeholder participation in projectsand programs; and accelerated its efforts in aidcoordination, especially in the past few years.These are major achievements.

Compliance, however, has been unevenacross and within areas of program and processemphasis. Governance only recently acquiredadequate weight in program priorities. There hasbeen only partial and halting progress in inte-grating critical private sector development (PSD),gender, and environmental considerations intocountry assistance programs. And neither pro-gram selectivity nor coordination with devel-opment partners has been achieved to the degreecalled for in the replenishment undertakings,although the Comprehensive Development

Framework (CDF) and the Poverty ReductionStrategy Paper (PRSP) initiatives have providednew momentum in the past two years.

Development OutcomesClearly, compliance with replenishment under-takings and development outcomes are not thesame. Nor is performance in the implementationof replenishment injunctions (which are statedlargely in terms of inputs and outputs rather thanoutcomes) the sole determinant of those out-comes. The global environment has been chal-lenging for countries with weak institutions andpoor governance, and poverty trends have beendisappointing. Against this background, theimproved development effectiveness of IDAoperations is notable, while the developmentoutcomes of IDA programs, assessed against itsambitious goals, have been partially satisfactory.1

Although IDA has done well in helping manycountries lay the foundations for economic growthand poverty reduction, the record of IDA coun-tries in sustaining growth at high enough levels,

O v e r v i e w

x x v

At the request of IDA’s donor governments and its executivedirectors, OED carried out a review of IDA’s implementation per-formance during IDA10–12. OED found the IDA10–12 replenishmentcommitments to be highly relevant and timely, but extraordinar-ily demanding for both IDA and its borrowers. The commitmentsinclude: (a) sharpening the poverty focus of support for countrydevelopment; (b) expanding access to social services, fosteringbroad-based growth, promoting good governance, and integrat-ing gender and environmental considerations into developmentefforts; and (c) increasing its development effectiveness throughmore selective, more participatory, and better-coordinated CASs.

Overall, OED finds IDA’s performance in relation to its com-mitments to be satisfactory, but with qualifications. IDA hasdone much to sharpen the poverty focus of its analytical work,policy dialogue, and lending. The quality of lending and non-lending services has improved. Recently, it has brought gover-nance to the fore. Yet it has made limited progress in integratingprivate sector development, gender, and environmental sus-tainability into its country programs. Taking account of a host offactors not under IDA’s control, the development outcomes of IDAprograms are rated partially satisfactory. Much remains to bedone by IDA and its partners to meet the multifaceted challengeof supporting sustained, pro-poor, broad-based growth.

In moving forward, IDA should focus on implementation ofits existing policy framework, align resources with strategicpriorities, and consolidate its mandates. IDA should concen-trate more on its areas of comparative advantage—work at thestrategic level in support of economy-wide, sector-wide, andgovernment-wide reforms—and make capacity building acore dimension of every aspect of its work. With the full sup-port of its shareholders, it should also play a more proactiverole in fostering aid harmonization and coordination—at theglobal and country levels—to reduce the high transactioncosts of aid.

This will require even greater country and program selectivity;expanded work in the critical areas of public sector reform andinstitutional development, a joint commitment with partners tomove from ad hoc to more structured coordination of aid pro-grams, and reinvestment in IDA’s analytical work and full fund-ing of approved Country Assistance Strategies.

IDA would also benefit from adjusting the replenishmentprocess. Three changes would improve the process: develop-ing a long-term vision focused on results; engaging developingcountries in setting priorities and monitoring replenishmentcommitments; and defining commitments in terms of monitorableand achievable objectives, with realistic costing.

T h e I D A R e v i e w a t a G l a n c eB o x 2

over long enough periods, and through measuressufficient to benefit the majority of the poor hasbeen mixed. This reflects a host of non-IDA influ-ences on development outcomes, includingregional conflicts, weak domestic capacities,mixed policy performance, variable quality ofaid, and exogenous shocks. But it also highlightsareas where IDA’s own performance, while gen-erally satisfactory and improving, could beenhanced still further and contribute more posi-tively to development outcomes.

IDA made important innovations in its organ-ization, lending instruments, and assistancestrategies to increase the responsiveness of itsoperations to country development priorities.Robust self-evaluation and policy recasting havebeen put in place, contributing to increaseddevelopment effectiveness. More important, theperformance of IDA lending has shown sub-stantial gains. But the linkage between institu-tional priorities and country programs has notbeen strong, and weaknesses in lending andnonlending activities that impede IDA’s opera-tional effectiveness at the project and country lev-els remain. Declines in the funding of analyticalwork, the substantial challenge of expandinggovernance work, the slow pace of improvementin institutional development impact and supportfor capacity building, inadequate selectivity, andinsufficient monitoring and evaluation (M&E) atboth the country and the project levels requirestronger remedial action, as acknowledged and,in some cases, pinpointed by management’sown self-assessments.

Why This Review?While IDA’s main goals and basic features haveremained consistent throughout its history, IDAhas progressively refined and broadened itsapproach. It reached an intellectual and pro-grammatic watershed with the publication of theWorld Development Report 1990: Poverty, whichadvocated a strategy of poverty reductionthrough economic policy reforms and produc-tive investments. The goal was to improve theincomes of the poor through labor-intensivegrowth and to expand their access to socialservices and social safety nets. The strategyreflected empirical evidence that demonstrated

the importance of growth and the central roleof human resource development in bringingabout significant and sustained improvements inliving standards.

The IDA10–12 replenishment undertakingsadopted the framework of the 1990 strategy, witha call to sharpen IDA’s poverty focus. Theyexpanded the agenda to include gender, theenvironment, and governance as part of a“broad-based framework for poverty reduction”(IDA 1998). They also urged IDA to increaseeffectiveness through increased country own-ership and improved development partnerships.The diversified and complex policy content ofthese commitments—a departure from the moregeneral guidance embodied in previous replen-ishment agreements—is the ultimate rationale forthis review. Has IDA acted responsibly to imple-ment its replenishment commitments? Has itchanged fast enough and in the right directions?As a result, is it well equipped to assist borrowercountries in today’s era of rapid technologicalchange, borderless private investment, andincreasingly fragile physical environment? Theseare the questions that motivated the IDADeputies’ call for an independent review ofIDA’s record.

The Program Dimension

A Sharpened Poverty FocusIDA has progressively strengthened the povertyorientation of its analytical work, investmentlending, and adjustment operations since adop-tion of the 1990 poverty strategy. Both momen-tum and specificity have increased oversuccessive replenishment periods. Internal pol-icy and institutional changes, combined withexternal contributions from IDA (and otherdonors), have moved many IDA borrowers intoa position to direct greater efforts toward povertyreduction goals than at the beginning of thereview period. But substantial and sustainableresults for the poor require consolidation anddeepening of development and aid reforms.

Redirecting Country Assistance Strategies.IDA undertakings have emphasized povertyreduction as their overarching objective, and


x x v i

this goal has been increasingly reflected in IDACASs, particularly since 1997. Broad-basedgrowth, human resource development, and pro-tection of vulnerable groups have become com-mon CAS objectives. CASs are also increasinglybased on comprehensive poverty diagnosis andconsultations with stakeholders. By FY00, agrowing number of IDA CASs included povertyreduction targets linked to intermediate objec-tives and benchmarks for tracking country andIDA performance.

To underpin the sharpened focus of the CAS,IDA has made major contributions in collectingand disseminating poverty data and analysis.Poverty Assessments (PAs) have been completedfor 90 percent of eligible IDA borrowers, andPublic Expenditure Reviews (PERs) have beenconducted for almost half. Still, linkages betweencountry programs and poverty outcomes arenot always well articulated. More needs to bedone to integrate broad-based growth povertyreduction strategies into macroeconomic andsector strategies and interventions. Improve-ments are needed in setting program prioritiesthat will have the greatest impact in reducingpoverty. For this, the quality and policy relevanceof economic and sector work (ESW) and themonitoring of linkages between IDA-supportedpolicies and programs and poverty outcomescontinues to need attention, with particularemphasis on strengthening borrower capacity tocollect and analyze poverty data, monitorprogress, and enhance program outcomes.

Expanding access to social services. IDA’scommitment to increase social sector lending hasbeen highly satisfactory over the IDA10–12period. Social sector investment lending is upfrom 20 percent of total IDA investment lend-ing in the late 1980s to a sustained level of 40percent since 1995. With the increase, IDAbecame the largest financier in human devel-opment and a major source of analysis andadvice, particularly in Africa. Through its ana-lytical work, lending, and technical assistance,IDA has helped countries strengthen their humandevelopment policies, increase their level ofsocial expenditures, expand access to basic ser-vices, and, particularly in the transition countries,

adjust pension and other social protection mech-anisms. Overall, IDA has shifted the focus of itsassistance from projects to sector-wide reformsand provided some form of social sector assis-tance in virtually all of its active borrowers.

On the whole, IDA has had more success inhelping countries to expand access to socialservices than in improving the quality of servicedelivery to the poor. The difficult challenge hasbeen to help countries deal with the politicallyand institutionally complex issues, particularlydeficits in public sector management and insti-tutional development, that hobble the sector-wide reforms and programs needed to producesustainable results. IDA has responded to thischallenge with several program innovations,including new sector-wide and country-wideprogrammatic and adaptable lending instru-ments, a stronger field presence, and greater pri-ority to institutional change, with a fullerrecognition of the importance of partnership andownership of policy reform. IDA has alsoincreased its participation in country-led, donor-coordinated, sector-wide programs in educationand health; given more attention to multi-country (regional or global) programs; and,most recently, increased its support for linkingsocial sector reforms more closely with povertyreduction goals through assistance to countriesin their preparation of PRSPs. Each of these inno-vations shows significant promise, but also pre-sents operational challenges that call for furtherresults-based process and procedural changeson the part of IDA, in conjunction with otherdonors.

The Challenge of Broad-Based GrowthIn promoting broad-based growth, the secondmajor component of IDA’s poverty reductionframework, IDA has traveled a considerabledistance in implementing its commitment toimprove the poverty and social dimensions ofstructural adjustment operations and to strengthenits support for PSD. Nevertheless, acceleration ofbroad-based, job-creating growth remains a majorchallenge.

Gains from adjustment lending. Bank-wideproject ratings indicate that adjustment operations

O v e r v i e w

x x v i i

have improved their development effectiveness.Satisfactory outcome ratings have risen from 65percent of completed projects in IDA9 to 80 per-cent in IDA11–12. In countries committed toreforms, IDA adjustment operations have con-tributed to macroeconomic stability and theremoval of key economic distortions, but evi-dence on income and employment generationfor the poor is less clear. Why has it been so dif-ficult for development partners, including IDA,to support countries in moving beyond the inter-mediate outputs of adjustment programs (newpolicies, legislation, privatization) to achievebetter poverty reduction results?

Many IDA countries have been inconsistentin their implementation of reform programs,short-circuiting their ability to sustain high growthrates and to implement the complex structuralreforms necessary for long-term poverty reduc-tion. It has also proved difficult to come upwith practical policy measures to achieve not justgrowth, but broad-based growth, and to addressthe factors that affect the ability of the poor toparticipate in the opportunities created bygrowth-oriented policies. Of concern is evalua-tive evidence, including testimony of stake-holders, that the links between policy change,sector strategy, and the expected pattern ofgrowth are weakly articulated in IDA CASs, andthe mechanisms that are to transmit the bene-fits of policy changes to the poor are not spelledout. There is an urgent need for work on thedeterminants of pro-poor growth in specificcountry circumstances and for greater clarityand specificity in how IDA assistance can bestsupport borrowers in setting and advancingcountry-specific priorities.

Lags in rural and private sector develop-ment. Lags in both rural and private sectordevelopment are also important contributing

factors to the weak broad-based growth out-come. IDA’s lending for agriculture and ruraldevelopment, largely overlooked in theIDA10–12 replenishment agreements, hasdeclined from 23.4 percent of commitments dur-ing IDA10 to 9.6 percent during the first year ofIDA12. It appears that IDA has withdrawn,appropriately, from unsuccessful efforts, such astop-down systems of extension services andcomplex rural development activities that hadhigher than average failure rates, but it has notput improved approaches in place. Adjustmentlending and related ESW have contributed to rel-evant policy reforms in a number of IDA coun-tries, improving agricultural price incentives andexports. Reform efforts, however, have notresolved other structural and institutional con-straints that impede agricultural productivity andmarketing and rural poverty reduction.

In the context of robust lending for socialfunds, rural infrastructure, and health and edu-cation, decreases in agricultural and rural lend-ing cannot be assumed, a priori, to be a badthing. However, given that most of the world’spoor will continue to live in rural areas well intothe twenty-first century, and that agricultureaccounts for a sizable share of poor countries’GDP, the lack of consensus regarding ruraldevelopment strategies among developmentpartners and the reduced priority of agriculturein aid programs are causes for concern. The sec-tor requires renewed attention on the part of theinternational development community, withIDA’s role to be determined in coordinationwith others.

In contrast, as efforts to stabilize macroeco-nomic conditions took hold in the 1990s, IDAincreased its PSD activities and, in certain sub-sectors, has had some success. But CASs still tendto lack well-articulated PSD strategies, which inpart reflects differences in perspective betweenIDA and its borrowers and continuing difficul-ties in linking IDA, International Finance Cor-poration (IFC), and Multilateral InvestmentGuarantee Agency (MIGA) activities into a coher-ent whole. In the past few years, IDA has forgeda closer working relationship with IFC. Consid-erably greater synergies, however, can be tappedand selectivity exercised by further delineating


x x v i i i

Why has it been so difficult to supportcountries in moving beyond the intermediate

outputs of adjustment programs to achievebetter poverty reduction results?

IDA’s role in improving the country policy frame-work, IFC’s connectivity to private corporations,and MIGA’s catalytic function in foreign invest-ment, in addition to shifting activities among insti-tutions and dropping activities better pursued byother actors. Although the Bank’s prior CAS ret-rospective did not analyze the treatment of PSD,the 2000 CAS report notes that by increasing theearly involvement of IFC and MIGA in CASpreparation, PSD programming has beenimproved. Thus, it proposes a deeper integra-tion of CASs and work programs in selectedcountries.

While many IDA projects have had success-ful outcomes, a selection of IDA Country Assis-tance Evaluations (CAEs) gives relatively lowratings to the effectiveness of IDA’s PSD workand underscores the need to focus more sharplyon the investment climate. These evaluations alsoreflect past weaknesses in IDA’s support forsmall and medium-size enterprises and ruralfinance—two areas with new strategies that aimto improve future efforts. In addition, inade-quate focus on the institutional and policy frame-work for privatization operations has meant thatefficiency gains have often not been widelyshared. In line with IDA12 commitments, suchoperations are now giving increased attention tosocial protection and environmental issues.Efforts to increase the private provision of infra-structure are more recent and show promise.

The PSD strategy paper now being preparedshould explicitly define the rationale for a cleardivision of labor and greater selectivity withinthe World Bank Group and vis-à-vis others. Itshould also identify ways to improve IDA’s sup-port for the policy and regulatory environmentfor private investment, the proper institutionalfoundations for privatization, the expansion ofthe private provision of infrastructure and socialservices, and the overall distributional impact ofPSD activities.

IDA’s strategic role. IDA’s comparative advan-tage in support of broad-based growth andpoverty reduction lies at the strategic level, notonly in lending for adjustment programs and keyinvestments in support of broad-based growth,but also in supporting the research and analyt-

ical work needed to identify the actions requiredfor the poor to share in the overall gains. ThePRSP initiative, launched after the negotiation ofIDA12 and based on CDF principles, representsa major step in realizing these goals. It offers anew way to link poverty analysis, country-ledpolicy design and program choice, partnerinvolvement, and monitoring. In this context, IDAshould increase its support for strengtheningborrower capacity to track programs, analyzeresults, improve program outcomes, and informbroad public debate. In addition, IDA manage-ment should continue to consider how toimprove the role of Sector and Thematic Strat-egy Papers (SSPs) in identifying knowledge andstrategic gaps, with a sharpened focus on povertyreduction, as a guide to clarifying the kinds ofactivities that make the most sense for the WorldBank Group.

Integrating Gender, the Environment, and GovernanceAlthough the IDA10–12 replenishment under-takings that called for integrating gender, envi-ronmental sustainability, and good governanceinto IDA’s country assistance programs werehighly relevant, progress has been constrainedby a lack of consensus, within IDA and amongits member countries, on the priority of theseconcerns, and on IDA’s appropriate role inadvancing them. While IDA has made importantcontributions in each of these areas, its successhas been limited relative to its commitments.

Slow progress in gender and environmentalmainstreaming. In implementing the commit-ment to reduce gender disparities in health andeducation, IDA assistance has achieved satis-factory results, in part because of significantcountry ownership and effective partnershipswith other development actors, and in partbecause interventions have been underpinnedby rigorous analytical work. In countries with

O v e r v i e w

x x i x

IDA’s comparative advantage in support ofbroad-based growth and poverty reductionlies at the strategic level.

large gender disparities, IDA has integrated gen-der into virtually all education projects, con-tributing to positive trends in girls’ schoolenrollments. It has also contributed to improve-ments in maternal health. IDA assistance, how-ever, has been weaker in promoting women’sparticipation in the economy and in improvingborrowers’ institutional frameworks for gender.These shortcomings have undermined the effi-cacy of the assistance and led to disappointingresults at the country level.

IDA’s implementation of environmentalundertakings has also been partially satisfac-tory. By the end of IDA11, National Environ-mental Action Plans (NEAPs) had beencompleted for nearly all IDA countries, but theirquality has been mixed and their use in CASs hasvaried greatly. To improve the treatment of theenvironment in country assistance programswhere environment is a priority issue, the Bank’s2000 CAS review identifies several “next steps.”Drawn from recommendations in OED, external,and self-evaluations, these steps call for a bet-ter integration of environmental considerationsinto ESW, a strengthening of the link betweenenvironmental considerations and poverty reduc-tion measures, and a better incorporation ofenvironmental indicators and trends in CAS diag-noses of country development issues. Thesesteps are essential to mainstreaming environ-mental considerations into IDA activities, but theyare only first steps.

While lending for environment projectsremains at about the level reached in IDA9, IDAhas increased the number of environmentalcomponents in projects in key sectors. Thisassistance has helped to increase public andgovernmental awareness of environmental issues,contributing to, for example, a reversal of landdegradation, improved incomes from arid lands,and strengthened environmental capacity. More-over, IDA has made efforts to help countriesimprove their environmental assessments,

although failures to implement environmentalsafeguard policies adequately in a few high-visibility projects have drawn public and Bankattention to the continuing need to furtherstrengthen the Bank’s environmental assessmentprocess. Safeguards and mitigation should be aminimum threshold, but not the main thrust ofthe environmental sustainability strategy. Projectsin all sectors need to be designed to ensure envi-ronmental quality and sustainability, but IDA hasyet to provide guidance to define, promote,monitor, or evaluate this broader objective ofmainstreaming.

Lack of country interest in borrowing fromIDA for gender and the environment has beena constraining factor, in part because of theavailability of grant funding from other sources.But IDA has also lacked clear accountability forgender and environmental mainstreaming. Inboth areas, IDA needs to concentrate on strength-ening borrower institutions and policies as amatter of priority. In addition, based on com-prehensive diagnoses, it should better integrategender and environment into CASs, and increasegender and environmental diagnoses in the eco-nomic and social analyses carried out in prepa-ration of IDA-supported projects, especially insituations where gender disparities are high orenvironmental threats serious. IDA also needs toclarify the scope of the Bank’s gender and theenvironmental mainstreaming policy, strengthenthe management for gender and environmentalissues, and establish M&E systems to regularlytrack and periodically evaluate IDA-wideprogress. New strategies to guide gender andenvironmental activities, currently under dis-cussion within the Bank, give IDA an opportu-nity to deal with these matters.

The critical factor of governance. Althoughlargely missing from the Bank’s 1990 povertystrategy, good governance was identified in theIDA10–12 period as a critical factor that affectsall other areas of program emphasis. IDA wasslow to comply with its IDA10 governanceundertakings, but over the past four years it hasgiven priority to public sector reform in coun-try assistance programs and openly addressedissues of corruption. To strengthen its capacity


x x x

Safeguards and mitigation should be aminimum threshold, but not the main thrustof the environmental sustainability strategy.

for expanded operational work, IDA hasincreased staffing, analytical work, and policyguidance. As a result of these efforts, country dia-logue and reporting on governance issues havepicked up, and lending for public sector reformhas grown. In addition, institutional developmentissues are increasingly addressed in lendingoperations and PERs, and more attention isfocused on wider issues of public expendituremanagement, although capacity building foreffective and accountable processes remains amajor task.

But there are shortfalls. The treatment of gov-ernance is not yet consistent across countries.Greater coordination among external agenciesin regard to governance conditionality and assis-tance is needed to make progress on the broadergovernance agenda. More support is needed, forexample, to bolster the rule of law, which is cen-tral to an environment conducive to investment,as well as to alleviate personal insecurity, aprincipal concern of the poor. IDA currentlysupports this work in only a handful of coun-tries, but it does not have the comparative advan-tage to deal with all aspects of judicial and legalreform (including law enforcement). IDA’sstrengthening of public financial accountabilityin borrower countries—a previously neglectedarea—has gained momentum, but it still requiresa considerably increased effort, focused oncapacity building in borrower countries at all lev-els of government. In addition, governance hasbecome a factor in determining CAS lendinglevels, with a “governance discount” applied toallocations, but this mechanism needs rethink-ing to ensure more effective treatment of gov-ernance performance.

The Process DimensionAs it has refocused the development agenda, IDAhas also been at the forefront of change in theaid “business.” IDA’s Replenishment Reportsput particular emphasis on greater selectivity inthe use of aid and on improving partnerships byincreasing participation and coordinating aidwith countries’ own development programs.

In line with replenishment commitments, thisreform agenda has focused on strengtheningfour key IDA processes: enhancing the role of

the CAS in setting program priorities, improvingthe performance-based allocation (PBA) systemto tighten the link between country policy per-formance and lending allocation, increasingstakeholder participation in projects and pro-grams, and strengthening aid coordinationamong donors. While the PRSP initiative hasgiven a significant impetus to these reforms,IDA, together with its development partners,still has a considerable distance to go in con-solidating and institutionalizing improved part-nership practices.

Improving Program and Country Selectivity

Enhancing the role of the CAS. There hasbeen an increasing focus on the country as theunit of account in the design and assessment ofIDA assistance. The CAS—as both a documentand a process—has become the main program-planning vehicle for this shift. Its self-evaluationcontent has improved considerably in recentyears. As a result of more comprehensive diag-nosis and greater participation in their design,CASs have contributed to program relevance,greater country ownership, and better aid coor-dination. As the Bank’s 1998 CAS retrospectivehighlighted, however, at the start of IDA12, lessthan 40 percent of CASs discussed selectivity inkey program areas, and fewer linked IDA’s strat-egy with its comparative advantages, includingprioritizing its activities across and within sec-tors and by instrument. FY00 CASs show someimprovement in program selectivity, whichreflects an increase in attention to this issue bymember governments and management, butprogress remains uneven. With the introduc-tion of PRSPs, the role of the CAS will changeagain. As management has indicated, as of July2002, CASs will normally be based on PRSPswhile retaining their identity as business plansfor the World Bank Group. Still, the challengeof program selectivity will remain. Moreover,consistency now needs to be achieved among

O v e r v i e w

x x x i

There has been an increasing focus on thecountry as the unit of account in the designand assessment of IDA assistance.

criteria for assessing the quality of PRSPs, CASlending triggers, and IDA’s PBA assessments.

Strengthening the system of performance-based allocations. IDA’s PBA system—theprincipal mechanism for achieving country selec-tivity—now better directs credits to countries withgood performance ratings than at the start ofIDA10. This improvement reflects increasedknowledge about the causes of growth andpoverty reduction, as well as specific replen-ishment recommendations, which include giv-ing greater weight to governance, environmentalsustainability, and nondevelopmental expendi-tures in assessing country performance. Sincethere are relatively few top performers (andmost are small economies), the bulk of IDAlending goes to countries in the middle per-formance range. Some shortcomings that remainin the design and implementation of the allo-cation system are related to two key issues:equitable treatment across countries and thestrength of the links between performanceassessment criteria and countries’ continuingpoverty reduction. These issues could beaddressed by rethinking the current “gover-nance discount” methodology, which has failedto capture some borrowers with serious gover-nance problems; further adapting the assess-ment criteria; and increasing transparency anddialogue with partners.2

Building More Effective Partnerships

From ad hoc to structured aid coordination.Few of the program and process objectives ofIDA’s Replenishment Reports can be satisfacto-rily achieved without enhanced aid coordination.During IDA10–12, the focus has moved beyondthe mechanics of donor agency coordination toimproved aid management, preferably led byrecipient countries, with donors directing theirsupport to sound, country-designed policies andprograms. These changes are reflected in suc-

cessive replenishment recommendations, cul-minating in IDA12, with its emphasis on coun-try-led partnerships that combine the objectivesof country ownership and donor coordination.But progress thus far has been limited, anduncoordinated aid programs continue to imposeheavy burdens on recipient countries and limitthe impact of aid programs.

Although progress was modest in IDA10–11,IDA has accelerated its efforts in aid coordina-tion, particularly in the past three years. In-country coordination has been reinforced bythe increased placement of IDA country direc-tors in the field. IDA has given greater empha-sis to harmonizing procurement procedures andevaluation processes among multilateral devel-opment banks. There has been a trend towardcountry-led coordination mechanisms. IDA hasincreased its participation in sector-wideapproaches (SWAps), an aid mechanism thatcombines government leadership in the designof a strategy, an agreed medium-term expendi-ture framework, external assistance providedwithin that strategic framework, and agreedprocesses and indicators for monitoring progresson the ground. These SWAps are one of the fewexamples of formally structured program coor-dination mechanisms focused on both designand implementation. As such they should serveas key building blocks to support recent CDF andPRSP initiatives. Overall, the shift in approachto country ownership and partnership is clear.Consolidation and institutionalization of thisshift, however, require the move from ad hoc tostructured arrangements at the country leveland a greater commitment to harmonization ofpolicies and procedures at the corporate levelby IDA and the other multilateral and bilateralagencies.

Moving beyond project participation. A sec-ond key feature of the move toward more effec-tive partnerships is increasing the participationof borrower country governments, civil society,and the private sector in IDA program design,implementation, and monitoring. The percent-age of IDA projects with at least some primarystakeholder participation nearly doubled from1994 to 2000, reaching 83 percent of all IDA proj-


x x x i i

Few of IDA’s program and process objectivescan be satisfactorily achieved without

enhanced aid coordination.

ects. PAs and PERs have recently become moreparticipatory, although clients remain dissatisfiedwith the extent of local capacity building. Theparticipation of stakeholders in CAS preparationshas also increased, but the impact on CAS designin all but a few cases is unclear. Moreover, themove beyond project participation to participa-tion in analytical work and strategy design hasintensified issues of representation, approach,and costs.

Getting to Results

The Corporate IssuesIDA has accelerated organizational changes sincethe start of IDA10. Its country focus and respon-siveness are stronger, with some 24 IDA coun-try directors now in the field. It has introducedinnovative lending instruments to increase flex-ibility and institutional development impact.Along with these transformations has come astrong focus on improving the quality of IDA per-formance at the project and country levels. Thisemphasis has begun to show significant resultsin improvements in the performance of thelending portfolio. Outcomes of completed proj-ects have risen to 70 percent satisfactory, closeto parity with the International Bank for Recon-struction and Development (IBRD, or WorldBank). The institutional development impactand sustainability of IDA projects have alsoimproved, although from very low levels.

Project-level performance is an important indi-cator of IDA’s contribution to country develop-ment efforts, but it tells only part of the story. Inrecent years, IDA has significantly shifted itsfocus from individual projects to the “higherplane” of country programs. OED CAEs andBank research results suggest that the two mostimportant determinants of country program out-comes are the level of borrower commitment tothe objectives laid out in the CAS and the extentto which the specific components of the strategy,instrument mix, and the efficacy of country pro-gram implementation, including partnerships,are adapted to country circumstances. Reflectingthese multiple factors, the 24 most recent CAEs(most of which cover assistance over the periodof the 1990s) found that the development out-

comes for a third of the country programs werefully satisfactory and the bulk of the remainderwere moderately satisfactory. Both the portfolioand CAE reviews highlight that developmentoutcomes are influenced by exogenous factors,borrower and partner performance, as well asIDA’s own performance. They also show adynamic of improvement over the period. Finally,they point to issues at the corporate level thatneed further work.

Accountability for policy compliance. IDA’spolicy framework remains highly relevant. Buta clarification of the rationale, intent, and scopeof IDA’s policy is needed in certain priority areas,particularly PSD, gender, and the environment.Recently, IDA management has responded toInspection Panel investigations by allocatingmore resources to quality assurance and com-pliance monitoring of safeguards policies. Butmore needs to be done to clarify the assignmentof accountabilities, improve staff training, andrealign staff incentives. Regular monitoring andperiodic evaluation also need strengthening at thecorporate, country, and project levels. Theseissues underscore the importance of rebalancingthe matrix toward the implementation of globalpriorities and strengthening the role of sectorstrategies in setting program goals, articulatingIDA’s role in meeting them, and helping tostrengthen implementation strategies.

Investing in analytical work and capacitybuilding. Broadening the range of lending instru-ments has increased IDA’s flexibility and respon-siveness. Both adaptable lending and new typesof adjustment lending facilitate IDA’s support ofinstitutional reforms and the tailoring of its assis-tance to country circumstances. But achieving fur-ther improvements in program outcomes stilldepends on ensuring adequate, high-quality ana-lytical work and enhancing the capacity-buildingdimensions of IDA’s assistance.

Deploying resources. Cutting across all otherissues is the matter of aligning resources topriorities. Three sets of issues are of particularrelevance. First is the availability of budgetresources for country priority lending and non-

O v e r v i e w

x x x i i i

lending services. As highlighted in the back-ground studies for this report and discussed inBank strategic direction papers and related budgetproposals, administrative resources declined forboth country lending and ESW over the periodof this review. Second, despite the Bank’s povertyreduction mission, country poverty level andperformance factors have been relatively minorconsiderations in budget allocations. Regional andcountry budget allocations do not explicitly dis-tinguish between IDA and IBRD countries. And,while corporate and intraregional decisions havebegun to provide more budget resources topoorer countries, consistent with performance,the impact to date has been small. There are com-plicating factors in linking budget to povertylevels—for example, the differences in operatingcosts among countries. However, poverty couldbe given greater weight in budget allocationsthrough the use of a country norm methodology,an approach that to date has only marginally influ-enced decisions. Third, the programming andbudget system has lacked adequate mechanismsfor reconciling commitments and budget allo-cations. This is a particular problem for CASs,which generally serve as a planning instrumentfor a two- to three-year period but are approvedseparately from annual country budgets. It ispotentially significant, therefore, that in this year’sbudget process Regions are not only carrying outdetailed costing of CASs, but also, for the firsttime, are preparing a three-year rolling budgetto minimize unexpected movements in Regionalfunding.

Recent management proposals. Over theperiod covered by this review, there has beena commendable buildup of management self-evaluation, learning, and proposed new actions,as identified throughout this report. While it istoo early to judge the efficacy and efficiency ofmany of the recent initiatives, their relevance toIDA’s evolving role is clear. In addition, inpreparing for this year’s budget exercise, man-agement has proposed further initiatives, includ-

ing several that are highly consistent with themajor findings and recommendations of thisreview. They are: (a) the establishment of amanagement committee whose principal role isto align corporate strategies, ensure institutionalselectivity, and manage tensions between cor-porate priorities and country programs; (b) thecontinued sharpening of the framework for cor-porate priority setting, and definition of criteriafor selecting those priorities; (c) the implemen-tation of more accurate tracking and manage-ment of nonlending services; (d) the introductionof budget process reforms, as noted above,involving the full funding of CASs within arolling three-year planning and budgeting hori-zon for FY02 and beyond; and (e) the furtherpromotion of decentralization and rebalancingof the management matrix, including additionalclarification of accountabilities.

Could IDA Have Done Better?IDA’s efforts to recast its mission; reach out toengage more broadly with its development part-ners; and reposition its country staff, country pro-grams, and lending instruments were all highlyrelevant adjustments that have contributed sig-nificantly to increasing development effective-ness. Moreover, the recent CDF and PRSPinitiatives provide important potential for furtherimproving program implementation and results.As a consequence, the IDA program at the mid-point of IDA12 is different in important waysfrom the program at the start of IDA10 in FY94.Its portfolio performance has steadily improved,and its contribution to development outcomesin the future has been enhanced.

The impact of IDA’s efforts would likely havebeen greater, however, if it had moved morequickly in the mid-1990s to increase attention togovernance and institutional impediments con-fronting country development, and directedmore of its analysis and dialogue to identifyingways to improve rural and private sector devel-opment and economic opportunities forwomen—all measures needed to stimulatebroad-based, pro-poor growth and povertyreduction. IDA’s impact would also have likelybeen enhanced if it had used its resources moreselectively, not only among countries based on


x x x i v

Cutting across all other issues is the matterof aligning resources to priorities.

demonstrated commitment to poverty reduc-tion, but also in its choice of program approachesand instruments, based on more adequate diag-noses of country circumstances and its owncomparative advantage in specific country con-texts. Finally, IDA’s effectiveness would likelyhave been further enhanced if it had foundways to improve capacity building in borrowercountries in partnership with others, developedthe necessary indicators and monitoringprocesses for a sharper focus on results, and,together with partners, moved considerably fur-ther in coordinating efforts at the country levelto support country-led programs. These chal-lenges have been described in numerous OEDreports, and many of them have been addressedin management’s self-evaluation reports andproposed future actions. The key to improveddevelopment outcomes in the future lies inaligning resources behind responses to thesechallenges in the context of the CDF/PSRPprocess at the country level.

Looking ForwardThe IDA10–12 review period has witnessed con-siderable change—in borrower countries, inIDA, and in the development system. TheIDA10–12 policy framework has encouragedIDA to move in directions that are relevant tothe needs of its borrowers and that remainbroadly germane for the future. But IDA and itspartners are still in the process of implementingthe core elements of the framework, reinforcedby new initiatives that were introduced duringIDA12. IDA now needs to deepen and broadenthe gains from the existing policy framework by: • Focusing on implementation• Adequately aligning resources to strategic

priorities• Consolidating the IDA mandates.

To achieve these goals, IDA will need toadjust further its role at both the country and theglobal levels, within the limitations set byresources available for its lending and non-lending services. In its primary role of support-ing country-based development efforts, IDAshould concentrate more on its areas of com-parative advantage, which, for the most part, lie

at the strategic level in supporting economy-wide, sector-wide, and government-widereforms, and on capacity building. At the sametime, IDA should play a more proactive role atthe global level to facilitate achieving the har-monization and coordination of external assis-tance needed to make greater progress towardthe overarching goal of poverty reduction.

Next StepsIDA could take several steps to build on theimprovements of the past seven years in thedevelopment effectiveness of its country andglobal roles: • Be more selective. IDA needs to do more to

increase its country, program, and corporateselectivity. This calls for further improvingIDA’s PBA system and ensuring consistencyamong the PRSP, CAS, and PBA processes. Italso entails strengthening sector strategies andbetter integrating them into CASs, based oncountries’ poverty reduction strategies, to helpdetermine where IDA should take a leading ora supporting role in coordination with otherdonors. IDA also needs to clarify its prioritiesand objectives across sectors and themes toidentify more clearly what it can (and cannot)commit to do. Because this corporate-levelselectivity cannot be determined effectivelyin isolation from other agencies’ strategic deci-sions, IDA needs to become part of a broad,agency-wide harmonization action plan. Inpractice, quantitative lending targets for sec-tors, lending instruments, or groups of coun-tries can serve as disincentives to greaterprogrammatic and country selectivity.

• Focus on governance and capacity build-ing. Every area of this review highlights thecentrality of governance and public sectorcapacity building, with a view toward encour-aging borrower governments to provide pub-lic goods and services more efficiently and tobe more transparent and accountable to their

O v e r v i e w

x x x v

IDA should play a more proactive role atthe global level to facilitate achieving theharmonization and coordination of externalassistance needed.

publics. To encourage more effective gov-ernment and improved program implemen-tation, every aspect of IDA’s assistance—fromanalytical work and design and supervisionof operations to monitoring for results—should have a capacity-building dimension.This may require expanding IDA grant financ-ing; it may also require instituting much morestructured coordination with grant-givingagencies. While the governance agenda setout in the IDA10–12 Replenishment Reportsremains relevant, recent experience suggeststhat IDA, its borrowers, and other develop-ment partners need to make a special com-mitment to enhancing public expendituremanagement and financial accountability.IDA should support, and link its future lend-ing to, the commitments of its borrowers toestablish time-bound plans for sound publicfinancial accountability systems.

• Commit to effective development part-nerships. IDA should work at the country andglobal levels to move to a new developmentarchitecture, including a shift from ad hoc tostructured arrangements of aid coordination.One potentially important way forward is touse the PRSP as a concrete way to apply theCDF principles. The implementation of thePSRP initiative is a priority for IDA12 andbeyond. Within this context, another priorityshould be to strengthen and broaden theapplication of sector-wide programs at thecountry level, following recent lessons onhow to make these effective, results-oriented,donor-coordinated approaches supportcountry-owned programs. At the global level,IDA donors should ask the Bank’s presidentto facilitate a time-bound process involvingthe heads of bilateral and multilateral assis-tance agencies to achieve greater harmo-nization in aid practices. The agenda shouldgo beyond procedural issues to strategic con-cerns of program and country selectivity.While IDA cannot and should not take thelead in all aid coordination efforts, it can fur-ther adjust its own processes, and encourageothers to do the same, to achieve the muchgreater coordination needed to reduce today’shigh aid transaction costs.

• Clarify program objectives and policies.IDA should articulate what it means by its“focus on poverty reduction” and clarify itsprogram objectives and policies in relation tothat goal. The record of the past seven yearsshows that it needs to accelerate broad-basedgrowth and governance reforms as key inter-mediate objectives of its overarching goal ofpoverty reduction. It also needs to link theobjectives and operational policies of thecrosscutting themes of gender, environment,and PSD directly to poverty reduction and toclarify the policy of gender and environ-mental mainstreaming.

• Better align resources with program pri-orities. Finally, in applying the lessons of thepast seven years, it is essential that IDA bet-ter deploy its resources relative to its com-mitments through its budget process and fullfunding of CASs. Poverty has been a relativelyminor factor in budget allocations, althoughthis could alter with the further application ofa country norm–based methodology. Effortsto cost CASs more realistically should beaccompanied by a process that ensures fullfunding of board-approved strategies. Also,ways are still needed to ensure funding ofappropriate levels of due diligence andprogram-specific ESW and to ensure a greaterfocus on capacity building.

The Replenishment ProcessThe findings of this review suggest that IDA’sreplenishment process itself might usefullybe fine-tuned. It offers a valuable opportunityto discuss, at a global level, the experienceand future direction of international devel-opment assistance and to mobilize funding tosupport key objectives. As currently consti-tuted, the process falls short of fully realizingthis potential.

Although consistent with an evolving con-sensus in the international development com-munity, replenishment undertakings have beenboth overdetermined and overloaded (for exam-ple, in specifying the shares of lending for sec-tors and countries and in the number and detailof the recommendations). They have sometimesbeen unrealistic about what IDA and its bor-


x x x v i

O v e r v i e w

x x x v i i

rowers could reasonably accomplish in a three-year period, having focused on inputs rather thanresults, without addressing budget resources.More important, IDA’s replenishment process hasbeen disconnected from its development part-ners, both borrowers and other assistance agen-cies, to the detriment of setting priorities andapplying its comparative advantage. Threechanges could improve the process and, thus,the impact of future undertakings. • Develop a long-term vision focused on

results. As part of the IDA13 negotiations,management should be asked to develop, inconsultation with borrowers, a long-term visionfor IDA, clarifying what is meant by IDA’spoverty focus, identifying ways to deploy IDA’sresources, and taking advantage of IDA’s globalrole as a complement to its country focus.

• Engage developing countries in settingreplenishment commitments. The far moretransparent and consultative process plannedfor the IDA13 negotiations goes well beyondthe measures taken in IDA12 and should helpto set achievable IDA goals and improve imple-mentation. Beyond the negotiations, IDA

should consider how to engage its partners inmonitoring and evaluating IDA13 perform-ance and results.

• Define commitments in terms of moni-torable and achievable objectives, withrealistic costing. IDA’s replenishment under-takings have tended to emphasize inputs and,too often, unrealistic output targets that do nottake adequate account of the need for coun-try ownership if reform efforts are to be sus-tained. It would be better for ReplenishmentReports to focus on development goals and onthe kinds of priority activities that IDA shouldsupport in advancing those goals, whileaccounting for IDA’s comparative advantageand reflecting realistic costings of replenishmentcommitments. In addition, ReplenishmentReports ought to agree on IDA performanceindicators and on a process of progress report-ing, recognizing that operating within CDFprinciples means moving even further awayfrom directly measuring the link between IDAactivities and country development outcomesto new ways of tracking IDA’s performance inpartnership with others.

1

The IDA Review andIts Context

Part I

The mission of IDA is to support efficient and effective programs to reduce povertyand improve the quality of life in its poorest member countries…. To achieve this,the focus must be on: sustainability—to achieve enduring development impact withan environmentally sustainable framework; and equity—to remove barriers andopen up opportunities for the disadvantaged.

—IDA12 Replenishment Report (p. iii)

3

Introduction

The International Development Association (IDA) is a unique instru-ment of development cooperation. Its establishment in 1960 markedan unprecedented commitment by the international community to

improve global welfare. IDA has made substantial and distinctive contributionsto growth and poverty reduction in low-income countries through conces-sional finance, global knowledge, and aid coordination services. The largestsource of concessional finance to low-income countries, IDA has providedan average of $6.45 billion annually (in constant 1995 dollars) to some 77eligible borrower countries through the 1990s.

The “development crusade” that gave rise toIDA has reached a new consensus in recentyears focused on poverty reduction and strongerdevelopment partnerships. While IDA’s basicfeatures—as a separately funded but integralcomponent of the World Bank Group—haveremained much the same, its support for coun-try development has broadened in line with thisevolving consensus and the changing circum-stances confronting its borrowers.1

Has IDA changed fast enough in recent yearsand in the right directions? Has it acted respon-sibly to implement its commitments? As a result,is it well equipped to assist its borrowers in anera of rapid technological change, borderless pri-vate investment, and an increasingly fragile nat-ural environment? These questions underlie themandate for this review as set out in the IDA12

replenishment agreement between IDA and itsdonors.2

IDA10–12 Replenishment UndertakingsBefore IDA10, agreements between IDA and itsdonors focused largely on resource concerns(size of the replenishment, eligibility criteria forborrowers) and included general endorsementof IDA’s ongoing and planned activities. Theextensive policy content of the IDA10–12 replen-ishment undertakings (IDA10, FY94–96; IDA11,FY97–99; and the first year of IDA12, FY00)makes this review of IDA’s compliance andeffectiveness especially important (see box 1.1).3

The IDA10–12 agreements called for majorprogram transformation to strengthen the povertyimpact of IDA’s assistance. In broad terms, theIDA10–12 accords endorse poverty reduction

11

as IDA’s overarching objective and instruct IDAto:• Sharpen the poverty focus of its country assis-

tance programs. • Direct assistance to expanding access to basic

social services, fostering broad-based eco-nomic growth through private sector devel-opment (PSD), integrating gender andenvironmental considerations into all activi-ties, and promoting good governance.

• Increase operational effectiveness by more selec-tive use of resources, more participatory designof country assistance programs, and better coor-dination with other development agencies.

These issues form the agenda of this assess-ment of IDA’s compliance.4 Overall, they arehighly relevant to the development challengesfacing IDA borrowers. The comprehensiveapproach to poverty reduction and higher-quality aid processes that they embody reflectaccumulated lessons of development experi-ence (see Dollar and Prichett 2000; Kanbur andSandler 1999; OED 1997a, 1998a, 1999a, 2000a;Tarp 2000). Individually, each of these areas ofemphasis makes sense. Together, however, theyhave proved extraordinarily demanding for IDAand its borrowers.

IDA’s Role in a Changing DevelopmentEnvironmentThrough the 1990s, total official developmentassistance to IDA-eligible countries amounted toabout 8 percent of their combined gross domes-tic product (GDP). Of that amount, IDA’s por-

tion accounted for some 14–18 percent. IDA’sassistance, therefore, cannot determine thechoices that governments make, although it cansupport and influence those choices. Nor canIDA alone be held accountable for countrydevelopment outcomes. The performance ofborrowers and other development partners, aswell as changes in the global economic envi-ronment, all influence those outcomes. IDA canand should be held accountable, however, forhow it deploys its resources in fulfilling its cor-porate commitments. (Did IDA do the rightthings, taking account of others?) It should alsobe held accountable for how effectively it shapesits assistance to meet borrowers’ priorities andcircumstances. (Did it do things right?) It is inthese terms that this review examines IDA’sefforts over the past seven years.

