W O R L D K O P E R A T I O NS EVA L U AT I O N DE P A R T M E N T kcm 21274 [Jg adesh: Wl.iruss Through Iv,finership é -*c "I 4 Flur- OPERATIONS EVALUATION DEPARTMENT ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Operations Evaluation Department (OED) is an independent unit within the World Bank; it reports directly to the Bank's Board of Executive Directors. OED assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. WORLD BANK OPERATIONS EVALUATION D EPARTMENT Bangladesh: Progress Through Partnership COUNTRY ASSISTANCE REVIEW Roger J. Robinson 1999 The World Bank www.worldbank.org/html/oed Washington, D.C. Copyright @ 1999 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First edition 1999 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 com- plete 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 ISSN 1011-0985 ISBN 0-8213-4293-2 Library of Congress Cataloging-in-Publication Data Robinson, Roger J., 1948- Bangladesh: Progress Through Partnership (country assistance review) / Roger J. Robinson. p. cm - (A World Bank operations evaluation study) ISBN 0-8213-4293-2 1. World Bank-Bangladesh. 2. Economic assistance-Bangladesh. 3. Bangladesh-Economic conditions. 4. Bangladesh-Economic policy. I. Title II. Series HG3881.5.W57R63 1998 338.95492-dc2l 98-36252 CIP Printed on recycled paper. Contents v Acknowledgments vii Foreword, Prefacio, Pr6face ix Executive Summary, Resumen, Resumi Analytique xvii Abbreviations and Acronyms 1 1. Achievements Despite Serious Constraints 2 The Development Challenge 2 What Difference Did IDA Make? 3 2. Relevance of IDA's Country Assistance Strategy 3 IDA Strategy in the Early 1980s: Getting a Handle on the Country's Problems 4 Assessment of the Relevance of IDA Strategy from 1980 to 1985: A Slow Start on the Development Agenda 5 IDA Strategy in the Late 1980s and Early 1990s: Trying to Adjust to Absorptive Constraints 6 Assessment of the Relevance of IDA Strategy from 1986 to 1990: Recognizing Some Lessons but Not Others 6 IDA Assistance Strategy from 1991 to 1996: Toughening Conditionalities 7 Assessment of the Relevance of IDA Strategy from 1991 to 1996: A More Ratio- nal Approach 7 Refining Country Strategy Relevance: Lessons for the Future 9 3. Strategy Implementation: Emphasis, Instruments, and Efficiency 9 Emphasis 11 Economic and Sector Work 12 Aid Coordination 12 Participation, Collaboration, and Awareness 13 The Efficiency of IDA's Assistance Program 15 4. Development Outcomes 17 Poverty Alleviation and Human Development 18 Family Planning, Health, and Nutrition 19 Education 20 Industry and Finance 21 Energy and Infrastructure 21 Agriculture, Food Security, and Flood Control 23 To Sum Up 25 5. Efficacy of IDA's Assistance Strategy: Overall Assessment 33 6. Lessons for the Future 34 Social Sector Development 36 Industrial and Financial Sector Development 37 Energy and Infrastructure 37 Agriculture and Flood Control 111 Bangladesh: Progress Through Partnership Annexes 39 Annex A: Executive Summary, Report No. 17455-BD, for Board Discussion 48 Annex B: Report from the Committee on Development Effectiveness (CODE) 50 Annex C: Summary of Bank-GOB Discussions of the CAR, June 2, 1999 51 Annex D: Government Response 53 Endnotes 55 Bibliography Figures 16 4.1 Economic Growth Comparisons 17 4.2 GDP per Capita (1980-96) 18 4.3 Exports (US$ millions), 1982-95 Boxes 27 5.1 Evaluation of IDA's Import Program Credits from 1980 to 1988 30 5.2 IFC Activity in Bangladesh Tables 8 2.1 Relevance of IDA's Country Assistance Strategy 10 3.1 Commitments (by Sector), 1980-95 11 3.2 Portfolio Ratings and Supervision Effort 11 3.3 Intensity of ESW 13 3.4 Comparative Data on the Efficiency of Bank Assistance Programs 14 3.5 Average Elapsed Time in Months for FY90-96 Approvals (IEPS to Board Presentation) 16 4.1 Macroeconomic Performance 19 4.2 Fiscal Performance (percent of GDP) 19 4.3 Social Indicators 20 4.4 Education Comparators 21 4.5 Manufacturing Performance (percent) 22 4.6 Comparative Overview of Infrastructure Provision iv Acknowledgments This report was prepared by a team led by Roger J. This study was published in the Partnerships and Knowl- Robinson. The team consisted of Homayoun Ansari, edge Group (OEDPK) by the Outreach and Dissemina- David Gisselquist, David Greene, Ashok Khanna, and tion Unit. The task team includes