Trends in development and aid. The periodcovered by this review is very much a transitionalone—in both country development and inter-national aid. At the start of IDA10, many devel-oping countries were plagued by slow ornegative growth, major economic dislocations,and rising official debt. Internal conflicts wereon the upswing, and the political liberalizationthat has come to mark the period was newlyunder way. Progress since then has not beenbroad or deep enough to significantly reducepoverty in most IDA countries. Although theshare of the population living on less than a dol-lar a day declined during the 1990s, the num-ber of poor people has remained roughlyconstant and inequality has worsened. But both


4

The IDA12 agreement called for “a review of the IDA program during the IDA10–11 period and an interim review of IDA12, includ-ing performance in implementing the recommendations of the Deputies set out in each of these Replenishment Reports.”

The specific undertakings related to these broad instructions center on: Six program priorities, and• Poverty reduction• Environmental sustainability• Gender• Governance• Private sector development• Social sector development

Four processes:• Aid coordination• Enhanced Country Assistance Strategy (CAS) design and

implementation• Participation• Performance-based allocations.

T h e I D A 1 0 – 1 2 R e p l e n i s h m e n t U n d e r t a k i n g sB o x 1 . 1

slow economic growth and conflict have beenmajor contributing factors to these trends (seebox 1.2 and Annex B for details on poverty,growth, and social trends). Governance problemsand incomplete social and economic reformshave also been serious impediments. Behindthese aggregate figures, however, stand a grow-ing number of borrower countries that haveimproved their economic management, increasedtheir rates of economic growth, and strengthenedtheir social development with IDA’s help (seeWorld Bank 2000d, 2001b for diversity withincountries). Some have also begun to tackle gov-ernance and public sector reforms.

In parallel, the development community hasreached a new consensus on aid policies andpractices, including a sharper focus on poverty,partnership, and policy performance. The paceof improvement in aid practices remains slow,but there is evident movement in useful direc-tions, reflecting the accumulated lessons ofdevelopment experience (see box 1.3). Thesepromising, though incomplete, transformationsrequire the deepening and broadening of reformsby both donors and recipients.

This evolving development agenda has poseddifficult challenges for countries with limitedresources and institutional capacities. It is nowwidely acknowledged that external assistance, tobe effectively implemented and sustained, needsto be directed to countries’ own strategies and pro-grams for change. Building consensus, aligningresources in support of major social and eco-

nomic reforms, and implementing complex reformprograms are formidable tasks—even where gov-ernment commitments to reform are clear.

An overview of IDA’s assistance. IDA hasbeen at the forefront of much of the renewal andchange in this period. • Its financial assistance totaled $42.3 billion in

the period. Assistance remained fairly constantthrough IDA10 and 11 and, after a drop in thefirst year of IDA12, is expected to increase.

• The bulk of that assistance went to the twoRegions with the largest number of poor peo-ple: Sub-Saharan Africa, which received 38.3percent of IDA commitments, and South Asia,which received 28.2 percent. (See box 1.4 forthree country examples of IDA’s assistance.)

• IDA also provided assistance for the unprece-dented challenges of countries in transitionand devised new approaches for assistingcountries emerging from conflict.

• Continuing a shift begun earlier in the decade,the largest share of IDA’s lending supportedthe efforts of countries to expand access tobasic social services. Another substantial sharewent to infrastructure development and

I n t r o d u c t i o n

5

Some 1.2 billion people—one of every five—live on less than a dollar a day. The share of the population living on less than a dol-lar a day fell slowly in developing countries during the 1990s, from 28 percent in 1987 to 23 percent in 1998, while population growthheld the number of poor people roughly constant. Social indicators improved more, particularly for health and education, but theaggregate gains are heavily influenced by rapid advances in China, which at the start of IDA10 accounted for a quarter of the world’spoor. (See Annex B for more detail on poverty and social indicators.)

Slow growth explains a large part of the weak poverty outcomes: poverty generally declined in countries whose economiesgrew rapidly and increased in those whose economies stagnated or contracted (Ravallion 2000; World Bank 2001b). The overalldecline in extreme poverty during the 1990s was driven by high rates of growth in countries with large numbers of poor people—although increasing inequality is now slowing the rate of poverty reduction in several countries.

Civil conflict, which afflicted 17 IDA countries in the 1990s, also explains part of the story. In Sub-Saharan Africa, for exam-ple, countries with better policy environments achieved average GDP gains of 5.2 percent during the 1990s, while countries expe-riencing civil strife or major political disruptions registered only 0.2 percent growth.

P o v e r t y , G r o w t h , a n d C o n f l i c tB o x 1 . 2

A growing number of borrower countrieshave improved their economic management,increased their rates of economic growth, andstrengthened their social development withIDA’s help.

improved economic management—areasessential to sustainable growth (see Annex Cfor IDA commitments and disbursements byRegion and sector).

• In support of its broadening developmentagenda, IDA introduced new lending instru-ments and program approaches to increaseits responsiveness, flexibility, and attention topolicy and institutional impediments in bor-rowing countries. It deepened its countryfocus by strengthening the role of CountryAssistance Strategies (CASs), increasing theprogram and budget authority of countrydirectors, and decentralizing staff to countryoffices. In addition, it has tightened the linkbetween country policy and institutional per-formance and levels of lending.

• Three major new initiatives involving a cen-tral role for IDA—the enhanced HeavilyIndebted Poor Country (HIPC) debt reliefinitiative, the Comprehensive DevelopmentFramework (CDF), and the Poverty ReductionStrategy Papers (PRSP)—reinforce therefocusing of the development agenda (seebox 1.5).

IDA’s performance. Along with these trans-formations in IDA’s way of doing business hascome a strong focus on improving the quality

of IDA’s performance at the project, country, andsector levels. This emphasis has begun to showsignificant results in improving the performanceof the lending portfolio (see figure 1.1). OEDevaluations of completed projects exiting theportfolio in the first half of FY00 suggest thatmore than 75 percent of IDA projects can berated satisfactory, reflecting improvements inIDA and borrower performance.5 Data from theBank’s Quality Assurance Group (QAG)6 confirmimprovements in the quality of the ongoingportfolio and a remarkable convergence in thequality of project preparation and supervision forIDA and IBRD countries.7

These gains have been achieved even thoughprojects are becoming increasingly complex anddemanding. Over 85 percent of IDA projectsevaluated since 1997 were substantially demand-ing for the borrower, and three-quarters weresubstantially complex. While this reflects, inpart, the broader scope of the reforms countriesare undertaking, projects that are too complexand demanding may overwhelm borrowercapacity. More than one in three IDA projectsshow unsatisfactory borrower implementation,including nearly half of the projects in Africa. Fur-ther progress will require increased capacitybuilding and further adaptation of assistance toborrower circumstances (OEDa).


6

Flows of official development assistance (ODA grants and loans with a grant component of 25 percent or more) to IDA countriesexhibited signs of fatigue through IDA10, declining from $50.0 billion in 1994 to $37.4 billion in 1997. A slight increase since thenhas brought ODA flows to $41.6 billion in 2000, still short of the decade’s earlier levels. In addition, key changes in aid programsover the past seven years have included: • An increase in the share of aid going to poor countries with improved policy performance ratings (Collier and Dollar 1999) • A more concentrated focus on poverty reduction as the objective of assistance programsa

• More explicit attention to matters of governance in both aid allocations and programs • Greater emphasis on country ownership of externally funded activities, and innovations in the provision of aid in support of country-

led efforts• Heightened attention to the still elusive objective of aid coordination • New interest among the international development community in the provision of global public goods to advance the fight against

poverty. These changes—in combination with improving country policies—hold promise for improved development outcomes in the

future. Still, greater strides are needed in both country economic and institutional reforms and aid coherence and coordinationamong assistance agencies.

a. As indicated in the recent articulation of international development goals and the linking of the enhanced debt reduction initiative with countries’ commitments toimproved poverty reduction and social sector expenditures.

T r e n d s i n O f f i c i a l D e v e l o p m e n t A s s i s t a n c eB o x 1 . 3

There is also some improvement in the sus-tainability (resiliency to risk) and institutionaldevelopment impact (the contribution of theproject to capacity building) of completed proj-ects.8 The share of IDA projects that substantiallymet their institutional development objectivesrose from less than 30 percent of projects dur-ing IDA9 and IDA10 to 35 percent since thebeginning of IDA11. Projects judged likely to besustained have increased to nearly half, up froma third during IDA9. But long-standing problemswith sustainability and institutional develop-ment continue to limit impact, more for IDA thanfor IBRD projects. (See Annex D for detailed

information on portfolio ratings of completed andongoing projects.)

Project-level performance is an importantindicator of IDA’s contribution to country devel-opment efforts, but it tells only part of the story.In recent years, IDA has significantly shifted itsfocus from individual projects to the “higherplane” of country programs. Evaluation andresearch results suggest that the two most impor-tant determinants of country program outcomesare the level of borrower commitment to theobjectives laid out in the CAS and the extent towhich the specific components of the strategy,instrument mix, and the efficacy of country


7

Mozambique, which became a member of the World Bank in 1984,emerged from conflict in 1992 and has since made significantand accelerating advances in implementing market-based eco-nomic policies, boasting one of the strongest privatization pro-grams in Africa. Following a period of post-conflict assistance,IDA’s current portfolio comprises 16 projects estimated at $737.3million in commitments. Recent projects include a fast-trackFlood Emergency Recovery operation, Railway and Port Restruc-turing, and efforts to support the government’s HIV/AIDS plan.Poverty reduction is the overall goal for the government and IDAin the country. The 2000 CAS, which supports the government’sFive-Year Strategy and the Interim Poverty Reduction StrategyPaper (IPRSP), focuses on three core objectives: (1) increasingeconomic opportunities, (2) improving governance and empow-erment, and (3) improving human capabilities. Donor assistanceplays a vital role in supporting the country’s economy andaccounts for approximately $600 million per year—more than 15percent of GDP. A Consultative Group (CG), for which IDA ischair, is the main coordinating mechanism for donor activities.IDA is a partner in sector-wide programs (which involve anagreed sector program and coordinated funding among donors)for health, education, transportation, and agriculture.

Bangladesh joined the World Bank in 1972. IDA has financedmore than 182 operations in the country, with loans totaling morethan $9.4 billion. IDA’s active lending portfolio for Bangladeshincludes 26 projects, valued at roughly $2.6 billion. A long-standingpopulation program has substantially contributed to increases infamily planning. Strong support in promotion of girls’ educationhas helped to boost secondary school enrollment rates. Effectivecooperation with the government, IDA, the African DevelopmentBank (AfDB), and the Japanese Overseas Economic CooperationFund has funded the Jamuna Bridge, which connects the north andsouth of the country. Efforts to build effective institutions under-

pin the current strategy and core objectives and include (a)improving macroeconomic management; (b) promoting privatesector–led growth; (c) reforming public sector management; (d)accelerating agricultural growth and rural development; and (e)promoting faster human development. IDA plays a key coordinatingrole by chairing the annual CG meeting, with local CGs focusedon specific sectors meeting periodically throughout the year. IDAalso supports a sector-wide approach (SWAp) in health that hasincreased access to health services, and donors are now work-ing together with the government to improve the quality of healthservices in the country. IDA’s lending levels in Bangladesh havebeen constrained in recent years due to absorptive capacity con-straints, slow progress on structural reforms, and problems of gov-ernance and weak institutions.

Ghana joined the World Bank in 1957 and IDA has sincefinanced some 94 operations in the country, accounting for morethan $3.5 billion in lending. The active lending portfolio of 23projects is one of the largest in Africa. Recent operations haveincluded a second Economic Reform Support Operation and Adapt-able Program Loans (APLs) for Community Water Supplies, UrbanDevelopment, and Agricultural Services. IDA’s current CAS issupportive of Ghana Vision 2020, the national strategy for reduc-ing hard-core poverty and achieving broad-based economic devel-opment. The CAS has two parts: Part I, which sets out thegovernment’s poverty reduction strategy, and Part II, which pre-sents IDA’s business strategy. External assistance included some$1.7 billion in commitments for 2000–01. The government has beenstrongly involved in aid coordination, and Ghana is credited withbeing one of the two Comprehensive Development Framework(CDF) pilot countries that have made the most overall progress infostering country-led, long-term development partnerships. Thewidely regarded, ongoing health SWAp currently involves 17donors and is characterized by strong government ownership.

C o u n t r y E x a m p l e s o f I D A ’ s A s s i s t a n c eB o x 1 . 4

program implementation, including partnerships,are adapted to country circumstances.9

On a Bank-wide basis, country program out-comes are not strongly associated with countryincome. Moreover, it appears that with skillfulstrategy design and effective Bank and partner per-formance, a country strategy can be relevant anda country program outcome can be satisfactoryeven in countries with weak policies or inadequateinstitutions. What is important, in addition to a gov-ernment’s commitment to reform and capacitydevelopment, is the extent to which IDA’s assis-tance strategy, projects, processes, and partnershipsare judiciously adapted to the country setting.

More specifically, as reported in the OEDAnnual Review of Development Effectiveness2000, reviews of completed OED Country Assis-

tance Evaluations (CAEs) identify several keydeterminants linking country-level program out-come with IDA performance:• Diagnosis. IDA’s diagnosis of major devel-

opment issues and constraints—what needsto be done—is generally good, but the qual-ity of diagnoses varies by issue and sector.Analysis of macroeconomic issues and con-straints is usually strong, but links betweenstabilization and growth are not always artic-ulated; and the overall impact of PublicExpenditure Reviews (PERs) and PovertyAssessments (PAs) is strongly correlated withtheir quality, which has been highly varied inthe past, although improving in recent years.

• Focus on poverty. As discussed earlier, IDA hassharpened the poverty orientation of its coun-


8

The Heavily Indebted Poor Country Initiative:The HIPC debt relief initiative was proposed by the World Bankand IMF and endorsed by 180 governments in the fall of 1996. Itwas the first comprehensive international effort to reduce theexternal debt—including multilateral debt—of the world’s poor-est and most indebted countries. A major review in 1999 producedan enhanced HIPC initiative that is “deeper, broader, and faster”than the original framework. As of February 2001, 22 countrieshad reached their decision point under the enhanced HIPC andare now receiving debt service that will amount to about $34 bil-lion over time, or a reduction of $20 billion in net present valueof their outstanding debt. In the long run, the policy implicationsof the HIPC initiative, which includes placing debt relief withinthe overall framework of poverty reduction and improving thepolicy dialogue between the poorest countries and their multi-lateral creditors, may be as important as the initiative’s imme-diate fiscal impacts.

Comprehensive Development Framework:Introduced by World Bank President James Wolfensohn in Jan-uary 1999, the CDF is a holistic approach to development, whichaims to balance macroeconomic concerns with social andstructural development requirements. The CDF is anchored in fourinterrelated principles that guide development assistance on thecountry level: a long-term holistic vision and strategy; enhancedcountry ownership of development goals and action; more strate-gic partnership among stakeholders; and accountability fordevelopment results. The CDF approach relies on country lead-ership, with development partners working selectively accord-

ing to comparative advantage. Since March 1999, the CDFapproach has been piloted in the West Bank and Gaza and in thefollowing 11 countries: Bolivia, Côte d’Ivoire, the DominicanRepublic, Eritrea, Ethiopia, Ghana, the Kyrgyz Republic, Morocco,Romania, Uganda, and Vietnam. The September 2000 progressreport found that 9 of the CDF pilot countries demonstrate goodprogress in determining a long-term vision and strategy, thoughoverall progress in fostering the development partnership envi-sioned by the CDF varies by country. The Bank’s current Strate-gic Framework puts the CDF at the center of the institution’sstrategic orientation for the coming years.

Poverty Reduction Strategy Papers (PRSP) Program:In September 1999, the Development and Interim Committees ofthe World Bank and the IMF endorsed PRSPs as the new frame-work for Bank and Fund efforts to achieve sustainable povertyreduction (World Bank 2001c). PRSPs have become the basis forconcessional lending from the World Bank and the IMF and fordebt relief under the enhanced HIPC; they are a key mechanismfor improving the policy performance of the poorest countriesthrough an emphasis on clearly defined, monitorable objectives.Built on the four CDF principles, PRSPs are to be country-ownedand grounded in participatory processes and should serve as theframework for other donor assistance. The PRSP benefits fromstrong coordination between the Bank and the Fund (with theBank taking the lead on the social and structural frameworkand the Fund leading on the macroeconomic framework) thatincludes oversight from a Joint Implementation Committee andjoint assessments of PRSPs by staff from both organizations.

T h r e e K e y I n i t i a t i v e s o f t h e L a t e 1 9 9 0 sB o x 1 . 5

try programs. Economic growth, a necessarybut not sufficient condition for poverty reduc-tion, is at the center of nearly all countrystrategies, as is the provision of basic socialservices. However, OED analysis and Bankpolicy research suggest that equity affectsthe poverty impact of growth and that safetynets and empowerment of the poor deserveincreased attention. More needs to be doneto integrate broad-based poverty reductionstrategies into macroeconomic and sectorstrategies and interventions. Also, inadequateattention by IDA and other donors to povertymonitoring and evaluation (M&E) has meantthat significant knowledge gaps persist regard-ing linkages between policies and programscommonly supported by IDA and povertyoutcomes.

• Balancing corporate and country priorities.IDA has improved its client focus. But clientownership does not eliminate divergencesbetween IDA and country priorities, nor con-flicting views on “right” policies. OED eval-uations have identified inadequate borrowercommitment as the most important reason forpoor policy or program implementation andhave emphasized the level of country con-

sensus about reform as critical to IDA pro-gram outcomes. CAEs have pointed to numer-ous past examples of lending to promotereforms, either through adjustment or invest-ment lending, before enough consensus hasemerged, thus incurring avoidable develop-ment risks.

• Assessing ownership. IDA and governmentsmay agree on broad objectives, but may dis-agree on the measures and timetables toachieve them. Concrete actions and trackrecords are better indicators of ownershipthan statements of intent. Instruments forassessing ownership are coming into wider use(and lending instruments evolving to better cal-ibrate with their findings), but such assess-ments are still not systematically mainstreamedinto project and program preparation.

• Diagnosing institutional development. IDA’scontribution to institutional development atthe country level has been modest, reflectingthe inherent difficulties of institutional changeand capacity building. IDA’s diagnosis ofinstitutional, political, and governance con-straints has traditionally been weaker than itseconomic and technical diagnoses. Improve-ments have recently begun to show at the


9

I m p r o v e m e n t s i n P r o j e c t O u t c o m e s ,I n s t i t u t i o n a l D e v e l o p m e n t , a n dS u s t a i n a b i l i t y i n I D A / B l e n d C o u n t r i e s

F i g u r e 1 . 1

IDA9 (1991–93) IDA10 (1994–96) IDA11+ (1997–00)

0

10

20

30

40

50

60

70

80

90

100

Perc

enta

ge

IDA Replenishment Periods

ID Impact(% substantial)

Sustainability(% likely)

Outcome(% satisfactory)

country level, but further progress is neededto articulate comprehensive institutional andgovernance strategies.

• Estimating capabilities. Several CAEs foundpatterns of over-optimism in IDA’s estimatesof the receptivity of borrower governmentsto IDA’s advice, their willingness to undertakedifficult reforms, and their capacity to imple-ment reform measures. In particular, pro-grams have often underestimated the politicaldifficulties and obstacles to rapid change oflong-term socioeconomic systems, especiallyin transition countries.

• Being selective. At the country level, selectiv-ity requires identifying interventions that areconsistent with IDA’s comparative advantagerelative to its partners. CAE reviews and theBank’s two recent CAS retrospectives suggestthat IDA is getting better at establishing pri-orities for country programs, but still has roomfor improvement. Justifying IDA’s presence inmany sectors by its convening role and capac-ity for policy analysis may underestimate thecapacity of other actors to take the lead in pro-gram interventions, or fail to take into accountthe need to use IDA administrative resourcesselectively to achieve results.

The remainder of this report focuses on IDA’sperformance in the aggregate on its sector andthematic replenishment undertakings, as well ason related process reforms. It draws on a widerange of OED country and sector evaluations andother sources to do this.

Overall, IDA has made major strides in recast-ing its mission—to make poverty reduction itsoverarching objective—and in redirecting itsoperations in line with its relevant but highlyambitious replenishment commitments. Imple-mentation of those commitments has been, onthe whole, satisfactory, with momentum havingincreased in IDA11. IDA has done much tosharpen the poverty focus of its analytical work,

policy dialogue, and lending. It has helped coun-tries lay foundations for renewed growth andaccelerated poverty reduction through improvedeconomic management and increased invest-ments in basic social services. In recent years ithas also brought governance to the fore as a crit-ical development issue. Compliance remainsuneven, however, across and within areas ofprogram and process emphasis, with limitedprogress in integrating PSD, gender, and envi-ronmental sustainability into country assistanceprograms.

Assessed against its ambitious goals, the devel-opment outcomes of IDA’s programs have beenpartially satisfactory.10 Although IDA has done wellin helping countries to lay foundations for eco-nomic growth and poverty reduction, the recordof IDA countries in sustaining growth at highenough levels, over long enough periods, and bymeasures adequate to benefit the poor is mixed.This reflects a host of non-IDA influences ondevelopment outcomes, including regional con-flicts, weak domestic capacities, mixed policyperformance, variable aid policies, and exogenousshocks. This said, IDA’s own performance, whilehaving significantly improved over the reviewperiod, could be enhanced still further and con-tribute even more positively to development out-comes. It needs to clarify corporate objectives andaccountabilities for implementing them, espe-cially in some key, crosscutting areas. It needs stillgreater selectivity in resource use among coun-tries and better prioritization of actions linked topoverty reduction within countries. Some famil-iar implementation challenges persist as well,especially in the area of institutional development,and the alignment of resources to program pri-orities remains incomplete.

Reviewing IDA’s PerformanceThis is the first independent report on the IDAprogram. Previous reports have been preparedby IDA management and have provided impor-tant background information for this review.11

The methods used by the OED review aredescribed in Annex G.

Questions. In evaluating IDA’s performance, thisreview asks four main questions:


1 0

IDA has made major strides in recasting itsmission—to make poverty reduction its

overarching objective.

• How has IDA responded to the recommen-dations set out in the IDA10–12 Replenish-ment Reports?

• Have these responses advanced core objec-tives at the country level, especially the over-arching objective of poverty reduction?

• Which factors have contributed to IDA’s per-formance and which have constrained it?

• What lessons can be drawn from the ex-perience of recent years in the major areas ofemphasis?

Challenges. These questions present severalchallenges. First, since neither the ReplenishmentReports nor follow-on instructions from man-agement provided specific compliance bench-marks,12 the review addresses whether IDA’sactions were consistent with the letter and thespirit of the recommendations. It looks at howwell IDA acted and institutionalized changes inits processes and programs. For example, on thecommitment to making poverty the overarchingobjective, the review examines what IDA hasdone to sharpen the poverty focus of its analyticalwork and lending and to link other programobjectives, such as PSD and environmental sus-tainability, to that goal. The review assesseswhether IDA has made its processes (such asresource allocations) more poverty focused.Because of the proliferation of instructions andthe absence of benchmarks, however, it hasbeen difficult for IDA to be consistent and per-suasive about its pursuit of objectives. Futurereplenishment undertakings would benefit fromgreater clarity and selectivity, as well as frommore specificity in management’s implementa-tion instructions.

Second, compliance is not the same as effec-tiveness. IDA’s performance on the replenish-ment recommendations (stated largely as inputsand outputs rather than outcomes) is not nec-essarily a good measure of its developmenteffectiveness. Because of the review’s time frame(FY94–00), the bulk of actions being assessed arestill under way, and their effects will continueover the medium to long term.13 The review,therefore, examines IDA’s development effec-tiveness largely in terms of the quality of itsunderlying analysis, the coherence of its program

at the sectoral (or thematic) and country levels,the fit of lending and nonlending services tocountry circumstances, the strategic selectivity ofresource allocations and choice of instruments,and, as far as possible, the impact on country poli-cies, institutions, and actions. The limited infor-mation on outcomes and results of IDA’s actions(even for a longer period than that covered bythis review) underscores the need for M&E at theproject, country, and corporate levels and a moreresults-oriented set of IDA commitments.

Third, many factors bear on outcomes in acountry context, making attribution difficult. Italso considered the influence of both internal fac-tors (such as institutional practices, accountabilitystructures, and alignment of resources with pro-gram priorities) and external events. Whereprogress has been less than expected, the reviewlooked at whether strategy, implementation, ora combination of the two contributed to theshortfall.

Approach. This review is based on major back-ground studies of each of the 10 program areasand process reforms emphasized in theIDA10–12 Replenishment Reports. It also drawson the findings from in-country consultations(with representatives of government, civil soci-ety, the private sector, and other assistance agen-cies) in nine “focus” countries; two internationalworkshops that included experts from borrowerand donor countries; and wide-ranging inter-views with Bank management and staff.14

The background studies and this report arebased on desk reviews of Bank documents,databases, studies, and evaluations;15 staff sur-veys; and country and international consultations.When evidence was available, the review useda results-based analysis to trace IDA inputs (poli-cies, alignment of resources), outputs (volumeand composition of lending and nonlendingservices), and reach (stakeholder participation,coordination with other donors) to outcomes andresults. It analyzed the quality of IDA’s lendingand nonlending services in terms of their rele-vance and fit, as well as evidence of outcomeperformance of ongoing and closed projects.Also, using the focus country consultations andOED CAEs, the review examined IDA’s


1 1

performance, taking account of its partners’performance.

In the following four sections, this evaluationexamines how well IDA has performed in imple-menting its IDA10–12 replenishment commit-ments within the context of changing globaldevelopment and aid trends, and how well IDAhas integrated its program priorities with its cor-porate policies and country assistance programs(Part II). It looks at how well IDA has institu-tionalized the recommended process reforms toimprove its overall effectiveness (Part III). Therecord reflects not only changing country cir-cumstances, but also IDA’s management of its

institutional capacities—its operational policies andprocedures, instruments, and budget resources(Part IV) and the consistency and realism of thereplenishment mandates themselves (Part IV).

The report concludes with suggestions forways to improve the alignment of future mandates,institutional capacities, and program design andimplementation to advance IDA’s contribution tothe overarching goal of poverty reduction. Thereport was presented to the Board of ExecutiveDirectors on May 29, 2001. A summary of the dis-cussion at that meeting is attached as Annex H.The Bank’s management response to the findingsof the review is attached as Annex I.


1 2

1 3

The Program Dimension

Part II

In line with IDA replenishment recommendations, IDA has done much to implementthe revised poverty reduction strategy adopted in 1990. IDA has contributed toimproved economic management, poverty data and analysis, and access to basic socialservices. Its adjustment programs have also become more attentive to poverty and socialissues. These actions have contributed to sounder foundations for economic growthand human development. IDA has had less success in aligning its instruments ofassistance to reinforce other measures for sustaining pro-poor, broad-basedgrowth, especially measures supporting agricultural and rural development anddevelopment of the private sector.

IDA has also responded to replenishment commitments to broaden its povertystrategy by taking steps to integrate gender, environmental sustainability, andgovernance in its country assistance programs. This has represented a demandingchange in IDA’s authorizing environment. While IDA has made important contributionsin each of these areas, the mainstreaming of gender and environmental issues hasbeen limited relative to commitments, and while the treatment of governance issueshas expanded markedly, the achievement of sustainable governance improvementsremains a daunting challenge.

1 5

Sharpening the Focuson Poverty Reduction

The IDA10–12 replenishment undertakings commit IDA to sharpen itsfocus on poverty reduction in accordance with the Bank’s 1990poverty strategy.1 They also urge IDA to integrate gender, environ-

ment, and governance into that strategy as a “broad-based framework forpoverty reduction” (World Bank 1998a) and to increase attention to povertyin its performance-based allocation (PBA) system (see Part III).

The replenishment commitments calling for asharper focus on poverty (box 2.1) instruct IDAto: • Strengthen the poverty focus of its Country

Assistance Strategies (CASs). • Build poverty analysis and monitoring into its

analytical work and policy dialogue. • Increase poverty-targeted and social sector

investment lending. • Promote broad-based economic growth

through poverty-focused adjustment lendingand support for private sector development(PSD).

Progress on these commitments has gainedconsiderable momentum since the start ofIDA10.2

Strengthening the Poverty Focusof Country Assistance StrategiesPoverty has received greater coverage in IDA’sCASs during the past seven years, especially

since 1997, encouraged by strengthened guide-lines and follow-up by management (OED2000d; IDA Review 2001b).• Broad-based growth, human resource devel-

opment, and protection of vulnerable groups,key elements of the Bank’s 1990 povertystrategy, are commonly stated objectiveswhere country circumstances warrant.

• CASs increasingly use data and analysis fromPoverty Assessments (PAs) and Public Expen-diture Reviews (PERs), which have gradu-ally improved their poverty focus.3

• Greater consultation with borrower govern-ments and other stakeholders in the prepa-ration of CASs has contributed to thestrategies’ greater relevance.

CASs have made less improvement, how-ever, in setting priorities and identifying meas-ures in support of country-specific, pro-poorgrowth strategies. Participants in country con-sultations commended IDA’s heightened atten-

22

tion to the poverty impact of its assistance, whileexpressing concerns about inadequate prioritysetting and tailoring to country circumstances andweak governance and institutions (box 2.2).OED Country Assistance Evaluations (CAEs) con-firm the relevance of the greater poverty focusin IDA CASs. But they also emphasize that moreneeds to be done to integrate poverty objectivesinto macroeconomic and sectoral operations andto identify the concrete measures needed not onlyfor accelerating growth but also for overcomingsocial and structural constraints to poor peo-ple’s participation in new economic opportuni-ties. This includes adequate attention to theinstitutional mechanisms through which policychanges are expected to benefit the poor.4

CAS monitoring has improved recently, aidedby the country program matrix introduced inIDA11.5 By FY00, more IDA CASs containedpoverty reduction targets linked to intermediateobjectives that could be monitored annually,6

although it is still too soon to know the effecton program outcomes.

Building in More Poverty Analysisand MonitoringIDA has made major contributions to collectingand disseminating poverty data and analysis,an essential underpinning for a sharper CASfocus on poverty and an important achieve-ment that might not have happened withoutIDA (box 2.3). Still, there is important unfinishedbusiness—from improving the quality and pol-icy relevance of much of the analysis to strength-ening countries’ capacities to collect and analyzepoverty data, monitor progress, and enhanceprogram outcomes.

Reviews of PAs,7 PERs,8 and other economicand sector work (ESW) highlight the impor-tance of working collaboratively with borrowergovernments and institutions. They also noteweaknesses that limit the strategic relevance ofthe work. PAs and PERs have had little impacton country policy where ownership bygovernment and other stakeholders has beenlacking.9 Recently, participatory PAs have begunto incorporate the views of poor and disadvan-taged groups, but it is difficult to trace theirinfluence on government policies or IDA assis-tance programs.10

The first year of Poverty Reduction StrategyPapers (PRSPs) exposed continuing weaknesses incountries’ poverty data collection and analysis andin public expenditure tracking, despite recent


1 6

IDA10• Complete Poverty Assessments.• Increase share of lending to social sectors and to poverty-

targeted investments.• Include specific poverty reduction measures in adjustment pro-

grams wherever feasible.IDA11• Incorporate participatory components in Poverty Assess-

ments and focus on the gender dimensions of poverty.• Strengthen poverty monitoring, including the impact of projects

with and without targeting mechanisms for reaching the poor.• Use Public Expenditure Reviews to assess spending in terms

of the development needs of the poor.• Increase attention to structural and fiscal measures to expand

opportunities for the poor.

IDA12• Develop a broad-based policy framework for poverty reduc-

tion emphasizing investment in people, broad-based growth,good governance, and environmental protection.

• Broaden participation in policy formulation and programimplementation.

• Direct about 40 percent of investment lending to the socialsectors.

• Mitigate inequalities. • Link assistance to poverty reduction targets, including the

international development targets.

I D A 1 0 – 1 2 C o m m i t m e n t s o n P o v e r t yB o x 2 . 1

IDA has made major contributions tocollecting and disseminating poverty data

and analysis, an essential underpinning fora sharper CAS focus on poverty.

efforts to increase country participation in PERs andto focus more on government financial manage-ment. Intensified efforts are needed to strengthenanalysis and monitoring capacities, ranging fromshort-term measures to credibly track public expen-ditures to long-term efforts to strengthen capaci-ties in poverty analysis, performance monitoring,and evaluation of program impacts.

Projects increasingly include impact evaluationframeworks to assess poverty-related results overtime, a sorely needed improvement.11 But mon-itoring and evaluation (M&E) by IDA and by bor-rowers remains fragmented and underfunded.That means that IDA and its partners are notbuilding a reliable database for assessing alter-native policy mixes and implementation

approaches to poverty reduction in differentcontexts. The PRSP and new forms of program-matic lending add new dimensions to the needto strengthen M&E. A key priority, as reportedin the early PRSP progress reports, is to supportthe buildup of borrowers’ capacities to trackprogress, analyze results, improve program out-comes, and inform broad debate on priorities.(See Part IV for further discussion of M&E.)

Management is aware of these problems inpoverty analysis and other ESW and has giventhem increasing attention in the last two years.Resolving the problems is critical to IDA’s ana-lytical, advisory, and aid coordination servicesand involves issues of accountability and resourceallocation (see Part IV).

S h a r p e n i n g t h e F o c u s o n P o v e r t y R e d u c t i o n

1 7

Strengths• Outstanding analytical work and dissemination of information• Ability to fold issues of poverty into broader discourses on eco-

nomic and social development• Strong human resource capacity and attractive lending terms• Aid coordination and sector coordination.

Source: OED 2000a, Bank data.

Weaknesses• Too little local ownership of ideas and interventions• A strategy that is too macro and connects poorly with micro-

economic issues and poverty• Little involvement in areas that matter most to the poor (agri-

culture, rural credit, small and medium-size enterprise devel-opment) and insufficient attention to governance issues

• Processes and procedures that are too rigid to permit true part-nership and full participation

• Insufficient focus on national capacity building for povertyanalysis, monitoring, and evaluation.

S t a k e h o l d e r s ’ V i e w sB o x 2 . 2

• PAs have been completed for most (90 percent) of IDA’s eli-gible borrowers. Although progress lagged in the overly ambi-tious IDA10 replenishment undertaking because of severedeficiencies in poverty data and the need to build governmentsupport and interest, many countries now collect and useincome and consumption data in developing policies. (As anindication of this increase in information, the World Devel-opment Report (WDR) 2000/2001’s global assessment of povertywas able to use data from 110 developing countries, comparedwith only 60 countries at the time of the previous povertyWDR in 1990.)

• Almost half of active IDA borrowers have a recent PER, and newapproaches in some countries are increasing the capacity for suchreviews and strengthening ownership. For example, Ghana andVietnam, assisted by IDA, have undertaken their own reviews.

• Other ESW, such as Social and Structural Reviews (SSRs) andsectoral analyses, has also strengthened the focus on poverty.

• Following IDA12 recommendations, IDA recently increasedanalytical work on other aspects of poverty reduction. A notableinitiative, responsive to IDA Deputies’ concerns, focuses onchild labor through the Global Child Labor Program, a catalystfor research and analysis, pilot studies, and other activities.

I D A ’ s C o n t r i b u t i o n t o P o v e r t y D a t aC o l l e c t i o n , A n a l y s i s , a n d D i s s e m i n a t i o n

B o x 2 . 3

In country consultations, stakeholders praised the World Bank’s broad strategic perspective on development issues, but expressed con-cern that the Bank sometimes promoted a “one-size-fits-all” approach. They cited several strengths and weaknesses of the Bank:

Increasing Poverty-TargetedInterventions and Social Sector LendingIDA has met its commitments to increase lendingto poverty-targeted and social sector interven-tions, focusing on efforts that benefit the poordirectly and expand their access to education andbasic health services. A greater challenge hasbeen to help countries with politically and insti-tutionally complex sector-wide reforms and pro-grams. For this, new program approaches—not justhigher levels of lending—have been critical.

Maintaining the share of poverty-targetedinterventions. In line with replenishment com-mitments since IDA10, the share of investmentlending classified under the Program of Tar-geted Interventions (PTIs)12 rose after IDA9 andthen leveled off (see Annex E). Monitoring of thepoverty impact of these investment credits haslagged, although it has been somewhat betterthan in other projects. The program remainsmainly a way of looking at inputs, however. Itsays nothing about outputs, outcomes, orimpacts, and its focus on direct targeting of thepoor may not always be the most effective wayof reducing poverty.13 To strengthen countries’

own poverty reduction strategies through bet-ter knowledge about effectiveness, tinkeringwith the PTI will be less important than sup-porting more and better outcome and impactevaluations, not just in projects but also at thecountry level.

Investing in social sectors. Investment in thesocial sectors has been a large part of IDA’spoverty strategy throughout the period underreview. This assistance has expanded access tobasic services, but the greater challenge hasbeen to help countries improve the quality andefficiency of service delivery to the poor throughsector-wide reform.

Lending in social sectors (health, nutrition, andpopulation; education; social protection; andwater supply and sanitation) rose from 20 per-cent of IDA investment credits in the late 1980sto 40 percent by 1995, where it has remained.That level satisfies the IDA10 and 11 commit-ments to emphasize the social sectors and the40 percent investment lending target of IDA12.

This assistance has varied across sectors andRegions, influenced by the presence of otherdonors and by country commitments and capac-ities to implement reforms (see box 2.4 andAnnexes C.1 and C.2).

The increase in social sector lending hasmade IDA the largest financier in human devel-opment and a major source of analysis andadvice, especially in Africa. Through its analyt-


1 8

Most of the increase in IDA investments since the mid-1990s has gone to health, nutrition, and population, whose share rose fromslightly more than $2.0 billion in IDA10 to some $2.3 billion in IDA11. Education commitments remained largely stable, increasingslightly from $1.85 billion in IDA10 to $1.91 billion in IDA11 (with the number of projects exceeding those in health). Social pro-tection lending was also largely stable, with a strong acceleration in lending in the Europe and Central Asia (ECA) Region; whileinvestments in water supply and sanitation increased in IDA11 after contracting significantly in IDA9 and IDA10.

South Asia and Africa dominate IDA’s social sector lending. In South Asia, where India and Bangladesh are sizable borrow-ers, social sectors account for about 45 percent of lending. In Africa, the social sectors account for just over 30 percent of lend-ing, with the amount declining slightly from IDA10–11, although the number of projects has increased. An FY00 HIV/AIDS initiativewill allocate up to $500 million to countries in Africa during the next three years.

The Regional distribution is affected by the concentration of lending in blend countries—as directed in IDA11–12—in socialand environmental matters. This influences sectoral patterns in countries such as China, India, and Pakistan. Also in the ECA Region,where countries are “nominal” blends, most of the social sector lending to date has entailed social protection provisions in thecontext of economic adjustment credits.

W h e r e D i d I n c r e a s e d I n v e s t m e n t s i n t h eS o c i a l S e c t o r s G o ?

B o x 2 . 4

The increase in social sector lending hasmade IDA the largest financier in human

development and a major source ofanalysis and advice, especially in Africa.

ical work, lending, and technical assistance, IDAhas helped countries strengthen their humandevelopment policies, increase social expendi-tures, expand access to basic services, and adjustpension and other social protection mecha-nisms, particularly in countries in transition.Overall, IDA has provided some form of socialsector assistance to virtually all its activeborrowers.

The focus of social sector assistance hasshifted from projects to sectors, with increasedattention on the quality and financing of servicedelivery for the poor. Through its analyticalwork and conditions attached to adjustmentlending and debt relief, IDA has encouragedcountries to protect—and even increase—socialexpenditures. Available evidence suggests a gen-eral upward trend in national expenditures,although often from very low levels, and someimprovement in the efficiency of that publicspending. This emphasis on countries’ ownexpenditure levels has reinforced IDA investmentcredits in support of inputs needed for the geo-graphic expansion of basic services (construc-tion of schools and clinics, provision of essentialdrugs); training for service providers; andimproved sector policies and service delivery,among others.

Program evaluation. Country consultationsfound that governments and civil society wel-comed IDA’s heightened emphasis on the socialsectors, its broadened strategic perspective, thestrong contribution of its analytical work, and itsincreasing coordination with other donors. OED

CAEs also stress the relevance of the increasinglysystemic focus of IDA assistance, especially onthe composition of social sector spending, itsregional distribution, and inefficiencies in pub-lic spending (see box 2.5).

Project portfolio review ratings also showedimprovements, particularly during IDA11, withoutcome ratings of completed projects risingfrom 61 percent in IDA10 to 78 percent in IDA11and the first year of IDA12 (see Annex D). Theratings showed some improvement in sustain-ability and the institutional development impactof investments, although both these resultsremain at inadequately low levels,14 reflectingwhat past evaluation findings have shown to bemajor challenges in implementing social sectorreform efforts.

When governments have demonstrated com-mitment and leadership and IDA has effectivelytailored its support to realities on the ground,project outcomes have been significant. Therecently evaluated Second Technician Educa-tion Project in India is a good example. State-level government commitment, ownership, andattention to execution made it possible to over-come the implementation challenges of a verycomplex project plan.15 However, reviews alsopoint to a number of frequently experiencedimplementation difficulties. For example, anOED evaluation of Bank assistance to health,nutrition, and population during the 1990s andmore limited education reviews (OED 1999b, d)point to:• Weaknesses in institutional analysis and insti-

tutional development impact, especially in


1 9

In Uganda, IDA adjustment and investment lending in the 1990s encouraged the Ministry of Finance to raise public expendituresin health and education. By documenting service performance problems and expenditure leakages, IDA helped strengthen the min-istry’s efforts to protect public spending.

In Burkina Faso, adoption of minimum budget allocations to health and education as a condition of adjustment lending and debtrelief reinforced IDA investments in support of the decentralized health system and a reorientation of the primary education sys-tem, resulting in higher school enrollments.

In Mozambique, IDA’s presence in the health sector strengthened the dialogue with the government on expanding public spend-ing on health and influenced the balance between investment and recurrent spending.

Source: OED 1999e, 2001b; Landau 1997.

E x a m p l e s o f I D A S u p p o r t f o r I n c r e a s e dS o c i a l S e r v i c e E x p e n d i t u r e s

B o x 2 . 5

efforts that aim beyond expanding access tosocial services to improving service quality tothe poor, a politically and institutionally com-plex task

• Uneven quality of sector work needed to tai-lor program priorities and implementationstrategies to country circumstances

• Unrealistic expectations about borrowers’institutional capacities and readiness forreform (OED 2000a; Johnson and Wasty 1993).

• Insufficient assessment of project impact andcost effectiveness in reaching the poor.16

The health, nutrition, and population studycovering operations through 1998 concludedthat, although the quality of institutional analy-sis has improved, the Bank has been more suc-cessful in expanding service delivery than inimproving service quality and efficiency andpromoting institutional change. Project designshave also tended to be more complex in coun-tries with weak institutional capacity.17

Moreover, investment projects have often hadto deal with systemic governance problems,such as weak budget management, weak finan-cial accountability, and the need for broad civilservice reform. The OED CAE for Cameroon, forexample, found that intended beneficiariesreceived no more than about half of publicexpenditures allocated to health and education.These problems cannot effectively be solvedproject by project. They require progress onpublic sector reform and institutional develop-ment on a broader programmatic basis and atthe national and provincial levels. While a newemphasis on private provision of social servicesmay alleviate some inefficiency in service pro-vision (IDA Review 2001g), broad governanceand public sector management challenges stillhave to be confronted directly and systematically.

Program innovations. To deal with theseimplementation challenges, IDA has introducedseveral program innovations. These include:new lending instruments (designed to deal withthe institutional complexities of improving ser-vice delivery and with getting better results fromsector reforms); increased participation incountry-led, donor-coordinated, sector-wide pro-

grams; more attention to multi-country (regionalor global) programs; and the PRSP process andits emphasis on linking social sector reformsmore closely to poverty reduction goals. Eachshows important promise, but also operationalchallenges that require further attention.

Two new adaptable lending instruments—Learning and Innovation Loans (LILs) and Adapt-able Program Loans (APLs)—were introduced in1997 to provide increased flexibility, especiallyfor investments involving complex institutionalchange.18 Experience shows the usefulness ofthese new instruments for testing operationalideas and facilitating adjustments and change asprojects unfold. But it also shows a need for moresupervision (and therefore more budget alloca-tions than anticipated) and better M&E in sup-port of these flexible implementation approachesthan have yet been put in place.19 In addition,two new forms of lending—the programmaticadjustment credit focused on medium-term socialand structural reforms and the more recentPoverty Reduction Support Credit focused onproviding support for countries’ medium-termpoverty reduction strategies—offer new ways ofreinforcing public sector reforms and institu-tional development.