Elizabeth Campbell- Carlos Reyes. Eneshi Irene K. Davis provided adminis- Pag (task team leader), Caroline McEuen (editor), Kathy trative support. Strauss and Lunn Lestina (graphic designers), and Juicy Qureishi-Huq (administrative assistant). Director-General, Operations Evaluation Department: Robert Picciotto Director, Operations Evaluation Department: Elizabeth McAllister Group Manager, Country Evaluations and Regional Relations: Ruben Lamdany Task Manager: Roger J. Robinson V  Foreword FOREWORD PREFACIO PRÉFACE This Country Assistance Review En este examen de la asistencia a Cette étude sur l'assistance au (CAR) presents the findings of an Bangladesh se presentan las conclus Bangladesh présente les conclusions OED team that visited Bangladesh in iones a las que Ilegó un equipo del d'une équipe de l'OED qui s'est August/ September 1996. This team Departamento de Evaluación de rendue au Bangladesh en held extensive discussions with Operaciones que visitó ese pais en août/septembre 1996. Cette équipe a government officials, representatives agosto-septiembre de 1996. El equipo eu des discussions approfondies avec of civil society and the academic sostuvo amplias conversaciones con des membres du Gouvernement, des community, nongovernmental orga- funcionarios gubernamentales, représentants de la société civile et des nizations (NGOs), other donors, and representantes de la sociedad civil y la milieux universitaires, des ONG, Bank staff in the resident mission. comunidad académica, ONG, otros d'autres bailleurs de fonds et les The valuable contribution and coop- donantes y funcionarios del Banco membres de la mission résidente de la eration of all involved is gratefully Mundial de la misión residente. Se Banque. Elle les remercie tous de leurs acknowledged. reconoce y se agradece la valiosa précieuses contributions et de leur This CAR covers a long period contribución de todos los que coopération. of IDA involvement with the participaron. La présente étude couvre la Government of Bangladesh and its El presente examen abarca un longue période d'assistance fournie development effort. The 16 years largo periodo de colaboración de la par l'IDA à l'appui de l'action de reviewed cover 93 projects, a AIF con el Gobierno de Bangladesh y développement du Gouvernement du massive volume of economic and sus esfuerzos en pos del desarrollo. El Bangladesh. Les 16 années sur sector work, and an ongoing examen del periodo de 16 afios abarcó lesquelles porte l'étude ont servi de policy dialogue that has examined 93 proyectos, un gran volumen de cadre à 93 projets, à un nombre all aspects of economic and social estudios económicos y sectoriales y un impressionnant d'études économiques life. This report represents a syn- diàlogo constante sobre politicas en el et sectorielles et à un effort de thesis of the key findings of an que se han analizado todos los concertation portant sur tous les evaluation encompassing a very aspectos de la vida económica y social. aspects de la vie économique et broad agenda. All materials, as En este informe se presenta una sínte- sociale. Ce rapport fait la synthèse des well as more detailed working sis de las conclusiones fundamentales principales conclusions de papers on each sector, are lodged de una evaluación muy completa. l'évaluation, qui était très ambitieuse. in the Bangladesh CAR files in Todos los materiales, asi como los Tous les documents, ainsi que des OED and are readily available. documentos de trabajo màs detallados rapports détaillés sur chaque secteur, The CAR's main conclusion is sobre cada sector, se encuentran en los figurent dans le dossier de l'étude sur that IDA's assistance to Bangladesh archivos del examen de la asistencia a l'assistance au Bangladesh à l'OED et over the past 16 years has been Bangladesh mantenidos por el peuvent être facilement obtenus. effective, and the partnership Departamento de Evaluación de La principale conclusion de forged with the government has Operaciones, y estàn a disposición de l'étude est que l'assistance fournie par brought some notable successes. los interesados. l'IDA au Bangladesh ces 16 dernières Good progress has been made in La conclusión principal del années a été efficace et que la the reduction of fertility and in key examen es que la asistencia que prestà collaboration établie avec le health indicators. Progress has also la AIF a Bangladesh en los Gouvernement a été fructueuse. Le been made in education, particu- últimos 16 afios ha sido eficaz, y que taux de fécondité a diminué et les larly primary education, and the la cooperación solidaria que se esta- principaux indicateurs de santé ont country has achieved self-suffi- bleció con el Gobierno ha dado lugar progressé. On a également enregistré vii Bangladesh: Progress Th rough Partnership ciency in food-grains in a nor- a éxitos notables. Se ha avan- des progrès dans le secteur de mal climate year. zado mucho en materia de l'éducation, en particulier dans As is readily understand- reducciôn de la fecundidad e l'enseignement primaire, et le able in a country with com- indicadores clave de la salud. pays peut satisfaire ses besoins plex economic, social, and También se han registrado pro- en céréales vivrières les années pohtical issues, progress has gresos en materia de educaciôn, où les conditions climatiques not been uniformly good. In sobre todo primaria, y el pais ha sont normales. general the reasons for disappoint- alcanzado la autosuficiencia en la pro- Comme on peut s'y attendre dans ing progress are deeply entrenched ducción de cereales alimentarios en un un pays en proie à de graves governance issues, often requiring a aio de condiciones meteorolôgicas problèmes économiques, sociaux et strong and sustained political normales. politiques, les progrès n'ont pas été commitment to effect change. Como cabe prever en un pais con uniformes. Le rythme décevant des The lessons drawn from this una complejidad de situaciones réformes tient, d'une façon générale, CAR are presented at a strategic económicas, sociales y políticas, el aux problèmes de fond qui se posent level, and are applicable in progreso no ha sido uniforme. En en matière de gouvernance et qui Bangladesh and other countries. general, cuando el avance no ha sido exigent souvent des pouvoirs publics The aim has been to develop satisfactorio se ha debido a cuestiones qu'ils témoignent d'une ferme volonté themes that cut across all sectors muy arraigadas de gestión de los de changement et de persévérance. and will thus help to refine IDA's asuntos públicos, que a menudo Les leçons tirées de cette étude sur Country Assistance Strategy and exigen un compromiso politico firme y l'assistance-pays sont présentées d'un future assistance efforts. sostenido para lograr el cambio. point de vue stratégique et Las lecciones extraidas de este s'appliquent non seulement au examen se presentan a nivel Bangladesh, mais également à d'autres estratégico y pueden aplicarse a pays. L'objectif était d'identifier des Bangladesh y a otros paises. El thèmes communs à tous les secteurs objetivo ha sido desarrollar temas pour aider l'IDA à affiner sa stratégie- comunes a todos los sectores y de esta pays et ses futurs efforts d'assistance. manera ayudar a afinar la estrategia de asistencia a los paises de la AIF y las actividades futuras de asistencia. Robert Picciotto Director-General, Operations Evaluation Department viii Executive Summary EXECUTIVE RESUMEN RÉSUMÉ SUMMARY ANALYTIQUE Despite the political traumas from Pese a las conmociones politicas que Malgré les troubles politiques qui ont 1977 to 1980, Bangladesh has tuvieron lugar desde 1977 hasta agité le pays entre 1977 et 1980, le made progress in economic and 1980, Bangladesh ha registrado Bangladesh a progressé dans la voie social development and has been avances en su desarrollo económico y du développement économique et transformed over the past 16 years. social y una transformación notable social et subi de profondes The country now approaches self- en los ùltimos 16 afños. El pais se está transformations au cours des 16 sufficiency in food-grains, and key acercando a la autosuficiencia en dernières années. Le pays a quasiment social indicators (fertility rates, cereales alimentarios y los atteint l'autosuffisance en céréales infant mortality, life expectancy, indicadores sociales clave (tasa de vivrières et les indicateurs sociaux de primary school enrollments, and fecundidad, mortalidad infantil, base (taux de fécondité, mortalité infantile, espérance de vie, scolarisation primaire et niveau d'instruction des adultes) ne cessent de s' améliorer. Malgré ces progrès, la sécurité alimentaire, les