IDA has also increasingly engaged in sector-wide approaches (SWAps), especially in healthand education. SWAps, which can be supportedby a wide variety of lending instruments, typi-cally involve multi-donor participation in sup-port of an agreed, country-led reform programand lending framework as part of a medium-termexpenditure plan. The rationale for this approachis dramatically illustrated by the situation inMozambique. Before an integrated educationsector program was set up, external assistanceto the Ministry of Education was being pro-vided under more than 150 different projects andsubprojects, funded by some 16 donor countries,6 U.N. agencies, and 3 multilateral financinginstitutions, including IDA.20

Consultations in Bangladesh, Ghana, andMozambique expressed strong support for theseapproaches, noting that, although they tooktime to organize, they markedly improved thedelivery of development assistance over the tra-ditional fragmented project approach. In a few


2 0

cases, such as the Ghana Health Sector Pro-gram (box 2.6), there has been sufficient agree-ment on objectives and approaches for donorsto pool their funding. Reviews and tracking ofsector-wide programs confirm that they havesignificantly improved project relevance andconsistency of sector expenditures with macro-economic conditions and institutional assistance;strengthened country ownership of the reformprogram; and improved country and donor dia-logue on priorities and approaches. Progresshas been slower on country budget management(at central and decentralized levels), supportfor improved implementation capacity, donorcoordination of procedures and processesthrough government mechanisms, and the estab-lishment of self-monitoring systems.

These SWAps have important potential asbuilding blocks in a country’s poverty reductionstrategy. Early experience with PRSPs has shownthe value of revisiting sector strategies in lightof explicit poverty reduction goals, to sharpenthe focus on equity and efficiency objectives andestablish indicators to assess performance inserving the needs of the poor. SWAps are ameans of aligning external assistance with thesestrategies to improve access, financing, and

accountability mechanisms in ways that benefitthe poor.21

New regional and global initiatives. AlthoughIDA usually provides repayable credits to mem-ber country governments, new regional andglobal initiatives on specific problems and sup-ported by grant funds have occasionally com-plemented country-based lending. An earlyexample was the Riverblindness Control Programin West Africa. More recent initiatives include theGlobal Forum for Health Research, which pro-motes research to improve the health of thepoor; the Stop TB Initiative; and the GlobalAlliance for Vaccines and Immunization, whichworks to boost childhood immunization ratesand vaccine research for major infectious diseasekillers in poor countries, including HIV/AIDS.The Bank has also joined a global effort toimprove the use of global data and strengthencountry monitoring capacity. While such pro-grams look promising for advancing the fightagainst poverty, expanding IDA’s participationposes challenges in setting priorities with part-ners, building partnership mechanisms for pro-gram implementation, and mobilizing funding,including grant funds, as needed.


2 1

IDA and 16 other assistance agencies have provided integrated support for the Ghana Ministry of Health’s five-year sector reform pro-gram. This support has included pooled funding arrangements and a common framework for M&E. Most donor funds, including theIDA credit, are provided as untied contributions to a common health account and managed under existing government procedures.The other resources are earmarked funds but are included in the government’s annual planning and budgeting cycle; they are chan-neled through the Financial Controller rather than given directly to program managers, apart from other funding. The health accountfunds, kept in a bank account under the control of the Ministry of Health and the Controller and Accountant General, are distributedto all eligible health units that meet minimum “readiness criteria,” including submission of annual budgets accompanied by quanti-fied targets and objectives and procedures for authorizing payments, maintaining accounts, and providing monthly reports.

Source: Johannson and Adams 1998.

P o o l i n g D o n o r F u n d s i n S u p p o r t o f H e a l t h S e c t o r R e f o r m i n G h a n a

B o x 2 . 6

2 3

PromotingBroad-Based Growth

In promoting broad-based growth, the second major component of IDA’spoverty reduction framework, IDA has supported economic policy andstructural reforms and productive investments to accelerate job-creating

growth. The IDA10–12 replenishment commitments on the importance ofgrowth for poverty reduction emphasize IDA’s role in improving the povertyand social impact of adjustment lending (box 3.1) and promoting private sec-tor development (PSD) (see box 3.6). IDA has made considerable changesin its adjustment operations over the period. It has also made progress onsome of its PSD replenishment undertakings. Nevertheless, the accelerationof sustained, broad-based, job-creating growth remains a major challenge ina large majority of its eligible borrower countries.

IDA’s Evolving Adjustment OperationsAdjustment lending has been IDA’s major tool forhelping countries improve their economic man-agement (box 3.2). As a share of total IDA lend-ing, adjustment credits fell from 25 percent onaverage during FY92–97 to 19 percent in FY98–99and 16 percent in FY00, as adjustment lendingto Africa and South Asia slowed. The share of sec-toral adjustment operations decreased signifi-cantly over this same period (see Annex F).

In the period under review, adjustment oper-ations have evolved in ways consistent with thereplenishment undertakings:• As countries moved from first-generation

macroeconomic reforms (trade, exchangerate, monetary policy) to second-generation

social and structural reforms (in the social sec-tors, financial and private sector develop-ment), adjustment lending evolved to supportinstitutional reforms and increasingly empha-sized poverty reduction and social sectorreforms. As the share of policy conditions sup-porting macroeconomic reform declined andthe share supporting public sector reforms,financial and private sector development,and social sector reforms increased, the focushas become more long-term, institutional,and microeconomic.

• In FY99, 15 percent of conditions on IDAadjustment credits related to social sectorreforms, an increase over previous years.“Poverty-focused” adjustment credits increased

33

from 33 percent of IDA10 operations to 46 per-cent of IDA11 operations (67 percent byvalue).1

A rising share of adjustment operations hasbeen rated satisfactory during the 1990s, accord-ing to Quality Assurance Group (QAG) qualityat entry ratings and OED quality at exit ratings.QAG found a decline in adjustment operationsat risk from 38 percent in FY95 to 16 percentin FY00. OED found an increase in outcome rat-ings of satisfactory or above from 65 percent inIDA9 to 80 percent in IDA11 and the beginningof IDA12. A quarter of the improvement isexplained by the lower share of evaluatedadjustment operations in the Africa Region,which has consistently ranked last amongRegions in adjustment lending performance(OED 2000a, p. 11). In addition, there have been

improvements in the design of operations andbetter Bank and borrower performance.

On the broader issue of impact on povertyreduction, two findings emerge clearly from arange of studies. One is that many countriesreceiving adjustment credits have been able toprotect (and even increase) social sector spend-ing (IDA Review 2001b). The increase, however,has not always improved results for poor or vul-nerable groups (Castro-Leal and others 1999;OED 1999c). Corruption, imbalances in alloca-tions between capital and current budgets, anddisparities in the distribution of expenditures toservices benefiting different income groupsundermine the impact of such spending onpoverty. The Bank’s recent retrospective onadjustment lending points out that the povertyand social focus of adjustment lending hasincreased over time, but it is still inadequate inmany cases and remains a major priority.2

The second finding is that adjustment oper-ations have done well, on average, in helpingcountries lay a foundation for acceleratinggrowth and poverty reduction. Countries haveimproved macroeconomic stability, loweredinflation, and removed economic distortions(Branson, Javarajah, and Sen 1996), and inmany countries trade, price, and regulatoryreforms have contributed to higher agriculturaloutput and productivity growth (Meerman 1997).Countries that have sustained reforms have hadat least modest increases in per capita incomegrowth and decreases in the incidence ofpoverty.3 Countries with good records of suc-cessful adjustment operations include India,Uganda, and Vietnam in the early 1990s andBangladesh in the mid-1990s. But the record onsustaining growth rates high enough that many


2 4

Replenishment recommendations related to theseoperations have encouraged IDA to include specialpoverty measures in the design of adjustment pro-grams, to protect social expenditures, and to seethat reform packages are compatible with produc-tive investment.

I D A 1 0 – 1 2C o m m i t m e n t so n I m p r o v i n g

t h e P o v e r t y I m p a c t o fA d j u s t m e n t L e n d i n g

B o x 3 . 1

Two Structural Adjustment Credits (SAC I and SAC II) in 1991 and 1994 and the Agricultural Sectoral Adjustment Credit in 1990 wereamong IDA’s most potent poverty reduction instruments. SAC I and II continued a focus on stabilization and market-oriented reformsbegun in the latter half of the 1980s, but were largely concerned with public sector management and private sector development.The adjustment and agricultural credits supported important reforms in coffee, for example, that have increased poor farmers’ incomes.The liberalization of coffee marketing allowed farmers to retain 65–80 percent of export earnings, compared with 30 percent beforethe reforms—increasing revenues by about $19 per capita per year.

P o v e r t y I m p a c t s o f A d j u s t m e n t L e n d i n gi n U g a n d a

B o x 3 . 2

Countries that have sustained reformshave had at least modest increases in

per capita income growth and decreasesin the incidence of poverty.

of the poor benefit is not broadly positive (seebox 3.3).

Why has it been difficult for countries to movebeyond new policies, legislation, and even action(privatization, for example) to better poverty out-comes? One reason is that many IDA countrieshave not consistently implemented their policyreforms, making it difficult to institutionalize thecomplex structural reforms needed to sustaingrowth rates high enough for long-term povertyreduction (Devarajan, Dollar, and Holmgren 1999;Killick, Gunatilaka, and Marr 1998). The recentcollaborative study Can Africa Claim the 21stCentury? notes that: “Many African countries havemoved in and out of compliance with economicand structural reform programs, so formally beingon a program has meant little for the policies actu-ally pursued over longer periods. And short-termreforms have failed to address some difficultunderlying institutional problems—and in somecases may have made them worse” (World Bank2000b, p. 35). Sound policies, the study notes, payoff in the medium run but require good economicmanagement over a sustained period, which hasbeen difficult to achieve.

Research on adjustment lending and reformin Africa also points to the intervening role ofthe domestic political economy and to the lim-itations of a conditionality-driven model of adjust-ment. Conditionality that is useful to governmentsin the early stages of reform may be too intru-sive during second-generation social and struc-

tural reforms. Typically, financing has tended tobe reduced in that later phase just as policyreforms are taking hold and resources are neededfor supporting policy and structural adjustments(Devarajan, Dollar, and Holmgren 1999). Reflect-ing these lessons, many adjustment credits sincethe mid-1990s have had fewer and simpler pol-icy conditions.4 “Floating” tranches have givengovernments more flexibility in timing reforms,within the discipline of a medium-term expen-diture framework (OED 1999c). Introduced in1998, programmatic adjustment lending prom-ises support for countries facing the institution-ally and politically complex demands ofsecond-generation reforms (see box 3.4). Coun-tries need strong budgetary and expenditurecapabilities to make these new adjustment cred-its work. Better monitoring of country and IDAperformance is needed as well.

Another contributing factor is the difficulty inidentifying practical policy measures for achiev-ing not just growth, but broad-based growth. Thepoor typically benefit from growth, but howmuch they benefit varies greatly (Ravallion 2000).The fight against poverty has to address the

P r o m o t i n g B r o a d - B a s e d G r o w t h

2 5

Despite dramatic achievements in macroeconomic stabilization and trade liberalization in 1986–91, growth and poverty reductionin Bolivia have fallen short of expectations. The OED Country Assistance Evaluation (CAE) cites five reasons: incomplete reforms,including in the financial sector; strong opposition by interest groups to privatization and legal, judicial, and regulatory reforms;regional imbalances in growth; inadequate social safety nets; and insufficient investment in human capital.

Boosting agricultural productivity has been on Tanzania’s reform agenda since the government agreed to a structural adjust-ment program in 1986.Yet despite a steady return to macroeconomic stability in the mid-1990s and a well-regarded reform pro-gram covering key areas of budget management, privatization, and incentives, agriculture has stagnated. Among the many reasonsfor the weak supply response are poor infrastructure, lack of credit, price variability, and weak public sector support for agri-culture, characterized by an uneven commitment to reform in the early 1990s. Noting the absence of a clear rural developmentstrategy, OED’s findings suggest that this may have contributed to the apparent neglect of policies for improving agriculturalgrowth and reducing rural poverty.

Source: OED 2000e, OED data.

W e a k B r o a d - B a s e d G r o w t h , D e s p i t eI m p r o v e d M a c r o e c o n o m i c S t a b i l i t y ,i n B o l i v i a a n d T a n z a n i a

B o x 3 . 3

Programmatic adjustment lendingpromises support for countries facing theinstitutionally and politically complexdemands of second-generation reforms.

factors limiting the ability of the poor to partic-ipate in the opportunities created by growth-oriented policies (Kanbur 2000; Ravallion 2000;Thomas and others 1999). The Bank’s 1999Annual Progress Report on Poverty (World Bank1999) notes weak attention in Country AssistanceStrategies (CASs) to the transmission mecha-nisms through which policy changes areexpected to benefit poor groups. The August2000 progress report on Poverty Reduction Strat-egy Papers (PRSPs) calls for more work on thedeterminants of pro-poor growth and for greaterspecificity about how to support borrowers insetting and advancing country-specific priorities.Weak progress in agriculture and rural devel-opment is one continuing factor.

Lags in Agriculture and Rural Poverty ReductionWorld Development Report 1990 and 2000/2001(World Bank 1990, 2000a) point out that reduc-ing poverty requires major investments in areaswhere the poor live and in activities they pur-sue. IDA’s experience confirms that investing inrural social and economic infrastructure canraise the incomes and living standards of thepoor (box 3.5).

Thus, the decline in IDA’s lending in agri-culture and rural development—from 24.3 per-cent of commitments under IDA10 to 19.1percent under IDA11 and to 9.6 percent duringthe first year of IDA12—raises some questions(see Annex C), especially in the absence ofcompensating increases by other assistance agen-cies.5 The decline is especially sobering in light

of a recent assessment of the “undercapitaliza-tion” of African agriculture (World Bank 2000b).Most of the world’s poor will continue to live inrural areas well into the first half of this century,and agriculture still accounts for a sizable shareof poor countries’ GDP.

To some extent, the decline in lending is theresult of a reclassification of assistance under dif-ferent sectoral headings (for example, lendingfor rural roads, which has increased, is now inthe infrastructure portfolio). But much of it alsocomes from IDA having withdrawn, appropri-ately, from unsuccessful efforts (such as top-down systems of extension services and complexrural development activities that had higher thanaverage failure rates), but without improvedapproaches having been put in place for help-ing countries make the comprehensiveimprovements needed to accelerate agriculturalproductivity growth, market development, andrural poverty reduction.6 Adjustment lendingand economic and sector work (ESW) have con-tributed in many countries to relevant policyreforms that have improved agricultural priceincentives and exports. However, the Africa col-laborative study notes that “recent reforms haveimproved agricultural price incentives, but theyhave not done as well at addressing other struc-tural and institutional constraints, including ruralinfrastructure (irrigation, roads, power, telecom-munications), agricultural research and extension,and farmer health and education—factors thatimpede agricultural productivity and output.”7

Relative decreases in rural or agricultural lend-ing over other sectors cannot be assumed to be


2 6

Introduced in 1998, programmatic structural adjustment loans and credits focus on medium-term structural and social reforms,providing support for a country-owned reform program through sequenced one- or two-tranche adjustment operations over threeto five years. Unlike traditional adjustment loans, in which conditionality is based on promised actions, these loans and creditsprovide funding for completed reforms.

Draft Interim Guidelines on Poverty Reduction Support Credits were proposed in April 2001 as an adaptation of programmaticadjustment lending. This instrument is conceived as a series of programmatic adjustment credits with a strong focus on poverty.Two to three annual credits in a series will be synchronized with the borrower’s annual budget and policy cycle to the extent prac-ticable. Each credit would support the borrower’s medium-term program of social and structural reforms and institution buildingset out in the country’s PRSP. The credits are also intended to facilitate coordination with the IMF. The first one is being preparedto support Uganda’s poverty reduction strategy and focuses heavily on public sector reform.

P r o g r a m m a t i c A d j u s t m e n t L e n d i n g t o I D AC o u n t r i e s

B o x 3 . 4

a bad thing a priori, especially since lending forsocial funds, rural infrastructure, and health andeducation have been robust. Lack of consensusamong development partners regarding ruraldevelopment strategies and the reduced prioritygiven to agriculture in aid programs is, however,a cause for concern. The sector requires revital-ized attention on the part of the internationaldevelopment community, with IDA’s role to bedetermined in coordination with others.

Strengthening Private Sector Development Lags in private sector investment are anotherimportant part of the weak growth story in IDAcountries. For example, the ratio of privateinvestment to GDP is not much higher than 5percent for Madagascar or Mali, although it is ashigh as 30 percent in Malaysia. As efforts to sta-bilize macroeconomic conditions took hold inthe 1990s, IDA increased its PSD activities, withsuccess in some areas. The IDA10–12 Replen-ishment Reports all urged greater support for PSDto accelerate job-creating, broad-based growth(box 3.6). However, PSD activities have not, onthe whole, been well integrated into IDA coun-try assistance programs.8

Some 250 IDA projects with PSD componentshave been approved since the start of IDA10. Atthe project level, OED assessments show that 87

percent of closed projects had satisfactory out-comes, 55 percent were judged likely to be sus-tainable, and 37 percent had at least substantialinstitutional development impact, about the sameratings as for IDA projects overall. While theoutcome ratings show satisfactory progress, thelow institutional development impact is of con-cern, given the importance of well-functioninginstitutions for achieving the boost to broad-based growth that can come from PSD. TheIDA-supported projects have entailed:• High compliance in microfinance (box 3.7) in

developing and disseminating best practiceglobally, including through IDA’s support forthe Consultative Group to Assist the Poorest(CGAP).

• Rapidly expanded support for private provi-sion of infrastructure since the start of IDA11,with 60 percent of such activities in Africa.

• Assistance for new private sector legal andregulatory frameworks (more support isplanned) and for some 131 privatizations (theaddition of the transition economies to IDA’sprivate sector work in the mid-1990s vastlyincreased the scope of this undertaking), butscale and effectiveness are not yet adequate.

• Modest progress in support of small andmedium-size enterprises and rural finance.IDA has moved away from using public


2 7

In Bangladesh, two of IDA’s most successful pro-poor investments have been in rural road rehabilitation and rural electrification.The rural electrification project, approved in FY90, sought to expand distribution networks to underserved areas and to rehabili-tate distribution systems taken over from state ownership. The project improved household lighting, increasing opportunities tostudy and earn income and raising the quality of life, and it led to more efficient energy use in industry and agriculture. At creditclosure, consumer connections far exceeded original estimates. A socioeconomic study showed that average household incomein the program villages is about 50 percent higher than that in non-electrified villages and attributed 22 percent of the differenceto electrification. The poverty rate in electrified villages is lower as well. The investment completion report notes that “althoughelectrification has not shrunk the gap between the richest and poorest,” the incomes of the poorest 10 percent of households inelectrified villages are much higher than those of their counterparts in non-electrified villages.

Although the forestry project in the Indian state of Madhya Pradesh did not explicitly focus on poverty, it led to significant improve-ments in livelihoods and living standards, especially for tribal peoples. While IDA financed activities in 2,500 communities withmore than 1 million people, the project expanded to more than 12,000 communities and 6 million people. It increased production(income) for the poor and created conditions for empowering poor communities on the forest fringe. Community protection fundshelped to increase public choice and to reduce debt, and thus reliance on moneylenders. The project also made substantialprogress toward environmental goals. Detracting from the project’s major social gains, however, were the conflicts over forest accessbetween villages during participatory planning and the limited participation by women in decisionmaking.

P r o - P o o r R u r a l D e v e l o p m e n t P r o j e c t si n B a n g l a d e s h a n d I n d i a

B o x 3 . 5

institutions to provide directed credit andservices to small enterprises and from usinggovernment-owned rural finance institutionsto provide targeted credit. But sound alter-natives have yet to be found.9 Thirty-five IDAcountries have no programs for increasingaccess to rural finance.

At the country level, many OED CAEs pointto a need for better choices of interventions,approaches, and instruments and for more atten-tion to the distributional impact of PSD activi-ties. The evaluations give low ratings to theeffectiveness of PSD work. Many express con-cern about weak institutions and the lack of acomprehensive, country-specific PSD strategy.10

The gains from increased efficiency are notalways widely shared, in large measure becauseof inadequate attention to the institutional andpolicy framework for PSD activities.11

A review of IDA’s PSD agenda shows a needfor improvements in several areas: • Business environment. IDA has focused on the

important first steps for improving the busi-ness environment through policies on foreigntrade, investment, and the financial sector. Ithas focused less on competition, consumerprotection, property rights, insolvency andbankruptcy reform, and corporate gover-nance, especially where institutions are weak.And while trade policy reform has altered pro-ducer incentives and promoted exports, fewprojects have explicitly supported exportdevelopment, and not enough use has beenmade of IDA’s analytical and advisory servicesfor addressing cross-border trade and invest-ment problems.

• Financial sector development. Since FY94,IDA has approved 70 operations with signif-icant financial sector development compo-nents, most of them in the Africa and Europeand Central Asia Regions. Evaluations point toa need to strengthen the institutional foun-dations for a fair and robust market-basedeconomy.12 Improvements in the policy andregulatory environment for financial marketsare needed to increase the access of ruraland urban micro, small, and medium-sizeenterprises to commercially viable financialservices.

• Privatization. Project ratings show that mostIDA-supported privatization efforts haveimproved results at the firm level, but macro-economic and distributional results have beenmixed. Problems with the methods andsequencing of privatization and weak insti-tutions are partly to blame. Where legal andregulatory institutions are weak, privatiza-tion has not improved (and has sometimesworsened) equity and has done little forshort-run productivity. But trying to reformstate enterprises rather than divest them hasnot been any more successful. Achievingbetter results will require a more intensive,country-specific diagnosis; a long-term view;proper institutional foundations; and a con-centration on key firms and sectors.

• Private provision of infrastructure and socialservices. Improving the efficiency of the deliv-ery of infrastructure and social services iscritical to growth and poverty alleviation. Inprivate provision of infrastructure, the recentfocus has been on helping governments setup policy and regulatory frameworks that


2 8

The IDA10 Replenishment Report emphasized improving the enabling environment for business—through legal and regulatory reforms,export promotion, and privatization—with private sector assessments as an analytical underpinning for countries’ efforts. The IDA11report, while reiterating the importance of the business environment, also emphasized private provision of infrastructure and ashift to a more poverty-focused agenda, including greater attention to small and medium-size enterprises. The IDA12 report inten-sified the focus on poverty reduction. It called on IDA to ensure that privatization activities were consistent with social equity andenvironmental sustainability—and took a broader systemic perspective, endorsing the piloting of partial risk guarantees to cat-alyze private investment and directing IDA to develop a comprehensive private sector development strategy by the end of 1999.

I D A 1 0 – 1 2 U n d e r t a k i n g s o n P r i v a t e S e c t o rD e v e l o p m e n t

B o x 3 . 6

will motivate the private sector to respond toconsumer needs and enable poor householdsand communities to share the benefits of bet-ter services. The emphasis has been on devel-oping market structures and regulatoryregimes appropriate to institutional and capac-ity constraints while improving governanceand expanding capacity. Private provision ofinfrastructure shows promise as a growtharea for IDA assistance.

A strategy paper prepared in 1999 laid out areorganization of World Bank Group PSD work.This step was accompanied by the preparationof subsector strategies for small and medium-sizeenterprises and the financial sector. But to datethere is no overall strategy for Bank support ofPSD linked to its mission of poverty reductionand based on World Bank Group areas of com-parative advantage. The strategy paper nowbeing prepared (for completion in 2001) willmeet this IDA12 replenishment commitment.

Clearly, PSD is a complex, multifacetedendeavor, and there is a need for IDA to exer-cise selectivity based on country commitmentand performance, demonstrated IDA compara-tive advantage, and partnership with other keyactors. With countries’ increasing economicopenness, there is demand for a greater focuson governance, environmental, and social issuesrelated to private investment. In this context, theforthcoming PSD strategy should propose selec-

tivity criteria reflecting the distinctive roles ofIDA/IBRD, IFC, and MIGA. Considerable syn-ergies can be tapped within the World BankGroup by combining IDA’s role in improving thecountry policy framework, IFC’s unique con-nectivity to private corporations, and MIGA’scatalytic function vis-à-vis foreign direct invest-ment. Development rewards would surely beenhanced if the three agencies were to workmore closely together and with their externalpartners to generate environmentally and sociallysustainable development. This would require

improving the delineation of respective roles,shifting activities among institutions, and drop-ping activities that are found to be better pur-sued by other actors. The sector strategy providesa good opportunity to make explicit the defini-tion and rationale for a clear division of laborand a greater selectivity within the World BankGroup and vis-à-vis others. It should also helpidentify ways to improve IDA’s support for thepolicy and regulatory environment for privateinvestment, the proper institutional foundationsfor privatization, the expansion of the privateprovisioning of infrastructure and social services,


2 9

The Poverty Alleviation Microfinance Project supports an ambitious expansion of microcredit activities in Bangladesh that has increasedthe outreach of the 171 participating nongovernmental organizations from 440,000 clients in 1996 to 1.8 million by mid-1999. Womenaccount for 90 percent of the clients, and loan recovery stood at 98.4 percent in June 2000. The program is executed by a nonprofitorganization and is governed by a general assembly representing the main stakeholders concerned with poverty alleviation and socialmobilization issues in the country.

Today, microcredit reaches 8 million households—60 percent of Bangladesh’s poor. An ongoing impact study finds that micro-credit has increased borrower incomes and asset accumulation; improved access to health, education, and sanitation facilities;and enhanced mobility and efficient time use. Through the Poverty Alleviation Microfinance Project, IDA helped to scale up theindustry’s operations and increase outreach during a time of declining grant financing. The Second Poverty Alleviation Microfi-nance Project will help the industry achieve sustainability by accessing financial markets through improved creditworthiness. Itwill improve industry disclosure standards, formalize the regulatory and supervisory system, and strengthen capital adequacy,portfolio quality, and governance systems.

Source: IDA Review 2001g.

B a n g l a d e s h : P o v e r t y A l l e v i a t i o n t h r o u g hM i c r o f i n a n c e

B o x 3 . 7

There is no overall strategy for Bank supportof PSD linked to its mission of povertyreduction and based on World Bank Groupareas of comparative advantage.

and a positive distributional impact fromincreased private sector–led growth.

Looking AheadTwo recent process initiatives—the CDF andthe PRSP—could significantly improve the effec-tiveness of IDA’s support for poverty reductionin borrower countries.• Application of the CDF principles, which

emphasize country ownership and coordi-nation of development assistance in supportof poverty reduction, should help IDA alignits country program priorities with borrowers’poverty reduction efforts through coordina-tion with other development partners.

• The PRSP initiative, by providing a structurefor operationalizing the CDF principles, isintended to achieve what the IDA replen-ishment recommendations have sought formany years: effective country strategies thatlink macroeconomic policy, sector strate-gies, and high-impact interventions toenhance opportunities and reduce vulnera-bilities of the poor.

Several lessons from the period under reviewshould inform these efforts: • Countries’ poverty reduction strategies and

IDA’s support of them must focus on accel-erating broad-based growth. Notwithstandingthe uneven country record, many IDA bor-rowers are in a better position than at thebeginning of the IDA10 period to focus onbroad-based growth and poverty reductiongoals thanks to their policy reforms and thecontributions of IDA and other developmentpartners. But rates of growth are projected toremain too low and inequality rates too highacross a large number of IDA countries to sig-

nificantly reduce poverty in the near future.And there is still much to learn about thedeterminants of sustainable, pro-poor growthin individual countries.

• IDA’s comparative advantage is strategic, notonly in lending for adjustment programs andkey investments in support of broad-basedgrowth, but also in supporting the researchand analytical work needed to ensure that thepoor share in the gains. Because of the impor-tance of private sector and rural develop-ment to economic growth and povertyreduction, these areas need redoubledefforts—with IDA’s role determined in coor-dination with its development partners.

• IDA’s contributions in the social sectors remainrelevant and in need of continued effort, par-ticularly the attention to sector-wide reforms.However, IDA and its partners need tosharpen their focus on results and on theinstitutional issues that impede their achieve-ment. Experience has demonstrated the ben-efits of working through sector-wideprograms. These coordinated, country-ledsector reform efforts ought to be more widelyused as the building blocks of IDA’s andother agencies’ support of countries’ povertyreduction strategies.

• Across all areas of activity, IDA has mademajor and widely regarded contributions tocountries’ efforts through its analytical work.This work needs to be sustained and adaptedto emphasize strengthening countries’ capac-ities to diagnose conditions, analyze andimprove policy and program outcomes, andinform broad debate. The availability of ade-quate resources for this aspect of IDA’s rolerequires priority attention, as recent man-agement reports have indicated.


3 0

3 1

Integrating Gender,Environment, andGovernance

The IDA10–12 replenishment agreements emphasized gender,environmental sustainability, and good governance as core elementsof IDA’s broadened poverty reduction framework. While these

undertakings were highly relevant, progress has been constrained by lackof consensus, in IDA and its member countries, on explicit priorities andon IDA’s role in advancing them.

IDA has made important contributions in eachof these areas, but progress has fallen short ofcommitments. In the areas of gender and envi-ronment, implementation of replenishmentundertakings has been partial, with greaterprogress on some matters than on others and,overall, slow progress on the integration, ormainstreaming, of gender and environmentalconsiderations in all IDA activities. Recently,good governance has received much height-ened attention as a critical factor in povertyreduction. Still, the effective treatment of gov-ernance issues, although expanding rapidly,remains a challenge, and unresolved issuesabout IDA’s role persist.

Although these three issues were largely over-looked in the Bank’s 1990 poverty strategy, theWorld Development Report 2000/2001 (WorldBank 2000a)—with its emphasis on opportunity,empowerment, and security—broadens the per-spective on poverty reduction in ways that sug-gest linkages to gender, environment, and

governance. IDA should build on those insightsto help strengthen domestic consensus, poli-cies, and institutional capacity for incorporatingthese issues into countries’ own developmentstrategies.

Addressing GenderThe Bank’s 1994 Operational Policy on “GenderDimensions of Development” (OP4.20) aims toreduce gender disparities and advance women’seconomic opportunities. It also requires the Bankto assist countries in strengthening institutions forincreased participation of women in develop-ment. The IDA10–12 commitments built on theseobjectives, giving added emphasis to integratinggender considerations into all IDA lending activ-ities (see box 4.1). In implementing those com-mitments, IDA assistance has achieved satisfactoryresults in education and health, in part becauseof significant country ownership and effectivepartnership with other development actors, andalso because operations were underpinned by

44

rigorous analytical work. IDA assistance hasbeen weaker, however, in promoting the eco-nomic participation of women and improvingborrowers’ institutional frameworks for gender.The shortfalls have undermined the efficacy ofthe assistance provided and led to disappointingresults at the country level. Moreover, the broaderintent of integrating gender in all IDA activitieshas yet to be implemented.1

Increasing attention to gender in analyticalwork and monitoring. Overall, the integra-tion of gender considerations into the full rangeof IDA economic and sector work (ESW) hasbeen weak. Analyses in the areas of health andeducation have increasingly considered genderdimensions, focusing on low enrollment of girlsand maternal and child health. Poverty Assess-ments (PAs) and Public Expenditure Reviews(PERs) have also increasingly included genderanalyses (Whitehead and Lockwood 1999; OED2000c), but again looking largely at educationand health and not at women’s participation ineconomic activities. Outside the social sectors,there has been some analysis of gender in agri-culture—where women account for a large partof the labor force in poor countries—and recentlyin transport, but still very little in these areas andeven less elsewhere.2 The lack of a consistentanalytical framework has impeded the integra-tion of gender considerations in general ESW.And gender-disaggregated data and analysisremain limited overall, even in the social sectors.The importance of these findings is underscoredby recent evidence on the links between analy-sis and lending, drawn from review of 12 coun-try cases, that confirms the oft-made argumentthat good analytical work leads to improvedgender integration in lending.

Both Bank policy and IDA replenishmentundertakings emphasize monitoring policyimplementation and impact on the ground, acommitment that needs stronger follow-up. In1987, the Bank became one of the first agenciesto establish a “women in development” ratingsystem to evaluate the integration of genderissues in projects. The system tracks projects afterthey have been approved, for the extent towhich inputs attend to gender commitments.Taking the next step of monitoring impact at theproject level has been more difficult. Only 20 per-cent of IDA projects use gender-disaggregatedmonitoring indicators, and 70 percent of theseare in health. And there is still no Bank-wide sys-tem to provide feedback on progress in main-streaming gender.

Integrating gender in country assistanceprograms. Overall, the diagnostic treatment ofgender in IDA Country Assistance Strategies(CASs) has improved significantly over the reviewperiod. Some 66 percent of a sample of IDACASs since 1997 include satisfactory gender analy-sis, compared with 7 percent in the two years pre-ceding IDA10. But even in CASs with good genderanalysis, the link between diagnosis and the pro-posed lending program remains mainly in theareas of education and health. Moreover, projectsand programs show no improvement in inte-grating gender considerations during the past 10years. While 46 percent of IDA projects under-taken (compared with 40 percent of IBRD proj-ects) contain at least some gender analysis, only25 percent include gender-specific actions. Again,most of these projects are in education and someareas of health.

Support for girls’ education, especially incountries where gender disparities are great,


3 2

The IDA10 and 11 replenishment commitments echoed the objectives of the Bank’s 1994 gender policy on reducing genderdisparities and advancing women’s economic opportunities, urging IDA to undertake regular gender analysis, expand support forfamily planning and social services for women (especially girls’ education), and integrate gender in all lending activities. IDA12called for intensifying and monitoring the mainstreaming of gender in IDA operations, expanding dialogue on gender, undertakinggender equality assessments, and helping countries implement national gender equality plans. It also instructed IDA to preparea gender strategy in 1999.

I D A 1 0 – 1 2 C o m m i t m e n t s o n G e n d e rB o x 4 . 1

stands out as an area of strong Bank commitment(see box 4.2). Evaluations of projects closing inFY95–00 indicate that where there are large gen-der disparities in enrollments, IDA has integratedgender concerns into virtually all education proj-ects and has contributed to positive trends infemale enrollment. Where disparities have beensmaller, gender-related issues in education havereceived little or no attention, a missed oppor-tunity to improve the situation for girls.

In the health sector, attention to gender hasfocused mainly on maternal health, throughcomponents to reduce illness and death amongwomen of childbearing age and among childrenby expanding maternal health and family plan-ning services (see box 4.3). Some of these com-ponents have increased women’s access togood-quality reproductive health services,3 butothers have been highly localized, and theirsustainability is uncertain. Most other projectcomponents (such as construction of health clin-ics) are gender-blind. Gender analysis has yet tobe undertaken for many health issues with amajor impact on women, including those relatedto environmental degradation. Many projectslack a sharp focus on results.

In the productive sectors, integration of gen-der concerns has been very limited, and ismainly seen in agriculture and transport. Oper-ations in other productive sectors have virtuallyignored gender issues. A review of projects ineight IDA countries designed to increasewomen’s economic participation finds that assis-

tance has been directed primarily to microlevel,ad hoc activities and that governments have notbeen able to scale up or replicate most of them,although a few notable successes demonstratethe benefits that can be achieved (see box 3.7,which describes the successful microfinanceproject in Bangladesh). In The Gambia, a long-standing national commitment to improvewomen’s economic status and to build the insti-tutional capacity for program implementationresulted in a country-led agenda through whichIDA provided program and institutional support.4

IDA assistance has given women a greatervoice in project implementation and helpedstrengthen institutions to deliver gender-awareprograms in selected sectors and in a few coun-tries. Overall, however, it has not worked withgovernments to strengthen institutional capacitiesfor designing and implementing gender policies.5

Clarifying the gender policy. Budget con-straints and lack of strong country interest havelimited progress in integrating gender into IDA’scountry assistance programs. But internal short-comings are to blame as well. Despite theincreased visibility of gender issues in the Bankfollowing the 1994 gender policy (formation ofthe External Gender Consultative Group in 1996,the Gender and Development Board in 1997, andnew gender thematic groups in 1998), a recentOED evaluation found no consensus in the Bankor in member countries on the scope ofthe Bank’s gender policy and implementation

I n t e g r a t i n g G e n d e r, E n v i r o n m e n t , a n d G o v e r n a n c e

3 3

IDA’s assistance in girls’ education has been a strong feature of its poverty reduction efforts in the past decade. The share of IDAeducation lending going to girls’ education rose from 39 percent in 1996 to 48 percent in 1999. While only 23 education projectstargeted girls at the start of IDA9, about half of the 98 projects in IDA’s education portfolio at the end of IDA11 did so.

The Bank’s 1994 gender policy focused explicitly on girls’ education, requiring that the Bank aim to reduce gender disparities ineducation and take gender into account in its operations in order to remove barriers to girls’ school attendance. In 1997, a girls’ edu-cation program identified 31 active borrowers as having significant gender disparities in education and targeted 15 of them (all inthe Africa and the Middle East and North Africa Regions) for exceptionally large enrollment gaps between boys and girls. The 1999education sector strategy maintained a focus on gender.

Analysis of a small sample of countries shows that IDA has supported both demand- and supply-based interventions, includ-ing construction of schools in remote areas, training of female teachers, and processes for reducing the opportunity cost of girls’schooling. In the cases reviewed, these efforts have had a significantly positive effect, except where exogenous factors impededproject implementation.

I n c r e a s i n g A s s i s t a n c e f o r G i r l s ’ E d u c a t i o nB o x 4 . 2

strategy. Many staff are unaware of even thosepolicy elements on which there is consensus; andthe Bank has yet to establish adequate supportand accountability processes for ensuring pol-icy implementation.6

There are two distinct but related problemswith the current policy. The first is the lack ofconsistency among international norms, Bankgender policy, and IDA12 replenishment com-mitments. Consistent with the platform of theBeijing Conference on Women,7 the Bankfocuses on reducing gender disparities and fur-thering women’s economic opportunities. Butwhile the Beijing platform places women’sadvancement within a gender analytical frame-work, the Bank has not fully made this shift.Even in the health sector, efforts remain focusedon the narrower goal of improving women’sreproductive health. The IDA12 agreement goesbeyond even the Beijing platform by instruct-ing IDA to undertake gender equality assess-ments and to assist countries in implementingnational gender equality action plans. The sec-ond problem is IDA’s focus on reducing gen-der disparities in education and health withoutsimultaneously addressing women’s economicparticipation, which limits the impact of healthand education improvements on women’s well-being, national development, and sustainableimprovements in gender relations.

Lessons. Characteristics of good practice inintegrating gender considerations into countryassistance are described in box 4.4. These lessonsand the findings from the recent OED genderevaluation lead to the following recommenda-tions for strengthening the integration of genderconsiderations into IDA activities:• Clarify the scope of the gender policy. As

a guide to IDA’s operations, the rationale,intent, and scope of the Bank’s gender pol-icy need to be clarified. The statement ofobjective should make clear that the policyis a core part of the institution’s overarchingobjective of poverty reduction.

• Strengthen borrower institutions andpolicies. Most borrowers have formulatednational policies and action plans, which aimto improve the well-being of women and areusually consistent with the principles statedin the Beijing Platform for Action. IDA shouldstrengthen institutions to support the imple-mentation of these policies and plans. Incountries where such policies and plans areweak, support for their strengthening (throughcountry dialogue and nonlending services)should be an IDA priority.

• Integrate gender considerations into CASs.Based on a comprehensive diagnosis, the CASshould explain how IDA assistance will takeinto account the linkages between poverty andgender. This assistance and its underlyingstrategy should be explicitly related to theborrower’s policy framework for gender.

• Integrate gender analysis into the designof IDA-supported projects. Gender diag-nosis should be integrated into the economicand social analysis carried out in the prepa-


3 4

The Fourth Population and Health project in Bangladesh adopted a “gendered” approach to maternal and child health. The OED GenderReview found that project outcomes clearly influenced such positive trends for women’s health in Bangladesh as sharply decliningfertility rates, rising women’s life expectancy, and lower rates of under-five mortality. The project resulted in the expansion of familyplanning–maternal and child health service delivery through satellite clinics in rural areas; a wide range of successful information,education, and communication programs that deal with the responsibilities of both men and women in family health; construction andstaffing of an Institution for Mother and Child Health, with women making up half the staff; establishment of a Gender Issues Officeheaded by a joint secretary for expanding women’s involvement in the Ministry of Health; increased efficiency of three intersectoralpopulation programs; and the recruitment and training of 4,500 female health assistants.

M a t e r n a l H e a l t h i n B a n g l a d e s hB o x 4 . 3

A recent OED evaluation found no consensusin the Bank or in member countries on the

scope of the Bank’s gender policy andimplementation strategy.

ration and design of IDA-supported projectsso that both men and women are able toaccess the benefits equitably. This is especiallycritical for countries with high genderdisparities.

• Strengthen management of the genderprogram and establish a monitoring andevaluation (M&E) system. The Gender andDevelopment Board should be provided withthe necessary authority to ensure gendermainstreaming across sectors and themes asappropriate, and actions should be taken toensure that institution-wide progress is reg-ularly tracked and periodically evaluated.

The Bank’s forthcoming Gender Strategy,soon to be released, provides an opportunity toclearly link gender activities with the overarch-ing goal of poverty reduction and provide aclear implementation strategy. Monitorablebenchmarks will be needed for measuringprogress and effectiveness in implementing thepolicy for mainstreaming gender into IBRD andIDA activities.

Enhancing Environmental SustainabilityThe creation of the Environment Department inthe 1987 reorganization marked environmentalsustainability as a major objective of the WorldBank’s work.8 Participation in the 1992 Rio EarthSummit and publication of World DevelopmentReport 1992: Environment and Development(World Bank 1992a) reinforced that commit-ment. Specific IDA10–12 replenishment under-takings emphasized four priorities for IDA’senvironmental work (see box 4.5):• Integrating environmental concerns into coun-

try strategies• Lending for environmental objectives • Enhancing the application of environmental

safeguard policies• Increasing attention to global environmental

issues.

IDA has broadened awareness of environ-mental issues in borrower countries, particu-larly at the project and program level. It hassupported innovative projects, extended the useof environmental assessments, and conducted


3 5

Influenced by assessments leading up to Rio and by negative public reaction to controversial IDA projects such as the Narmadadam in India, the IDA9 Replenishment Report elevated environmental sustainability to the same level of importance as poverty reduc-tion and broad-based growth. The IDA10 Replenishment Report reiterated that emphasis, while the IDA11-12 reports revised theemphasis slightly, putting poverty reduction at the center and environmental sustainability and broad-based growth in essentialsupporting roles.

More specifically, the Replenishment Reports recommended:

I D A 1 0 – 1 2 C o m m i t m e n t s o n E n v i r o n m e n t a lS u s t a i n a b i l i t y

B o x 4 . 5

• Completion of national environmental action plans (NEAPs)in all IDA countries and their incorporation in CASs and coun-try policy dialogue

• Continued lending for environment projects and integrationand mainstreaming of environmental considerations in allIDA activities

• Improved implementation of safeguard policies, including byassisting borrowers to improve the quality and scope of envi-ronmental assessments

• Support for projects addressing global environmental issues(climate change, biodiversity, desertification) and integratingglobal concerns in country dialogue.

Country Ownership, including visible and actively implemented country gender policy or strategy and strong commitment reflectedin domestic institutions, and the design of assistance consistent with countries’ own action plans.

Analytical work that underpinned and laid a strong technical foundation for IDA’s operations.A holistic approach to gender issues within a country assistance program, including attention to institutional issues.Strong partnerships with other donors and civil society organizations and systematic monitoring of results.

C h a r a c t e r i s t i c s o f G o o d P r a c t i c e si n I n t e g r a t i n g G e n d e r

B o x 4 . 4

significant research and analysis. Direct lendingfor environmental projects increased in IDA9 andthen leveled off, while environmental compo-nents of other projects increased significantly(table 4.1). In many countries, project workdeveloped out of NEAPs, usually prepared withIDA assistance. Still lacking, however, is the fullintegration of NEAPs and environmental sus-tainability concerns into CASs and country dia-logue, and the establishment of criteria formainstreaming environmental concerns in othersectors and linking them to poverty reduction.Weak borrower interest and tight budgets playa part, but so do inadequate guidelines, account-abilities, and incentives for implementing IDAcommitments.

Integrating environmental concerns intocountry strategies. Following replenishmentrecommendations, NEAPs have been completedin nearly all IDA countries, although more slowlythan intended and with mixed quality and follow-up. The combination of high-quality NEAPs,strong government interest, and IDA supportresulted in strong environmental programs insome countries (see box 4.6), but overall the pic-ture is mixed.

In 1999, as throughout the decade, only abouthalf of the CASs reviewed treated environmen-tal sustainability adequately.9 CASs were notrequired to include environmental performanceindicators, nor was environmental sustainabilitya priority in management reviews or a major fac-tor in the policy dialogue for many countries.

IDA’s underlying ESW and other analysis ofenvironmental issues have been good to excel-lent, but environmental analyses for most IDA

countries are more than five years old. Staffworking on environmental issues say that theylack the time, resources, and incentives to applythe analysis. As a result, IDA’s environmentalactions have been less strategic than they mighthave been.10

Lending for environmental objectives. Directenvironmental lending shows no discernibletrend since increasing in IDA9, although thenumber of projects has increased slightly (table4.1). However, environmental components inagriculture, urban, water supply, and energyprojects increased, both in the number of proj-ects and the estimated level of financing, inIDA10 and then leveled off.11

Guidelines for adjustment lending (Opera-tional Directive 8.60) suggest that environmen-tal factors be considered, but they do notencourage environmental reforms through adjust-ment operations. A recent management reviewfound little increase in environmental coveragein adjustment operations over the period and hasnow provided the basis for follow-up action(Reed 1996; World Bank 1997a). A workinggroup (including the Environmentally andSocially Sustainable Development Network,ESSD, and other units) is currently preparing theconversion of the Operational Directive (OD8.60)that will provide guidance on treating the envi-ronment in adjustment operations.

Beyond direct lending, replenishment agree-ments have increasingly stressed the need toaddress environmental concerns in all projects.This emphasis on mainstreaming goes beyondthe mitigation (“do no harm”) actions of theBank’s safeguard policies to encompass positive


3 6

When the NEAP in Bangladesh was found to be a long shopping list of concerns without adequate priorities, the country depart-ment mounted a further mission with the government to establish priorities and translate them into projects. Initially a singleenvironmental project was planned, but the parties decided that it would be more effective to incorporate most of the priority actionsidentified in the NEAP into other projects. IDA did this in several of its own projects, and the United National DevelopmentProgramme also conducted a follow-on project to the NEAP to improve environmental management. As a result, the NEAP wastranslated into an effective program that was mainstreamed into the IDA portfolio.

Source: IDA Review 2001h.

U s i n g t h e N E A P t o D e v e l o p S t r o n gP r o j e c t W o r k s

B o x 4 . 6

(“do good”) aspects of project design in all sec-tors. To date, there are no clear guidelines forpromoting, monitoring, or evaluating main-streaming, suggesting that it is not yet a func-tional priority, despite the replenishmentrecommendations. The recent IDA report onmainstreaming finds that sector strategies(notably energy and urban development) haveincorporated significant environmental inputand committed to better mainstreaming of futuresectoral operations. The report finds only mod-erately satisfactory mainstreaming in CASs, how-ever, and no integration as yet in PovertyReduction Strategy Papers (PRSPs): “While insome cases environment may be missing fromclient priorities for a good reason, there is muchmore to be done to ensure that appropriateattention is paid in the country dialogue to thelong-term sustainability of natural resource useand to the impacts of environmental degrada-tion on development” (World Bank 2001e, p. 4).

Where environmental concerns have been aborrower priority and figured prominently in IDAcountry strategies, results have been impres-sive. IDA lending has appropriately addressed

natural resource management (the emphasis inIBRD lending is on pollution abatement). Proj-ects have helped reverse land degradation (SodicLands Project in India, Loess Plateau Project inChina), strengthened environmental manage-ment (through a series of environmental man-agement projects in Madagascar with coordinateddonor assistance, and in China), and increasedincome generation on arid lands (Arid Lands Proj-ect in Kenya). Environmental project perform-ance and supervision have been on a par withBank lending in general.12

Enhancing safeguards. OED and manage-ment reviews in 1996 and 1997 concluded thatsafeguard policies and procedures were essen-tially sound and had often helped to mitigatepotentially adverse effects of IDA-funded proj-ects and boost environmental awareness. Thereviews also noted major shortfalls: environ-mental assessments have come too late in theproject cycle to influence design, alternativeswere not adequately considered, and supervisionof actions recommended in environmentalassessments has often been weak.


3 7

Estimated IDADirect Direct Environmental financing of

environment environment components, other environmental lending lending sector projects componentsb

Fiscal year (number of projects) ($ millions) (project no.)a ($ millions)

1990 2 191 13 155

1991 1 179 19 290

1992 5 229 34 554

1993 4 202 23 235

1994 4 172 26 581

1995 3 69 33 552

1996 5 349 34 1,063

1997 12 257 25 435

1998 4 149 28 556

1999 5 213 21 437

2000 7 60 27 608

a. Based on sectoral reviews of the agriculture, urban, water supply and sanitation, and energy portfolios. This includes environmental components in sector projects that closed over

1990–00.

b. The value of IDA financing for environmental components has been calculated at the same proportion as the overall IDA commitment in total project costs, as this review did not under-

take a component-by-component breakdown. This approach is conservative and likely underestimates IDA’s support for environmental components.

D i r e c t I D A L e n d i n g f o r t h e E n v i r o n m e n ta n d E n v i r o n m e n t a l C o m p o n e n t s , F Y 9 1 – 0 0T a b l e 4 . 1

Assessments by management and staff haveattributed shortfalls in part to inadequateresources for implementing safeguards. Also,decentralization of responsibility for environ-mental assessments to Bank Regions has createdan awkward conflict of interest for Regionaland network staff, making them both promotersof environmental issues and policemen of envi-ronmental safeguards.

The April 2000 Inspection Panel report on theWestern China Poverty Reduction Project broughthigh-visibility attention to these implementationproblems. It also revealed widely divergentviews among managers and staff on how toapply safeguard policies. Reviews of safeguardpolicies and environmental assessment processesare under way. They seek to improve imple-mentation and find ways to extend safeguardsto adjustment and programmatic lending. Moreresources have been allocated to the safeguardprocess in the FY01 budget, and a new centralunit dedicated to safeguard policy quality assur-ance should impose a more integrated man-agement approach.13

But the larger issue of efficacy remains.Despite a positive impact at the project level, theeffects at the country level have been modestbecause of the limited coverage of IDA-fundedenvironment projects, the lack of follow-up pro-cedures after project closure, and the slow adop-tion of effective environmental assessment

processes in most countries. Many countrieshave established their own environmental assess-ment regulations, but implementation remainsweak.14 Moreover, there has been almost nomove toward a more strategic, sectoral approachto environmental assessments. The ongoingreview of safeguard policies should identifyways to build greater country capacity (box 4.7describes a good case example).

Increasing attention to global environmen-tal issues. IDA countries stand to suffer mostfrom the adverse impacts of global concerns,such as climate change. IDA has provided infor-mation on the local costs and benefits of address-ing global environmental issues, and GlobalEnvironment Facility (GEF) funds have beenused to finance such projects.15 Analysis hasidentified where local and global benefits can bemutually reinforcing and tradeoffs minimized(Hughes and Lvovsky 1999). The emphasis onlocal impacts of global issues has helped indeveloping relevant projects and programs. TheClean Fuels Initiative, for example, seeks toreduce locally harmful particulate emissions aswell as carbon dioxide emissions.16 In addition,IDA has begun to help countries mitigate neg-ative environmental impacts and should expandthose efforts where circumstances and interestmake this a priority.17

IDA also has great potential for addressingregional and transboundary issues, but the strongcountry orientation of its internal organizationmakes it difficult to address such issues. Budgets,programs, and incentives are not designed tosupport this work. Some GEF projects, such asthe Lake Victoria project and the Central Amer-ican Mesoamerican Biodiversity Corridor, haveeffectively addressed regional transboundary


3 8

In 1995 Eritrea issued a National Environmental Management Plan. IDA provided support for its implementation, with an IDF grantto prepare environmental assessment guidelines, legislation, and training. Although the 1998–00 border conflict delayed formalenactment of legislation, the government’s guidelines have been applied. In 2000 IDA undertook a pilot initiative in which the Eritreanenvironmental guidelines would be applied to IDA projects in lieu of Bank guidelines. Three projects have been processed suc-cessfully in this manner.

E r i t r e a n E n v i r o n m e n t a l G u i d e l i n e s : D o i n g t h e R i g h t T h i n g R i g h t

B o x 4 . 7

IDA also has great potential for addressingregional and transboundary issues, but

the strong country orientation of itsinternal organization makes it difficult

to address such issues.

problems with global implications. There has alsobeen IDA lending follow-up to the Lake Victo-ria project. While the GEF and other grantresources offer opportunities for solving trans-boundary problems, these resources are lim-ited; consequently, IDA should do more tosupport cross-border programs when there aresubstantial benefits to neighboring countries. Inthese cases, IDA should modify, as needed, itsheavily country-focused programming and budg-eting in order to bring the countries together tosolve shared problems (in watersheds or otherecosystems that span national boundaries). Theseissues are important to development, and fewother agencies can address them effectively.

Challenges and constraints. Environmentalsustainability is among IDA’s most challenginggoals and a source of tension within the Bank andwith borrower countries. Using environmentaland natural resources is essential to growth andpoverty reduction. Abusing them is a threat tohealth, economic sustainability, and long-termecosystem integrity. Many environmental impactsare indirect and long-term, which makes draw-ing the line between use and abuse difficult.Debates rage between parties interested in short-term gains and those with long-term horizons, andbetween those interested in local and national pri-orities and those interested in global priorities.Opportunities for enrichment from exploitationof natural resources can lead to corruption andimpede sound policymaking.

IDA’s primary external constraint in pursuingits environmental commitments is the lack ofinterest and commitment among member gov-ernments. Many countries do not want to bor-row from IDA to address environmental issueseither because they do not view them as a pri-ority18 or because they prefer to use grant funds.In several cases, the GEF has been the onlyway to bring environment projects into countryprograms.19 But these projects are no substitutefor IDA operations, because the GEF has a verydifferent mandate—-to provide incremental fund-ing for cross-border or global issues, not povertyreduction.

Internal factors also affect IDA’s perform-ance. The environment has been treated con-

ceptually and operationally more like a sectorthan a crosscutting theme, and the links to IDA’spoverty reduction work are still undefined. Fre-quent reorganizations have confused account-abilities for environmental goals. The Bank’snetwork role in setting, supporting, and moni-toring objectives needs clarification and strength-ening. While high-profile crises attract attentionand galvanize action (as in the follow-up to theWestern China project), over the long-term crisesappear to make borrowers and managers risk-averse in program design.

Recommendations. The review of IDA’s assis-tance in ensuring environmental sustainabilitypoints to a number of lessons, some broadly sim-ilar to those that have emerged from the evalu-ation of IDA’s work on gender. The lessonsinclude the importance of strengthening coun-tries’ own policies and institutions; better inte-grating environment in CASs and sectorstrategies; making environmental sustainabilitya central feature of policy dialogue, with par-ticular attention to the links between poverty andenvironment;20 integrating environmental analy-sis into project design to foster environmentalquality and sustainability (with safeguards andmitigation serving as a minimum threshold, nota first line of defense); and further strengthen-ing accountability within the Bank for environ-mental and safeguard policy compliance.

Specifically, the findings and lessons lead tothree main sets of recommendations (these areoutlined in more detail in the environment back-ground study for this review and are taken intoaccount in the design of the new environmentsector strategy):• IDA should build on its comparative advan-

tage and analytical capacity to demonstrate thecritical role of the environment in sustainabledevelopment and poverty reduction. It shouldincorporate environmental objectives into its


3 9

IDA’s primary external constraint inpursuing its environmental commitments isthe lack of interest and commitment amongmember governments.

country assistance and sector strategies andenhance its efforts at country capacity build-ing, focusing on strengthening institutions,policies, and regulatory enforcement.

• IDA should strengthen its accountability forincorporating environmental goals into itsactivities where the environment is a prior-ity issue and for complying with the Bank’senvironmental safeguard oversight systemand process. In parallel, it should modernizeand adapt its policy framework to the chang-ing practices and instruments being used,and place greater emphasis on helping bor-rowers build their own capacities to formu-late and implement environmental policiesand manage environmental resources andrisks.

• In helping to advance the global environ-mental agenda, IDA should concentrate itsassistance on the local and national benefits,and increase its attention to regional (trans-boundary) issues in its analytical work andprograms.

Overall, IDA should concentrate its resources,in partnership with other actors, on assisting theefforts and strengthening the institutions of coun-tries that demonstrate by their policies and

actions a commitment to environmental sus-tainability. Where countries are not committedto addressing environmental problems, IDAshould work toward informing policymakers ofthe importance of environmental considerationsto strategic development priorities through bet-ter treatment of the environment in the diagno-sis underlying CASs, collaborative analyticalwork, and policy dialogue.21 Moreover, experi-ence around the world has shown that govern-ment intervention—through regulations andmarket-based incentives—is essential to safe-guarding the environment. Governments havea great deal of control over the allocation of nat-ural resources, so linkages with IDA’s efforts tosupport effective governance are also importantto environmental objectives.22

Improving Governance With poverty reduction front and center ofour agenda, our work at the “rock-face” mustbe on governance, institutions, and capacitybuilding.

—James D.Wolfensohn

1999 Annual Meetings Speech

IDA was slow to implement replenishmentundertakings in IDA10, not only on such sensi-tive matters as the rule of law, transparency, andaccountability, but also on capacity building,institutional development, and public sectormanagement (see IDA Review 2001e). Butmomentum accelerated in the latter half of the1990s (see box 4.8). World Bank President JamesWolfensohn gave new prominence to gover-nance issues, emphasizing public sector reformas an integral part of the Bank’s poverty reduc-


4 0

Recognizing that governance intersects all areas of IDA’s support for countries’ development efforts, replenishment undertakingshave urged IDA to assist borrower governments in strengthening the rule of law and public accountability systems and becomingmore transparent and participatory. Each CAS is to show how governance is being addressed, and PERs are to be used tostrengthen public expenditure management and to assess whether nondevelopmental spending (such as on the military) iscrowding out social expenditures. Replenishment undertakings also urged IDA to assist judicial reform and anticorruptionmeasures, encourage broader participation in governance reform, and tighten the connection between lending levels and gover-nance performance.

I D A 1 0 – 1 2 R e p l e n i s h m e n t C o m m i t m e n t so n G o v e r n a n c e

B o x 4 . 8

IDA should concentrate its resources,in partnership with other actors, on

assisting the efforts and strengthening theinstitutions of countries that demonstrate a

commitment to environmental sustainability.

tion mission and instructing country teams toaddress corruption openly. A major effort fol-lowed to expand capacity for operational workin governance through increased staffing, ana-lytical work, and policy guidance; to increaselending for public sector reform and other ele-ments of governance; and to link overall lend-ing more closely to countries’ governanceperformance (discussed in Part III).

Building Bank competencies. Efforts havebeen under way during the past three years tobuild staff capacity in public sector reform. ThePublic Sector Group, established in 1997, quicklyachieved a strong record in developing infor-mation and practical guidelines on best practices.The Bank recently recruited experts in publicaccounting and auditing and is building up itsteam of legal and judicial reform specialists.

Analytical and policy work on governance hasalso expanded in the past three years.23 The mostrecent policy statement, Reforming Public Sec-tor Institutions and Strengthening Governance,proposes a more proactive approach to gover-nance and institution building. Three BankRegions—Africa, Europe and Central Asia, andLatin America and the Caribbean—have recentlyproduced special regional studies on gover-nance.24 The Bank’s Development EconomicsVice Presidency has done valuable analyticalwork,25 and the World Bank Institute (WBI) hasmade helpful contributions.26 The Bank hasincreased its analytical work on public financialaccountability and related policy guidelines,although it has been slow to take a systematicapproach to assessing and strengtheningborrowers’ public financial accountability sys-tems, which have repeatedly been shown to beimportant for achieving desired outcomes ingovernance.27

Fuller treatment of governance in CASs andlending. Governance now figures prominentlyin CASs, which are widely discussed with stake-holders in each country. Governance consider-ations have also been given greater weight incountry performance-based lending allocations,though not as effectively as they need to be (seePart III).

The coverage of governance and public sec-tor reform in lending programs is still partial, heldback by resource constraints and limited coun-try ownership. Nonetheless, the number of pub-lic sector reform projects and multi-sector projectswith significant public sector reform compo-nents more than doubled from 45 in FY97 to 96in FY99.28 Structural adjustment operationsaddressing public sector reform also more thandoubled from FY98 to FY99, to 22.

Satisfactory ratings for completed public sec-tor management projects rose as well, from 54percent in FY95 to 94 percent in FY98 (dropping

slightly to 91 percent in FY00). In FY98–00 morethan 75 percent of the rated projects were con-sidered likely to be sustained, and two-thirds tohave a substantial institutional developmentimpact (all ratings exceed the average for Bankprojects overall). And because successful reformrequires strong ownership, broad participation,modest investment, long implementation, andtimely technical assistance, the Bank has exper-imented with adapting lending instruments forpublic sector reforms.29

Still, in most countries the dialogue on gov-ernance reform is just beginning. IDA has yet toput governance squarely on the table in a con-sistent way for all borrower countries. It does notalways give core elements of public sectorreform—from improved budgeting, auditing,and accountability to the rule of law and trans-parency in public sector activities—the attentionthey deserve in country assistance programs.And it still faces a challenge in strengtheningcountry ownership and political commitmentfor reform (box 4.9). Governance is not an easyagenda—for IDA or for borrowers. Buildingdomestic consensus for reform is critical. Becausetransparency is a key ingredient, IDA needs todo more to “walk the transparency talk.” Andprogress will require close coordination with theIMF and other development agencies.


4 1

Coverage of governance and public sectorreform is still partial, held back by resourceconstraints and limited country ownership.

Emerging issues. Assessment of the countrywork relating to governance raises four mainissues: • Consistency of treatment. IDA’s reporting and

country dialogue on governance haveexpanded and become increasingly candid,but country treatment is not consistent. Suchinconsistencies are hard to explain to bor-rower countries and risk undermining thedialogue in cases where a firmer stance hasbeen adopted. In addition, better harmo-nization of donor assistance strategies ongovernance is needed to avoid underminingefforts to build national ownership and con-sensus in support of effective government.

• Rule of law and judicial reform. Many stud-ies show that the rule of law is important notonly in creating an environment conductiveto investment but also in alleviating personal

insecurity, a principal concern of poor peo-ple (Dollar and Kraay 2000, 2001; World Bank2001d). Until recently, IDA viewed assistancein establishing the rule of law largely as partof promoting a conducive environment forprivate investment.30 In a few countries ithas begun to support broader reform of legaland judicial systems, and has started to pre-pare more such efforts. Especially given thatsome key areas of this work (such as lawenforcement) are not part of IDA’s compar-ative advantage, IDA should, consistent withCDF principles, closely coordinate its legal andjudicial reform activities with its develop-ment partners.

• Institutional capacity. IDA has increasinglyincorporated institutional concerns in its activ-ities, from public expenditure work to civil ser-vice reform and tax administration reform.31


4 2

The May 1998 OED Country Assistance Note on Kenya concludedthat IDA assistance had not been effective in promoting reforms.Given weak compliance with IDA conditionality, and the poorgovernance situation, the note recommended limiting lending tosmall, poverty-targeted interventions and shifting the focus tononlending activities.

The September 1998 CAS envisaged Kenya to be in a low-case lending scenario for the three years of FY99–01. Base-caselending levels including budget support were directly linked toimprovements in governance. The focus on governance to trig-ger the base-case scenario was a first in IDA history. If the basecase were triggered, the volume of lending in the first yearwould be around $100 million, but strong progress in imple-mentation would justify increasing annual lending toward thehigher end of the range. IDA would also invest in nonlendingservices.

IDA’s stance of suspending new lending to Kenya (coordi-nated with other donors) led the government to appoint theChange Team in July 1999 and to initiate economic governanceand policy reforms. In August 2000, IDA approved a $150 mil-lion budget assistance loan (the EPSRC), although the condi-tions for such support, as specified in the September 1998 CAS,were not fully met. Following the EPSRC, two AIDS projects andan Emergency Energy Project were also approved. Total IDAcommitments between FY99 and December 2000 have been$362 million, of which $322 million was approved in the last sixmonths of 2000.

Although some progress has been made in the design ofgovernance reforms, there has not been much progress in termsof effective implementation. Improvements in governance werenot sufficient to merit a move to large-scale lending, becauseit was predicated on strong progress in implementation. Therewere also several additional factors. One is the continuing riskof policy reversals, as in the past. This risk has now material-ized with a delay in the sale of Kenya Telecom (a central trig-ger for IDA assistance in the 1998 CAS); with the declaration ofKenya’s Anticorruption Authority as unconstitutional, strippingit of investigation and enforcement powers; and with the pas-sage of a bill in the Parliament to cap commercial bank inter-est rates, undoing liberalization in the financial sector. Second,the impact of the governance reforms at the central level had notbeen felt at the local level of communities. And third, importantlaws, bills, and circulars expected to be passed have not yet beenapproved by Parliament.

IDA’s approval of fast-track emergency projects also raisesthe issue of IDA’s role in a crisis situation in a country with a poorgovernance environment. There are good reasons to engage incrises, but rapid-disbursing program lending should not antic-ipate reforms, but rather should depend on solid and evidentprogress in the implementation of the governance reforms, espe-cially where institutional changes are required. For this reason,IDA has held back the disbursement of both the second trancheof the EPSRC and further disbursements under the EmergencyEnergy Project.

K e n y a : A C a s e S t u d y o f W i t h h o l d i n gL e n d i n g t o F o s t e r G o v e r n a n c e R e f o r m s

B o x 4 . 9

Still, mainstreaming governance concerns in allsectoral work and strengthening public expen-diture management remain major challenges.Tackling crosscutting and systemic institutionalproblems calls for long-term, integrated, multi-sectoral CASs for institution building that areprepared with development partners.

• Public financial accountability and trans-parency. Accountability is a key part of goodgovernance. Weak public financial accounta-bility systems in many IDA countries under-cut program implementation. Both fiduciaryconcerns and development effectiveness callfor more attention to this matter. As an under-pinning for future IDA lending, borrowersshould commit themselves to setting up soundfinancial accountability systems and agree toput in place—with external assistance ifneeded—monitorable plans for achievingthem. Since establishing such systems willtake time, the commitment should be to steadysteps toward a robust and transparent systemof public financial accountability. Transparencyis also important, but IDA (and other assistanceagencies) does not demand that project reportsfrom borrowers be made public. As part of abroader effort to harmonize donor reportingrequirements, IDA and its development part-ners should require the public agencies theyassist to prepare and make public regularreports on their activities, to inform the pub-lic about their performance, and to set mon-itorable performance targets.

Lessons The three issues addressed here—gender, envi-ronment, governance—have been highlighted askey components of the IDA10–12 “broad-basedframework for poverty reduction.” In drawingtogether the lessons of development experienceover the past decade, the World DevelopmentReport 2000/2001: Attacking Poverty (WorldBank 2000a) confirms their relevance in itsemphasis on opportunity, security, and empow-erment as essential elements of a comprehen-sive attack on poverty. Experience in the pastseven years offers three sets of lessons for IDA.

First, IDA has provided important support tocountries’ development efforts in these cross-

cutting areas, but real progress has usuallyhinged on the existence of a consensus forchange in the borrowing country. Positive out-comes have occurred where assistance hasresponded to or encouraged country policies andaction plans. This implies an approach of coun-try selectivity, which concentrates resourceswhere the greatest opportunity for gains exists,and support directed to strengthening coun-tries’ own institutions as the entry points for IDAsupport. In countries where national policies areweak, IDA should support their strengtheningas a priority through collaborative analyticalwork, policy dialogue, and institutional capac-ity building. Strategic selectivity also means leav-ing the field to other agencies better able to movethe agenda forward in specific areas.

Second, IDA needs to clarify the intent andscope of its policies for “mainstreaming” bothgender and environmental sustainability, linkingthem clearly to the goal of poverty reduction. Italso needs to further strengthen the managementof its gender and environment programs. For this,Networks need to articulate strategies that setachievable objectives for which IDA is preparedto commit the necessary resources and developindicators and processes for tracking perform-ance. Regions need to ensure that these issuesare incorporated into CASs where diagnosisshows the issue to be a development priority.QAG and OED need to give more attention tothe mainstreaming of these issues in their eval-uative frameworks.

Third, good governance affects all of a coun-try’s development efforts and all areas of IDA’sassistance. Moving governance and institutionaldevelopment issues to the center of the IDAprogram has been a major accomplishment ofthe past several years. But achieving significantreform remains difficult. IDA and its borrowersand other development partners must accept jointresponsibility for this agenda and distinct


4 3

Real progress has usually hinged on theexistence of a consensus for change in theborrowing country.

accountabilities for improved outcomes. IDAneeds to sustain the recent momentum in atten-tion to governance, while broadening the cov-erage and increasing the consistency of treatmentacross countries. With other assistance agen-cies, IDA needs to improve and deliver on itssupport for institutional development and acceptthat it is a long-term process. In particular, IDAand its borrowers and other partners need tocommit together to a major drive to improve pub-lic expenditure management and public finan-cial accountability, reinforced by increasedtransparency.

IDA’s support of a broad-based approach topoverty reduction in borrower countries hasbecome increasingly relevant in the period under

review. But IDA is still learning how to set pol-icy priorities that promote equitable and sus-tainable growth that include the poor, bothwomen and men, and learning how to helpgovernments become more effective andaccountable in delivering public services requiredfor a well-functioning economy. The main taskover the next few years is to consolidate suchlessons. Much of what needs to be done involvessetting sharper priorities clearly linked to theoverarching objective of poverty reduction andstrengthening implementation strategies to bet-ter take account of institutional and policy weak-nesses, country circumstances, and the actionsof development partners. These are largely mat-ters of process, which Part III addresses.


4 4

4 5

The Process Dimension

Part III

Along with refocusing its development agenda, IDA has been at the forefront offundamental change in the “business” of aid. IDA replenishment commitments,aimed at improving development effectiveness, have emphasized three aspects ofaid quality: greater selectivity, increased participation, and donor coordinationin support of countries’ own development efforts.

IDA’s progress was moderate during IDA10–11, but the ComprehensiveDevelopment Framework and Poverty Reduction Strategy Paper initiativesintroduced after IDA12 negotiations have given significant impetus to this aidreform agenda. Consolidating and institutionalizing these new processes willrequire more-structured aid coordination at the country level, in support ofrecipients’ poverty reduction strategies, and greater harmonization of processesand procedures at the corporate level. Equally, IDA needs to further adjust its owninstruments and procedures and should be encouraged by its donors to facilitatea time-bound process among agencies to achieve the long-overdue harmonizationof aid practices so as to reduce the transaction costs borne by IDA borrowers.

4 7

Improving Programand Country Selectivity

To improve IDA’s development effectiveness, the IDA10–12Replenishment Reports urged IDA to enhance the role of CountryAssistance Strategies (CASs) and strengthen selectivity in resource

allocation (see box 5.1).1

Enhancing the Role of the CountryAssistance StrategyIDA has increasingly focused on the country—not just the project—as the unit of account in thedesign and assessment of assistance. CASs havebeen the main program-planning vehicle forthis shift. To better align IDA assistance programswith replenishment commitments, donor gov-ernments proposed during IDA9 replenishmentnegotiations that IDA prepare a multi-year CASfor each borrower, which would be discussedwith the Board of Executive Directors.2 Sincethen, replenishment recommendations havegiven increasing attention to the CAS as a plan-ning document and as a process for sharpeningIDA’s poverty focus and increasing its pro-grammatic selectivity.

IDA has largely complied with the recom-mendations to make CASs more poverty focused,more comprehensive in their diagnosis of devel-opment challenges, and more participatory as aprogramming process. The CAS has increas-ingly become a vehicle for strengthening coun-try ownership and enhancing aid coordination.But CASs have not yet achieved the degree of

strategic selectivity called for in IDA11 and 12,which is based on greater specialization andcoordination with other assistance agencies at thecountry level.3

CAS content and lending programs. Since thestart of IDA10, internal CAS guidelines haveevolved broadly in line with replenishment rec-ommendations. Current guidelines call for a com-prehensive diagnosis of countries’ povertychallenges and identification of measures to speedimprovement. They stress selectivity and part-nerships and encourage stakeholder participationin CAS preparation. As of July 1, 1999, they havemandated the public release of IDA CASs.

Since 1998, management has issued threeretrospectives on how well the CAS processmeets those guidelines.4 These and other reviewsfind notable improvements in the coverage andcontent of CAS documents. Analytical work isoften reflected in the CAS,5 though analyticalweaknesses still constrain the policy relevanceof the diagnosis: The poverty focus and diagnosisof CAS documents have been strengthened (asnoted in Part II).

55

• Governance has received significantly moreattention in recent CASs.

• Attention to the role of other donors has alsoimproved.

• Problems remain with the timeliness of PERsand the uneven treatment of gender, envi-ronment, and private sector development(PSD) (see box 5.2 on the increasing use ofjoint IDA-IFC CASs, which should improveattention to PSD).

Most CASs are now prepared through a par-ticipatory process involving the borrower gov-ernment, and usually civil society representativesand donors. Consultations carried out for thisreview found that participants in some countriesviewed the process as an effective effort that hadinfluenced the CAS design. In other countries,participants expressed frustration with the appar-ent lack of impact beyond greater informationon the IDA program.6

A review of lending programs in a sample ofcountries (including the countries visited in IDA

review consultations) showed a reasonably closecorrespondence between the program discussedin the CAS and actual lending, although manyprojects were not tightly linked to the underly-ing CAS analysis (see Part II).7 The CAS processwas also found to be flexible enough to dropor delay lending when circumstances change orgovernments prove reluctant to carry out agreedreforms, or to introduce new operations whenpriorities change, especially in emergencies orwhen unexpected opportunities emerge, such asa change in government. Nonlending services areless well discussed and prioritized, both the“due diligence” economic and sector work (ESW)and the program-specific analytical work thatunderpins lending.

Strategic selectivity. Improvements in cover-age and consultation have not yet been accom-panied by much increased strategic selectivity inthe design of country programs. Many CASspresent criteria for choosing among options,but proposed lending operations do not con-sistently reflect those criteria.8

The CAS retrospective analyses of 1998 and1999 found that a majority of CASs discuss coor-dination with other donors and comparativeadvantage in broad terms, but fewer than 40 per-cent cover selectivity in program design for keyareas. Recent CASs have been more attentive to


4 8

IDA10–12 replenishment recommendations urged IDA to be moreselective within countries by concentrating on areas of its com-parative advantage, and more selective among countries by direct-ing more assistance to borrowers with sound policy environments.

To achieve greater program selectivity, the recommenda-tions encouraged IDA to enhance the role of its CASs in settingprogram priorities, taking account of country needs and theactivities of other assistance agencies. They urged that CASsfocus more selectively on actions to speed poverty reduction;provide a comprehensive diagnosis of the country’s developmentchallenges—drawing on the results from major building blocksof IDA’s analytical work, such as Public Expenditure Reviews(PERs), Poverty Assessments (PAs), and environmental actionplans—and be finalized only after discussion with the Board.They also urged that the borrower government, civil society, and

other donors participate in preparing the CAS to better align theIDA program and the country’s development priorities and thatthe CAS be made public.

To achieve greater country selectivity, recommendationsencouraged IDA to adjust its performance-based allocation(PBA) system to link lending allocations more closely to coun-tries’ policy performance. They also called for assigning greaterweight to governance, environmental factors, and nondevelop-mental expenditures; applying performance criteria more trans-parently and more consistently across countries; and openingthe system to more public scrutiny.

To forge more effective partnerships, the recommendationscalled on IDA to engage greater stakeholder participation in proj-ects and programs and improve coordination with other agen-cies and strengthen country-led management of aid.

I D A 1 0 – 1 2 R e p l e n i s h m e n t C o m m i t m e n t s o nI m p r o v i n g D e v e l o p m e n t E f f e c t i v e n e s st h r o u g h S e l e c t i v i t y

B o x 5 . 1

A majority of CASs discuss coordination withother donors and comparative advantage in

broad terms, but fewer than 40 percent coverselectivity in program design for key areas.

the activities of other donors, but few link IDA’sstrategy with its comparative advantage or pri-oritize activities by sector and instrument (anIDA12 recommendation). As the 1999 retro-spective noted, while many CASs use the vocab-ulary of strategic selectivity—mentioningcomparative advantage, track record, choice ofinstruments, and so on—few actually use selec-tivity itself as an analytical tool for focusing oncertain sectors or themes and exiting from oth-ers. Furthermore, very few spell out why par-ticular lending or nonlending instruments wereselected.9

Several factors have contributed to this lim-ited progress in programmatic selectivity:• A reluctance to withdraw from activities and

to rely on partners to assume responsibil-ity.10 Once significant staff and budgetaryresources have been invested in identifyingand preparing a new area of operation, thatactivity tends to remain in the program evenif it is no longer fully consistent with strate-gic selectivity. Borrowers also often prefer thatIDA stay and that it take the lead across a widerange of issues.

• Contradictory directions set by replenishmentcommitments. Endorsements of overall selec-tivity and selectivity within country programsare accompanied by exhortations for IDA toengage in the full range of development issues.Recommendations of new areas of emphasis,such as governance or the financial sector, arenot offset by specific recommendations forreduced attention elsewhere. The Replenish-ment Reports have adopted the rhetoric ofpartnership but have not yet consolidatedmandates or explicitly recognized where IDAcould “comply through others.”

Programmatic selectivity improved in FY00CASs, although progress is uneven. The Mozam-bique CAS notes, for example, “to the extent thatother partners have a strong presence and com-parative advantage [in the social sectors] and arewilling to finance activities on a grant basis, theBank will scale back IDA’s financing contribu-tion.” The proposed lending and nonlendingprograms seem to apply this principle well. Thelatest CASs also show a marked increase in per-formance benchmarks, which should enhanceattention to development results. In three self-evaluation pilots (Bolivia, Mali, and Yemen),progress benchmarks that clearly differentiatebetween country and IDA performance indica-tors will be used in ongoing evaluations of pro-gram impacts by the country teams. Improvedmonitoring at the country, sector, and project lev-els will also be needed if benchmarks are to beused more often to encourage strategic selectivityfavoring the areas in which IDA is demonstra-bly contributing to results on the ground.

Building in the CDF and PRSP processes. TheComprehensive Development Framework (CDF)and Poverty Reduction Strategy Paper (PRSP) ini-tiatives have major implications for the CAS.The PRSP process is expected to become thevehicle for applying the CDF principles to coun-try assistance efforts and for focusing IDA andother donor support on countries’ own devel-opment priorities. In addition, PRSPs and jointBank-Fund assessments of them will providecomprehensive diagnosis for the bulk of IDA

I m p r o v i n g P r o g r a m a n d C o u n t r y S e l e c t i v i t y

4 9

Joint IDA-IFC CASs, introduced in FY97 to foster closer coordination and cooperation with IFC, may lead to more strategic atten-tion to private sector development. More than half of IDA CASs in FY00 were prepared jointly with IFC. These CASs feature a shareddiagnosis and agenda for action for private sector development. The 1999 CAS retrospective study notes that joint CASs have promoteddialogue on IFC’s role and strategy in the country beyond individual operations. The Azerbaijan CAS (FY00) specifies how IFC activ-ities will complement IDA assistance to the financial sector and identifies a number of priority areas for IFC involvement in pri-vate sector development. A review of FY00 IDA CASs found that joint CASs generally contain a more coherent, comprehensive treatmentof PSD and financial sector issues and often include a discussion of past experience and lessons learned in these areas.

J o i n t I D A - I F C C o u n t r y A s s i s t a n c eS t r a t e g i e s H i g h l i g h t P r i v a t e S e c t o rD e v e l o p m e n t

B o x 5 . 2

The latest CASs also show a marked increasein performance benchmarks.

country lending and nonlending programs. Theparticipation built into the PRSP process willlessen the need for participation in CAS prepa-ration. Management has indicated that as of July1, 2002, CASs will normally be based on PRSPsand will become business plans for the WorldBank Group.

While CAS content will change, the chal-lenge of achieving greater selectivity in pro-gram priorities and instruments and of settingperformance benchmarks will not. Countries’presentations of their own strategic prioritiesand monitorable targets in their PRSPs shouldmake this easier. But CASs will still need tomake strategic choices based on IDA’s corporategoals and comparative advantages and the activ-ities of other assistance agencies.

Experience shows that IDA’s comparativeadvantage is mainly at the strategic level—help-ing countries to improve their overall economicmanagement and policy framework, set sectorstrategies, and implement sector-wide reforms.This view of IDA’s role was widely endorsed incountry consultations. New programmatic lend-ing in support of countries’ PRSPs conforms tothis strategic role, although to be effective itrequires a major investment in improving thepublic expenditure management systems andfinancial accountability of member countries.As discussed below, IDA should also make moreuse of sector-wide programs as building blocksof the PRSP process, basing its involvement onits corporate objectives and implementationstrategies as set out in Bank Sector StrategyPapers.

Converting the CAS to an IDA business planthat supports a country’s poverty reduction strat-egy also has implications for PBA. The PRSP, CAS,and IDA-PBA process all need to be reconciledwith each other, as discussed below.

Strengthening the System ofPerformance-Based AllocationsIDA’s PBA system for rationing lending resourcesis its principal mechanism for achieving countryselectivity. The system has evolved over theIDA10–12 period to reflect new knowledge andIDA replenishment recommendations. Thesechanges have strengthened the link betweencountries’ policy and institutional performancesand lending levels. Further adjustments are neededto fully ensure equitable treatment of countries,to strengthen the links between actions that mat-ter for poverty reduction and CAS lending scenariosand lending allocations, to apply performancecriteria transparently and consistently, and to openthe PBA system to more public scrutiny.11

Since 1977, the Bank has had a system thatgenerates an indicative allocation of all IDAfunds available for the next three years based oncountry performance. Under the system’s currentterms, the evaluation of borrowers’ perform-ance is based on the implementation of soundpolicies and institutional arrangements con-ducive to sustainable economic growth andpoverty reduction within a framework of goodgovernance, as assessed through the Country Pol-icy and Institutional Assessment (CPIA), and theeffective implementation of their IDA portfolio(see box 5.3).12

Today, the PBA formula directs moreresources to good performance than it did at thebeginning of the period.13 For every $1 percapita lent to the median IDA borrower inFY93–95, about $1.20 went to those ranked inthe top quintile by performance and $0.85 tothose in the lowest quintile. By FY98–2000, thespread had widened, with $2.10 per capita goingto the best performers and $0.60 to the poorest.But there is relatively little difference in lendingper capita among the middle three quintiles, and-because many in the top quintile are relativelysmall, the bulk of IDA lending, in absoluteterms, goes to the middle-rated performers.

Complying with design and implementationrecommendations for the PBA system. Con-sistent with replenishment recommendations,the design and implementation of the PBA sys-tem have been improved:


5 0

IDA’s comparative advantage is mainly at thestrategic level—helping countries to improve

their overall economic management andpolicy framework, set sector strategies, and

implement sector-wide reforms.

• Policy performance criteria have beenamended to reflect increased knowledgeabout the causes of growth and povertyreduction and instructions in IDA replenish-ment agreements.

• Staff have worked to improve the consis-tency of assessment and review processesacross countries.

• Allocations take into account borrowers’ per-formance on existing IDA credits.

Country allocations may be raised above nor-mal PBA levels under certain circumstances,such as for post-conflict borrowers (see box5.4). In FY00, allocations above the norm alsocontinued for many transition economies in

Europe and Central Asia and for a few better-performing Sub-Saharan African countries. Forpost-conflict countries, a modified process withexplicit criteria for eligibility has been estab-lished. In other cases, the CAS will need toexplain the rationale for the deviation.

Improving the PBA system. Notwithstandingthe many improvements, shortcomings remainin the design and implementation of the PBA sys-tem. These shortcomings relate to two issues thatare fundamental to the system’s relevance andcredibility: the extent to which the systemensures reasonably equitable treatment acrosscountries, and the strength of the links betweenCPIA performance criteria and countries’ further


5 1

The PBA system annually allocates all IDA funds available overthe following three years in a three-step process.

Step one: assessing country performance. Staff prepare twoannual ratings of each borrower’s performance, the CPIA ratingand the IDA portfolio performance rating, currently combined inan 80 to 20 ratio. CPIA ratings derive from an unweighted aver-age of 20 indicators that relate to four sets of concerns: economicmanagement, structural policies, policies for social inclusion,and public sector management and institutional arrangements.Policies, not actual outcomes, are key considerations. The com-bined performance rating is reduced by a third for borrowersfound to have severe governance problems—the “governancediscount.”

Step two: determining the indicative allocations. Perfor-mance ratings are factored into a formula to arrive at a three-year indicative, or “norm,” allocation of IDA resources for each

country. The formula ensures that the resource allocation risesfaster than the performance rating, allowing a considerablereward for countries at the upper end of the performance scale.Population size and GNP per capita are also factored in. IDA man-agement determines the approved allocations in light of thenorm allocations and other relevant factors. Maximum and min-imum lending conventions apply, and allocations for some spe-cial case borrowers, including effective IDA-blend and eligiblepost-conflict countries, are not driven by the formula.

Step three: relating the allocations to CASs. The approvedallocation generally forms the base case lending scenario for eachCAS (or CAS update) presented to the Board for review. But CASsdo not provide information on the country performance ratings orthe norm and approved allocations underlying the base caselending scenario. There is no link between the PBA system andthe alternative CAS lending scenarios (the high and low cases).

T h e P e r f o r m a n c e - B a s e d A l l o c a t i o n S y s t e mB o x 5 . 3

The PBA system has been modified to better align IDA allocations with the exceptional financial needs of post-conflict IDA countrieswhile still maintaining a focus on policy performance. The modified allocation process includes: • Eligibility criteria for countries warranting exceptional treatment• Reengagement criteria to determine the size of the initial allocation• Post-conflict progress indicators to assess performance in post-conflict countries eligible for exceptional treatment• Limits on the duration and volume of exceptional allocations.

The post-conflict progress indicators consolidate CPIA indicators, refined and regrouped to reflect the realities of countriesaffected by conflict, with a new cluster of indicators relating to public security, reconciliation, and demobilization and disarmament.

A N e w A p p r o a c h t o I D A A l l o c a t i o n s t oP o s t - C o n f l i c t B o r r o w e r s

B o x 5 . 4

poverty reduction. To improve the handling ofthese two issues, further adaptations of the sys-tem are suggested.14

• Establish a written record. Staff ratings shouldcontain written rationales for each CPIA cri-terion. Without them, it is difficult to ascertainwhether criteria have been interpreted uni-formly across countries and regions andapplied consistently (despite the significantefforts of IDA management to improveconsistency).15

• Improve the design of the rating systems. TheCPIA rating criteria, which account for 80percent of the overall performance rating,have evolved to reflect increased knowledgeof the determinants of growth and povertyreduction and IDA replenishment recom-mendations. There is room for further devel-opment of the CPIA and IDA portfolioperformance rating systems (as suggested inbox 5.5).16

• Rethink the governance discount. The gov-ernance discount, introduced in 1998, is notworking as intended, except to show thatIDA takes governance seriously. By the sec-ond year of operation, only borrowers withthe very lowest overall CPIA ratings wereaffected. Yet the discount did not capturesome borrowers whose governance has raisedserious concerns, sometimes to the point ofcurtailing IDA lending. Further study is neededto identify ways to achieve both equitable out-

comes and a more substantial cut in lendingto countries that fail to improve governance,even if their other performance ratings areaverage. In particular, more work is neededon establishing better governance indicators,such as that under way within the Bank,which involves discussions with researchersoutside the Bank and IDA’s partners.

• Relate the PBA system to the CAS and PRSPprocesses. Although the approved IDA allo-cations that emerge from the PBA process arereflected in most CAS base case lending sce-narios, two improvements are needed. First,as recommended in the IDA12 Replenish-ment Report, each CAS should discuss therationale for the base case lending scenario.This is especially needed when the approvedallocation differs from the norm allocation orwhen it is made without applying the allo-cation formula (as in the case of blend coun-tries). Second, the triggers17 recommended inthe CAS for moving to higher levels of lend-ing need to be clearly linked to CPIA criteriafor measuring performance. And as the CASbecomes focused on support of countries’PRSPs, consistency will be needed betweenthe CPIA criteria and those suggested in theguidelines for Bank-Fund joint staff assess-ments of PRSPs.

• Increase disclosure. More than anything, theprocess needs greater transparency anddialogue with partners. Recently, the Bank dis-


5 2

For the CPIA system: • Remove criteria not shown by research to be important for

growth and fund research to identify policies that lead topoverty reduction (beyond that achievable by growth alone).

• Assign appropriate weights to the criteria.• Ensure that each CPIA criterion consists of just one element

that can be appropriately rated and reviewed.• Develop consistent definitions for each rating level for each

criterion, with similar interval values between equivalentratings for each criterion.

• Ensure that each criterion rates policy performance rather thanlevel of development, rewarding equally all borrowers doingeverything feasible in their situation.

• Ensure that each criterion can be objectively assessed andthat IDA has the capability to assess it.

• Redefine the governance indicators in light of ongoing con-sultations and research.