futures tendances démographiques, la dépendance nationale à l'égard de l'aide étrangère -- et la fragilité de l'écosystème du Bangladesh continuent de poser e /problème. Le pays traverse une période de mutation politique, avec la mise en place d'un système démocratique multipartite et d'une administration civile qui a connu de J nombreux remaniements. Le premier gouvernement a nationalisé la plupart - ldes secteurs (industrie, finances et grandes entreprises commerciales) et résolument opté pour un adult literacy) have shown steady esperanza de vida, matricula en la développement directement orchestré improvement. escuela primaria y alfabetizaciôn de par l'État. Les gouvernements qui se Despite these achievements, adultos) han registrado una sont succédé ont bien tenté de questions remain as to food secu- constante mejoria. désengager l'État des activités rity, future demographic trends, Pese a estos logros, sigue économiques, mais en vain ; ces the country's aid-dependency, and habiendo interrogantes en torno a la efforts, contrariés par l'opposition the fragility of its ecosystem. seguridad alimentaria, las tendencias politique ou les intérêts en jeu, n'ont Bangladesh is in the midst of a demogràficas futuras, la dependencia en effet pas été soutenus. political transformation and devel- de la asistencia y la fragilidad de su La Banque mondiale est opment of a multiparty democra- ecosistema. Bangladesh se encuentra intervenue dans le développement tic system, encompassing also the en medio de una transformación économique du Bangladesh dès la establishment of a civil adminis- politica y del desarrollo de un création du pays. Cette association a ix Bangladesh: Progress Through Partnership tration structure. The first sistema democràtico revêtu la forme de prêts, de government nationalized multipartidista que abarca conseils sur les mesures à most of the industrial, finan- también el establecimiento de prendre et de coordination des cial, and larger commercial una estructura de bailleurs de fonds. Au cours des sectors, and espoused a administración pública. El deux dernières décennies, l'IDA strong, direct public sector primer gobierno nacionalizô la a financé près du quart de tous role in the development mayoria de los sectores indus- les engagements d'aide process. Successive governments triales, financieros y comerciales màs étrangère, couvrant tous les secteurs have sought to reverse the role of grandes y optó pot una función économiques, y compris le soutien de the state in direct economic activi- fuerte y directa del sector público en la balance des paiements. Un nombre ties, but this effort has not been el proceso de desarrollo. Los impressionnant d'études économiques consistently sustained and has gobiernos sucesivos han procurado et sectorielles a été réalisé pour faltered in the face of political or revertir la función del Estado en las faciliter les réformes et la coordination vested interests. actividades económicas directas, pero de l'aide. The World Bank has been este esfuerzo no se ha sostenido en involved in Bangladesh's economic forma constante y los intentos han Le défi du développement : vers un development from the country's fracasado por razones politicas o de équilibre réaliste entre les besoins et beginnings. This partnership has intereses creados. les attentes encompassed lending, policy El Banco Mundial ha L'une des grandes faiblesses de advice, and donor coordination. participado en el desarrollo l'économie du Bangladesh tient à son Over the past two decades, the económico de Bangladesh desde los taux d'investissement et d'épargne International Development Associ- albores del pais. Esta cooperación intérieure. Le pays est par ailleurs ation (IDA) has financed about a ha abarcado financiamiento, fortement tributaire des flux d'aide quarter of all foreign aid commit- asesoramiento en materia de concessionnelle. Ainsi, le Bangladesh a ments, covering all sectors of the políticas y coordinación con los reçu une aide plus importante pendant economy, including balance of donantes. En las últimas dos les trois années qui ont suivi sa payments support. A massive décadas, la AIF ha financiado création que durant les 24 années volume of economic and sector alrededor de una cuarta parte de précédentes, lorsque le pays était work has contributed