For the IDA portfolio performance rating system:• Separate the responsibilities of the borrower and of IDA for

portfolio performance (a task now being worked on by IDAmanagement).

• Remove the counterintuitive (but substantial) positive effecton ratings resulting from the cancellation of poorly perform-ing projects.

• Adjust the rating scale to make it compatible with the CPIArating scale, with which it is combined.

S u g g e s t e d C h a n g e s f o r t h e P e r f o r m a n c eR a t i n g S y s t e m s

B o x 5 . 5

closed the PBA methodology on its Websiteand has discussed it with some developmentpartners. But disclosure has not yet led to thetwo-way dialogue recommended in the IDA12Replenishment Report. Individual perform-ance ratings are not disclosed, although coun-tries are publicly listed by performancequintile. A recent Board decision authorizeddisclosure of ratings to individual borrowersand asked for further analysis of disclosure.Fuller disclosure is needed to ensure equitabletreatment of countries, to share the knowledgeof outside parties in the continuing appraisaland evolution of the system, and to use thesystem as a diagnostic tool for enhanced part-nership rather than simply a means ofrationing IDA resources.18 This change fits inwell with efforts to build more effective devel-opment partnerships through increased coun-try ownership, improved donor coordination,and distinct donor and recipient account-abilities for effective use of aid.

Assisting poor performer countries. Asintended, greater selectivity has resulted inlower lending allocations for countries withpoor policy performance. Nonetheless, IDA’score objective of poverty reduction requires acontinuing engagement with these countries.Poor policy performance can be due to weakcountry capacity as much as to weak commit-ment by leaders. While adjustments in someCPIA criteria could make this distinction clearer,the main issues are to identify assistance instru-ments that can be used effectively in weak pol-icy environments and to ensure the adequacyof budget resources for nonlending serviceswhere lending is limited.

The replenishment recommendations encour-aged IDA to remain engaged in countries withpoor policy performance through well-tailoredcountry strategies and quality nonlending ser-vices, complemented by targeted lending where

appropriate. They also urged IDA to respondpromptly to changes in performance and todevelop a comprehensive approach for activi-ties in post-conflict countries. IDA has broadlycomplied with these guidelines, although proj-ect outcomes have been persistently lower incountries with poor performance. A recent QAGreview of ESW has also found, based on a rel-atively small sample of activities over a limitedtime, that the quality of ESW has also beenlower in countries with poor performance rat-ings, while CAEs and interviews with countrystaff emphasize the inadequate stock of analyt-ical work in these countries.

OED evaluations emphasize five main lessonsfor assisting poorly performing countries:• Give primary attention to governance issues

and weak institutional capacity, which canundermine project implementation.

• Avoid complex project designs and overesti-mating implementation capacity—simple,social fund–type projects may be best.

• Use innovative lending instruments (such asAPLs and LILs) that sequence assistance overtime and use delivery mechanisms that canmore directly reach intended beneficiaries.

• Ensure close supervision and good aid coor-dination, suggest the value of having a pres-ence in the field.

• Apply a differentiated approach to ESW inweaker policy environments, including strongsupervision and enhanced quality assurance(OED 2000a).

One implication of these lessons is the needto have a high level of budget resources rela-tive to lending in weak policy and institutionalenvironments.


5 3

The quality of ESW has also been lower incountries with poor performance ratings.

5 5

Building MoreEffective Partnerships

Building more effective partnerships with borrowers and otherdevelopment agencies is the second major process reform emphasizedin the IDA10–12 replenishment recommendations. Increasing the

participation of borrower countries’ governments, civil society, and theprivate sector and improving coordination in development partnerships arekey features of that reform.

Moving Beyond Project Participation As urged by replenishment recommendations(see box 6.1), participation in IDA projects hasincreased considerably since the start of IDA10,although staff are still learning how to mosteffectively engage stakeholders. Participation inanalytical and policy work is increasing, butquestions of representation, approach, and costsremain.1

Record on participation. Stakeholder partic-ipation in IDA-financed projects increased sub-stantially over the IDA10–12 period: • In a random sample of 88 IDA projects, the

share with participation by primary stake-holders increased from just under 45 percentto 83 percent and the share with a high levelof participation (beyond consultations)increased from 29 percent to 50 percent.

• The share of IDA CASs prepared with par-ticipation by members of civil society groupsrose from 20 percent to nearly 70 percent.

Yet the Bank has no overall policy on partic-ipation and has produced no major policy state-ment since its 1994 report The World Bank andParticipation. Although some 15 OperationalDirectives and Operational Policies address par-ticipation, the lack of clear policy guidelines isparticularly evident in economic and sector work(ESW) and in support to governments to broadenparticipation in policy and sector program design.Thus, progress has been more limited in newareas of emphasis in the replenishment mandates:• Participation has been greatest in Poverty

Assessments (PAs), although the share hasremained at 60–70 percent since IDA10,despite repeated directives to do even more.There has been less participation in PublicExpenditure Reviews (PERs) and other ESW,although borrower governments, civil society,and other donors have been more involvedin recent initiatives.

• Efforts have increased to encourage govern-ments to include broad participation in their

66

own activities and to create an enabling envi-ronment for participation. However, there islittle guidance on this matter and little evidenceon what has been done.2 The Povety Reduc-tion Strategy Paper (PRSP) process representsa significant departure by including broad-based participation in its formulation, imple-mentation, and outcome-based monitoring asone of the five core PRSP principles. Experi-ence with this enhanced participation at thestrategic level should be tracked and evalu-ated with IDA’s partners.

Quality and effectiveness have not alwayskept pace with participation in projects, CASs, andunderlying diagnostic work. Best-practice par-ticipation efforts in projects have often remainedislands of success. Government participation inpreparing CASs has become routine, but thequality and impact of civil society engagementremain highly variable. In the best cases, both IDAstaff and stakeholders have noted an improvedfocus in the CAS. But examples of best practiceare still infrequent, and participants receive lit-tle feedback. Much the same can be said aboutparticipation in ESW. When efforts have beenmade to work collaboratively, quality and capac-ity building have benefited, but the costs are highand examples of best practice are few.

Constraints on effective participation. Per-haps the key obstacle to effective participationis the lack of understanding by IDA and gov-ernment agencies of what needs to be done tohelp communities organize to contribute mean-ingfully to project design and to build institutionsto implement projects that represent diverseinterests. Weak governance also plays a role.Government resistance remains a major con-straint to participation in countries, while staffincentives and resources are the biggest con-straint in the Bank.

Costs are another significant factor. Wide con-sultation is expensive, for borrowers and IDA.For stakeholders (including government andcommunities), the costs of participation averagedbetween 2 and 12 percent of total costs in eightcase study projects.3 Only limited information isavailable on Bank costs, which occur mainly dur-ing the preparation of operations, but theyappear to be substantial, although less thanthose to governments. Three case study projectshad preparation costs of $22,000–$75,000, muchof it covered by trust funds. Costs of participa-tion in CAS preparation ranged from 4 to 30 per-cent of the total (as high as $50,000–$60,000when participation was extensive).

These factors suggest a need to build and dis-seminate best practices in participation and toguard against requirements or expectations thatare too high. Particularly for institution buildingreforms, the strategy articulated by governmenthas to be internalized at the administrative andsubnational levels. Consultation with the pri-vate sector and civil society needs to reinforce,not conflict with, democratic political processes,including those of parliaments.


5 6

IDA10 recommendations emphasized increasing the participation of primary stakeholders in IDA-assisted projects, environmental assess-ments, and Country Assistance Strategies (CASs); giving more attention to governance; and building staff capacity to promote partic-ipation. IDA11 and 12 added increasing participation in ESW, particularly the analytical work to underpin a sharpened poverty focus(the Poverty Assessments, social sector reviews, and Public Expenditure Reviews) and encouraging borrower governments to broadenpublic participation in their own projects and programs. IDA10 also emphasized better coordination among donors, while IDA11 andIDA12 recommended greater programmatic selectivity based on IDA’s comparative advantage. IDA12 emphasized country-ledpartnerships that combine country ownership and donor coordination.

I D A 1 0 – 1 2 R e p l e n i s h m e n t C o m m i t m e n t s o nI m p r o v i n g D e v e l o p m e n t E f f e c t i v e n e s st h r o u g h P a r t n e r s h i p s

B o x 6 . 1

Perhaps the key obstacle to effectiveparticipation is the lack of understandingby IDA and government agencies of what

needs to be done to help communitiesorganize to contribute.

Implications for IDA policy. While efforts tobuild participation into IDA-financed projectscontinue, more emphasis is needed on buildingcountries’ capacity to make their own policy andprogram processes more participatory. It is notclear that this is an area of IDA’s comparativeadvantage and thus its support in this regard mayneed to be worked out in closer partnership withothers.

Improving Aid Coordination Few of the program and process objectivesemphasized in the IDA Replenishment Reportscan be achieved without better aid coordination.During IDA10–12, the focus expanded beyondthe mechanics of donor agency coordination toimproved aid management, led by recipientcountries, with donors directing support to soundcountry-designed policies and programs. Replen-ishment recommendations reflect this evolvingperspective. Still, progress has been limited, anduncoordinated aid programs continue to imposeheavy burdens on recipient countries and tolimit the impact of aid agencies’ programs.4

Compliance in building partnerships. Untilthe past few years, IDA had made only modestprogress in complying with commitments thatentailed actions within its own control. But themomentum has accelerated recently with thePartnership Initiative of 1998 (World Bank 1998c)and the Comprehensive Development Frame-work (CDF) and PRSP processes. IDA’s initialtranslation of these initiatives into operationalpolicies, procedures, and organizational changesand its work with other multilateral developmentbanks on harmonizing processes and proce-dures are encouraging developments. • Most of the progress in strengthening coor-

dination with bilateral aid agencies hasoccurred in the field. Weak commitment hasslowed procedural and policy reforms at theagency level.5 Posting IDA country directorsin the field has improved in-country coordi-nation, and greater use of in-country coordi-nation groups of donors and government hasshown some positive results. Still, complaintspersist about IDA’s inadequate consultationwith other donors on country programs and

its lack of transparency and timeliness insharing information and providing feedback.

• IDA management has conveyed to staff theneed to address programmatic selectivity atthe country level in consultation with bor-rowers and other assistance agencies. Man-agement has also provided increased guidancethrough directives on the CASs and the sec-tor strategies. A recent report on the eightcountry CDF pilots notes progress, though stilllimited, in achieving greater selectivity onthe ground.

• IDA-led Consultative Groups (CGs) and localcoordination forums are moving towardcountry-led partnerships. Some CG meetingsare now held regularly in borrower countries(Bolivia, Ghana, Tanzania, and Uganda).

• The Strategic Partnership with Africa (SPA),launched in 1987 and chaired by the Bank,has continued to evolve and deepen coordi-nation processes. While its initial focus wason quick-disbursing resources to support eco-nomic reforms, sector programs are nowreceiving significant emphasis.6

• Work with other agencies is moving forwardon a comprehensive approach to guide assis-tance to post-conflict countries, and a newoperational policy statement has been issuedbased on extensive consultations with otherdonors, civil society, and NGOs (IDA Review2001c, para. 43–46).

• As discussed in Part II, IDA has increased itsparticipation in sector-wide approaches(SWAps), which combine several reinforcingelements: government leadership in the designof a strategy set within a medium-term expen-diture framework; external assistance in sup-port of the strategy objectives; and agreedindicators for monitoring performance andprogress.7 Donor support takes variousforms—from pooled financing in a few casesto separate investment projects within the

B u i l d i n g M o r e E f f e c t i v e P a r t n e r s h i p s

5 7

Uncoordinated aid programs continue toimpose heavy burdens on recipientcountries and to limit the impact of aidagencies’ programs.

sector-wide framework—and the open dis-cussion of differences in approach gives gov-ernments an opportunity to choose amongalternatives. SWAps also provide an importantmeans of advancing sector reforms andaddressing implementation capacity prob-lems in countries’ poverty reduction strategies.Both governments and donors support thisapproach, but harmonization of donor pro-curement and other procedures remains aproblem (World Bank 2000e; Jones 1999,2001). The PRSP process has increased coor-dination between the Bank and the IMF inlow-income countries. A Joint IMF–WorldBank Implementation Committee was estab-lished in May 2000 to oversee the HeavilyIndebted Poor Countries (HIPC) debt relief ini-tiative and the PRSP program. This commit-tee’s primary concern has been to establisha common approach to the HIPC initiative andPRSP program and to ensure that countriesreceive consistent messages from both organ-izations. PRSP missions are conducted jointly,and a joint staff assessment accompanies thedocuments to the Bank and IMF Boards. ThePRSP process, which is strongly supported byborrowers and other assistance agencies,could become the major vehicle for increas-ing donor coordination at the country level.

Overall, the principles of country ownershipand partnership now inform IDA’s strategiesand priorities on many fronts. Still, IDA has notyet fully aligned its own practices and procedureswith these new approaches. Nor have IDA andother assistance agencies adequately harmo-nized their policies and procedures to fit thisframework.

Implications for IDA’s future role. To moveforward, IDA, its borrowers, and other devel-opment partners need to accept mutual respon-sibility and distinct accountabilities for achieving

development outcomes. Borrower countriesmust take responsibility for implementing soundpolicies and building effective institutions, under-pinned by transparent, accountable systems offinancial management and evaluation. Devel-opment agencies need to take responsibility forthe predictable delivery of quality aid throughcoherent support for agreed objectives, bettercoordination and division of labor along lines ofcomparative advantage (including a willingnessto gear up to provide diagnostic work in areasof leadership), and efficient aid delivery that isaligned with capacity building and country-ledefforts.

More concretely, IDA should concentrate onthree interrelated measures:• Carrying through with the IMF on institu-

tionalizing joint support of the PRSPs, includ-ing strengthening collaboration with otheragencies supporting countries’ poverty reduc-tion strategies.

• Promoting expanded use of sector-wide pro-grams as building blocks of the PRSP process,drawing on experience to refine andstrengthen core processes, encouraging alead donor model to reduce transaction costsfor recipients and donors, and adjusting itsown processes and procedures to facilitateeffective participation.

• Pursuing strategic selectivity at the corporatelevel, based on sound sector strategies, andthe integration of those strategies into CASsthat are consistent with countries’ own povertyreduction strategies.

In addition, this review recommends thatIDA donors ask the institution to assume astronger role in accelerating donor coordinationto reduce transaction costs for recipients of aidfrom multiple sources and to strengthen coun-try capacity for managing and coordinating aidresources. The forces working against donorcoordination and harmonization are strong. Asthe preeminent multilateral development financeinstitution focused on poverty reduction, IDA hasthe potential for facilitating change. The agency-level agenda goes beyond important proceduralissues, such as procurement and reporting, to thepolicy and strategic issues of division of labor,


5 8

IDA should be encouraged to initiate andmanage a conference of bilateral and

multilateral agency heads to agree to a time-bound, monitorable plan of action.

aid delivery mechanisms to enhance ownershipand capacity, and PBAs. While the time for realprogress is long overdue, there is still no stand-ing mechanism to ensure high-level agencycommitment. IDA should be encouraged to ini-tiate and manage a conference of bilateral andmultilateral agency heads to agree to a time-bound, monitorable plan of action.

Directions for the FutureThese process reforms—in program and coun-try selectivity and partnerships based on greater

participation and better aid coordination—areessential for greater development effectiveness.These reforms need to be pursued simultane-ously to maximize their impact. Most of theimprovements are needed at the country level,in the design and implementation of assistanceprojects and programs. But changes are alsoneeded at the agency level to adjust policies andprocedures to support improved country assis-tance programs. Part IV turns to the internaladjustments that IDA needs to make to supportthese development effectiveness measures.

B u i l d i n g M o r e E f f e c t i v e P a r t n e r s h i p s

5 9

6 1

Getting to Results

Part IV

IDA replenishment commitments have been broadly consistent with an evolvingconsensus in the international development community [see Development CommitteeCommuniqué 2001] and have encouraged IDA to move in directions relevant tothe needs of borrower countries. As a result, IDA in FY00 is very different from IDAin FY94. Still, translating key objectives into monitorable programs and verifiableresults remains unfinished business. Issues of accountability for policy compliance,instrument choice and program monitoring, and alignment of resources with pri-orities need closer attention. Further organizational change, as recent managementreviews and proposals indicate, is required if the broad policy framework laidout in the IDA Replenishment Reports is to bring desired results.

While the period covered by this review has been one of considerable change—in borrower countries, in IDA, and in the international development system—IDAand its partners are still putting core elements of the IDA10–12 framework intopractice, reinforced by the new initiatives that have come into play in IDA12. Forthe immediate future, the principal needs are to consolidate mandates, focus onimplementation strategies, and direct resources to agreed new program prioritiesand processes.

6 3

Corporate Issues

Over the seven-year period covered by this review, IDA managementhas repositioned IDA and recast its mission statement in line withits replenishment commitments. It has:

• Refocused and broadened the developmentagenda. As discussed in Part II, IDA has sharp-ened its poverty focus, although helping coun-tries accelerate equitable and sustainablegrowth remains a challenge. IDA has alsoexpanded its agenda, giving more attention togovernance and institutional impediments,devising new approaches for aiding post-conflict countries, and assisting in the unprece-dented challenges facing transition economies.

• Increased responsiveness to borrowers. TheWorld Bank Group has moved program staffto the field, including 24 of 51 country direc-tors. This has enhanced the coordination ofaid at the country level and improved dialoguewith government and civil society about poli-cies and priorities.1 In implementing theenhanced Highly Indebted Poor Country(HIPC) debt relief initiative, Bank leadershiphas also responded to a key developmentimpediment for many IDA countries. In addi-tion, project preparation, appraisal, andapproval have been accelerated, thus dealingwith a long-standing complaint of borrowers.

• Diversified instruments. IDA has introducednew lending and nonlending products and

services, including Adaptable Program Loans(APLs), Learning and Innovation Loans (LILs),new programmatic adjustment credits, moresector-wide programs with development part-ners, and due diligence analytical work. It hasalso stepped up efforts to build more effectivepartnerships in support of country develop-ment through the Comprehensive Develop-ment Framework (CDF) and Poverty ReductionStrategy Paper (PRSP) initiatives, although bothare too new to assess. In addition, severalnew global and regional initiatives are impor-tant to IDA countries, including the new $500million African HIV/AIDS program.

• Improved quality of the IDA portfolio. Betteroutcomes have raised the overall performanceof IDA’s portfolio close to that of the IBRD.This is a major achievement. Of course, qual-ity at entry remains below 100 percent, andthe long-standing issues of uncertain andunlikely sustainability and limited institutionaldevelopment continue to impede long-termimpact. Performance in Africa has improvedbut has not yet reached the level found inother Regions; and some sectors remain prob-lematic, including rural development.

77

While these changes have positioned IDA toadvance core commitments, implementationremains a challenge. Three main implementationshortfalls—in strategic selectivity, support forcountry ownership and capacity building, andfocus on and accountability for results—need tobe addressed through adjustments in the fol-lowing corporate management areas.

Accountability for Policy ComplianceA recent management report to the DevelopmentCommittee stated “while the country must ‘own’its vision and program, the Bank must ‘own’ andbe accountable to shareholders for its diagno-sis and the program it supports” (DevelopmentCommittee 2000). For the Bank, the relevanceand implementation of corporate policies is a keystarting point. Yet policy objectives are not enun-ciated clearly enough to guide country pro-grams, and mechanisms of accountability forpolicy compliance are weak. These shortcom-ings leave a gap between corporate policies onIDA commitments and country programs.

Lack of clear policy objectives has particularlyconstrained effective integration of cross-sectoralthemes such as gender, environmental sustain-ability, and private sector development intoCountry Assistance Strategies (CASs). Sector andThematic Strategy Papers (SSPs) are intended toarticulate policy objectives and guide their imple-mentation at the country programming level.However, OED evaluations show weak integra-tion of sector and thematic strategies into coun-try assistance programs (OED 2000b). Througha thorough review of the role and effectivenessof past SSPs, management is now seeking tosharpen these documents’ strategic role andstrengthen their linkages to the CAS process.

Strategies are now in place for nine sectorsand thematic areas, but as of this writing they

are still being prepared for three areas that sig-nificantly lag in IDA compliance: gender, envi-ronment, and private sector development.Typically, SSPs assess the development ration-ale for Bank participation in the sector and iden-tify best-practice approaches. They are weakerin giving direction to Bank activities based onthe Bank’s record and comparative advantage.OED has emphasized that SSPs should specifythe policy framework, taking account of globalexperience and best practice; assess the per-formance of Bank programs to improve devel-opment effectiveness; describe the futureprogram of assistance in the aggregate, includ-ing monitorable global goals, countries of con-centration, deployment of instruments, and skilland budget requirements; and assess ongoingand planned activities of partners, the need fornew alliances, and the comparative advantageof the Bank (OED 2000a). Management’s reviewof SSPs is focusing on these issues and willidentify cost-effective ways to address them.Consistent with CDF principles, the review willrecommend wide-ranging consultation with part-ners to assess past performance, define com-parative advantage, and set out the vision andstrategic directions for sectors and thematicareas. With these reforms, SSPs can and shouldplay more of a role in helping to guide replen-ishment undertakings as well.

Weaknesses in accountability procedures havealso impeded compliance with IDA commit-ments. A gender study, for example, found norequirements to assess how gender issues aretreated at any stage in the project cycle, includ-ing supervision. Nor have accountabilities for themainstreaming of environmental issues beenclear—a constraint likely to be addressed in thenew environment sector strategy.

Accountability for quality, which has receivedextensive management attention, also requiresimprovement, although the Quality AssuranceGroup (QAG), since its creation in 1997, hasdone pioneering work in tracking the quality ofongoing Bank operations.

More rigorous scrutiny of quality safeguardsupervision (for both social and environmentalaspects) was introduced in last year’s QAGassessments, and safeguard aspects are now


6 4

Lack of clear policy objectives hasparticularly constrained effective integration

of cross-sectoral themes such as gender,environmental sustainability, and private

sector development.

widely addressed in Bank operations. Yet, whileserious neglect appears to be rare, the qualityof Bank oversight during appraisal and super-vision could be improved in about 20 percentof cases. To support this objective, the Bank isnow in the process of revising its managementsystems for safeguards policies, with a view tomaintaining decentralized responsibility whilestrengthening central coordination. This processshould lead to an improvement in the account-ability mechanisms for safeguards compliance,although some issues remain to be addressed.2

In addition, both background studies for thisreview and a Bank task force on the cost of doingbusiness have called for a shift to a long-term,country-based capacity-building approach toaccountability for safeguards and public finance.This would entail a systematic rather than aproject-by-project emphasis on capacity build-ing. It would also require a broad discussion ofthe harmonization of environmental safeguardand financial accountability policies amongclients and donors, beginning with assistance tocountries to meet their own national standards(IDA Review 2001e, 2001h).

Instrument ChoiceIDA uniquely combines the provision of globalknowledge with financing and advisory services.As reported in the Annual Review of DevelopmentEffectiveness 2000 (OED 2000a, hereafter referredto as ARDE 2000), the choice and mix of instru-ments—matched to the development objectives,consensus on reform in the borrower country,and the country’s policy, institutional, and polit-ical context—are key determinants of effectiveBank assistance.

ARDE 2000 finds that lending instrumentsperform better in countries with improved pol-icy and institutional environments (as measuredby the Bank’s CPIA ratings), but the sensitivityvaries by instrument. Technical assistance loans,which support targeted capacity building, performwell in countries with above-average performanceratings, but only half are satisfactory in countrieswith lower ratings. Investment loans fare slightlyworse, likely reflecting what project evaluationsshow about the influence of weak policy envi-ronments on investments, as well as the inade-

quate adaptation of projects to weak institutionalcontexts. Sector adjustment loans, which supporteconomy-wide policy changes, perform well inboth high- and medium-rated countries, but con-siderably worse in low-rated countries. This maybe due to weaker sectoral ministries and capac-ities, the complexity of sectoral reforms in poorlyperforming countries, and a multiplicity of donorsproviding different and overlapping advice andassistance. Studies of adjustment lending suggest,however, that some policy reforms, particularlyfirst-generation macroeconomic reforms involv-ing primarily the central ministries, can besuccessfully implemented in countries with aweak track record when there is a turnaroundand adequate government commitment. But theconditions for successful, complex second-generation structural and sectoral change are farmore demanding.

Instrument choice raises several importantissues:• Analytical work. IDA’s contribution to coun-

try policy analysis is widely regarded as amajor area of comparative advantage.3 Yeteach thematic background paper prepared forthis review raised concerns about the qualityand adequacy of analytical work, includingthe core due diligence and country-specificsectoral and thematic work that underpinsoperations (table 7.1).4

The declining resources and quality prob-lems of economic and sector work (ESW) arecause for concern. QAG’s recent findings—thatthe value added by country managementexceeds that by sector management, and thatthe quality of ESW is especially weak in the-matic areas—shows a need to further rebalancethe management matrix (OED 2000a). TheNetworks are supposed to lead ESW, as wellas support it. Structural obstacles that preventthis need to be addressed. Ways also need tobe found to fund good due diligence ESWwithout crowding out country-specific projectand program analysis. Two core products,Poverty Assessments (PAs) and Public Expen-diture Reviews (PERs), continue to receivelower than average QAG ratings, and key fidu-ciary due diligence assessments are lagging insome areas (table 7.2). Yet staff are concerned

C o r p o r a t e I s s u e s

6 5

that due diligence work has squeezed resourcesfor project and program analytic work as thetotal resources for ESW have declined.5

• Adjustment, programmatic, and investmentlending. Each of these lending instrumentsserves specific functions and has its ownimplementation challenges (see box 7.1).Selectivity in lending instrument choice is astrategic matter and ought to be treated thatway. This becomes particularly importantwith the sharper division of labor amongdonors that the CDF promises. At a mini-mum, the rationale for instrument choiceshould be made explicit in CASs and effec-tiveness carefully monitored and evaluated.

• Multi-country programs. The World Devel-opment Report 2000/2001 (World Bank 2000a)makes a strong case for the poverty reduction

benefits of increased regional and global pro-gramming. Using IDA resources for multi-country programs would require setting updecision processes for determining prioritiesand could entail setting aside an agreedamount of IDA funds for multi-country usethat would not be allocated through the PBAsystem.

• The special needs for enhanced capacity build-ing. As this review has frequently noted, greaterdevelopment progress could be made byenhanced support for national capacity build-ing in IDA operations. A key question iswhether IDA possesses the right instrumentsfor this. Its instruments may need to be broad-ened to include grants or be linked moreeffectively with grant resources available else-where. Grants will be more attractive to coun-


6 6

PER CPFA CFAA CPARPlanned Planned Planned Planned

Region 98–99 2000 98–99 2000 98–99 2000 98–99 2000

Africa 20 10 29 – 2 7 7 6

East Asia and Pacific 2 2 2 1 – – – 1

Europe and Central Asia 3 2 – – 2 2 – 1

Latin America and the Caribbean 1 – – – – 1 1 4

Middle East and North Africa 1 – – – – – – 1

South Asia 4 2 12 – – 7 – 4

Total 31 16 43 1 4 17 8 17

Number of IDA borrower countries

with item completed 28 33 4 8

Note: In addition to the number of completed ESW items in this table, there are a number of reports under preparation and listed as “planned” for FY01.

Source: World Bank data.

D u e D i l i g e n c e E c o n o m i c a n d S e c t o r W o r ki n I D A C o u n t r i e s b y R e g i o n ( F Y 9 8 – 0 0 )

T a b l e 7 . 2

Use FY93 FY96 FY00

Lending 150 142 118

Economic and sector work 99 64 72

Supervision 111 119 152

Othera 50 66 96

a. Includes TA and aid coordination, country program support, and client training.


B r e a k d o w n o f O p e r a t i o n a l C o u n t r y S e r v i c eC o s t s b y A c t i v i t y ( D i r e c t C o s t s O n l y ,i n U S $ m i l l i o n s )

T a b l e 7 . 1

tries reluctant to borrow for this purpose andwill enable a needed long-term perspective.6

Monitoring for ResultsIDA’s broadened agenda has increased thedemandingness of its objectives and interventionsand put heavy pressure on human and budgetresources, implying both higher risks and poten-tially higher rewards. Thus, IDA needs improvedperformance indicators and monitoring and eval-uation (M&E)—at project, country, and corpo-rate levels—with a clear focus on results.

The quality of M&E has improved since thelate 1990s, but ratings still fall short of expecta-tions: a 1999 assessment found that less than oneproject in two had adequate M&E in place. GoodM&E systems are crucial for systematically assess-ing progress in meeting core targets and pro-moting development effectiveness throughresults-oriented management. Evaluation reportsand a 1999 Bank working group on M&E see themain weaknesses at the project level as weakincentives, diffuse accountabilities, and weakcapacity—in both Bank and borrower countries.

There are three key elements for develop-ment effectiveness and M&E systems. They are:(a) the need to build incentives or demand forM&E data for decisionmaking at all Bank and bor-rower levels, with a focus on development out-comes and results rather than inputs anddeliverables; (b) ownership by governments andother stakeholders, reinforced by evaluationcapacity development; and (c) harmonizationamong external assistance agencies of their M&Erequirements as well as coordination in securingadequate funding.

Multiple measurement systems in the Bankhave created a related problem of lack of con-sistency and coherence in performance meas-ures that limits IDA’s (and its partners’) abilityto track and assess results and adjust policiesand programs.7 The Bank is working to har-


6 7

Adjustment program lending. Adjustment lending can support pol-icy changes and reforms economy-wide or at the sectoral level,but only when a critical mass of stakeholders becomes convincedof the need and direction for reform. Conditionality can be effec-tive as an instrument of mutual commitment, shared objectives, andreciprocal obligations, but not as a coercive tool. Cash-strappedpoor countries may be coaxed into accepting externally imposedconditions, but policies are likely to be poorly implemented andsubsequently reversed. Adjustment lending has traditionally sup-ported “stroke of the pen” reforms involving limited public con-sultation, but it is increasingly used to support more complexsecond-generation reforms, involving many stakeholders, asreported in the Bank’s recent adjustment lending retrospective.Bank efforts to use adjustment lending to promote civil servicereform have been mostly unsuccessful, but the Bank has begun toexperiment with more flexible and consultative programmaticapproaches. These new programmatic efforts support a borrower’smedium-term program of reforms, often through a series of single-tranche operations based on agreed actions taken.

Investment lending. Investment lending can allow experi-mentation and joint problem solving of specific development con-

straints and is more appropriate for certain kinds of targetedcapacity building. But investment projects are more likely to behampered by shortfalls in counterpart funds and a weak publicsector. The procurement and other due diligence requirementsassociated with investment lending can strengthen borroweraccountability and capacity, but are not always well adapted toborrower procedures. A multiplicity of uncoordinated investmentloans in a single sector is a particular problem of IDA countries.

Instrument mix. Combining adjustment lending with otherinstruments—such as analytic and advisory services, technicalassistance, and investment loans—may improve the policy envi-ronment and, hence, the outcomes. Complementary investmentsto build capacity, improve service delivery, or develop support-ive infrastructure are often necessary to achieve the benefits ofpolicy change. Adaptable program lending may be appropriatefor long-term public sector reform, depending on commitmentto reform and the quality of the trigger points and evaluationbenchmarks selected.

Source: Branson and Hanna 2000; Devarajan, Dollar, and Holmgren 2001; Girishankar1999; Killick, Gunatilaka, and Marr 1998; OED 2000a.

F e a t u r e s o f A l t e r n a t i v e L e n d i n gI n s t r u m e n t s

B o x 7 . 1

Multiple measurement systems in the Bankhave created a problem of lack of consistencyand coherence in performance measures.

monize and strengthen performance measuresinternally and with partners. PRSPs and sup-porting programmatic adjustment credits willrequire an overall approach for selective M&Eat the project level, greater M&E at the sectorand country levels, and enhanced country capac-ity for both. Future Replenishment Reportscould reinforce these efforts to advance M&Ereforms by requesting that management focusmore on outcomes and results in reporting onthe IDA program.

Alignment Between Resources andStrategic Program PrioritiesCutting across all other issues is how budgetresources are deployed. Network and Regionalstaff are concerned that current budget alloca-tions fall seriously short of the funding neededto meet IDA commitments. Budget processeshave changed significantly over the past fiveyears in response to the matrix structure intro-duced in the mid-1990s, other alterations in theBank’s way of doing business, and the StrategicCompact (FY98–01).8 The Compact and internalCost-Effectiveness Review anticipated thatimprovements in the quality of Bank services andgreater resource availability would lead to anincrease in lending. However, many of theexpected efficiency gains have yet to beachieved. Moreover, resources were allocated forimplementation of the CDF, PRSPs, and othernew initiatives, including the enhanced HIPCdebt relief initiative, within a budget set by theStrategic Compact commitment level. So whilethe Bank’s development agenda expanded,expenditures for lending and ESW continued todecline, in part because of the increased budgetrequirements of the “knowledge bank” (seetable 7.1).

To better align resources with strategic pri-orities, two sets of key issues are relevant:poverty-efficient budget allocations and the linkbetween programs and budgets.

Poverty-efficient budget allocations. Despitethe Bank’s poverty reduction mission, a coun-try’s level of poverty gets relatively minor explicitconsideration in budget allocations. Regionaland country budget allocations do not distinguishbetween IDA and IBRD. While corporate andRegional decisions have begun to channel moreprogram funding to poorer countries, consistentwith performance, the impact has been small.In FY00, the budget allocation to IDA countrieswas, on average, $0.54 per capita, and to IBRDcountries, $0.45.

There are complicating factors in makingsuch a link (for example, the differences inoperating costs among countries). However,poverty could be given greater weight in budgetallocations by applying the country normmethodology9—an approach that, to date, hasonly marginally influenced budget decisionsbecause of unresolved methodological and pol-icy issues.

The link between program and budget. TheBank’s programming and budget system lacksadequate mechanisms for reconciling budgetallocations and commitments to shareholders,bondholders, donors, partners, and countries.CASs, which generally serve as two- to three-yearplanning instruments, are approved separatelyfrom annual country budgets. CASs are sup-posed to estimate costs associated with pro-posed lending scenarios and to present programrecommendations that correspond to realisticbudget levels. But unanticipated resourcedemands, including new mandates or crises,can arise over the two- to three-year CAS period,and Regional vice presidents are not committedto the CAS funding levels. A projected CAS pro-gram for a particular year may not be adequatelyfunded, and there are current examples of sig-nificant budget shortfalls.

A more automated CAS costing tool thathas recently been endorsed should facilitatemore realistic estimating of indicative costs. Butit will not guarantee that a CAS—even afterBoard review—will be fully funded across allyears. A system that combines CAS costingwith a bottom-up, zero-based corporateresources allocation process on a multi-year


6 8

While the Bank’s development agendaexpanded, expenditures for lending and

ESW continued to decline.

basis would get closer to full funding. It is apositive step, therefore, that in this year’sbudget process a three-year rolling budget isbeing prepared, for the first time, to minimizeunexpected movements in Regional funding.This step should be accompanied by reportingto the Board when there are changes in pre-viously discussed CAS programs due to alter-ations in budget allocations.

IDA commitments should also be costed.Although none of the many IDA10, 11, and 12replenishment commitments were explicitlycosted, many of them intersect ongoing orplanned processes, including CASs, sector strate-gies, and other major initiatives. The cost ofrecommendations should be presented to IDADeputies before the conclusion of a replenish-ment negotiation, and to the Board with the IDAReplenishment Report. Where a replenishmentrequirement implies actions beyond ongoing oralready planned efforts, it should have a sepa-rate implementation path, with the necessary pol-icy and budgeting changes.

A related issue is the Bank’s heavy depend-ence on trust funds.10 Those funds that provideadvisory services and technical assistance for spe-cific development-related activities amount toabout a third of the Bank’s administrative budget,and the percentage is even higher for somecountries and sectors.11 The multiplicity of fundscan cause distortions and high transaction costs.A large amount of staff time is spent acquiringand administering these resources, but theirimpact on country program budgets is not clear.Given these difficulties, there needs to be a sig-nificant rationalization of trust fund arrange-ments, an issue that has been highlighted in theBank’s Strategic Directions planning.

Overall, a sizable implementation agenda liesahead to ensure effective implementation of theIDA program. Improved accountability systems,instrument choice, and M&E are central. But lit-tle of this will be possible without a better align-ment of budget resources and program priorities,the major goal of the budget reform currentlyunder way.


6 9

7 1

Implications forthe Future

While implementation of replenishment undertakings has movedthe IDA program in the right direction, IDA and its partners havenot yet fully institutionalized the changes or entirely absorbed

the lessons of experience. What is needed now is to deepen and broadenthe gains from the existing policy framework by:

• Focusing on implementation in areas ofemphasis

• Aligning resources to strategic priorities• Consolidating the IDA mandates.

For this, IDA will need to further adjust its roleat the country and global levels, within the lim-itations set by resources available for adminis-trative costs, lending, and nonlending services.In its primary role of supporting country-baseddevelopment efforts, IDA should concentrate onits areas of comparative advantage—most ofthese are at the strategic level of supportingeconomy-wide, sector-wide, and government-wide reforms through analytical work and lend-ing—and on capacity building. IDA should alsoplay a more active role at the global level to facil-itate the harmonization and coordination of exter-nal assistance needed to speed progress towardthe overarching goal of poverty reduction.

More specifically, IDA could build onimprovements of the past seven years in the fol-lowing ways.

Be More SelectiveAlthough there is consensus on this point, moreneeds to be done to increase IDA’s country,program, and corporate selectivity.• Country selectivity. IDA’s performance-based

allocation (PBA) system has brought increasedselectivity to the allocation of resources at thecountry level, based on assessments of thesoundness of countries’ policies and institu-tional arrangements. Further advances couldbe made by continuing to improve the Coun-try Policy and Institutional Assessment (CPIA),rethinking the governance discount, and ensur-ing consistency with the Poverty ReductionStrategy Paper (PRSP) and Country AssistanceStrategy (CAS) processes. Also, broader dis-closure of the rating system would improveconsistency of treatment and allow the systemto benefit from open scrutiny and to serve asa diagnostic tool for strengthening developmentpartnerships.

• Program selectivity. Countries’ poverty reduc-tion strategies provide a new way of setting

88

priorities for IDA and other donor countryassistance programs. More effective programselectivity, and instrument choice, depends inpart on better integration of strengthenedsector strategies into CASs. This should helpIDA define the focus of its lending and non-lending services, including where to take aleading or a supporting role with otherdonors, and its choice of instruments.

• Corporate selectivity. IDA needs to clarify pri-orities and objectives by sectors and themesand identify more precisely what it can (andcannot) commit to do, selecting the highestpayoff interventions for advancing povertyreduction, based on its comparative advan-tage. This corporate-level selectivity needsto be established in concert with the strate-gic decisions of other agencies as part of abroadened harmonization action plan.

IDA quantitative lending targets for sectors,instruments, or groups of countries serve as dis-incentives for effective program and countryselectivity.

Focus on Governance and Capacity BuildingEvery area of this review highlights the central-ity of good governance. IDA should continue tointensify its support of effective governance andpublic sector capacity building, increasing itscoordination with other external assistance agen-cies to encourage borrower governments tobecome more efficient in their provision of pub-lic goods and services and more transparentand accountable to their publics. • Every aspect of IDA’s assistance—from ana-

lytical work, to design and supervision ofoperations and monitoring for results—should have a capacity-building dimension.The implications for time, costs, and adjust-ments in ways of doing business mean thatIDA may need to consider options forexpanded grant financing for nonlendingservices and for much greater coordinationwith grant-giving agencies in support ofcapacity building.

• The sectoral impacts of systemic governanceand institutional weaknesses in borrower

countries suggest the need for institutionaldevelopment strategies that identify cross-sectoral concerns, prioritize areas of assis-tance, and create synergies in the handling ofproblems in different sectoral programs.

• While the governance agenda set out in theIDA10–12 Replenishment Reports remainsvalid, recent experience suggests that IDA, itsborrowers, and other development partnersneed to make a special commitment toimproving countries’ public expenditure man-agement and financial accountability. Thisreview recommends that IDA link its futurelending to time-bound plans for establishingsound public financial accountability systems.To foster that accountability, IDA should alsoencourage greater transparency by requiringthat agencies and programs supported byIDA issue regular public reports that set per-formance targets and record progress onachieving them.

Commit to Development PartnershipsIDA should work actively at the country andglobal levels to move aid coordination processesfrom ad hoc to structured arrangements. • The PRSP process provides a concrete way

of applying the Comprehensive Develop-ment Framework (CDF) principles in sup-port of countries’ development strategies.Institutionalizing this support, in coordinationwith the IMF and other assistance agencies,should define the core of IDA’s future pro-gram. The phasing-in of the new PovertyReduction Strategy Credits in step with coun-tries’ capacities for effective use and moni-toring should be handled with care given itsstrategic relevance.

• Expanded use of sector-wide approaches(SWAps) should be a building block of IDAsupport for countries’ development efforts,taking into account lessons of recent experi-ence on how to use these results-orientedapproaches to support country-owned pro-grams. Even where other agencies are betterpositioned than IDA to take the lead, IDA ana-lytical work or lending, or both, may still beneeded. As necessary, internal project andprocurement procedures should be adjusted


7 2

to enable flexible participation in structuredaid coordination arrangements.

• IDA Deputies should use IDA to achievegreater coherence and harmonization in aidpractices, which are so long overdue. This ini-tiative should involve the heads of bilateraland multilateral assistance agencies andshould set a clear schedule for achievingresults. The global agenda should go beyondprocedural issues—as important as they are—to policy and strategic issues of program andcountry selectivity and the processes for bol-stering country-led aid management.

Clarify Objectives and StrengthenAccountabilities• The process introduced in IDA12 for track-

ing implementation of IDA commitmentsshould be maintained and strengthened.Clear guidelines on compliance should beprovided to staff at the start of the replen-ishment period, and Regional managementshould be responsible for tracking countrycoverage of relevant commitments, takingaccount of countries’ priorities and activitiesby others.

• Clearer objectives and guidelines are neededin the pursuit of the IDA policy framework,especially for private sector development(PSD) and the mainstreaming of gender andenvironmental sustainability in IDA activities.Networks need to devise clear and achiev-able aims and indicators, strengthen theirinvolvement in key sectors, and track main-streaming in country programs. Regional man-agement needs to ensure adequate treatmentin CASs, reflecting appropriate selectivity.QAG and OED reviews ought to give moreattention to mainstreaming.

• Recent internal recommendations for M&Eprocesses should be followed up to under-pin a focus on results, especially at the coun-try and sector levels, with attention toimproving capacities in borrowing countries.Of particular immediate importance is thedevelopment of processes for evaluating per-formance and tracking progress on PRSPsand the supporting PRSCs and wider CASprograms.

Better Align Resources with PrioritiesA better deployment of resources relative toIDA commitments is essential to applying theother lessons of the past seven years. The rela-tion between the budget process and the fullfunding of CASs is of particular importance. • To make poverty a more prominent factor

in the Bank’s budget allocation process, acountry norm methodology has recentlybeen applied as a supplementary measure.Resolving methodological and policy issuesthat are holding up the move to full coun-try norm–based budgeting would supportthe Bank’s overall emphasis on poverty andperformance.

• Recent efforts toward more realistic costingof CASs should be accompanied by a processfor ensuring full funding of Board-discussedCASs across their two- to three-year period.This year’s preparation of a three-year rollingbudget is a step in this direction.

• Ways need to be found to ensure appropri-ate levels of due diligence and program-specific ESW and greater attention to capacitybuilding. Options for grant financing forcapacity-building activities need to beexplored.

This review also suggests that the IDA replen-ishment process itself might usefully be re-examined to facilitate reforms in the overalldevelopment architecture. The process providesa unique opportunity to discuss, at a globallevel, past experience and future directions forinternational development assistance and tomobilize funding in support of key objectives andto evaluate the development effectiveness ofthe overall aid system. As currently constituted,the process falls short of fully realizing thispotential.

Although broadly consistent with an evolvingconsensus in the international developmentcommunity, the replenishment undertakingshave been both overloaded and overdetermined

I m p l i c a t i o n s f o r t h e F u t u r e

7 3

The IDA replenishment process itself mightusefully be re-examined to facilitate reformsin the overall development architecture.

(for example, in the specification of lendingshares going to individual sectors and coun-tries, and in the number and detail of therecommendations). Undertakings have oftenbeen unrealistic about what IDA and its bor-rowers could reasonably accomplish in three-year periods, have focused more on inputs thanresults, have neglected to consider IDA budgetresources, and have failed to consider the impli-cations for the overall development assistancebusiness.

Specifically, the replenishment process hasbeen too disconnected from IDA’s developmentpartners, both borrowers and other assistanceagencies, so that priorities have not always beenas relevant as they might have been, and IDA’scomparative advantage has not been sufficientlyemphasized in the agreed undertakings. Threechanges could improve the process, and thus,the impact of future undertakings.

Develop a Long-Term Vision, Focused on ResultsAs part of the IDA13 negotiations, the IDADeputies ought to ask management to develop,in consultation with borrowers, a long-termvision for IDA beyond the replenishment period.This vision ought to clarify what is meant by IDA’spoverty focus, identify how IDA’s resources canbest be deployed, set up an internal managementsystem, and consider IDA’s global role as a com-plement to its support for country programs.

Engage Developing Countries in Settingand Reviewing Replenishment CommitmentsOpening communication channels that allowfor interaction and feedback from country

experiences will help to set IDA priorities andimprove implementation. The consultationsteps taken in the IDA12 negotiations were astart. The transparent and consultative processplanned for the IDA13 negotiation—includingconsultations with borrower governments andcivil society and the posting of papers on anIDA Website—is a solid second step. Follow-ing the negotiations, IDA should consider waysto ensure continuous feedback and to engageits partners in M&E of IDA13 performance andresults.

Define Commitments in Terms ofMonitorable and Achievable Objectives and Take Account of TheirRealistic CostingIDA replenishment undertakings have tended toemphasize inputs and output targets that do nottake account of the need for country ownershipif reform efforts are to be sustained. It would bebetter for Replenishment Reports to focus ondevelopment goals and on identifying the kindsof activities that IDA should support in advanc-ing those goals, as a matter of priority, takingaccount of IDA’s comparative advantage andrealistic costings of replenishment commitments.In addition, Replenishment Reports ought toagree on IDA performance indicators and onobjective reporting on progress toward desiredoutcomes. This process needs to recognize thatoperating within CDF principles means movingeven further away from a direct measure of thelink between IDA activities and country devel-opment outcomes to new ways of tracking IDA’sperformance in partnership with others. One wayto facilitate this is to strengthen the role of sec-tor strategies in defining IDA’s comparativeadvantage and the objectives to which IDAwould commit itself—in terms of program strat-egy, targets, and deployment of resources—asthe basis for commitments agreed to in IDAnegotiations, subject to overall resourceavailabilities.


7 4

The replenishment process has been toodisconnected from IDA’s development

partners, and IDA’s comparative advantagehas not been sufficiently emphasized.

7 5

Established as an integral part of the WorldBank: IDA was established in 1960 to lend tolow-income member governments of the WorldBank that could not afford to borrow on theBank’s near-market terms. Though a separatelegal entity, IDA operates as an integral part ofthe Bank—sharing the same staff and head-quarters; reporting to the same president; andapplying the same standards for loan appraisaland approval.

Lends only to low-income countries: IDA lendsonly to countries that have a per capita income(in 1999) of less than $885 and lack the financialability to borrow from the Bank. At present, 78countries are eligible to borrow from IDA.Together, these countries are home to 2.3 billionpeople, comprising 53 percent of the total pop-ulation of the developing countries. Today, 1.5 bil-lion of these people survive on incomes of $2 orless a day. Since 1980, IDA lending has amountedto 12–18 percent of total official developmentassistance to eligible borrower countries.

Some countries, such as India and Indonesia,are eligible to borrow from IDA due to their lowGNP per capita, but they are also creditworthyfor some IBRD borrowing. These countries areknown as “blend” borrowers. Countries thatonce borrowed from IDA but are now too pros-perous to qualify include Chile, China, CostaRica, and Egypt.

Provides loans on concessional terms: IDAloans (called credits) have maturities of 35–40years with a 10-year grace period on repaymentof principal. There is no interest charge, butcredits carry a small service charge, currently 0.75percent on disbursed balances. Since 1960, IDAhas lent $120 billion to 106 countries. It lends,

on average, about $5–6 billion a year for differenttypes of projects. Though the major proportionof its lending has always been investment lend-ing, IDA also provides forms of quick-disburs-ing adjustment and program loans conditionedon agreed macroeconomic, sectoral, or publicmanagement reforms. In recent years it has alsoprovided, on a pilot basis, partial risk guaran-tees in support of private investments. Thoughits Articles of Agreement permit IDA to providefunding on a grant basis, it has done so only ina limited number of special circumstances, insupport of particular global programs, certainpost-conflict situations, and debt relief efforts.

IDA funds are allocated to borrowing coun-tries in relation to their population size, level ofincome, and record of success in managing theireconomies and their ongoing IDA projects. Overtime, IDA has made modifications in its alloca-tion policies: for example, putting a ceiling onlending to its two largest borrowers, India andChina, and giving increased weight to coun-tries’ policy and institutional performance. InFY00, 47 percent of new commitments went toSub-Saharan Africa, 27 percent went to SouthAsia, 11 percent to East Asia and the Pacific, 7percent to Eastern Europe and Central Asia, andthe remainder to poor countries in the MiddleEast and Latin America.

Funded largely by government contribu-tions: Whereas the Bank raises most of its fundson the world’s financial markets, IDA is fundedprimarily by contributions from member coun-tries. Their cumulative contribution since thestart of IDA totals US$96 billion equivalent.Additional funds come from transfers of Bankincome and from IDA borrowers’ repayments ofearlier IDA credits.

ANNEX A1: IDA’S FEATURES

ANNEXES

Donors agree on a three-year cycle on thereplenishment of IDA funds. The 12th replen-ishment will finance projects over the threeyears starting July 1, 1999, and will allow IDA

to lend about $20 billion, of which donor con-tributions will provide a little over half. Todaythere are 37 IDA donors, two of which were onceIDA borrowers.


7 6

A n n e x e s

7 7

Throughout its 40-year history, the purpose ofIDA’s assistance has remained the same: to sup-port low-income-country economic develop-ment and poverty reduction. However, itspolicies and programs have evolved continuouslyin response to changes in the challenges itsborrowers have faced and in thinking aboutdevelopment.

The 1960s focus on infrastructure: In itsearliest days, IDA followed the lead of the IBRDand concentrated on financing roads, ports, andother infrastructure projects. The approachreflected economic thinking in the 1950s and1960s, which emphasized the accumulation ofcapital as the engine of growth, and growth asthe key to poverty reduction. As food deficits andpayment imbalances increased in developingcountries, it became clear that both sets ofassumptions were too simplistic. In response, inthe mid-1960s, IDA increased its agriculturelending and began lending for education.

The 1970s decade of expansion: IDA under-went a decade-long period of transformation andgrowth in the 1970s. Despite improved rates ofeconomic growth, millions of people in devel-oping countries were still living in poverty, andgrowing inequities were increasingly evident. Asa result, confidence in the prevailing broad“trickle-down” approach to development hadwaned, and attention then turned to more directefforts at poverty alleviation, including ruralactivities aimed at increasing the productivity ofthe rural poor as well as increased lending forhuman development. In support of this evolv-ing agenda, development research and ESWwere expanded and IDA lending increased froman annual average level of commitments of $229

million in 1961–68 to $3.8 billion in 1980.Although a part of this growth reflects inflation,in constant dollars annual commitments morethan doubled from 1970 to 1980, as did thenumber of projects.

The 1980s advent of policy-based lending: Bythe end of the 1970s, it had become clear thatdevelopment progress and IDA project per-formance were being seriously constrained byproblems in the policy and institutional envi-ronments in many borrower countries. But IDA’smain instrument of project lending was unableto provide adequate support for the economicpolicy and structural adjustment reforms deemednecessary. The onset of the debt crisis in the early1980s lent urgency to this situation, creating thecircumstances for the introduction and rapidexpansion of adjustment lending conditioned onborrowers’ commitments to undertake specifiedpolicy reforms. Focused initially on distortionsin the macroeconomy, adjustment lendingbecame a major feature of IDA lending in sup-port of both structural and sectoral reforms andcontinues to account for roughly one-quarter($1.5 billion) of IDA lending today. Structuralreform underscored the importance of two mainpoints: (a) along with renewed growth, explicitpro-poor policies and programs were needed inthe many countries where poverty remainedpervasive and deep; and (b) sustainable reformcould be supported but not induced by exter-nal assistance. Both lessons became central tofurther changes in IDA’s program in the 1990s.

The 1990s broadened agenda of develop-ment concerns: From the 1990s onward, IDA’sefforts have been marked by an expanding pol-icy agenda along with a drive for increased

ANNEX A2: EVOLUTION OF IDA’S PROGRAM

development effectiveness. Environmental sus-tainability, gender equity, and the improvementof governance became core policy priorities insupport of the overarching objective of povertyreduction. At the same time, IDA was confrontedwith increasingly diverse and complex borrowercircumstances as China and the newly inde-pendent countries of Central Asia joined itsranks and numbers of IDA countries experi-

enced serious internal conflicts that called fornew forms of post-conflict assistance. IDA under-took several key reforms during the decade to“change the way it does business,” shifting itsfocus from individual projects to the country asthe main unit of account; strengthening anddiversifying its analytical work; and placinggreater emphasis on stakeholder participation,transparency, and aid coordination.


7 8

A n n e x e s

7 9

ANNEX B: POVERTY AND SOCIAL INDICATORS

Category 1987 1990 1993 1996 1998 (est.)Percentage below $1/day PPP 1993 prices (%)IDA-only countries 33.86 39.28 39.92 38.22 33.60IDA-only countries, less Bangladesh 38.34 41.34 41.37 41.44 37.70Blend countries 37.06 37.74 36.17 30.82 30.03All IDA/blend countries 36.59 37.97 36.75 31.96 30.59All IDA/blend countries, less China 43.09 42.60 41.87 41.63 39.43Absolute number below $1/day PPP 1993 prices (million)IDA-only 278IDA-eligible 1,063China 290

Note: PPP = Purchasing power parity.

Source: Institutional Business Warehouse.

P o v e r t y i n I D A C o u n t r i e s , 1 9 8 7 – 9 8( $ 1 / d a y P P P )

T a b l e B . 1

Indicator 1980 1990 1997 1980–90 (%) 1990–97 (%)Life expectancy (years)All IDA 58 61 63 6 3IDA without China 53 57 60 8 5Africa 47 48 50 3 3Infant mortality rate (%)All IDA 86 65 59 -24 -11IDA without China 112 85 74 -24 -12Africa 117 101 92 -14 -9Primary school enrollment (%)All IDA 66 86 98 30 14IDA without China 66 60 69 -9 15Africa 55 47 48 -15 2Female illiteracy (%)All IDA 62 50 44 -18 -13IDA without China 71 62 56 -12 -10Africa 76 64 54 -17 -15

All indicators are population weighted.


I D A C o u n t r i e s — S o c i a l I n d i c a t o r sT a b l e B . 2


8 0

ANNEX C1: IDA COMMITMENTS AND DISBURSEMENTS BY REGION

IDA10 IDA11 FY 2000Region Commitment Disbursement Commitment Disbursement Commitment Disbursement

Africa (%) 39.8 44.0 34.4 37.1 47.2 34.2

East Asia and Pacific (%) 19.2 19.5 13.6 19.0 11.0 16.7

Europe and Central Asia (%) 6.3 3.5 11.5 6.4 7.1 9.9

Latin America and

Caribbean (%) 5.4 4.6 5.7 6.3 3.8 4.9

Middle East and

North Africa (%) 2.4 1.9 4.2 3.3 3.6 3.5

South Asia (%) 26.5 26.2 30.3 27.5 26.9 30.5

Total (%) 100 100 100 100 100 100

Total (US$ million) 19,136 14,355 18,946 14,072 4,366 3,942

I D A C o m m i t m e n t s a n d D i s b u r s e m e n t sb y R e g i o n — I D A 1 0 T O F Y 0 0 ( p e r c e n t a g e a n d t o t a l , U S $ m i l l i o n )

T a b l e C 1 . 1

A n n e x e s

8 1

C o m m i t m e n t s b y R e g i o nF i g u r e C 1 . 1

0

10

20

30

40

50

Perc

enta

ge

IDA 10 IDA 11 FY 2000

Africa East Asiaand Pacific

Europe and Central Asia

Latin Americaand Caribbean

Middle Eastand North Africa

South Asia

D i s b u r s e m e n t s b y R e g i o nF i g u r e C 1 . 2

0

10

20

30

40

50

Perc

enta

ge

IDA 10 IDA 11 FY 2000

Africa East Asiaand Pacific

Europe and Central Asia

Latin Americaand Caribbean

Middle Eastand North Africa

South Asia


8 2

ANNEX C2: IDA COMMITMENTS AND DISBURSEMENTS BY SECTOR

IDA10 IDA11 FY 2000Sector Commitment Disbursement Commitment Disbursement Commitment Disbursement

Agriculture (%) 24.3 29.6 19.1 21.5 9.6 23.1

Economy-wide (%) 9.0 9.4 14.2 10.2 18.9 10.7

w/o Environment (%) 4.1 3.7 4.0 4.0 1.6 3.7

Private sector development (%) 2.4 0.7 2.2 1.9 4.5 2.3

Infrastructure (%) 18.7 20.5 19.8 19.6 15.6 19.6

w/o Transportation (%) 12.1 12.6 15.3 13.3 13.9 13.8

Industry/finance (%) 9.9 8.5 10.4 12.3 9.7 7.5

Social sector (%) 37.9 31.7 36.3 36.2 44.6 38.9

w/o Education (%) 12.9 12.7 12.5 14.8 12.7 13.8

Health, nutrition, population (%) 14.2 8.9 15.3 10.6 15.7 15.2

Total investment (US$ million) 14,390 10,632 15,252 12,264 3,685 3,548

Total adjustment (US$ million) 4,746 3,723 3,694 1,809 681 395

Total (US$ million) 19,136 14,355 18,946 14,072 4,366 3,942

I D A C o m m i t m e n t s a n d D i s b u r s e m e n t sb y S e c t o r — I D A 1 0 t o F Y 0 0 ( p e r c e n t a g e a n d t o t a l , U S $ m i l l i o n )

T a b l e C 2 . 1

A n n e x e s

8 3

C o m m i t m e n t s b y S e c t o rF i g u r e C 2 . 1

Agriculture Economy-wide Private sector development

Infrastructure Industry/finance Social sector

0

10

20

30

40

50

Perc

enta

ge

IDA 10 IDA 11 FY 2000

D i s b u r s e m e n t s b y S e c t o rF i g u r e C 2 . 2

Agriculture Economy-wide Private sector development

Infrastructure Industry/finance Social sector

0

10

20

30

40

50

Perc

enta

ge

IDA 10 IDA 11 FY 2000


8 4

ANNEX D: IDA PORTFOLIO PERFORMANCE

IDA 9 (1991–93) IDA 10 (1994–96) IDA 11+ (1997–00)a

ID ID IDSust % impact Sust % impact Sust % impact

Projects Share Out likely or % sub. Projects Share Out likely or % sub. Projects Share Out likely or % sub.# % % Sat better or better # % % Sat better or better # % % Sat better or better

Sector GroupAgriculture 115 36 55 29 25 101 28 64 36 34 107 23 65 45 39Economic-wide 39 12 56 31 13 74 20 66 38 20 96 21 76 54 25Human-resource related 51 16 78 45 34 72 21 58 29 21 123 27 73 44 32Industry 34 11 50 26 29 20 6 45 30 25 38 8 61 46 42Infrastructure 81 25 67 37 35 89 25 63 39 27 97 21 71 44 44NetworkEnvironmentally & Socially

Sustainable Development 115 36 55 29 25 101 28 64 36 34 122 26 66 45 39Finance, Private Sector &

Infrastructure 126 39 63 33 34 120 34 58 37 27 156 34 65 41 41Human Development 40 13 78 50 33 61 17 61 30 20 102 22 78 50 33Poverty Reduction & Economic

Management 39 12 56 31 13 74 21 66 38 20 81 18 77 56 24Lending TypeAdjustment 37 12 65 38 22 63 18 75 49 24 70 15 80 54 29Investment 283 88 61 33 29 293 82 59 32 27 391 85 69 45 37RegionAfrica 180 56 52 29 26 209 59 55 27 22 236 51 61 36 32East Asia and Pacific 33 10 79 61 42 32 9 84 69 28 47 10 89 72 62Europe and Central Asia n/a n/a n/a n/a n/a 8 2 100 88 50 41 9 95 70 39Latin America and Caribbean 14 4 79 50 36 18 5 72 44 50 37 8 73 40 35Middle East and North Africa 21 7 57 33 25 13 4 69 31 46 12 3 75 33 17South Asia 72 23 74 28 26 76 21 64 28 26 88 19 74 56 32Lending SourceIDA 303 46 61 33 28 332 46 62 34 27 421 48 71 46 36IBRD 341 52 70 58 35 368 51 71 53 32 408 47 76 63 44Blend 17 2 59 35 24 24 3 58 54 21 40 5 68 50 33Grand Total (IDA & Blend) 320 100 61 33 27 356 100 62 35 26 461 100 71 46 35Note: Percentages exclude projects not rated. Sectors represent IDA Replenishment Report groupings as of the end of FY00.

a. The data for FY00 exits represent a partial sample (120 out of 269) and reflect the processing of all ICRs received through August 2000. The processing of the remainder of the FY00 sample

is ongoing, expected to be completed by spring 2001.

O u t c o m e , S u s t a i n a b i l i t y , I n s t i t u t i o n a lD e v e l o p m e n t ( I D ) I m p a c t , a n d A g g r e g a t eR a t i n g s b y S e c t o r , N e t w o r k , R e g i o n , a n dL e n d i n g T y p e / S o u r c e b y E x i t Ye a r s 1 9 9 1 – 9 3 ,1 9 9 4 – 9 6 , 1 9 9 7 – 0 0 ( b y P r o j e c t )

T a b l e D . 1

A n n e x e s

8 5

IDA 9 (1991–93) IDA 10 (1994–96) IDA 11+ (1997–00)a

ID ID IDDisburse Sust % impact Disburse Sust % impact Disburse Sust % impact

$ Share Out likely or % sub. $ Share Out likely or % sub. $ Share Out likely or % sub.millions % % Sat better or better millions % % Sat better or better millions % % Sat better or better

Sector GroupAgriculture 4,882 35 62 31 28 4.440 26 79 42 42 5,389 26 73 59 48Economic-wide 2,199 16 65 29 30 4,153 24 81 42 21 4,532 22 83 55 24Human-resource related 1,829 13 93 56 50 2,880 17 63 41 20 4,908 23 77 52 36Industry 2,078 15 56 39 28 1,336 8 55 36 37 1,489 7 75 58 57Infrastructure 3,155 22 75 52 36 4,163 25 73 45 18 4,696 22 76 57 42NetworkEnvironmentally & Socially

Sustainable Development 4,882 35 62 31 28 4,440 26 79 42 42 6,075 29 72 57 46Finance, Private Sector &

Infrastructure 5,570 39 69 46 33 6,056 36 68 44 24 7,249 34 70 52 43Human Development 1,491 11 94 61 51 2,322 14 65 38 16 3,844 18 87 59 38Poverty Reduction & Economic

Management 2,199 16 65 29 30 4,153 24 81 42 21 3,847 18 86 58 23Lending TypeAdjustment 3,357 24 66 32 33 6,403 38 78 46 25 4,871 23 88 60 35Investment 10,785 76 69 42 33 10,569 62 71 40 28 16,144 77 74 55 41RegionAfrica 5,323 38 54 30 25 7,867 46 72 33 29 7,892 38 69 44 32East Asia and Pacific 2,408 17 88 75 45 2,215 13 90 85 16 4,503 21 92 80 71Europe and Central Asia n/a n/a n/a n/a n/a 427 3 100 90 43 1,085 5 95 66 34Latin America and Caribbean 385 3 84 73 41 649 4 73 34 51 869 4 85 39 36Middle East and North Africa 615 4 84 57 10 187 1 88 19 36 358 2 91 35 43South Asia 5,413 38 72 30 37 5,627 33 66 36 25 6,308 30 70 56 27Lending SourceIDA 11,692 22 68 40 34 13,317 24 76 40 28 17,228 25 78 56 40IBRD 37,903 73 74 65 40 39,531 70 76 60 34 48,771 70 82 70 51Blend 2,451 5 71 39 29 3,655 6 64 51 24 3,787 5 70 55 38Grand Total (IDA & Blend) 14,143 100 69 40 33 16,972 100 73 42 27 21,015 100 77 56 39Note: Percentages exclude projects not rated. Sectors represent IDA Replenishment Report groupings as of the end of FY00.

a. The data for FY00 exits represent a partial sample (120 out of 269) and reflect the processing of all ICRs received through August 2000. The processing of the remainder of the sample FY00

is expected to be completed by spring 2001. Active portfolio data reflects projects active as of July 1, 2000, and it is taken from QAG materials.

O u t c o m e , S u s t a i n a b i l i t y , I n s t i t u t i o n a lD e v e l o p m e n t ( I D ) I m p a c t , a n d A g g r e g a t e b yS e c t o r , N e t w o r k , R e g i o n , a n d L e n d i n gT y p e / S o u r c e b y E x i t Ye a r s 1 9 9 1 – 9 3 , 1 9 9 4 – 9 6 ,1 9 9 7 – 0 0 ( b y R e a l D i s b u r s e m e n t s , F Y 9 6 U S $ )

T a b l e D . 2


8 6

ANNEX E: LENDING FOR POVERTY-TARGETED INTERVENTIONS (PTIs)

Volume ofNumber of commitments

Category Lending Year(s) projects (%)

IDA Investment 1992–93 37 43

1994–96 45 53

1997–99 41 48

2000 28 48

Blend Investment 1992–93 21 19

1994–96 14 41

1997–99 43 68

2000 33 4

D A / B l e n d P o v e r t y - T a r g e t e dL e n d i n g ( p e r c e n t a g e P T I )

T a b l e E . 1

A n n e x e s

8 7

IDA employs a range of lending instruments,but specific investment loans dominate IDAcommitments. Moreover, their relative sharehas increased over time, while sector adjust-ment loans and sector investment/mainte-

nance loans have declined in importance.Structural adjustment loans show a marginalincrease in terms of the distribution of amountof commitment. APLs and LILs started inIDA11.

ANNEX F: LENDING BY INSTRUMENT

IDA10 IDA11 FY2000Amount of Percent total Amount of Percent total Amount of Percent totallending, IDA lending, IDA lending, IDA

Instrument US$ million commitments US$ million commitments US$ million commitments

Adaptable Program

Loan 0 0 515 3 568 13

Debt Reduction Loan 0 0 85 0 0 0

Emergency Recovery Loan 359 2 843 4 81 2

Financial Intermediary

Loan 248 1 137 1 5 0

Learning and Innovation

Loan 0 0 113 1 76 2

Program Structural

Adjustment Loan 0 0 0 0 0 0

Rehabilitation Loan 283 1 60 0 0 0

Sector Adjustment Loan 2,281 12 980 5 73 2

Sector Investment/

Maintenance Loan 1,312 7 699 4 110 3

Specific Investment Loan 11,784 62 12,540 66 2,588 59

Structural Adjustment

Loan 2,183 11 2,571 14 608 14

Technical Assistance Loan 688 4 409 2 257 6

Total 19,138 100 18,952 100 4,366 100

D i s t r i b u t i o n o f I D A C o m m i t m e n t sb y I n t e r v e n t i o n

T a b l e F . 1

The distribution of commitments by numberof loans by type of intervention presentedabove indicates a larger percentage share ofAPLs and LILs compared to their share in theamount of lending. This highlights the rela-tively small average size of these interventions.

While the relative share of Specific InvestmentLoans by amount was increasing, there was adecline in terms of the relative share by num-ber of loans, which indicates a declining aver-age size of commitment for Specific InvestmentLoans.


8 8

IDA10 IDA11 FY2000Number Percent Number Percent Number Percentof IDA of total of IDA of total of IDA of total

Instrument credits number credits number credits number

Adaptable Program Loan 0 0 19 5 14 11

Debt Reduction Loan 0 0 2 1 0 0

Emergency Recovery Loan 6 2 20 5 3 2

Financial Intermediary

Loan 3 1 4 1 1 1

Learning and Innovation

Loan 0 0 26 7 18 14

Program Structural

Adjustment Loan 0 0 0 0 0 0

Rehabilitation Loan 5 1 2 1 0

Sector Adjustment Loan 19 6 11 3 2 2

Sector Investment/

Maintenance Loan 22 6 13 3 1 1

Specific Investment Loan 219 65 220 58 64 51

Structural Adjustment

Loan 28 8 35 9 7 6

Technical Assistance Loan 37 11 28 7 16 13

Total 339 100 380 100 126 100

D i s t r i b u t i o n o f I D A C o m m i t m e n t s b yN u m b e r o f L o a n s b y I n t e r v e n t i o n

T a b l e F . 2

A n n e x e s

8 9

OED evaluates development interventions byassessing how their results stack up against theirown stated objectives.1 From an accountabilityperspective, this goal-based approach is attrac-tive because it relates results to objectives agreedto by the Board of Executive Directors.2

Specifically, OED evaluates outcomes by con-sidering three factors: • Relevance of the intervention’s objectives in

relation to country needs and institutionalpriorities.

• Efficacy, the extent to which the develop-mental objectives have been (or are expectedto be) achieved.

• Efficiency, the extent to which the objectiveshave been (or are expected to be) achievedwithout using more resources than necessary.

The assessment of relevance is especiallycritical. When done well, it nets out excessivelyor inadequately ambitious objectives. Combin-ing these three factors, overall outcome is ratedon a six-point scale, ranging from highly satis-factory to highly unsatisfactory. (See box G.1 fordetails.) Outcome is not a simple average of thefactors, because the factors interact: they arecumulative and interdependent (for example,satisfactory relevance is useless if efficacy ispoor).

How Did OED Evaluate IDA10–12? Compliance. The review concentrated on IDA’scompliance with replenishment undertakingsand development contributions in six thematicdevelopment priorities: poverty reduction, socialsector development, private sector develop-ment, governance, environmentally sustainabledevelopment, and gender. It also addressed fourpriority process reform objectives: PBAs, enhanced

CAS design and implementation, aid coordina-tion, and participation.

Outcomes. OED assessed the extent to whichIDA-supported projects, programs, and otheractivities were relevant to these concerns, thedegree to which they have been or are expectedto be achieved (efficacy), and at what cost(efficiency).

Design. OED’s review used a multifacetedevaluation design, including: • Desk reviews of Bank documents and work-

ing papers, project operations databases,extant studies, and OED CAEs

• Staff surveys, with more than 200 respondents• In-country and international consultations

with representatives of government, civil soci-ety, the private sector, and other assistanceagencies in nine focus countries involvingmore than 680 participants

• Two international workshops bringingtogether nearly 60 experts from both bor-rower and donor countries

• Wide-ranging interviews with relevant Bankmanagers and staff

• Where the evidence allowed, results-basedanalyses tracing IDA inputs (policies, align-ment of resources), outputs (volume andcomposition of lending and nonlending ser-vices), and reach (stakeholder participation,coordination with other donors) to outcomesand results.

This work was done with the support of anadvisory group of six prominent internationaldevelopment experts, three each from devel-oping and developed countries, including anumber with experience in high-level govern-ment, private sector, civil society, and multilat-eral development bank positions.

ANNEX G: METHODOLOGICAL NOTE: RATING IDA’S DEVELOPMENTPERFORMANCE

Because few activities initiated under IDA10,11, and 12 are complete, the evaluation had toconsider a number of proxies for assessing IDA’sperformance: the quality of the analysis under-lying the program, its coherence at the countryand sector (or thematic) levels, the responsive-ness of lending and nonlending services to thecountry context, the selectivity of resource allo-cations and choice of instruments, and, wherepossible, IDA’s impact on country policies, insti-tutions, and actions. Box G.2 highlights themethodological challenges.

How Well Did IDA Meet AgreedUndertakings? Overall, IDA’s compliance with the replenishmentundertakings has been satisfactory, with impor-tant qualifications. During the IDA10–12 period,

IDA made significant advances in sharpeningthe poverty focus of investment and adjustmentlending and analytical work, established a strongpresence in the social sectors, brought gover-nance issues to the fore, and enhanced processesrelated to Country Assistance Strategies, PBAs,participation, and aid coordination. Compliancewas uneven across and within areas of programand process emphasis. With regard to under-takings on social sector development and CASenhancement, IDA did very well. Though recentlyaccelerating, IDA’s implementation of undertak-ings related to gender, environment, and PSD wasmodest in depth and pace in relation to thereplenishment agreements.

The compliance rating is based on the find-ings of IDA review background studies, whichidentified the extent and pace of implementa-


9 0

Highly satisfactory: All relevant developmental objectives are(or are expected to be) achieved and/or exceeded efficiently, withno shortcomings.

Satisfactory: Most of the relevant development objectivesare (or are expected to be) achieved efficiently with only minorshortcomings.

Partially satisfactory: Significant shortcomings are observed,even though most of the major relevant objectives, on balance,are (or are expected to be) met.

Partially unsatisfactory: Many of the major relevant objec-tives are not (or are not expected to be) met; major shortcom-ings are observed.

Unsatisfactory: Most major, relevant objectives are not (orare not expected to be) met and/or most objectives are notrelevant.

Highly unsatisfactory: None of the relevant objectives is (oris expected to be) met or is not relevant.

O E D ’ s O u t c o m e R a t i n g S c a l eB o x G . 1

Conducting evaluations is almost always complicated, but thisevaluation was especially demanding and complex.

First, neither the Replenishment Reports nor managementinstructions provided specific benchmarks against which com-pliance or outcomes could be measured. OED, therefore, lookedat the extent to which IDA took actions to change its processesand programs and to institutionalize those changes.

Second, the temporal scope of the evaluation was limited. Atthe request of the IDA Deputies, the evaluation focused on per-formance under IDA replenishments 10 and 11 and the first year ofIDA12. Many of the programs supported during this period are stillunder way, and their development outcomes are not fully known.

Finally, since many factors affect country program outcomesin a country context, attributing success or failure to IDA’s

actions posed a particular difficulty. To deal with this, OED tookaccount of external factors that might have affected IDA’s devel-opment performance, such as fluctuations in the world economy,borrowers’ and partners’ performance, war and civil distur-bances, and natural disasters.

In addressing this last issue, OED conducted in-country con-sultations in nine focus countries, chosen according to the fol-lowing criteria: size of IDA program, inclusion of blend andIDA-only countries, regional diversity, country performance rat-ings (that is, countries across the CPIA spectrum, excluding thevery lowest, where the IDA program was small because of verylow performance), availability of an OED CAE, and special cir-cumstances (for example, post-conflict).

W h a t W e r e t h e M a i n C h a l l e n g e s o f t h e I D A E v a l u a t i o n ?

B o x G . 2

tion to date of more than 150 separate IDA10–12undertakings. The rating reflects the overallextent of substantive compliance, rather than theeffectiveness or outcomes of IDA’s actions.The number and scope of undertakings madethis an ambitious challenge. The individualreplenishment undertakings varied widely, rang-ing from broad redirections in operations tocalls for specific reports. Findings on the degreeof IDA’s compliance made use of a system ofratings on individual undertakings, which wasreviewed with management. This system alsoserved as input into the substantive discussionof implementation summarized in the back-ground paper entitled IDA10–12 Replenish-ment Undertakings Implementation Matrix (IDAReview 2001f).

What Were the Development Outcomes? OED finds the development outcomes of IDA pro-grams—influenced by exogenous factors andborrower and partner performance, as well as IDAperformance—to be partially satisfactory, withnotable improvements over the period. Muchprogress can be seen in project-level outcomesas well as in quality indicators compiled by QAG.For example, as the figure shows, projects com-pleted in the IDA11 period (mostly initiated in ear-lier periods) are significantly more likely than

those completed in IDA9 or 10 to be rated by OEDas having satisfactory outcomes, likely sustain-ability, and substantial institutional developmentimpact. (Too few cases are available from theIDA12 exiting year for analysis.) This is a welcomedevelopment, and the hypothesis of increasinglysatisfactory outcomes is well supported by otherevidence displayed throughout the IDA review.In particular, QAG active portfolio data showsubstantial improvement in project quality.

Project-level performance is an importantindicator, but does not tell the full story of IDA’scontribution to development outcomes. In recentyears, IDA has shifted its focus to the “higherplane” of country programs. The results fromOED’s CAEs for 23 IDA and IDA blend countries(adjusted to match the rating scale in box G.1)found that 13 of these country programs wererated partially satisfactory, compared to 8 thatwere rated fully satisfactory.3 (Two others wererated unsatisfactory.) While CAEs cover a longerperiod than the IDA review (about 40 percentcover periods through 1999 or later and another30 percent through 1998), most make use of up-to-date QAG data, assess efficiency using coun-try budget coefficients, and draw lessons forimproving program performance. These resultssupport the finding that the developmentoutcomes of IDA programs, which also reflect

A n n e x e s

9 1

Tr e n d s i n P e r f o r m a n c e o f I D AP r o j e c t s C l o s i n g i n t h e Ye a r s C o v e r e db y R e p l e n i s h m e n t s 9 , 1 0 , a n d 1 1

F i g u r e G . 1

� �

�

� �

�

� �

�

IDA 9 IDA 10 IDA 11

0

10

20

30

40

50

60

70

80

90

100

Perc

enta

ge

Outcomes (percent satisfactory)

Institutional development impact (percent substantial)

Sustainability (percent likely)

factors outside of IDA’s control, have been par-tially satisfactory.

An important indicator of performance acrosscountries is the allocation of aid. Recent workby DEC (box G.3) suggests that the allocationof IDA resources across countries is highlyaligned with ratings on poverty reduction andpolicy performance.

Given the terms of reference of the review,of particular importance to this assessment is per-formance against the objectives embedded in thereplenishment undertakings. The results of thatanalysis are shown in table G.1 in terms of theirrelevance, efficacy, and efficiency.

To assess relevance, OED evaluated the cor-porate strategies in each of the priority areas andtheir translation in country assistance programs.Overall, the relevance of IDA’s efforts in mostareas targeted by IDA undertakings has been sat-isfactory or highly satisfactory.

The bottom line on development, however, iswhat happens on the ground and this calls for anassessment of the efficacy of IDA’s activities. Tothis end, OED asked the extent to which, giventhe corporate strategy, implementation is likely toachieve program objectives. As table G.1 shows,in 6 of the 10 priority areas likely progress againstobjectives is only partially satisfactory, whileaggregate efficiency has fallen short of the targetsset forth in the Strategic Compact, although notspecifically measured against IDA programs.

On the most important dimension—povertyreduction—evaluations of IDA’s performance at

the country level point to a generally positiverecord in assisting countries to lay the founda-tions for economic growth and poverty reduc-tion. Yet the record of IDA countries in sustaininggrowth high enough and long enough to bene-fit the majority of the poor has been disap-pointing in most IDA countries due in large partto many factors beyond IDA’s control. The reviewfound that IDA’s efforts to foster private sectordevelopment, enhance opportunities for women,promote environmentally sustainable develop-ment, use its resources more selectively at thecountry level, and promote aid coordination fellshort of admittedly demanding objectives.

As noted earlier, since many IDA-supportedprojects and programs from the period underreview are still ongoing and their results not fullyknown, OED considered a number of proxiesfor performance that informed its assessment.This analysis found many areas of progress, butsome significant shortcomings as well: • Quality of analysis underlying the program.

IDA’s analytical work is one its strengths,highly regarded by borrowers and otherdevelopment partners. The increase in thequantity of data and analysis, especially onpoverty, is a major achievement. But thequality of the data and in-country capacitybuilding on data collection and analysis (forexample, on poverty and gender) remainimportant challenges (as borne out in thePRSP progress reports), often because coun-try ownership is lacking.


9 2

A major methodological difficulty is distinguishing the contri-bution of IDA from all other activities. OED’s approach is toassess CASs down to the project level and see whether theresults of lending and nonlending operations were satisfactory.This approach has limitations because of fungibility.

An additional approach, which has different limitations, isthe aid effectiveness analysis that has been done in the Bank’sresearch department (“Aid Allocation and Poverty Reduction,”forthcoming in European Economic Review). It estimates the con-tribution of aid to poverty reduction econometrically across coun-tries. This approach evaluates major aid programs—bilaterals and

IDA—for its allocation across countries, but cannot distinguishdifferent effects of “rival” aid programs within a country. This iscomplementary to OED’s approach, which is much better in-country than cross-country. Specifically, DEC finds that IDA’sallocation is superior to any of the bilateral programs and, indeed,that its allocation rule is good in absolute terms. IDA deviates fromthe poverty-policy rule (direct aid to low-income countries withreasonable policies) on an ad hoc basis, but not systematically.Its deviations usually reflect additional knowledge.

Source: Development Research Group.

E v a l u a t i n g A i d E f f e c t i v e n e s s A c r o s sC o u n t r i e s

B o x G . 3

• Coherence at country and sector (or thematic)levels. Although IDA has strengthened itspoverty orientation, more needs to be doneto integrate that orientation into macro-economic and sectoral priorities and policies.While CASs have shown substantial improve-ment in their coverage of poverty during thepast seven years, with most occurring since1997, they have made less improvement in set-ting priorities and identifying country-spe-cific, poverty-focused assistance strategies.

• Responsiveness of lending and nonlendingservices to country context. While the innova-tions in lending instruments are an importantfeature of the period, there also has been adecline in support for ESW, and continuingweakness in M&E, which denies IDA and itspartners adequate knowledge on cost-effectiveinterventions in support of core objectives.

• Strategic selectivity. OED found—and man-agement agrees—that strategic selectivityremains a challenge for IDA and its partners.

• Impact on country policies, institutions, andaction. Portfolio performance shows improve-ments on outcomes, and QAG ratings—whichare useful leading indicators—are improving.But institutional development impact, thoughimproving, still remains too low. In addition, the

Annual Review of Development Effectiveness for1999 and 2000 (OED 1999a, 2000a) and manyOED evaluations stress weaknesses in institu-tional analysis and capacity development.

ConclusionsOverall, then, OED finds that IDA has compliedsatisfactorily with the replenishment undertakings,which were ambitious. Development outcomeswere partially satisfactory, in part because offactors outside IDA’s control. These ratings reflectaccelerating improvements in IDA and countryperformance, but still limited progress in achiev-ing the overarching goal of poverty reduction.

Recent shortcomings remain, but implemen-tation trends augur positively for the future. Torealize this potential, however, the next IDAreplenishment discussion should develop a long-term vision focused on results, engage devel-oping countries in setting replenishmentconditions, and define those commitments interms of monitorable and achievable objectiveswith realistic costing. More immediately, todeepen and broaden the gains from the exist-ing policy framework IDA needs to:• Focus on implementation in areas of emphasis • Align resources to strategic priorities • Consolidate the IDA mandates.

A n n e x e s

9 3

OutcomesDevelopment objective Compliance Relevance Efficacy/efficiency

Poverty reduction Satisfactory Highly satisfactory Partially satisfactory

Social sector development Highly satisfactory Highly satisfactory Satisfactory

Private sector development Partially satisfactory Partially satisfactory Partially satisfactory

Governance Satisfactory Satisfactory Satisfactory

Environmentally sustainable

development Partially satisfactory Partially satisfactory Partially satisfactory

Gender Partially satisfactory Satisfactory Partially satisfactoryProcess reform objective

CAS enhancement Highly satisfactory Highly satisfactory Satisfactory

Performance-based

allocations Satisfactory Highly satisfactory Partially satisfactory

Aid coordination Satisfactory Satisfactory Partially satisfactory

Participation Satisfactory Satisfactory Satisfactory

C o m p l i a n c e a n d O u t c o m e s R e l a t e dt o I D A U n d e r t a k i n g s

T a b l e G . 1


9 4

The Board of Executive Directors welcomedOED’s review of IDA. They found it both timelyand important, as well as a signal of the signif-icance that both the Executive Board and Man-agement attach to independent evaluation.Directors generally agreed with the findings andrecommendations. They viewed it as a usefulinput into the IDA13 replenishment. Directorscongratulated OED on a comprehensive andwell-written review and noted, in the context ofthe first such evaluation, the challenges OEDfaced in undertaking this exercise. They con-gratulated IDA Management on its performance,as documented by OED. They also appreciatedthe guidance provided by the Board’s Commit-tee on Development Effectiveness (CODE) asembodied in the statement of its chairman.

Directors welcomed the finding that IDA per-formance was satisfactory, with qualifications,with regard to compliance with IDA undertak-ings. They observed that this performance wasespecially noteworthy because of the breadthand complexity of these undertakings. Theyalso agreed with OED that IDA performance hasimproved over the period, recognizing that IDAtoday is very different from IDA in 1994.

With respect to overall development out-comes, most Directors accepted OED’s rating ofpartially satisfactory, while stressing that the glasswas more than half full with respect to IDA’s con-tribution. Directors stressed that IDA’s bottom lineis poverty reduction, and that poverty is stillrampant in IDA countries. They discussed themany factors that affect development outcomesin IDA countries, including, of course, the actionsof the countries themselves; the role of devel-opment partners, including IDA; and exogenousfactors such as trade and price movements. In thiscontext, they discussed the difficulty of isolating

IDA’s specific contribution to outcomes, notingthe divergence of views between Managementand OED on the outcome rating. Directors agreedthat the focus on outcomes reinforced the impor-tance of the development community’s redou-bling its efforts under IDA13 to promote povertyreduction, to develop better indicators of devel-opment effectiveness, to enhance its evaluationmethods, and to build evaluation capacity inIDA countries.

Directors welcomed IDA’s work onperformance-based lending allocations. Theynoted that, as with any system of this kind, addi-tional improvements were possible, and theywelcomed OED’s recommendations for furtherstrengthening. They suggested that IDA con-tinue to work to refine the system and the under-lying methodology, especially with respect to thetreatment of governance, transparency, and linksto the CAS and PRSP processes. They discussedthe issue of weak performers and the need toassist these countries in building capacity.

Looking ahead, Directors welcomed the broadagreement between Management and OED onpriorities, especially the focus on implementa-tion, consolidation, alignment of resources tostrategic priorities, and the related costing of IDAmandates. They stressed the importance of build-ing on the CDF/PRSP process in IDA countries,notably with regard to ownership, participation,partnership, and harmonization of operationalpolicies and procedures. Many Directors calledfor a strengthening of the ESW program tounderpin the CAS and as a basis for lending,including programmatic lending, and as a toolfor capacity building, notably in the areas of pub-lic financial accountability and fiduciary man-agement. Directors agreed with OED andManagement that more needed to be done to

ANNEX H: CHAIRMAN’S SUMMING-UP—MEETING OF THE EXECUTIVEDIRECTORS (MAY 29, 2001)

incorporate gender, environment, and PSD intoIDA programs and welcomed the steps thatManagement was taking to do so. With regardto private sector development, they specificallynoted the need for the IFC to work in a moreintegrated fashion with IDA, and asked that spe-cific modalities and instruments be explored.

Directors agreed that the OED review andbackground documents should be made avail-able publicly, together with the ManagementResponse and OED’s supplementary note on rat-ing methods. Directors also agreed that thissumming up of today’s discussion be includedin the OED review.

A n n e x e s

9 5


9 6

I. IntroductionCreated in 1960 to provide financing to addressthe development needs of the poorest countries,IDA has long been a vehicle for articulating thedevelopment agenda, gaining global consensusaround it, and supporting its implementation.IDA9 (1990–93) set out much of today’s devel-opment agenda—with its focus on poverty reduc-tion, gender, participation, private sectordevelopment, and environmental protection. Sub-sequent IDA agreements have extended anddeepened the framework. The IDA10 period,1993–96, emphasized country performance, sup-port for the social sectors, and assistance to Sub-Saharan Africa. The IDA11 period, 1996–99,focused on promoting the sustainability of reformsby strengthening institutions, improving gover-nance and addressing corruption, and buildingownership at national and local levels. IDA12,which began in 2000, built on these emphases,giving increased attention to addressing the effectsof conflict; helping countries build and implementcomprehensive, long-term, and monitorablepoverty reduction strategies; empowering thepoor; and strengthening IDA’s PBA system.

IDA Performance. Throughout the period ofthe review by the OED, Management has focusedon operationalizing the IDA framework throughcontinuous learning, which in turn has translatedinto steady improvement in the quality of IDAactivities, especially its country focus and povertyreduction strategy. But, of course, there is stillmuch room for improvement in a number ofareas, as indicated in the OED review and in self-evaluations, including the recent Strategic Com-pact Assessment, which is now being reflectedin Management follow-up actions. Indeed, theIDA evolution reflects a strong commitment to

independent evaluation and self-evaluation,designed to understand and improve IDA’sdevelopment effectiveness. The Bank has useda variety of self-evaluation vehicles, most notablyproject and ESW reviews by the Bank’s QAG, aswell as retrospectives of Country AssistanceStrategies (CASs), ESW, and lending volumesand instruments. Systematic self-evaluation hasincreasingly underpinned and complementedthe evaluations carried out by OED. Managementhas used self-evaluation and internalized theresults of independent evaluation to learn, tochange, and to improve its assistance to low-income countries.