to policy todos los compromisos de asistencia encore la province du Pakistan reform and aid coordination. exterior, en todos los sectores de la oriental. economia, incluso para atender De nombreux observateurs, au The Development Challenge: Toward necesidades de balanza de pagos. Bangladesh et ailleurs, ont émis la Effective Balance Between Needs and Un gran volumen de estudios crainte que le recours à l'aide Expectations económicos y sectoriales ha étrangère ne pose des problèmes de A fundamental weakness of the contribuido a la reforma politica y souveraineté nationale et ne Bangladesh economy has been the a la coordinaciôn de la asistencia. compromette l'autonomie du pays low rate of investment and domes- dans le processus de développement. tic saving. Dependency on El desafio del desarrollo: Les ressources fournies au titre de external concessional flows has Hacia el equilibrio eficaz entre las l'aide étrangère étant fongibles, elles been high. In the country's first 3 necesidades y las expectativas ont en fait contribué à financer les years, Bangladesh received more Una de las deficiencias fundamentales déficits budgétaires et les entreprises aid than in all its 24 years as East de la economia de Bangladesh ha sido publiques inefficaces, ce qui est peut- Pakistan. la baja tasa de inversión y de ahorro être allé à l'encontre de la nécessité Many observers, both domes- interno. La dependencia de flujos d'améliorer la rentabilité et le tic and foreign, have expressed exteriores en condiciones recouvrement des coûts dans les concern that the country's use of concesionarias ha sido elevada. En los compagnies de services publics. Cela foreign aid bas infringed on primeros tres afños de existencia del étant, les institutions du secteur national sovereignty issues and pais, Bangladesh recibió más ayuda public n'auraient probablement pas undermined self-reliance in the que en los 24 afños en que el territorio mieux fonctionné en l'absence d'une x Executive Summary development process. Given se Ilamó Pakistán Oriental. aide étrangère. Sans concours the fungibility of resources, Muchos observadores, extérieur, bon nombre de aid indeed funded budget nacionales y extranjeros, han programmes et de projets deficits, inefficient state- expresado inquietud por el auraient été abandonnés ou bien owned enterprises (SOEs), hecho de que el pais, al recurrir leur exécution aurait tardé and perhaps reduced the a la asistencia extranjera, haya davantage, avec les profondes pressure to improve effi- infringido cuestiones de répercussions économiques et ciency and cost recovery in public seguridad nacional y haya socavado les importants coûts sociaux que cela utilities. However, it is unlikely la confianza del pais en su propio aurait entraînés. Tel aurait été le cas, that public sector institutions proceso de desarrollo. Dado que los par exemple, du programme de would function more efficiently in recursos son intercambiables, la population, du programme de the absence of aid. Without exter- asistencia se dirigió en realidad a vaccination infantile, du nal support, many programs and financiar déficit presupuestarios, développement des petits ouvrages projects would have proceeded empresas estatales ineficientes y d'irrigation ou de l'enseignement more slowly, if at all, and this quizá redujo la presión para elevar la primaire. delay would have had serious eco- eficiencia y la recuperación de costos nomic ramifications and adverse en los servicios públicos. No Bien-fondé de la stratégie-pays social costs. Examples abound, obstante, es poco probable que las such as the population program, instituciones del sector público 1980-85: Le programme de child immunization, minor irriga- funcionaran más eficientemente si no développement tarde à démarrer tion development, and primary se contase con asistencia. Sin Au début des années 80, le pays est passé education. respaldo externo, muchos programas d'un régime de loi martiale au y proyectos hubieran avanzado más gouvernement civil du président Zia ur- Relevance of the Country Assistance lentamente, y este retraso hubiera Rahman. Le nouveau gouvernement avait Strategy tenido graves ramificaciones d'ambitieux objectifs de développement, económicas y costos sociales tels que parvenir à un taux de fécondité de 1980-85: A Slow Start on the