Management Perspective. In this spirit oflearning and evaluation, Management very muchwelcomes the OED IDA review. Managementappreciates the comprehensive nature of thereview—and the work of OED managementand staff—and agrees with the broad thrust ofmost of the recommendations. Management isparticularly pleased that the OED review con-firms and validates many of the conclusionsIDA itself has drawn and the actions it hastaken—actions that have already made a differ-ence, notably in terms of poverty focus, greaterselectivity both within and across countries, andsubstantial improvements in the quality of lend-ing and ESW. At the same time, Managementacknowledges that there have been shortcom-ings in certain aspects of the Bank’s performanceover the period of the OED review, particularlyin the early part of the IDA10–12 period, and thatimportant challenges remain. However, Man-agement is concerned with the “partially satis-factory” rating OED gives IDA for developmentoutcomes, as it does not adequately captureIDA’s dynamic progression over the period. IDA

ANNEX I: MANAGEMENT RESPONSE

in 2001 is clearly very different from IDA in1993. We believe that more explicit attention tothis critical time dimension would provide amore resonant assessment of IDA’s achieve-ments to date, and remaining challenges.

Organization of This Note. This note dis-cusses IDA from three perspectives: povertyand country focus; key thematic and sectoral pri-orities; and selectivity and replenishment issues.

II. IDA’s Poverty and Country FocusIDA’s performance in both doing the right thingsand doing them right has improved over thereview period, as indeed is acknowledged by theOED review. Strategically, the emergence of thelow-income-country business model grounded inthe PRSP and the CAS is a seminal achievementof recent years, providing a strong basis for coun-try ownership, partnership, and selectivity in thepursuit of poverty reduction. This approach hasbuilt on the lessons from research on poverty anddevelopment assistance, from the 1990 WorldDevelopment Report on poverty to the 1998 pol-icy research report Assessing Aid, and the WorldDevelopment Report (WDR) 2000/2001: AttackingPoverty.1 IDA’s performance also has changeddramatically, improving across a broad front overthe period—in CASs, whose quality and povertyfocus have improved sharply; in lending; in pol-icy implementation; and in ESW. Country selec-tivity has improved too, reflecting significantenhancements to IDA’s PBA system. While, asnoted by OED, further progress in strengtheningthe links between CASs and poverty outcomes isstill needed, the work of recent years hasenhanced IDA’s development effectiveness.

A. IDA FY94 It is worth recalling the IDA of 1994 (“50 yearsis enough”). The CAS, a relatively new instrument,was evolving from the purely internal documentthat it had been initially. IDA CASs were not dis-closed and were not participatory—we wereonly beginning to share them with clients. Thefirst in-depth review of CASs was four yearsaway. Client surveys were not yet in IDA’s toolkit.OED evaluations showed that project qualitywas low, and there were few signs of improve-

ment. QAG did not exist. Nor did the HeavilyIndebted Poor Countries (HIPC) initiative. Con-cern about compliance with operational policieshad just led to the creation of the InspectionPanel. The Bank often over-promised—includ-ing on IDA commitments—and thus, even withthe best of efforts, frequently under-delivered rel-ative to its ambitious undertakings. Even so,Bank research shows that the poverty reductionimpact of IDA was far higher than the impact ofofficial development assistance on average andthat IDA’s impact improved in the 1990s.

B. IDA TodayToday IDA has become significantly morepoverty-focused and effective. This is not to saywe are—or should be—satisfied. Much workremains to be done, especially in getting actualresults measured in terms of reduced poverty andprogress toward the other international devel-opment goals. Drawing on lessons from evalu-ation (including early work by OED on thisreview) and research, Management has devel-oped a strategy for making IDA more effectivein the future. In September 2000 in Prague, theDevelopment Committee reviewed the paper“Supporting Country Development: World BankRole in Low- and Middle-Income Countries.”2

With the overarching goal of poverty reductionas the starting point, the paper set out IDA’scountry programming cycle, elaborating on therelationship between the PRSP and the CAS,reinforcing the emphasis on country ownership,and highlighting and clarifying the role andselectivity of IDA. The report recognized theimportance of support for country capacitydevelopment, strong diagnostic and analyticwork, and M&E. Subsequently, these same issuesfigured prominently in the Bank’s StrategicFramework Paper, discussed by the Bank’s Exec-utive Board in January. This way of doing busi-ness in IDA countries, endorsed again at theSpring Meeting of the Development Committee,consists of four components: vision, diagnosis,programming, and results.

1. Country VisionThe most important lesson the developmentcommunity has learned during the 40 years

A n n e x e s

9 7

since IDA was created is that development can-not be done to a country; if it is to be success-ful, it must be done by the country. Thefoundation of IDA’s work thus has to be acountry-led and country-owned policy frame-work covering the range of structural, social,environmental, institutional, and macroeconomicpolicies that make for successful development;country commitment is essential. In IDA-eligiblecountries, the associated program to implementthe policy framework is articulated in a PRSP.

Poverty Reduction Strategy Papers. Build-ing on the lesson of the central importance ofcountry ownership, early in IDA12 the Bankand IMF introduced the PRSP3 approach to theprovision of concessional assistance and debtrelief under the Enhanced HIPC Initiative. ThePRSP approach is an important step in opera-tionalizing the principles of the CDF,4 buildingon the emerging findings of the WDR, andsharpening IDA’s poverty focus. The PRSPapproach has been extended to other (non-HIPC) IDA countries, gaining wide acceptanceby these countries and by other donors as a basisfor improving country strategy formulation andaid coordination. Although it is too soon toexpect development impacts, implementation ofthe PRSP approach has accelerated rapidly. Bythe end of March 2001, 32 countries had pre-sented Interim PRSPs (most in connection witha HIPC Decision Point), and four countries hadpresented their first full PRSPs.

2. DiagnosisIDA’s assessment of a country’s policies, insti-tutions, and private sector strength provides thebasis for determining both how best, and by howmuch, IDA should support the country’s efforts.Indeed, while the country must “own” its visionand program, IDA must “own” and be account-able to shareholders for its diagnosis and the pro-grams it supports. In most cases, the country’svision, priorities, and analysis and the Bank’sdiagnosis are mutually interactive, supporting andcomplementing each other. ESW, prepared andshared with clients and partners, provides keyinputs and an analytic basis that the country canuse in developing its vision and its own diag-

nosis. In addition, ESW underpins CAS prepa-ration, the overall dialogue with client countries,and the lending program.

Rebuilding ESW. The OED review stresses thecritical importance of IDA’s ESW and the needto ensure adequate resources for it. Managementagrees. While the quality of ESW has been ris-ing, budget pressures have contributed to adepletion of the stock of this important basis forthe policy dialogue. In the period FY96–01, onlyabout 60 percent of active IDA borrowers(excluding the smallest countries) were cov-ered by PERs and 20 percent by ESW on pro-curement and financial management systems.Recent Bank-Fund analysis reinforces the needfor such ESW, especially in HIPC countries,where major gaps have been identified in sys-tems for tracking budgetary expenditure andfor measuring outcomes. But stronger fiduciaryESW also is needed more broadly, including tounderpin both the IMF-Bank Joint Staff Assess-ments (JSAs) of PRSPs and lending. Going for-ward, Management is preparing an ESWimplementation note, focusing on filling thegaps in ESW coverage in core areas (includingpoverty analysis) and continuing the improve-ment in quality, especially in poorly performingcountries, where lending has tended to be lesseffective; rebuilding the pipeline of CAS-basedESW to underpin future lending and the policydialogue; and improving participation in and dis-semination of ESW in client countries.

3. Country ProgrammingThe Bank’s diagnostic work, in turn, providesthe basis for its country business strategy, inwhich—as a committed long-term partner—itaims to help the country carry out its vision forgrowth and poverty reduction. Within this strat-egy, set out in the CAS, IDA’s program of lend-ing and nonlending assistance reflects countrypriorities, needs, performance, and institutionalcapacity. The specifics of the program takeinto account IDA’s track record, its compara-tive advantage, the actual and potential con-tributions of other partners, and, where relevant,Bank efforts in the provision of global publicgoods.


9 8

Country Assistance Strategy. In 1990, when theCAS was first adopted, it was exclusively a doc-ument for Board discussion of IDA country pro-grams, with country authorities having access tothe CAS only when their Executive Director sharedit with them. Since then, the CAS and the CASprocess have evolved. All CASs are now pre-pared in consultation with the government, whichhas resulted in a more comprehensive countrydiagnosis, closer alignment between the IDA pro-gram and government priorities, and a betterassessment of implementation capabilities. Therealso has been improved treatment of participation,governance, poverty reduction, human develop-ment, selectivity, and partnership. In line withIDA agreements, all IDA CASs are now disclosed.Going forward, the CAS country vision will comeincreasingly from the PRSP, with the CAS itselfbecoming more of a business plan setting out IDA’scomparative advantage vis-à-vis its partners.

HIPC Initiative. For many IDA countries, HIPChas become an important element of the coun-try program; however, the OED review mentionsit only briefly. Launched by the World Bankand the IMF in the fall of 1996, the HIPC initia-tive was the first comprehensive approach toreducing the external debt of the world’s poor-est countries.5 A major review of the programin 1999 resulted in a significant enhancement ofthe original framework to provide broader andfaster debt relief. As of February 2001, 22 IDAcountries had begun benefiting from HIPC assis-tance that will provide some US$34 billion indebt relief over time. To date, the Bank hasdelivered some US$1.5 billion in debt relief,including US$225 million in IDA grants, under-pinning the policy dialogue in a major way ina number of countries.

Lending Instruments and Mix. As part of thediscussion of the Bank’s country programmingcycle, Supporting Country Development [Devel-opment Committee 2000] suggested adaptingIDA lending instruments to better supportcountry-led poverty reduction strategies, with theparticular instrument mix reflecting country con-ditions. Under this approach, investment lend-ing (including through SWAps) is expected to

remain a substantial share of IDA financial sup-port in most country programs. But over time,for countries with strong programs, the povertyreduction support credit (PRSC) is also expectedto become an important vehicle of IDA finan-cial support and the policy dialogue. In mostcases, the resources and conditionalities asso-ciated with the PRSC will complement those ofthe Fund’s Poverty Reduction and Growth Facil-ity (PRGF), with IDA supporting the policy dia-logue on the social and structural agenda andthe Fund supporting the policy dialogue onmacroeconomic issues.

Fiduciary Policies. IDA is required to ensurethat its funds are used for the purposes intended,and its financial management and procurementpolicies are designed to this end.6 But with theincreasing recognition of the fungibility of aid (seeDollar and Pritchett 1992) and of the importanceof good governance, IDA also has sharpened itsattention to countries’ public expenditure, pro-curement, and financial management systems asa development issue, focusing on the overall useof public resources and supporting policy andinstitutional development as needed. Mindful ofthe fact that it will take considerable time andeffort to show results in this critical area, IDAencourages borrowers to develop efficient andtransparent systems for the allocation, execution,reporting, and auditing of budgetary resources,working with them and other development part-ners through institutional development grants,credits, and fiduciary ESW.

Safeguard Policies. OED recommends accel-erated work on the effort to recast Bank and IDAsocial and environmental safeguard policies,with specific attention to building country capac-ity. Management agrees that this is a priority, andpreparatory work is under way. But at the sametime, Management also accords priority to imple-mentation issues, to which we have progressivelydevoted increased resources and attention. Overthe past year, for example, Management hasadopted several measures to enhance the imple-mentation of safeguard policies, including bystrengthening systems for management of safe-guard policies at the central and Regional levels.

A n n e x e s

9 9

Actions entailed clearly assigning accountabilitiesfor safeguard implementation, ensuring thatsafeguard clearance has independent funding,providing clear and timely guidance to taskteams, and strengthening training activities. Thecentrally based safeguard quality assurance unitis now providing special advisory support andtraining for headquarters and field staff toimprove their knowledge of safeguard policies.Similar activities are being taken with govern-ment counterparts, with expansion proposedfor the next fiscal year—with the understanding,however, that building safeguard expertise alongwith broader regulatory infrastructure in clientcountries will take time.

4. ResultsAt the end of the day, it is results on the groundthat matter. This means that governments mustset clear targets, monitor and evaluate per-formance against those targets, and use theresults to shape current and future programs. ThisCDF principle is embedded in the PRSP, whichincludes country performance benchmarks andoutcomes as central elements. But judging Bankperformance is also essential; thus, work tostrengthen the M&E of the Bank’s contributionto poverty reduction and other country devel-opment objectives is very much a part of effortsto improve the CAS process.

Monitoring Progress. OED draws attention tothe need to continue improving the quality ofIDA M&E and strengthening borrower capacityin this area. This is clearly a very importantissue. A three-year program to promote effec-tive use of results-based M&E practices wasendorsed by the Board’s Committee on Devel-opment Effectiveness (CODE) in September2000 and was subsequently launched. The pro-gram has two strategic thrusts. Capacity is beingbuilt both in country and in the Bank tostrengthen the understanding and use ofperformance-based M&E through an integratedprogram that puts country- and sector-level M&Eas a central focus, supported by relevant project-level M&E. The first year of implementation hasfocused on five country pilots (three of them IDAcountries) and internal capacity building. In the

pilots, the foundations for sustainable, results-based M&E are being tested for possible repli-cation in other countries. This includesdemonstrating how M&E can be used to supportthe CDF/PRSP process and CASs. The secondyear of the program will see the addition of threemore pilot countries. Meanwhile, capacity-building activities within the Bank have includedworkshops, development and dissemination oftools, good practice examples, and advisoryservices to task teams.

Portfolio Quality. IDA also has paid consid-erable attention to improving the quality of itsportfolio, looking to improve results measuredin terms of development outcomes. OED andQAG indicators show how these efforts havemade a difference. The OED results showstrongly improving trends (see figure 1.1 in theOED Review); however, they cover projects thatare exiting the portfolio, with a majorityapproved by the Board prior to FY95. QAG’sassessments, which cover the active portfolio (bydefinition, projects of a younger vintage), showan even stronger uptick (see figure I.1). As QAGratings have been found to be a good predictorof development outcomes as later measured byOED, this suggests that OED ratings may rise asthe portfolio matures and these projects also exit.In the meantime, Management attention to theimplementation of IDA’s fiduciary and safeguardpolicies also has produced significant results.QAG indicators of the quality of the supervisionof fiduciary/safeguard aspects of IDA lendinghave increased substantially, from 82 percent sat-isfactory or better in FY97 to 92 percent in FY00.

III. Thematic and Sectoral IssuesOED raised several thematic and sectoral issuesin its review. Management is using strategydevelopment, implementation, and monitoringas one of the tools to address these issues. Sincethe launch of the Networks to link staff acrossRegional units working in the same sector or inthe same thematic area, the Bank has used Sec-tor and Thematic Strategy Papers (SSPs) as a toolto clarify the Bank’s role in these areas. In prin-ciple, the SSP is to a sector or theme what theCAS is for a country—it reviews the lessons of


1 0 0

experience (notably drawing on self-evaluationand OED evaluation), identifies issues, definesstrategic directions, and recommends actions. Inpractice, the first SSP was issued in 1997, and atotal of nine have been prepared to date.7 Man-agement has now initiated an evaluation of thestrategic relevance, analytic quality, businessimplications, and process efficiency of SSPs; this

stocktaking is expected to be discussed at CODEin early FY02. The early results of this SSP self-evaluation are already helping to shape thedesign of new ones. SSPs have been preparedor are under preparation for all of the major sec-toral and thematic areas raised by OED. Theyhelp to respond to the issues that OED posed,starting with institutions and governance.

A n n e x e s

1 0 1

Q A G D a t a o n I D A P r o j e c t Q u a l i t yF i g u r e I . 1

CY 97 CY98 CY99

0

20

40

60

80

100

Perc

enta

ge

Quality at Entry (Share judged satisfactory or better)

7383

88

FY97 FY98 FY99 FY00

0

20

40

60

80

100

Perc

enta

ge

Quality of Supervision (Share judged satisfactory or better)

58

7582

90

FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 FY98 FY99 FY00 FY01 Latest

0

10

20

30

40

50

Perc

enta

ge

46 4743

40 40

3328 30

27

20

15 14

Projects at Risk

A. Governance and Capacity BuildingManagement believes that weak governanceand ineffective institutions undermine develop-ment effectiveness across the board, hence theimportance it attaches to public sector reform andcapacity building. This also was highlighted inSupporting Country Development [DevelopmentCommittee 2000], which called on the Bank andthe countries’ other partners to work tostrengthen local capacity and institutions. OEDcalls for a broader country-wide or sector-wideapproach, rather than a project-by-projectapproach, to address such problems. Manage-ment supports this OED finding, which is alsoin line with the conclusions of the Bank’s recentSSP, Reforming Public Institutions and Strength-ening Governance. The latter emphasizes theimportance of institutional reform and providesa plan of action for working with client coun-tries to strengthen public institutions.8 Indeed,the Poverty Reduction and Economic Manage-ment Network (PREM) and sector families inother Networks are increasingly working togetherto mainstream institutional concerns in Bankwork and integrate them in country settings asappropriate. Overall, IDA’s work on governancetranslates OED’s general recommendation intospecific actions tailored to country needs.

Capacity-Building Work. IDA has been increas-ingly active in capacity building in support ofgood governance. During the last three years,31 IDA countries have received or have begunto prepare loans with significant public expen-diture or financial management components.As noted earlier, the coverage of PERs, CountryFinancial Accountability Assessments (CFAAs),and Country Procurement Assessment Reports(CPARs) in IDA countries, while less than desir-able, is significant. The Bank’s Legal Departmentand PREM have taken the lead in supportinglegal and judicial reform operations under prepa-ration or execution in more than a dozen IDAcountries. The Legal Department is building upits legal and judicial reform unit and, in wide col-laboration with other units, is developing a lawand justice network. Both PREM and the LegalDepartment have worked closely with the WorldBank Institute on anticorruption assistance to IDA

countries, and IDA has implemented in-countryworkshops or surveys on anticorruptionapproaches in some 23 client countries. AlthoughManagement agrees that more needs to be done,it cautions that the complexity of institutionalchange requires an adaptable, sustained, andmedium-term approach. In that context, theintroduction of PRSCs gives the Bank anotheruseful tool to support broad reforms for bettergovernance and stronger institutions. Other lend-ing and nonlending efforts, based on countryneeds, will build domestic capacity to supportthe detailed implementation of the country’smedium-term program of policy and institu-tional reforms, in areas ranging from civilemployment practices to judicial reform to localgovernment strengthening. Fiduciary assess-ments, and the underlying work on publicexpenditure management, financial accounta-bility, and procurement, will help to identifywhere support for capacity building is mostcrucial.

B. Gender MainstreamingOED confirms IDA’s substantial support for gen-der actions in health and education, especiallyin maternal health and girls’ education. However,OED concludes that during the review period,IDA did not make sufficient progress in incor-porating gender into its other activities, partic-ularly those supportive of women’s economicadvancement. Management agrees and has madegender mainstreaming one of its corporate pri-orities. The Bank and IDA have moved todevelop the analytic basis for gender main-streaming and put together a practical strategy,recently endorsed by CODE.

Gender Strategy. In early 2001 the World Bankcompleted a policy research report that exploredthe links among gender issues, public policy, anddevelopment (see World Bank 2001a). Thisreport has served as an important building blockin the preparation of the updated strategy for theBank and IDA. The draft gender strategy, Inte-grating Gender into the World Bank’s Work: AStrategy for Action, is designed to clarify theaccountabilities and responsibilities for gendermainstreaming and to direct resources to main-


1 0 2

streaming efforts in the Regions. The draft SSPalso calls for establishing an effective system fortracking and assessing gender mainstreaming,including developing new indicators to monitorimplementation. It proposes a proactive effortover the next three to five years to ensure thatanalysis of the impact of gender in povertyreduction is available for countries with an activeBank program; and that policy consultations(for instance, on PRSPs and CASs) benefit fromthe input of women’s groups in client countriesand systematically incorporate the gender vari-able. Demonstrating the purposeful nature of thestrategy, it is fully costed.

C. Environmental MainstreamingOED calls for a corporate program to bettermainstream environment into IDA country pro-grams, clarify accountability, and track progress.IDA has made real progress in environmentalmainstreaming, but more is needed. The Bank’sdraft SSP, Making Sustainable Commitments: AnEnvironment Strategy for the World Bank, dis-cussed at CODE on May 2, takes up all of theissues raised by OED and sets out tangible stepsto address them operationally.

Environment Strategy. The draft SSP focuseson encouraging a stronger country dialogue onkey environmental issues, linking environmentwith poverty issues more explicitly—for exam-ple, the long-term sustainability of naturalresource use and the impacts of pollution andenvironmental degradation on poor people andon the prospects for economic growth anddevelopment. In fact, Making Sustainable Com-mitments proposes measures to improve thequality of life (addressing today the environ-mental issues that affect the health and pro-ductive assets of poor people), the quality ofgrowth (ensuring sustainable use of naturalresources to assure that future generations willbe able to enjoy these assets), and the qualityof the global commons (by proposing incentivesfor country programs to incorporate global con-cerns appropriately)—requiring, as in the caseof gender, strengthened analytic work and clearerincentives and accountabilities in the Bank. TheSSP recommends introducing into the CAS indi-

cators and targets based on country priorities. Itnotes the importance of building country capac-ity for environmental management. The envi-ronment strategy, like the gender strategy, costsout its recommendations, estimating theresources needed to implement the strategyover the next three years and beyond. However,it is important to note that during the reviewperiod IDA had already increasingly main-streamed environmental activities across sec-tors, particularly in areas such as natural resourcemanagement and urban development. Thisprogress is documented in IDA’s Environmen-tal Mainstreaming Report (World Bank 2001e).Altogether, the portfolio of core environmentalprojects and projects in other sectors with envi-ronmental components has grown considerably.There are about 250 projects under implemen-tation, representing about $6 billion in IDAfinancing.

D. Private Sector DevelopmentOED notes that PSD activities have increased inthe 1990s, and cites its project reviews showingthat, in general, these projects have had suc-cessful outcomes. OED’s criticism in the IDAreview is really at the country level, where it hasidentified problems with the integration of pri-vate sector activities and noted insufficientattention to institutional development. Manage-ment believes that IDA, like other developmentassistance agencies and governments, has haddifficulty in finding solutions to the weakness ofthe private sector and market institutions inpoor countries. However, the Bank and IDAhave had an active agenda of work to draw les-sons of experience as a basis for building aneffective strategy.

Strengthening IDA Support for Private Sec-tor Development. The Bank’s strategic frame-work for the fight against poverty is based ontwo pillars: building the climate for investment,jobs, and sustainable growth; and investing inpeople and empowering them to participate indevelopment. The private sector is one of thecentral features in the life of poor people acrossthe world, and must figure prominently in eco-nomic growth and development. Empowerment

A n n e x e s

1 0 3

means expanding private sector activity andthus creating opportunities for poor or less pow-erful people—in particular by fostering a market-friendly investment climate that encouragesgrowth of enterprises, whether small or large—in part by making credit more widely available.Management has moved to strengthen the inte-gration of the PSD agenda into IDA programs,while doing the preparatory work in support ofa renewed strategy. A key step has been bettercoordination of IDA and IFC/MIGA activities inIDA CASs, including ten joint IDA/IFC/MIGACASs since the beginning of FY2000. The out-line of the Bank Group’s new PSD strategy wasdiscussed by the CODE subcommittee on March26, and there was a technical briefing for Exec-utive Directors on May 23 on the basis of anote on issues and options for the strategy andtwo comprehensive background papers. Exten-sive internal and external consultations on thesedocuments will soon begin. The final SSP willbe discussed by Executive Directors in Decem-ber. The strategy will focus on actions thatenhance the attractiveness of the investment cli-mate—to promote faster “catch-up” growth—andapply private sector incentives and marketplacesolutions to better delivery of basic services(infrastructure, education, health) to poor peo-ple. Also, over the next few years, the work pro-gram of the Bank’s research group, DevelopmentEconomics, will include efforts to answer ques-tions about what works in improving the invest-ment climate, notably in low-income countries.

E. Rural DevelopmentOED calls for renewed attention to agricultureand the rural sector, citing the downturn in IDAlending in these areas. Because so many of theworld’s poor people live in rural areas, Man-agement believes that broad-based growth mustinclude support for rural development. However,addressing rural poverty raises real dilemmas.The right balance must be struck between lend-ing to agriculture and investment in the broaderrural sector, for example, in rural education orinfrastructure, including roads and electricity.Similarly problematic is the balance betweenproject lending and indirect support throughmore economy-wide market- and trade-

enhancing reforms. Both of these balances haveto be struck at the country level and will beaddressed in an updated rural strategy.

Updated Strategy. An update to the Bank’s1996–97 rural SSP, now under preparation, is ben-efiting from the compilation of “bottom-up”Regional analyses and from donor and client con-sultation.9 The exercise analyzes the downturnin rural/agriculture lending but also looks atother sector operations and adjustment lendingthat have had important impacts on rural areas.It is worth noting, for instance, that if the areasof education, health, nutrition, and communityinfrastructure are included, about 33 percent oftotal FY99–00 lending for Africa directly sup-ported activities in rural areas.10 The strategy alsoreviews another difficult issue—the poor per-formance in many past rural operations, also doc-umented by OED reports. The past few yearshave brought substantial improvements in theperformance of the rural portfolio across Regions,as QAG reviews have confirmed. This is theresult of a systematic closure of failing projects(beginning in 1996) and greater attention tonew project design and implementation.11 Inparticular, the introduction of community devel-opment projects has provided a new model—amore participatory approach to address ruralneeds. Now nearly all rural projects in Africa, forexample, are designed with much more partic-ipation of beneficiaries and emphasis on decen-tralization than in the past. All of this points tothe inadequacy of focusing on a narrow defini-tion of the rural sector and new lending num-bers as the sole metric. The challenge is torebuild the rural portfolio in a way that learnsfrom past experience, using recently successfulmodels, especially community-based participa-tory approaches, while recognizing that theseefforts will still be very difficult and resource-intensive. At the working level, this means ensur-ing a multi-disciplinary focus and cross-sectorteamwork in expanding economic opportunitiesin rural areas, rural input into PAs, and contin-ued support for participatory activities. In addi-tion, PRSCs may be used to support rural actionin countries that have developed strong ruralstrategies and articulated them in PRSPs. The


1 0 4

upcoming rural SSP will set out the options foran expansion of lending for rural development,broadly defined (including development of non-farm activities), but with better targeting of thepoor. It will suggest targets and indicators forcontinued work in improving the quality and per-formance of rural operations. The new SSP willalso speak to the importance of greater attentionto women’s needs in agriculture, expanded sup-port for rural education and infrastructure, tradeliberalization (with countries in the Organisationfor Economic Co-operation and Development[OECD] and also regional trade and market inte-gration in Africa), and adaptive research—inaddition to direct support for agriculture.

IV. Selectivity and Replenishment IssuesOED calls for greater selectivity. Managementagrees that this is a priority; it has provided sig-nificant attention to this issue in recent months.The Strategic Framework Paper notes that greaterselectivity in what we do will result in betterleverage of the Bank’s and IDA’s scarceresources. IDA is tackling selectivity across threedimensions: within countries through the CAS,across countries, and at the regional or globallevel. A framework of corporate priorities isnow in place as a guide to selectivity.

A. Balancing Country and Sector PrioritiesIn the context of country vision and partnershipwith the client country and other donors, IDAintends to focus on key social and structuralreforms, with an emphasis on capacity building.OED recommends strengthening the role ofsector strategies and integrating them betterinto the CAS process. Management agrees. Asnoted earlier, Management is now taking stockof experience with SSPs since 1997, aiming tostrengthen their design and implementation.The goal is to increase their impact on countryprograms. However, this should not happen atthe expense of country ownership and countryfocus. We recognize the natural tension betweenbottom-up country-driven priorities and sectorgoals, and we are working to enhance the effec-tiveness of sector strategies to meet corporatepriorities and to sharpen selectivity, while retain-ing country focus.

Linkages between SSPs and CASs. To enhancethe linkages between SSPs and CASs, SSPs willaim to provide clearer guidance on country andintrasector emphases, and look more closely atinstruments—which ones are most effective andunder what circumstances. There has alreadybeen progress in linking the sector and countryfoci. SSPs are increasingly building on workdone by Regional units and country teams them-selves. A good example is the draft rural SSP,cited above, which factors in the demands andconstraints faced at the country level, based onbottom-up Regional rural development strategies.In the other direction, sector teams already pro-vide inputs into CASs, drawing on SSPs, and Man-agement is looking at how best to move thissector participation more upstream in the CASprocess. The enhanced two-way exchange willincrease the usefulness and implementability ofSSPs and the rigor of the choices made in theCAS process.

B. Selectivity among Countries—Performance-Based AllocationsThe OED review gave extraordinary attention tothe PBA system. Management agrees with thereport’s conclusion that “IDA’s PBA system hasbrought increased selectivity in the allocation ofresources at the country level” (para. 8.3). Man-agement has actions under way with regard toa number of OED’s recommendations: (a) fur-ther strengthening the CPIA; (b) tightening thelink between the PBA system and the CASprocess; and (c) establishing a more completerecord of the rating process to the extent pos-sible, given task managers’ budget and timeconstraints.

Quality of the IDA Allocation Process. Man-agement has launched a review of the CPIAprocess, drawing on internal evaluations and stafffeedback as well as OED’s review. Overall, how-ever, Management finds that much of the review’sdiscussion of the CPIA and, by extension, thebasis for IDA’s country allocations, is more crit-ical than warranted. IDA’s allocation system hasbeen a pioneering activity—one that is a demon-strated success and is widely seen in the devel-opment community and in research studies as

A n n e x e s

1 0 5

a model for linking aid to performance. It seemsshortsighted to fault this approach simplybecause there are still ways to improve it, as iscontinually happening in any event.

Governance Discount. Management notes thatthe workings of the governance discount, intro-duced in 1998, need to be kept under review andfine-tuned with experience. However, Manage-ment finds OED’s suggestion of “rethinking” it,which seems to go further, counterproductive.Management believes that the governance dis-count serves two important purposes. It sharplyreduces lending to countries with very seriousgovernance concerns, where resources mightwell be wasted. And the fact that it exists bringsattention to the governance indicators in theCPIA and highlights governance weaknesses incountries that do not receive a discount. The PBAsystem, including specifically the governancedimension, was discussed with the IDA Deputiesat their Paris meeting in February; our sense fromthat meeting is that the donors broadly agree withthe concept, but would value further discus-sion of the mechanism.

Link with Budget. OED recommends a strongerlink between IDA’s poverty reduction mandatesand country unit budget allocations, pointing toan expansion of mandates without explicit resourc-ing. Management notes that the Strategic Direc-tions Paper, which provides the framework for thebudget, discusses the expanded activities in IDAcountries over the past three years—includingPRSP activities, the Enhanced HIPC Initiative, andpost-conflict assistance—and calls for increasedbudgetary resources for these activities, as well aswork to close ESW gaps as noted above. The Strat-egy Directions Paper also introduces a three-yearrolling budget framework to help ensure the con-sistency of country programs and available budgetresources over the CAS period.

C. Partnership Aid partnerships are fundamental to the PRSPapproach. IDA is already working to strengthencountry-based, government-led aid coordina-tion with explicit division of roles, based on com-parative advantage, among external partners.

Ghana and Vietnam, for example, have recentlyput this into practice in PRSP exercises. Butequally, Management believes that the donorcommunity as a whole must address the issuesof harmonization and coordination on a broadfront, including through the Development Assis-tance Committee of the OECD (OECD/DAC)and the regional development banks.

Partnership at the Country Level. The PRSPis becoming the basis for better-structured aidpartnerships at the country level. It providesenhanced arrangements for the Bank and theFund to take responsibility for distinct aspectsof the policy dialogue with borrowers, and seta common framework in which the two insti-tutions jointly and separately assess and moni-tor progress. An important context is the IMF’sdesire to focus more on its core macroeconomicand related structural areas of responsibility,looking to the Bank to take the lead in the dia-logue on social, structural, and sectoral issues.The complementary but distinct roles of thePRSC and the Fund’s PRGF are expected to bea significant element in enhancing this partner-ship. The PRSP also provides a vehicle for struc-turing partnerships with other multilateraldevelopment banks (MDBs) and bilateral donors,thus aiding selectivity. There is increasingcooperation with MDBs on analytic and diag-nostic work, for example, with the Asian Devel-opment Bank on fiduciary work.

Coordination and Harmonization. Severalinitiatives to improve coordination among mul-tilateral institutions and with bilateral donorsare under way. The Bank has been very activein all major initiatives on harmonization of poli-cies and procedures, selectivity, and aid man-agement, and is also engaged with MDBs andothers in discussions of broader institutionalcoordination and reform. The Bank also partic-ipates in the work of the recently establishedOECD/DAC Task Force on Donor Practices,which focuses on harmonization among bilateralagencies, with the Bank acting as the represen-tative of MDBs in this work. In addition, the Bankhas proposed a prioritized agenda for harmo-nization work at the agency, country, and global


1 0 6

levels, covering procurement, financial man-agement, and environmental assessments, whichwas endorsed by the Development Committeeat its meeting on April 30, 2001.12

Regional and Global Partnerships. Selectiv-ity has also motivated a recent Bank review ofpartnerships to focus on those in which IBRD’sor IDA’s participation reflects the institution’scomparative advantage, and those in whichother institutions should be looked to for lead-ership. A new governance structure for part-nerships, discussed with Executive Directors inJanuary, is now in place.

D. Replenishment IssuesOED highlights the value of borrower views inthe replenishment process. At the request ofthe IDA Deputies, IDA has moved in recentyears to incorporate more borrower views. InIDA13, the June 2001 IDA Deputies’ meeting willbe held in Ethiopia to allow for borrower con-sultations organized and led by the EconomicCommission for Africa. Beyond this, the Deputieshave invited borrowing countries to nominate sixrepresentatives to join the IDA13 replenishmentmeetings—a significant broadening of the IDAprocess. Replenishment documents are alsobeing made publicly available for the first time.The Bank’s Resource Mobilization Departmenthas compiled a survey that seeks feedback fromborrowers on IDA’s policy framework and oper-ations. The survey has been returned by over 200respondents in ten countries, and its results willbe shared with IDA Deputies in June. ExternalAffairs and the Africa Regional Office also con-duct periodic IDA-country client surveys, whichare inputs to the IDA13 process.

Outreach and Nongovernmental Organiza-tions. Individual donors consult quite closelywith their own nongovernmental organizations(NGOs). But Management has been trying toreach out to NGOs in the South. During IDA12negotiations, for example, the Deputies heldconsultations with a number of southern NGOs,and in the current replenishment that process willbe continued. Work is under way in ExternalAffairs to identify southern NGO representa-

tives to participate at the third IDA13 meeting(in early October).

Grant Funding. OED proposes that IDA expandits use of grant funding and forge better part-nerships with other donors to support capacity-building activities. The issues of IDA’scomparative advantage vis-à-vis other partnersand expanded grant funding are part of a broaderconsideration of IDA terms in which the IDADeputies are already engaged in the IDA13replenishment discussions.

Replenishment Focus on Results. OEDobserves that there is a mismatch between IDA’slong-term development goals (and the impor-tance of tracking outcomes) and the periodicreplenishments that are politically and finan-cially necessary for donor governments. Duringreplenishments, donors are focused on estab-lishing the framework for the upcoming replen-ishment period under which their governmentscan take decisions to commit resources. They aretypically less focused at this time on imple-mentation and long-term results. To addressthis, in the IDA13 replenishment the IDADeputies are looking at two key long-termissues—debt sustainability and growth prospectsin IDA countries.

Midterm Reviews. The IDA Deputies may alsowant to consider using future midterm reviews—meetings held between replenishment years—to focus explicitly on IDA’s results and progresstoward outcomes, looking at country programsand follow-up to PRSPs, recently closed projects,and projects under implementation. Manage-ment will raise this suggestion with the IDADeputies, building on the experience of thesuccessful review in Lisbon last year.

Consolidation of Mandates. OED recom-mends that replenishment mandates be consol-idated because they have become overloadedand overdetermined. It does not provide detailedguidance on how to consolidate, other thancautioning against specification of sector andcountry/Regional shares. This is an issue the IDADeputies may want to take up.

A n n e x e s

1 0 7

Costing. OED proposes costing of IDA man-dates. Management believes that the costing ofmandates would provide the Deputies with anunderstanding of what the costs of implemen-tation would entail and inform their views on thecost-benefit tradeoffs. This would also providea basis for assessing whether IDA was ade-quately resourced and for examining tradeoffsand the implications for the budget process.This is a complex and staff-intensive exercise,but exploratory work along these lines is alreadyunder way.

V. ConclusionThe OED review has provided an importantopportunity for the development community tostep back and assess IDA’s effectiveness in thelarge, and to draw conclusions and lessons forfuture action. On this basis, Management con-cludes that the impact of the IDA replenish-ment agreements on IDA’s policies, strategies,and development outcomes has been highlysignificant. The overall IDA program during theperiod under review has been big, complex, andchallenging. Although not without implemen-tation problems, false starts, and disappoint-

ments, it has been increasingly successful. Con-ceived during the debates about the Wapen-hans Task Force Report in 1992 and born beforethe 1994 Annual Meetings in Madrid, theIDA10–12 era has witnessed a major turn-around—in strategy and partnership, in portfo-lio performance and development effectiveness,and in institutional credibility and leadership.Meanwhile, the face of development assistancehas changed, in no small measure because ofIDA. But the turnaround is very much a workin progress, and fragile; and the unfinishedagenda is large. We must nurture the criticaldrivers that have sustained the turnaround todate, including accountability, selectivity, andtransparency. These have spurred IDA’s institu-tional renewal and must underpin the replen-ishment process going forward. More important,these same drivers must also be applied to ourcountry and sectoral programs. Clearly the chal-lenge ahead is to focus our renewed institutionalenergies on achieving results measured in termsof progress toward reduced poverty and theother international development goals. These arethe benchmarks by which we must measureIDA’s performance in the future.


1 0 8

Overview1. Management does not agree with OED’s assess-

ment of the development outcomes of IDA’s work as“partially satisfactory.” This difference in judgmentappears to come from two factors. First, the OED rat-ing is an average for the whole FY94–00 period and,as such, does not give sufficient attention to thestrong improvement in IDA performance on all frontsin recent years (although these improvements arehighlighted in the OED review). For example, usingindependent and self-evaluation as a basis for action,IDA has significantly improved the quality of CASs,lending, implementation of policies, and economic andsector work (ESW). OED puts heavy weight on a sam-ple of 24 IDA-country CAEs. By far the majority ofyears covered by these CAEs fall before or in the firsthalf of the review period. The second factor is the lackof clarity on what OED is measuring in terms ofIDA’s impact on development outcomes. While OEDcites a long list of factors outside the scope and con-trol of IDA programs—including the fact that IDAaccounts for substantially less than 20 percent of totaldevelopment assistance going to IDA countries andfar less as a share of overall investment in IDA coun-tries—OED appears not to have given these factorssufficient weight in determining its rating. Indepen-dent research by the Bank’s Development Econom-ics Department shows that the poverty reductionimpact of IDA was far higher on average than that ofofficial development assistance and that IDA’s impactimproved in the 1990s. That said, there is clearlyroom for IDA to do better. Indeed, many of the“weaknesses” identified by OED (for example, fund-ing for ESW, monitoring and evaluation, further workon governance, and institutional development andcapacity building) are precisely the issues that man-agement itself has identified through self-evaluationand launched actions to further enhance IDA’s devel-opment effectiveness.

2. Management questions the analytic underpin-nings to the OED conclusions and notes that thefacts in the review and in the background paper sup-port a conclusion that IDA’s allocation system has con-sistently functioned effectively, achieving an allocationthat is significantly more performance-based thanconcessional assistance as a whole. While it notes thereis no room for complacency, management points outthat the system has improved over time and fulfillsthe donor mandate specified in IDA12. In particular,management believes that the governance discounthas served a very useful function. Lending has beenreduced sharply in countries with very weak gover-nance. In addition, the process has served to high-

light governance weaknesses in countries that donot actually receive a governance discount. While allparts of the performance-based allocation system arereviewed systematically for possible improvements,including the governance discount, management doesnot believe it needs to be substantially altered. Man-agement will continue to work to refine the system,in consultation with IDA Deputies and others, includ-ing donors that have adopted similar systems.

Chapter 11. See Annex A for a description of IDA’s basic fea-

tures and the evolution in its program.2. IDA depends for most of its financial resources

on contributions from its wealthier member countries,usually provided on a three-year cycle referred to asa replenishment period. At each funding interval,IDA’s management and donor countries (representedby their IDA Deputies) agree on a set of terms andconditions, which are set out in a ReplenishmentReport. Once endorsed by the World Bank’s Boardof Executive Directors and adopted by its Board ofGovernors, the recommendations become formalcommitments of the institution. The current IDA12agreement was endorsed by the Board in January1999.

3. A matrix summarizing IDA’s undertakings relatedto the key recommendations in the three Replenish-ment Reports is available on request.

4. Other replenishment recommendations per-taining to administrative, financial, and membershipissues are not covered.

5. OED defines borrower performance as theextent to which the borrower assumed ownership andresponsibility to ensure quality of preparation andimplementation, and complied with covenants andagreements, toward the achievement of developmentobjectives and sustainability; and Bank performanceas the extent to which services provided by the Bankensured quality at entry and supported implementa-tion through appropriate supervision (including ensur-ing adequate transition arrangements for regularoperation of the project). Portfolio cleanup efforts mayhave initially contributed to declining IDA portfolioperformance in FY98–99, as problem projects exitedthe portfolio.

6. Established in 1997, QAG conducts yearlyreviews of the quality of project preparation (qualityat entry) and supervision, based on a random sam-ple of projects and using standards it developed.

7. The percentage of IDA projects at risk of notmeeting their development objectives—an indicatorof subsequent OED ratings—has declined from 1990

1 0 9

ENDNOTES

to 1999. Two major findings emerge from annualQAG assessments of quality at entry and quality ofproject supervision. First, both quality at entry andquality of supervision have improved markedly forIDA projects from 1997 to 1999. Quality at entry forIDA projects improved from 73 percent satisfactoryto 88 percent, and quality of supervision increasedfrom 61 percent satisfactory to 84 percent. Moreover,while IDA projects initially rated lower for both theseindicators, the quality gap between IDA and IBRDprojects appears to have been eliminated, based onthe 1999 cohort.

8. Institutional development impact (IDI) is definedas the extent to which a project improves the abil-ity of a country or region to make more efficient, equi-table, and sustainable use of its human, financial, andnatural resources through: (a) better definition, sta-bility, transparency, enforceability, and predictabil-ity of institutional arrangements or (b) betteralignment of the mission and capacity of an organ-ization with its mandate, which derives from theseinstitutional arrangements, or both (a) and (b). IDIincludes both intended and unintended effects of aproject.

Sustainability assesses the durability of net proj-ect benefits—that is, the likelihood that they will bemaintained or exceeded over the long haul. While theoutcome rating is the best estimate of whether an oper-ation justifies the use of scarce resources, the sus-tainability criterion reflects the evaluator’s judgmentof its ability to continue producing net benefits in theface of risk and uncertainty.

Outcome, the primary measure of project per-formance, summarizes the likelihood that the projectwill achieve its major relevant objectives (do the rightthings) efficiently (do things right). The relevancecheck ensures that achievements rated satisfactoryaddress the country’s current development needsand the Bank’s operational priorities. Outcome cap-tures both accomplishments and the expected levelof future net benefits.

9. Of the 24 OED IDA Country Assistance Evalu-ations, most of which cover assistance over the decadeof the 1990s, about 33 percent were rated fully satis-factory, 60 percent moderately satisfactory, and 8percent unsatisfactory.

10. Management does not agree with OED’s assess-ment of the development outcomes of IDA’s work as“partially satisfactory.” This difference in judgmentappears to come from two factors. First, the OED rat-ing is an average for the whole FY94–00 period and,as such, does not give sufficient attention to thestrong improvement in IDA performance on all fronts

in recent years (although these improvements arehighlighted in the OED review). For example, usingindependent and self-evaluation as a basis for action,IDA has significantly improved the quality of CASs,lending, policy implementation, and ESW. OED putsheavy weight on a sample of 24 IDA-country CAEs.By far the majority of years covered by these CAEsfall before or in the first half of the review period. Thesecond factor is the lack of clarity on what OED ismeasuring in terms of IDA’s impact on developmentoutcomes. While OED cites a long list of factors out-side its scope and control—including the fact that IDAaccounts for substantially less than 20 percent of totaldevelopment assistance going to IDA countries andfar less as a share of overall investment in IDA coun-tries—OED appears not to have given these factorssufficient weight in coming to its rating. Independentresearch by the Bank’s Development EconomicsDepartment shows that the poverty reduction impactof IDA was far higher than that of official develop-ment assistance on average and that IDA’s impactimproved in the 1990s. That said, there is clearlyroom for IDA to do better. Indeed, many of the“weaknesses” identified by OED (for example, fund-ing for ESW, monitoring and evaluation, further workon governance, and institutional development andcapacity building) are precisely the issues that man-agement itelf has identified through self-evaluation andlaunched actions to further enhance IDA’s develop-ment effectiveness.