perniciosos. Abundan ejemplos, remplacement et éliminer les importations Development Agenda como el programa demográfico, la alimentaires d'ici à 1985. The early 1980s witnessed a inmunización infantil, el fomento del transition from martial law to the riego en pequeia escala y la 1986-90 : Les leçons n'ont été qu'en civilian government of President educación primaria. partie mises à profit Ziaur Rahman. Its development L'intégration dans la SAP des goals were ambitious, such as Relevancia de la estrategia de enseignements tirés de l'expérience du replacement fertility levels and asistencia al pais début des années 80 a finalement elimination of food imports by permis de mieux adapter la stratégie à 1985. 1980-85: Un inicio lento del la situation. En 1986, l'IDA avait programa de desarrollo reconnu la nécessité de s'attaquer aux 1986-90: Recognizing Some A principios de la década de 1980 se contraintes sectorielles et Lessons but Not Others produjo una transición de la ley macroéconomiques, en imposant aux Eventually lessons learned in the marcial al gobiemo civil del pouvoirs publics des conditions plus early 1980s influenced the Coun- presidente Ziaur Rahman. Sus metas de précisément axées sur l'action try Assistance Strategy toward desarrollo eran ambiciosas, como el logro gouvernementale. L'IDA a également greater relevance. By 1986 IDA de niveles de fecundidad de reemplazo y la reconnu qu'il était vain de s'attaquer had recognized the need to eliminaciôn de las importaciones de au problème des entreprises publiques address sector and macroeco- alimentos para 1985. inefficaces sans un ferme engagement nomic constraints with sharply de la part de l'État. focused policy conditionality. IDA 1986-90: Se reconocen also recognized that trying to deal algunas lecciones pero no otras with inefficient public enterprises Con el tiempo, con las lecciones de los xi Bangladesh: Progress Through Partnership in the absence of a strong primeros afios de la década de 1991-96: Une approche plus government commitment was 1980, la estrategia de asistencia rationnelle fnot productive. al pais llegó a ser más pertinente. Les documents relatifs à la Hacia 1986, la AIF había stratégie de l'IDA du début des 1991-96: A More Rational reconocido la necesidad de 4 années 90 témoignent de Approach resolver las limitaciones l'approche plus rigoureuse de A review of IDA's strategy sectoriales y macroeconómicas, l'Association et de la nouvelle documents from the early 1990s con una condicionalidad claramente importance accordée à l'exécution et à discloses a hardening of IDA's enfocada en las politicas. Ademàs, la l'engagement des pouvoirs publics. Les strategy and a sharper focus on AIF reconoció que no era productivo prêts à l'ajustement ont été ramenés à implementation and government tratar de resolver el problema de las environ 15 % du programme de commitment. Adjustment lending empresas pùblicas ineficientes si financement prévu et les opérations was reduced to about 15 percent faltaba el firme compromiso del sectorielles entreprises de manière plus of the proposed lending program, Gobierno. sélective, compte dûment tenu de la and sector operations were under- volonté de l'État de respecter le taken in a selective manner, in line 1991-96: Un enfoque más racional calendrier fixé pour le vaste programme with assessment of the govern- Si se examinan los documentos sobre de réformes. ment's willingness to proceed with la estrategia de la AIF desde principios a broad reform agenda in a timely de esta década, se percibe que la Affiner la stratégie-pays manner. institución ha endurecido su enfoque On peut tirer plusieurs enseignements y se ha centrado más en la ejecución y majeurs de la stratégie-pays suivie par Refining Country Strategy Relevance en el compromiso del Gobierno. El l'IDA ces 15 dernières années: A review of IDA's Country financiamiento para fines de ajuste se Assistance Strategy (CAS) over the redujo a aproximadamente el 15% del • Critères de performance past 15 years highlights and programa crediticio propuesto, y las mesurables : Une stratégie-pays reinforces key generic lessons: operaciones sectoriales se iniciaron de est d'autant mieux adaptée et utile manera selectiva de conformidad con au plan opérationnel qu'elle • Monitorable Performance la evaluación de la disposición del