11. See World Bank 1997c. In addition, the Bankpublished an IDA retrospective in the early 1980s: seeWorld Bank 1982.

12. Following the IDA10 negotiations, IDA man-agement brought the report to the attention ofRegional managers, but it issued no explicit staffguidelines on how to implement the requirements.After the IDA11 negotiations, management circulateda note to staff summarizing the commitments in aselective rather than in a comprehensive way; the noteincluded the statement that “Regions are responsiblefor ensuring that our Country Assistance Strategies takeexplicit account of these commitments.” In mid-FY00,IDA management prepared a matrix to track imple-mentation of the IDA12 agreement—a useful inno-vation for raising staff awareness, particularlyconsidering that staff surveys undertaken for thisreview revealed that few staff were familiar with mostIDA commitments.

13. By July 2000, the portfolio cutoff date for thisreview, less than a third of expected IDA12 commit-ments were made and less than 5 percent were dis-bursed; of IDA11 commitments, 34 percent were


1 1 0

disbursed; and of IDA10 commitments, 70 percentwere disbursed. The number of projects that haveclosed and have completed reports are even fewer.

14. The major thematic studies and reports ofcountry and international consultations are listed in thereferences and are available on request. The analysisof gender and participation was undertaken as part ofa concurrent OED evaluation, with IDA-specific infor-mation disaggregated from the broader Bank-wideassessment. The nine focus countries are Bangladesh,Bolivia, Cambodia, Ghana, India, the Kyrgyz Repub-lic, Mozambique, Uganda, and Vietnam. The selectionof a subset of countries was designed to: (a) deepenthe analysis of the program and process reviews, (b)test the findings of those assessments against per-ceptions in the countries, and (c) gain a sense ofIDA’s overall role and performance by taking intoaccount the performance of the borrowing country andits other development partners. The global consulta-tions took place in two workshops, in Bonn in Decem-ber 1999 and in Washington in November 2000.

15. These included both self-evaluations under-taken by management and independent evaluationsby OED, as well as references to literature publishedoutside the Bank.

Chapter 21. Based on the 1990 World Development Report on

Poverty, the Bank’s strategy advocated a two-prongedapproach: (a) economic policy reform and productiveinvestment aimed at increasing the incomes of thepoor through labor-intensive growth; and (b) expan-sion of their access to basic social services and socialsafety nets.

2. This section on sharpening IDA’s focus onpoverty reduction draws on the background study,Review of Poverty Reduction in IDA10–12 (IDA Review2001b), which is available on request, and PovertyReduction in the 1990s: An Evaluation of Strategyand Performance (OED 2000d).

3. As noted further on, shortfalls in the strategy con-tent of such analytical work have limited its impacton country policies and programs.

4. OED 2000d. To encourage increased attentionto these concerns, management has devised morestringent criteria for evaluating the poverty focus ofCASs. These criteria include whether poverty is at thecenter of the policy dialogue, whether the CASassesses the impact on poverty of Bank-supported pro-grams, and whether the CAS includes benchmarks andmonitoring indicators. Further efforts to improve thisperformance is under way in the context of thePoverty Reduction Strategy Paper (PRSP) process.

5. This new feature of CASs lists objectives, therelated diagnoses and actions to be taken, and bench-marks for tracking country and IDA performance.

6. Two notable examples of best practice in thisregard are the FY00 CASs for Honduras and Mozam-bique, both of which identify clear monitorablepoverty reduction targets.

7. See OED 2000c, which found many PAs want-ing as an aid to country strategies. Nearly half did notadequately address the individual elements of broad-based growth, social service provision, and safety nets.Nor did they explain what worked and why andprovide adequate justification for the priority rankingscontained in strategy recommendations. Treatment ofgovernance issues was patchy, including the respon-siveness of institutions to the poor. The lesson drawnwas that PAs are much more influential when demandfor data is linked directly to domestic policyformulation.

8. On PERs, the most recent evaluation from 1998concluded that quality had improved since the mid-1990s. However, analyses of spending policies wereoften still too dated to have more than a modestimpact on IDA lending, country policy, or aid coor-dination. It recommended a stronger demand orien-tation, use of more selective and sequenced analyses,and sharper focus on service delivery. Recent PERsshow many of these features. It is important to note,however, that the total average cost of four key duediligence products—PAs, PERs, Country EconomicMemoranda, and Social and Structural Reviews—isapproximately $650,000 per country. Also see IDAReview 2001b.

9. According to a recent OED review of PAs, in Tan-zania, for example, despite a fairly extensive partic-ipatory PA exercise in 1996 the dissemination of thePA within the government was limited and its impli-cations for policy are only now—some four yearslater—being discussed in the context of the PRSP. SeeOED 2000c.

10. Important examples of where participatoryPAs have worked well are noted in Robb 1999.

11. The PRMPO (Poverty Division) thematic grouphas recently produced a handbook on impact eval-uation, with guidance on poverty monitoring andevaluation and a monitoring and evaluation techni-cal note for the PRSP sourcebook. See Prennushi,Rubio, and Subbaro Poverty Net (available at www.worldbank.org/poverty/strategies/srcbook/m&e0925.pdf).

12. Investment projects are classified under the Pro-gram of Targeted Interventions if they include spe-cific targeting mechanisms for reaching the poor,

E n d n o t e s

1 1 1

such as a food security component or public worksscheme.

13. As reported in the OED study of the Bank’simplementation of its poverty strategy, poverty-targetedinterventions have been neither a particularly reliableindicator of the poverty orientation of Bank lendingnor a particularly accurate measure of who benefitsfrom project investments. See OED 2000d.

14. See OED 1999a and 2000a. It should be noted,however, that the 2000 review points out that increasedattention to these matters by the Human DevelopmentNetwork may not yet be captured in the most recentfigures.

15. India Second Technician Education Project,Credit Number IDA 22230.

16. OED’s HNP (health, nutrition, and popula-tion) study (OED 1999d), for example, found that fewproject documents presented a coherent analysis ofhow project interventions would be translated intoimproved health outcomes for the poor, despite oftenbeing classified as poverty targeted interventions. Aneven smaller percentage of projects were able, on clos-ing, to demonstrate clear results for the poor. Whilethis does not mean that such interventions did notreach their target groups, it does mean that adequateprocesses are not in place to assess the extent of proj-ect impact and cost-effectiveness in reaching thepoor. Similarly, the OED review of education assis-tance in Bangladesh noted that 11 of 12 projectslacked benchmarks or monitoring indicators.

17. Project portfolio ratings that show greaterincrease in the demandingness and complexity of IDAprojects compared with IBRD projects underscorethis point (see Annex D).

18. In FY98–99, these two new types of loansaccounted for 23 percent, 22 percent, and 21 percent,respectively, of all Bank lending in HNP, education,and social protection.

19. LILs and APLs, which together are called “adapt-able lending instruments,” were introduced in 1997.LILs support small-scale, innovative projects that havethe potential to lead to larger projects and are usedto test new approaches, often in start-up situations andwith new borrowers. APLs provide phased supportfor long-term development programs and are usedwhen sustained changes in institutions, organiza-tions, or behavior are key to successfully imple-menting a program.

20. The government was simply unable to moni-tor this large number of projects. In addition, therewere serious inequities under the projects in resourceallocations to schools, provinces, and districts. Thenumerous project implementation units and inde-

pendent, parallel activities such as audits and super-visions drew heavily on limited government capac-ity. This was compounded by the donors’ use ofdifferent procedures and financial management sys-tems in parallel operations. For a discussion of theexperience with sector-wide programs as imple-mented in Africa, see Johannson and Adams 1998;Schacter 2001.

21. World Bank 2000a. In particular, see Chapter11, “Reforming Development Cooperation to AttackPoverty.” The importance of SWAps as mechanismsof country-led enhanced aid coordination is furtherdiscussed in Part III.

Chapter 31. See OED 2000d. The classification of operations

as “poverty focused” has been used since 1992 andrelates to adjustment operations that support gov-ernment efforts to reallocate public expenditures infavor of the poor, eliminate distortions and regulationsthat disadvantage the poor, and support safety netsto protect the most vulnerable.

2. According to the QAG’s 1999 quality at entryreview, 64 percent of a sample of Bank and IDAadjustment operations gave adequate treatment topoverty issues. In addition, only 22 percent of the sur-veyed FY98–00 operations included poverty andsocial indicators, though many of those operations cov-ered poverty issues.

3. See Meerman 1997. In addition, an OED reviewof adjustment lending showed that two-thirds of a sam-ple of countries that implemented reforms achievedper capita income growth and reductions in the inci-dence of poverty of some 1 percent in most cases. Alsosee Jayarajah, Branson, and Sen 1996.

4. The retrospective points out, however, thatthere are still too many and too complex conditions.

5. According to IFAD, between 1988 and 1998, totalODA to agriculture fell by almost two-thirds in realterms (IFAD 2001, p. 229).

6. The reasons underlying this decline are currentlybeing assessed as part of a review and updating of theBank’s rural sector strategy, which should feed into aheightened emphasis on broad-based growth in IDA13.OED CAEs tell a mixed story, with a number notingsignificant IDA contributions to improved agriculturalgrowth through adjustment lending and others notingthe lack of a country-specific strategy for promoting agri-cultural productivity. OED CAEs for Albania,Bangladesh, Togo, and Uganda found that IDA helpedto increase agricultural productivity and rural incomesby supporting improvements in the countries’ macro-economic conditions during the 1990s. Several other


1 1 2

CAEs noted, however, that IDA had been less activeand effective in supporting specific investments and sec-tor strategies in agriculture. For example, CAEs forEthiopia, Cameroon, Tanzania, and the CFA countriesconcluded that IDA’s development effectiveness hadbeen hindered by the lack of an adequate country ruraldevelopment strategy.

7. See World Bank 2000b, p. 187. Based on its find-ings, this report proposes a “business plan” thatfocuses on the huge investments required by both theprivate and public sectors to capitalize African agri-culture, increase its competitiveness, and harness thepotential of agricultural growth and rural development.

8. This section draws on the background studyReview of Private Sector Development in IDA10–12(IDA Review 2001g), which is available on request.

9. This new strategy, along with the new depart-ment (Bank/IFC) for small and medium-size enter-prises, both recently launched, promises a moredynamic response, and an ongoing rural financestudy could provide needed guidance.

10. Several CAEs have noted, in particular, thatwhile stabilization and liberalization support hadimportant effects, more attention should have beenpaid to linkages among financial sector reform, pri-vatization, and judicial reforms as well as underlyingconstraints due to poor corporate governance.

11. A joint CAS program between IDA and IFC wasintroduced in FY97 to foster closer coordination andcooperation between the organizations. See box 5.2in this volume.

12. An OED review of Bank Group work on thefinancial sector found that individual project ratingsdid not capture what was actually happening in thesector as a whole. While the relevance of Bank sup-port was high, with lending operations strongly tar-geted on sector reforms, efficacy was mixed andoverall the assistance was said not to be strong. Thestudy (World Bank 1998b) covered 23 countries, 10of them IDA borrowers.

Chapter 41. This section draws on a two-part OED evalua-

tion of the integration of gender in Bank assistance,including 12 in-depth country analyses, 9 of whichwere IDA countries, and the disaggregation of datarelated to IDA provided specifically for this review.

2. Two recent research reports have, however,given particular attention to the links between growth,poverty, and gender. See Blagden and Bhanu 1999;World Bank 2001a.

3. For example, the Bangladesh Fourth Populationand Health project (Credit No. 2259) required the gov-

ernment to fill 4,500 out of 5,000 health assistant jobswith women because experience had shown thatwhen assistants were women, family planning andother services were delivered more effectively.

4. Evaluation findings also show that the poten-tial adverse impact of gender-blind projects wasaverted in countries with strong policies or with com-paratively greater gender equality. However, in coun-tries with large gender disparities, women wereusually unable to access project benefits equitably andthe project impact on men and women was different.

5. In particular, IDA has worked primarily throughsector ministries in supporting education and health;it has not sought to engage the gender agencies orministries that now exist in most countries.

6. A review points to the absence of Regional gen-der units other than Latin America and the Caribbeanas a particular impediment to fuller integration ofgender in country programs, especially where col-laborative diagnostic work and dialogue are neededto build national consensus.

7. The Beijing Platform for Action, the main doc-ument adopted by the Fourth World Conference onWomen, seeks to promote the advancement of women.The document is available at http://www.un.org/womenwatch/daw/beijing/platform.

8. This section draws on the background studyReview of Environmental Sustainability Issues in IDA10–12, 2001 (IDA Review 2001h), which is availableon request, and the OED Review of the Bank’s Per-formance on the Environment (OED 2001d). Thesetwo reports were prepared in conjunction with oneanother and share a set of recommendations.

9. See Hamilton and Shyamsundar (2000) for areview of 51 CASs from 1993 to 2000. Using a priority-based standard, the 1998–99 CAS retrospective ratedcloser to 67 percent satisfactory or above. Manage-ment’s 2000 CAS review also finds that the numberof Bank-wide CASs with unsatisfactory treatment hasdeclined to 16 percent, although the IDA managementreport on environmental mainstreaming, discussedbelow, finds treatment of environment in IDA CASs“moderately satisfactory.”

10. Based on its recent study of Bank-wide ESW,management has indicated that the Bank needs to bediligent in ensuring that environment issues are ana-lyzed for all countries and that ESW-based knowledgeshould be current where environment is an importantdevelopment issue. For this purpose, managementproposes broadening the due diligence category toincorporate environmental reviews as the first step.

11. From 1996 to 1999, China accounted for about41 percent of environment component lending. With

E n d n o t e s

1 1 3

China no longer a concessional borrower in IDA12,the aggregate IDA lending level can be expected todecline.

12. This finding is based on reviews in the 15selected countries of IDA’s environmental projectscompleted during the period under review, and ofsupervision reports of projects under implementation.The focus countries for the IDA review included: Azer-baijan, Bangladesh, Bolivia, Cambodia, China, Côted’Ivoire, Ghana, India, the Kyrgyz Republic, Mozam-bique, Nepal, Tanzania, Uganda, Vietnam, and Yemen.

13. See Part IV of this report for a broader dis-cussion of design and implementation of safeguardpolicies.

14. One of the more valuable follow-ups to NEAPshas been the establishment by individual countries oftheir own environmental assessment procedures;however, this is happening slowly. For example, lessthan half of the IDA borrowers in Africa have adoptedany such procedures and in many, the procedureshave yet to become effective.

15. GEF projects, which are grant-financed andcarry their own preparation funding, have oftenproven to be critical instruments used by environ-mental staff to convince country directors and coun-tries to include environment projects. Unfortunately,these projects address only a limited number of issuesand often substitute for, rather than complement,IDA’s environment lending in support of countrydevelopment.

16. Among IDA members, only China and Indiaare significant contributors to greenhouse gas emis-sions, but nearly all IDA countries will be affected bysome combination of rising temperatures, rising sealevels, and more-violent weather patterns.

17. The Bank and IDA supported several projectsto help countries minimize damage from the last ElNiño, and IDA is working with countries to takeanticipatory actions to changes in rainfall, sea level,and other expected effects of climate change.

18. Most countries have by now established envi-ronmental agencies or ministries, often after com-pleting a NEAP and usually with IDA or other donorsupport. However, many of these agencies remainweak, environmental regulations are often not wellenforced, and the ministries lack input in their coun-tries’ overall development strategy and planning.

19. GEF projects are funded by grants and carrytheir own funds for preparation and supervision, sothey do not use IDA credits or budget resources.

20. DfID’s recent (2000) paper, Achieving Sus-tainability, Poverty Elimination and the Environ-ment, makes a strong case for this linkage as essential

for achieving the 2015 International DevelopmentTargets.

21. Consistent with this and other independentevaluations, along with some self-evaluations, theBank’s 2000 CAS retrospective has recently proposeda set of “next steps” that include better integration ofenvironmental considerations into economic and sec-tor work, a strengthened link between environmen-tal considerations and poverty reduction measures, andbetter incorporation of environmental indicators andtrends in CAS diagnoses.

22. Compare, for example, the experiences inChina, which is making an effort, and in Bangladesh,where rent-seeking activities in the ministry respon-sible for the environment are an obstacle to anyprogress.

23. Two early papers include the 1983 WorldDevelopment Report (WDR): Managing Development(World Bank 1983) and From Crisis to SustainableGrowth—Sub-Saharan Africa: A Long-term Perspec-tive Study (World Bank 1989), which was the first Bankreport to raise issues of governance explicitly. TheBank’s first policy on governance work was set outin a 1992 report entitled Governance and Development(World Bank 1992b), which was updated in 1994 asDevelopment in Practice: Governance—the WorldBank’s Experience (World Bank 1994). Since the startof IDA10, further guidance was provided in a policypaper entitled Helping Countries Combat Corruption:The Role of the World Bank (World Bank 1997b),which was updated in April 2000. The 1997 WDR onthe role of the state (World Bank 1997d) was a majoreffort to bring institutional economics into the Bank’scountry analysis.

24. See World Bank 2000b, in which a centraltheme is “improving governance, managing conflictand rebuilding states.”

25. Notably, WDR 2000 (World Bank 2000e) andWDR 2001 (World Bank 2001d) are both heavily ori-ented to the task of institution building. This year’sfocus is on institutions to tackle poverty; next year’sis on nurturing market institutions.

26. Key WBI inputs have included skills develop-ment for staff in the Bank and in borrower governments(on governance generally and in the preparation of anti-corruption action plans specifically), pioneering the useof corruption perception surveys and public servicedelivery surveys, and improving the measurement ofthe impact of corruption on development.

27. For example, OED’s CAEs for Bangladesh,India, and Indonesia and Country Assistance Note forKenya underscore the issue of public financialaccountability.


1 1 4

28. This includes public sector reform compo-nents only in public sector management and multi-sector projects.

29. The Uganda Poverty Reduction Strategy Credit(PRSC) is expected to be the first in a series of pro-gram loans to support public sector reform. In India,a program credit has supported the first phase of acomprehensive public sector reform program in UttarPradesh. In Tanzania, a Programmatic StructuralAdjustment Credit (PSAC) supports the government’smedium-term strategy for policy and institutionalreform aimed at bolstering private sector developmentin the country. And in Indonesia, the Bank has helpedto establish a governance partnership involving civilsociety, the government, the private sector, and thedevelopment partners.

30. For example, it has provided technical assis-tance, investment credits, and included elements inadjustment loans to a variety of countries for suchthings as regulatory reform and privatization, devel-opment of legal and regulatory frameworks relatedto energy and water reform projects, banking, andtelecommunications.

31. For example, the African Region collaboratedwith the African Executive Directors to launch acapacity-building initiative (Partnership for CapacityBuilding in Africa, or PACT, 1996). Its evolution wasprotracted, but the outcome, which includes inte-gration with the Africa Capacity Building Foundation,seems promising.

Chapter 51. Replenishment Report comments about the

CAS, and later about the PRSP, are really about theunderlying strategy IDA is following in each country.

2. Prior to that decision, the Bank had routinelyprepared Country Strategy Papers (CSPs), but thesewere confidential documents that were shared neitherwith the Board nor, in most cases, with the borrowergovernments. From 1991 to 1994, CASs were preparedfor Board discussion on a parallel track with CSPpreparation for management. Thereafter, the CASbecame the single strategy document for both man-agement and Board review.

3. This assessment is based primarily on the firsttwo CAS retrospectives undertaken by the Bank. Inaddition, IDA Review team members conductedreviews of CAS coverage in specific program areas ofemphasis as part of thematic background reports. Athird retrospective has recently been completed.

4. See Bank Procedure (BP) 2.11 Annex A, Con-tent of Country Assistance Strategy Document, Janu-ary 1995.

5. The improvement noted here refers to the CASdocument in particular; many of these practices pre-viously occurred in the Country Strategy Papers thatdid not go to the Board or the countries.

6. An internal review found that even as recentlyas the first half of FY99, only 40 percent of CASs had“extensive” civil society consultation.

7. For that purpose, the IDA Review Team exam-ined the CAS documents and the actual lending pro-grams for nine focus countries in order to identifydifferences between the programs as proposed in theCASs and the actual approvals of lending operationsin subsequent years.

8. There is an important distinction to be made herebetween areas of IDA involvement in underlying strate-gic work, on which not only its own but other lendingoften depends, and its involvement at a major level inlending. IDA tends to be needed in all critical areasbecause other agencies are not equipped or willing todo the sharp analytical work needed to map out a sec-tor strategy, but it need not follow that IDA play amajor financing role in all of the sectors. This, however,raises budget resources issues, discussed in Part IV.

9. A later review further confirms that this remainsan issue that needs greater attention.

10. The recent OPS CAS retrospective found thatonly 38 percent of CASs cited donor involvement asthe reason for no Bank activity in a given area. Thishas in some cases reflected the express preference ofclients that IDA stay involved, in other cases it reflectedcompeting sectoral perspectives, and in still others itresulted from the failure of a partner agency to deliverin an area of assumed leadership.

11. Management questions the analytic underpin-nings of the OED’s conclusions and notes that the factsin the review and in the background paper supporta conclusion that IDA’s allocation system has con-sistently functioned effectively, achieving an alloca-tion that is significantly more performance-basedthan concessional assistance as a whole. While thereis no room for complacency, the system has improvedover time and fulfills the donor mandate specified inIDA12. Management believes, in particular, that thegovernance discount has served a very useful func-tion. Lending has been reduced sharply in countrieswith very weak governance. In addition, the processhas highlighted governance weaknesses in countriesthat do not actually receive a governance discount.While all parts of the performance-based allocationsystem are reviewed systematically for possibleimprovements, including the governance discount,management does not believe the system needs to besubstantially altered. Management will continue to

E n d n o t e s

1 1 5

work to refine the system, in consultation with IDADeputies and others, including donors that haveadopted similar systems.

12. This section draws on the background studyReview of the Performance-Based Allocation System,IDA10–12 (IDA Review 2001a), which is available onrequest.

13. Some recent Bank research has indicated thetotality of official development assistance (ODA) hasmoved considerably toward better performance.According to this work, IDA’s efforts in 1990 to linkaid to performance were essentially undermined byother aid flows that did not appear to make that link-age. By 1997–98, however, the situation had changedmarkedly for the better: an increase of $1 million ofaid allocated on total ODA terms would have liftedan estimated 284 more people out of poverty, whilethe corresponding number for IDA was 434. Whileindicating that the numbers need to be used with cau-tion, the work suggests that the differentials are quiterobust. See Collier and Dollar 1999.

14. Included here are suggestions of broadestimplication; the background report also recommendsmore detailed adjustments and a discussion of someadditional issues that might usefully be considered thatwould provide for: (1) alternative allocation mecha-nisms and (2) ideas on how the PBA might need tobe adapted in light of the PRSP process now beingintroduced for all IDA borrowers.

15. Moreover, it is impossible to claim that the sys-tem is transparent (a goal urged by the ReplenishmentReports); in disclosing the PBA process, as discussedbelow, IDA would be exposed to a credibility risk inthe absence of such records. Though this suggestionwas not formally costed, a rough estimate indicatesthat implementation of the system costs about $700,000per year—a reasonable figure for a system that allo-cates some $20 billion annually—and establishing awritten record might add 50–100 percent to thatamount.

16. IDA management intends to undertake a majorreview of CPIA design in the coming year. This reportrecommends that close involvement of IDA’s partnersmight usefully be built into that review.

17. Triggers define the specific conditions for mov-ing to higher or lower lending scenarios. Most com-monly stated triggers are macrofiscal performance,structural reform, and portfolio triggers.

18. There are legitimate obstacles to fuller disclo-sure, including the need for a written record and theIBRD borrowers’ concern that disclosure of their rat-ings could negatively affect their financial market rat-ings, but these considerations could and should be

addressed in the review of the system planned bymanagement in the coming year.

Chapter 61. An OED review of IDA-specific participation

issues is reported here.2. The PRSP process represents a new and signif-

icant departure, including as one of five “core prin-ciples” a broad-based participatory process informulation, implementation, and outcome-basedmonitoring.

3. Governments bear most of the costs, oftenfinancing them with resources from the project loan.Communities also bear significant costs for project par-ticipation, contributing labor, materials, cash, andtime.

4. This section draws on the background studyReview of Aid Coordination in an Era of PovertyReduction Strategies, IDA10–12 (IDA Review 2001c)and The Drive to Partnership: Aid Coordination andthe World Bank (OED 2001a).

5. With other multilaterals, improvements, thoughlimited, have come at the headquarters level, whileresults in the field have been mixed.

6. This informal forum was originally known as the“Special Program of Assistance for Africa” and largelyinvolved discussions among donors, but both purposesand processes have changed over time. See OED1998b.

7. OED 1998b, para. 59–64. New SWAp initiativesbeing pursued by IDA in Cambodia and Vietnam areexplicitly justified by expectations that they will simul-taneously result in enhanced country capacity andownership, more coherent donor support to countrysector strategies, and, once established, greatly reduceaid delivery transaction costs.

Chapter 71. In-country consultations conducted for this

review strongly commended the increased represen-tation in the field, especially noting improvements inthe content and timeliness of policy analysis and dia-logue as well as enhanced coordination with otheragencies.

2. Under the revised system, the Regions willretain responsibility and accountability for safeguardcertification and compliance for most (“normal risk”)projects; the responsibility will be shared between theESSD Networks and the Regions, with outstandingissues to be resolved by the relevant managing direc-tor for “special risk” projects; and a quality assurancerole is exercised by the central Quality Assurance andCompliance Unit in ESSD. These arrangements offer


1 1 6

some improvement, but still face three problems: thelack of defined criteria to identify “special risk” proj-ects; the absence of a provision for above-Regionalresolution of unresolved issues for “normal risk”projects; and only partial resolution of the potentialconflict of interest in funding and dual reportingarrangements; the Regional safeguards complianceteams will still depend on cross-support from proj-ect task managers for a substantial portion of theirwork.

3. The value of IDA’s ESW was emphasized incountry consultations both by borrowers and by otherassistance agencies.

4. “Due diligence” ESW has been identified by man-agement as a core set of underlying analyses that needto be kept up-to-date for work in borrowing coun-tries, including: Poverty Assessments, Public Expen-diture Reviews, Country Economic Memoranda,Country Profile of Financial Assessments, CountryFinancial Accountability Assessments, Country Pro-curement Assessment Reviews, Social and StructuralReviews.

5. A specific concern, as noted in Part III, is thelow allocation of resources and low quality of eco-nomic and sector work in poorly performing coun-tries. If lending to weak-performing countries is tobe limited and good-quality ESW does not fill in, itis hard to see what role IDA can claim in thesecountries.

6. The Bank’s support for the African-led Partner-ship for Capacity Building (PACT) aims, over the nextfive years, to mobilize $1.0 billion for a capacity-building trust fund. The further step proposed hereis to make a portion of IDA resources available on agrant basis for use in mainstreaming capacity build-ing in IDA operations where needed and/or to poolbilateral trust funds for capacity building.

7. Among regions and sectors, between QAG andOED, and among country-level assessments for theCDF, PRSPs, performance-based resource allocation,and Country Assistance Evaluations.

8. The Compact’s objectives were to be financedthrough savings and redeployments as well as throughadditional budgetary resources of $250 million overFY98–00, with the administrative budget returning (inreal terms) in FY01 to its FY97 level—as it has nowdone.

9. In the country norm methodology, three mainvariables—population, poverty, and performance(CPIA index)—are used in a regression analysisagainst the historical trend of budget allocations bycountry, to derive a formula that yields a “countrynorm budget” to which additional resources are

added to ensure funding for portfolio supervision.Management can choose to place greater weight onpoverty, or on performance, by adjusting the for-mula’s coefficients for these variables. The averagebudget dollars per capita (with GNP per capita inparentheses) are as follows: Africa $1.96 ($491),East Asia $0.57 ($737), ECA $0.90 ($819), LAC $0.70($1,772), MNA $0.23 ($731), and South Asia $0.23($467).

10. The relevant Bank Operational Policy (OP14.40)defines a trust fund as “a fund administered by theBank in accordance with the terms of agreementwith a donor.” Not all “supplementary funds” are,strictly speaking, trust funds in this sense. The Bankacts as a fiscal agent for some funds, channelingthem to other entities to administer. There is also “com-plementary funding” that supports cofinancing ofloans and credits, and debt-service funds. Discussedhere are those arrangements under which the Bankadministers donor-entrusted funds that provide advi-sory services and technical assistance for a specific,development-related activity or activities.

11. For work in certain sectors or themes, thedependence on trust funds to deliver core lending andESW is now particularly strong. Notably, if averagedover the entire health, nutrition, and population port-folio for FY97–00, trust funds financed about 45percent of the total cost of loan preparation andappraisal.

Annex G1. The results may represent substantial progress

relative to the status quo ante (or what would havehappened in the absence of Bank/IDA intervention)but still lead to less than fully satisfactory outcomesif they had not met their stated objectives or had doneso inefficiently. This is especially likely when theobjectives are particularly demanding, as was thecase with the IDA replenishments.

2. The alternative—goal-free evaluation—is usedby OED whenever it is feasible to weigh the net ben-efits of the intervention in relation to the opportunitycost of the resources used. Thus, economic and finan-cial rates of returns are routinely used to assess thejustification of projects where appropriate data can beobtained. Unfortunately, cash flow analysis is notpractical for the complex programs funded by IDA.Even for projects, the implementation of policy andcapacity-building objectives (which are often themost crucial) cannot be quantified precisely enoughto allow cash flow analysis.

3. The methodology used in the CAEs has not beenagreed with management.

E n d n o t e s

1 1 7

Annex I1. See World Bank 1990, 2000a; Dollar and Pritchett,

1998.2. “Supporting Country Development: World Bank

Role and Instruments in Low- and Middle-IncomeCountries” (DC/2000-19).

3. See Poverty Reduction Strategy Papers: Opera-tional Issues, Joint IMF/World Bank Paper (R99-241),December 10, 1999, available at http://www.worldbank.org/poverty/.

4. For information on the Comprehensive Devel-opment Framework, access http://cdf.worldbank.org/.

5. See The Initiative for Heavily Indebted PoorCountries—Review and Outlook (DC/98-15), Sep-tember 2, 1998. It is important to note, too, that start-ing in the late 1980s, the World Bank established anumber of mechanisms for reducing IDA countries’debt levels, including the Debt Reduction Facility forIDA-Only Countries, which has now extinguished $6billion in commercial debt, and the Fifth DimensionProgram, which has allocated resources totaling $1 bil-lion to offset IBRD interest payments owed by IDA-only countries.

6. Article V, Section 1(g) of IDA’s Articles of Agree-ment states: “The Association shall make arrange-ments to ensure that the proceeds of any financingare used only for the purposes for which the financ-ing was provided, with due attention to considerationsof economy, efficiency and competitive internationaltrade and without regard to political or other non-economic influences or considerations.”

7. The SSPs are as follows: Reforming Public Insti-tutions and Strengthening Governance: A World BankStrategy (R2000-91), July 11, 2000; Rural Development:From Vision to Action (R97-25), March 11, 1997; Fuelfor Thought: Environmental Strategy for the Energy

Sector (R99-131), July 20, 1999; Health, Nutrition, andPopulation (R97-168), July 22, 1997; Population andthe World Bank: Adapting to Change (SecM2000-154),March 24, 2000; Education Sector Strategy (R99-68), May18, 1999; Social Protection Sector Strategy: From SafetyNet to Springboard (R2000-160), September 7, 2000;Strategy for the Financial Sector (R2000-169), October19, 2000; Cities in Transition: A Strategic View of Urbanand Local Government (R99-191), November 2, 1999;World Bank Group Private Sector Development Strat-egy (R99-175), December 14, 1999; Corporate Gover-nance: A Framework for Implementation (SecM99-590),September 13, 1999; and Small and Medium Enterprises:Sub-Sector Strategy (R2000-67), May 25, 2000. Seehttp://www.worldbank .org/, search term: Sector Strat-egy Papers.

8. See Reforming Public Institutions, note 7.9. See “Reaching the Rural Poor: The Rural Devel-

opment Strategy of the World Bank,” forthcoming.10. South Asia provides another Regional exam-

ple. While the portfolio of the rural unit has declined,estimates of rural commitments by other sector units(including health, education, etc.) suggest an almost40 percent increase in lending for the rural areasover the decade.

11. In 1990, for example, 64 percent of rural devel-opment projects under implementation in Africa wereat risk; by 1999–2000, that measure was reduced toabout 21 percent. Actual problem projects dropped from33 percent to 17 percent. The quality of projects at entryand during supervision has also improved, the latterrising from 50 percent in 1997 to 80 percent in 1999.

12. See Development Committee Communiqué,April 30, 2001, at http://wbln0018.worldbank.org/DCS/devcom.nsf/(communiquesm)/6FF3986EFC7595BD85256A3E006A6411?OpenDocument.


1 1 8

IDA Review Background PapersAll Background Papers are available uponrequest.

IDA Review. 2001a. Review of the Performance-Based Allocation System, IDA10–12: OED,World Bank: Washington, D.C.

——. 2001b. Review of Poverty Reduction inIDA10–12. OED, World Bank: Washington,D.C.

——. 2001c. Review of Aid Coordination in anEra of Poverty Reduction Strategies, IDA10–12.OED, World Bank: Washington, D.C.

——. 2001d. Proceedings Note IDA Review ExitWorkshop. OED, World Bank: Washington,D.C.

——. 2001e. Review of Governance—The Criti-cal Factor, IDA10–12. OED, World Bank:Washington, D.C.

——. 2001f. IDA10–12 Replenishment Under-takings Implementation Matrix. OED, WorldBank: Washington, D.C.

——. 2001g. Review of Private Sector Develop-ment in IDA10–12. OED, World Bank: Wash-ington, D.C.

——. 2001h. Review of Environmental Sustain-ability Issues in IDA10–12. OED, World Bank:Washington, D.C.

——. 2001i. Report on Country Consultations.OED, World Bank: Washington, D.C.

Other ReferencesAdams, Isaac, and Richard Johannson. 1998.

Evaluation Capacity Development in Africa:Selected Proceedings from a Seminar in Abid-jan. Washington, D.C.: World Bank.

Blagden, Mark, and Chitra Bhanu. 1999. Gender,Growth, and Poverty Reduction: Special Pro-gram of Assistance for Africa, 1998 StatusReport on Poverty in Sub-Saharan Africa.World Bank Technical Paper 428. Washington,D.C.

Branson, William, and Nagy Hanna. 2000. Own-ership and Conditionality. OED WorkingPaper No. 8. Washington, D.C.

Castro-Leal, Florencia, J. Dayton, L. Demery, K.Mehra. 1999. “Public Social Spending in Africa:Do the Poor Benefit?” World Bank ResearchObserver 14:49–72.

Collier, Paul, and David Dollar. 1999. Aid Allo-cation and Poverty Reduction. World Bank Pol-icy Research Working Paper. Washington, D.C.

Devarajan, S., David Dollar, and Torgny Holm-gren, eds. 1999. Aid and Reform in Africa.Washington, D.C.: World Bank.

Dalimi, Mansoor, A. Dhareshwar, D. Kaufmann,N. Lopez, V. Thomas, and W. Yan. 1999. TheQuality of Growth. Washington, D.C.: WorldBank.

DfID (Department for International Develop-ment, U.K.). 2000. Achieving Sustainability,Poverty Elimination and the Environment.London.

Development Committee. 2001. “Communiqué.”April 30, 2001. Available at http://wbln0018.worldbank.org/DCS/devcom.nsf/.

——. 2000. Supporting Country Development:World Bank Role and Instruments in Low-and Middle-Income Countries. Washington,D.C.

Dollar, David, and Aart Kraay. 2001. PropertyRights, Political Rights, and the Developmentof Poor Countries in the Post-Colonial Period.Washington, D.C.: World Bank.

——. 2000. Growth Is Good for the Poor. Washing-ton, D.C.: World Bank.

Dollar, David, and Lant Pritchett. 2000. Assess-ing Aid: What Works, What Doesn’t, and Why.World Bank Policy Research Report. NewYork: Oxford University Press.

Girishankar, Navin. 1999. Civil Service Reform:A Review of World Bank Assistance. Opera-tions Evaluation Study, Report No. 19599.Washington, D.C.: World Bank.

Hamilton, Kirk, and Priya Shyamsundar. 2000. AnEnvironmental Review of 1999 Country Assis-tance Strategies—Best Practice and LessonsLearned. Washington, D.C.: World Bank.

Hughes, Gordon, and K. Lvovsky. 1999. Envi-ronmental Challenges of Fuel Use: PollutionManagement in Focus. Washington, D.C.:World Bank.

IFAD (International Fund for Agricultural Devel-opment). 2001. The Challenge of Ending RuralPoverty, Rural Poverty Report 2001. New York:Oxford University Press.

Jayarajah, Carl, William Branson, and Binyak Sen.1996. The Social Dimensions of Adjustment:

1 1 9

REFERENCES

World Bank Experience, 1980–93. Washing-ton, D.C.: World Bank.

Johannson, Richard, and Isaac Adams. 1998.“Sector-wide Approaches to M&E in theHealth Sector in Ghana.” In Evaluation Capac-ity Development in Africa: Selected Proceed-ings from a Seminar in Abidjan. Washington,D.C.: World Bank.

Johnson, John, and Sulaiman Wasty. 1993. Bor-rower Ownership of Adjustment Programsand the Policy Economy of Reform. WorldBank Discussion Paper No. 199. Washing-ton, D.C.

Jones, Stephen. 2001. Increasing Aid Effective-ness in Africa? The World Bank and SectorInvestment Programs. Washington, D.C.:World Bank.

——. 1999. Tracking of Support Provided underSector Programs. Washington, D.C.: WorldBank.

Kanbur, Ravi. 2000. Economic Policy, Distribu-tion and Poverty: The Nature of Disagree-ments. Ithaca, N.Y.: Cornell University.

Kanbur, Ravi, and Todd Sandler. 1999. The Futureof Development Assistance: Common Poolsand International Public Goods. OverseasDevelopment Council, Policy Essay No. 25.Washington, D.C.

Killick, Tony, Rumani Gunatilaka, and Ann Marr.1998. Aid and the Political Economy of Pol-icy Change. London: Routledge.

Landau, Luis. 1997. Rebuilding the MozambiqueEconomy: Assessment of a Development Part-nership. OED Study Series. Washington, D.C.:World Bank.

Meerman, Jacob. 1997. Reforming Agriculture:The World Bank Goes to Market. Washing-ton, D.C.: World Bank.

OED (Operations Evaluation Department). 2001a.The Drive to Partnership: Aid Coordinationand the World Bank. OED Study Series. Wash-ington, D.C.

——. 2001b. Uganda Country Assistance Eval-uation. Washington, D.C.

——. 2001c. Annual Review of DevelopmentEffectiveness. Washington, D.C.

——. 2001d. OED Review of the Bank’s Per-formance on the Environment. Washington,D.C.

——. 2000a. Annual Review of DevelopmentEffectiveness,Washington, D.C.

——. 2000b. A Review of the World Bank’s 1991Forest Strategy and its Implementation. Wash-ington, D.C.

——. 2000c. Poverty Assessments: A Follow-upReview. Washington, D.C.

——. 2000d. Poverty Reduction in the 1990s: AnEvaluation of Strategy and Performance.Washington, D.C.

——. 2000e. Tanzania Country Assistance Eval-uation. Report No. 20902-TA. Washington, D.C.

——. 1999a. Annual Review of DevelopmentEffectiveness. Washington, D.C.

——. 1999b. Education in Bangladesh: A Country-Specific Sector Review. Washington, D.C.

——. 1999c. Higher Impact Adjustment Lending,Vols. 1 & 2. Washington, D.C.

——. 1999d. Investing in Health: DevelopmentEffectiveness in the Health, Nutrition, andPopulation (HNP) Sectors. OED Study Series.Washington, D.C.

——. 1999e. Burkina Faso Country AssistanceEvaluation. Report No. 20704. Washington,D.C.


——. 1998b. The Special Program of Assistancefor Africa (SPA), an Independent Evaluation.Washington, D.C.


——. 1997b. Reforming Agriculture: The WorldBank Goes to Market. Washington, D.C.

Prennushi, G., G. Rubio, and K. Subbaro. “Mon-itoring and Evaluation: Technical Notes.”Poverty Net. Available at http://www.worldbank.org/poverty/strategies/srcbook/m & e0925.pdf.

Ravallion, Martin. 2000. Growth, Inequality andPoverty: Looking beyond Averages. WorldBank, Development Research Group. Wash-ington, D.C.

Reed, D. 1996. Structural Adjustment, the Envi-ronment and Sustainable Development. Lon-don: Earthscan, World Wide Fund for Nature.

Robb, C. 1999. Can the Poor Influence Policy:Directions in Development. Washington, D.C.:World Bank.


1 2 0

R e f e r e n c e s

1 2 1

Schacter, Mark. 2001. Sector Wide Approaches,Accountability and CIDA: Issues and Recom-mendations. Hull: Canadian InternationalDevelopment Agency.

Tarp, Finn, ed. 2000. Foreign Aid and Develop-ment: Lessons Learnt and Directions for theFuture. London: Routledge.

Thomas, Vinod, Mansoor Dalimi, AshokDhareshwar, Daniel Kaufmann, Nalin Kishor,Ramon Lopez, Wang Yan. 1999. The Qualityof Growth. Washington, D.C.: World Bank.

Whitehead, Ann, and Matthew Lockwood.1999. Gender in the World Bank’s PovertyAssessments: Six Case Studies from Sub-Saharan Africa. Discussion Paper 99:1–51.Geneva: United Nations Institute for SocialDevelopment.

World Bank. 2001a. Engendering Development:Through Gender Equality in Rights, Resources,and Voice. Policy Research Report. New York:Oxford University Press.

——. 2001b. Global Economic Prospects andthe Developing Countries 2001. Washington,D.C.

——. 2001c. Update on the HIPC Initiative andthe PRSP Program, February 6, 2001. Availableat http://www1.worldbank.org/prsp/PRSP_Policy_Papers

——. 2001d. World Development Report 2001.New York: Oxford University Press.

——. 2001e. How Far into the Mainstream? AReview of Environmental Issues in IDA Activ-ities. Washington, D.C.

——. 2000a. World Development Report 2000/2001: Attacking Poverty. New York: OxfordUniversity Press.

——. 2000b. Can Africa Claim the 21st Century?Washington, D.C.

——. 2000c. “Monitoring and Evaluation: Tech-nical Notes.” Poverty Net. Available at http://www.worldbank.org/poverty/strategies/chapters/monitoring/m&e0925.pdf.

——. 2000d. Poverty Reduction and the WorldBank: Progress in Fiscal 1999. Washington, D.C.

——. 2001e. Education and Health in Sub-Saharan Africa: A Review of Sector-wideApproaches. Washington, D.C.

——. 1999. Annual Progress Report on Poverty.Washington, D.C.

——. 1998a. Additions to IDA Resources: TwelfthReplenishment, A Partnership for PovertyReduction. Washington, D.C.

——. 1998b. Financial Sector Reform: A Reviewof World Bank Assistance. Report No. 17454.Washington, D.C.

——. 1998c. Partnership for Development: FromVision to Action. Partnership Group. Wash-ington, D.C.

——. 1997a. Five Years after Rio: Innovations inEnvironmental Policy. World Bank ESSD Stud-ies and Monographs Series. Washington, D.C.

——. 1997b. Helping Countries Combat Cor-ruption: The Role of the World Bank. Wash-ington, D.C.

——. 1997c. IDA in Action 1993–1996: The Pur-suit of Sustained Poverty Reduction. Wash-ington, D.C.

——. 1997d. World Development Report 1997:Role of the State. New York: Oxford Univer-sity Press.

——. 1994. Development in Practice: Gover-nance—the World Bank’s Experience. Wash-ington, D.C.

——. 1992a. World Development Report 1992:Environment and Development. Washington,D.C.

——. 1992b. Governance and Development.New York: Oxford University Press.

——. 1990. World Development Report 1990:Poverty. New York: Oxford University Press.

——. 1989. From Crisis to Sustainable Growth—Sub-Saharan Africa: A Long-term PerspectiveStudy. Washington, D.C.

——. 1983. World Development Report: Manag-ing Development. Washington, D.C.

——. 1982. IDA in Retrospect: The First TwoDecades of the International DevelopmentAssociation. Washington, D.C.

IDA’s Partnership for Poverty Reduction · Pooling Donor Funds in Support of Health Sector Reform...

Documents

Transcript of IDA’s Partnership for Poverty Reduction · Pooling Donor Funds in Support of Health Sector Reform...