prévoit des critères spécifiques Benchmarks: Country Gobierno para proceder con un permettant de mesurer Assistance Strategies become amplio programa de reforma en las l'avancement des réformes et les much more relevant and fechas previstas. résultats. operationally useful if there are • Flexibiliassé : Avant de définir une specific benchmarks in both Aumento de la pertinencia de la nouvelle stratégie-pays, il faudrait policy reform and outcomes. estrategia de asistencia: Lecciones s'attacher davantage à déterminer • Flexibility in the Lending para el futuro l'efficacité de la précédente. Program: Strategies should Tras un examen de la estrategia de • Évaluer la capacité institutionnelle de allow considerable flexibility asistencia a los paises de los últimos façon réaliste: La SAP sous-estime to adjust the lending program quince afños se destacan y se refuerzan souvent le temps nécessaire à la in response to progress in lecciones genéricas fundamentales: réforme des institutions et au identified key reform areas. renforcement des capacités, ainsi que la • Realism of Risk Assessments: • Puntos observables de complexité de la tâche. The validity and relevance of referencia para medir los • Enseignements à tirer pour CASs are greatly enhanced by resultados: Las estrategias de asis- l'avenir realistic assessments of the tencia a los paises cobran mucha • Le bilan de l'aide fournie par difficulties of achieving màs pertinencia y son màs útiles l'IDA au Bangladesh ces dernières progress. These make the risk desde el punto de vista operacional années est globalement positif et assessments more meaningful, si incluyen puntos de referencia la collaboration avec les pouvoirs and expectations more especificos para medir la reforma publicassé: Avant de définir une realistic. de las politicas y los resultados. nouvelle stratégie-pays, il faudrait xii Executive Summary • Experience Needs a • Flexibilidad del programa s'attacher davantage à Thorough Assessment: crediticio: Las estrategias déterminer l'efficacité de la Before embarking on defin- deberán ser lo suficientemente précédente. ition of a future Country flexibles como para permitir •Évaluer la capacité institutionnelle Assistance Strategy, more el ajuste del programa de façon réaliste: La SAP sous- attention should be directed crediticio en respuesta a los estime souvent le temps nécessaire à to determining the effec- progresos alcanzados en la réforme des institutions et au tiveness of the last strategy. materia de reformas clave renforcement des capacités, ainsi Realism in Institutional identificadas. que la complexité de la tâche. Capacity Assessments: In • Realismo de las evaluaciones de general, CASs underestimate riesgos: Las estrategias de Enseignements à tirer pour l'avenir the complexity and the time asistencia a los paises son mucho Le bilan de l'aide fournie par l'IDA au necessary to bring about insti- màs vàlidas y pertinentes cuando Bangladesh ces dernières années est tutional change and capacity se efectúan evaluaciones realistas globalement positif et la collaboration enhancement. de las dificultades que impiden el avec les pouvoirs publics a produit des avance. Esto hace que las résultats palpables. Le pays s'est Lessons for the Future evaluaciones de riesgos sean màs transformé, en dépit d'un climat Overall, IDA assistance to Bangla- significativas y las expectativas, politique très difficile. Comme on desh has been effective, and the màs realistas. pouvait s'y attendre, compte tenu de partnership forged with the gov- • La experiencia requiere una la complexité des enjeux ernment has brought about some evaluaciôn integral: Antes de économiques, sociaux et politiques, les notable successes. The country has iniciar la formulación de una progrès ne sont pas uniformes. Si les been transformed, and this has estrategia futura de asistencia a changements s'opèrent à un rythme been achieved in a very difficult los paises, deberà determinarse décevant, c'est parce que des political climate. As might be con màs atención la eficacia de la problèmes de fond se posent en expected with such a complex última estrategia. matière de gouvernance et que les array of economic, social, and • Realismo de las evaluaciones de la puissants intérêts en jeu cherchent à political issues, progress has not capacidad institucional: En general, maintenir le statu quo. been uniformly good. The disap- las estrategias de asistencia a los La future stratégie d'assistance au pointing progress is the result of paises subestiman la complejidad y Bangladesh devra tenir compte des deeply entrenched governance el tiempo necesario para el cambio leçons générales énumérées ci-après, issues and powerful vested inter- institucional y el fortalecimiento de qui s'appliquent à tous les secteurs. ests that have sought to maintain la capacidad. Redéfinir le rôle du secteur the status quo. public. Le secteur privé, les ONG et The following are the generic Lecciones para el futuro les collectivités locales pourraient lessons that cut across all sectors, En términos generales, la asistencia de prendre une part beaucoup plus active to guide the future Country Assis- la AIF a Bangladesh en los ùltimos à la prestation de services. Il faudra tance Strategy. afios ha sido eficaz y la relaciôn s'attacher davantage à promouvoir la Redefining the Role of the solidaria establecida con el Gobierno participation des collectivités locales Public Sector. There is much more ha dado lugar a algunos éxitos et des municipalités à tous les aspects scope for the private sector, notables. El pais se ha transformado, du développement et de l'entretien des nongovernmental organizations y esto se ha logrado en el entorno de infrastructures, ainsi qu'à l'éducation (NGOs), and local governments to un clima politico muy dificil. Como et aux soins de santé primaires. be involved in the provision of cabe esperar de una complejidad tal Mettre davantage l'accent sur la services. Much greater efforts de problemas econômicos, sociales y responsabilisation. Les futurs projets must be made to encourage a politicos, el progreso no ha sido devront être conçus de façon à larger role for local and municipal uniforme. Los casos de avance poco responsabiliser davantage les individus governments in all aspects of alentador se deben a cuestiones muy chargés de fournir un service ou de réaliser infrastructure development and arraigadas de gestiôn de los asuntos un projet. Toutes les parties prenantes xiii Bangladesh: Progress Through Partnership maintenance, education, and pùblicos y a poderosos intereses doivent être associées à la conception primary health care. creados que han procurado des projets pour définir les Increasing Empbasis on mantener el statu quo. paramètres de contrôle et le cadre Accountability. Future A continuación se presentan permettant de les appliquer project design rnust seek to las lecciones genéricas para el Nécessité d'accorder une improve the accountability of futuro, que abarcan todos los plus grande attention aux those responsible for sectores, a fin de orientar las questions de viabilité. Il n'existe providing a service or completing estrategias futuras de pas de formule applicable à tous les a project. Project design must asistencia al pais. secteurs ou à tous les projets, mais on involve all stakeholders, to define Redefinición de la funciôn del doit veiller de beaucoup plus près au the parameters of accountability sector público. Existe un amplio financement récurrent des dépenses de and set a framework for alcance para la participaciôn del fonctionnement et d'entretien, au enforcement. sector privado, las ONG y los recouvrement des coûts (même s'il est Need to Pay Greater Atten- gobiernos locales en la prestación de partiel au début du projet) et aux tion to Sustainability Issues. No servicios. Deberá desplegarse un structures institutionnelles, qu'il single approach will work in all esfuerzo màs intenso por alentar una s'agisse d'organismes d'exécution ou sectors or projects, but much more función más preponderante de los de cadres organisationnels. emphasis must be given to recur- gobiernos locales y municipales en Une preuve concluante: rent expenditure requirements for todos los aspectos del desarrollo y el l'appropriation du projet par operations and maintenance, to mantenimiento de la infraestructura, l'Emprunteur. Il faut s'employer cost recovery (if only partial in the asi como de la educaciôn y la atención davantage à renforcer la volonté beginning), and to institutional primaria de la salud. politique d'introduire les changements structures or implementing Mayor énfasis en la rendicidn de institutionnels nécessaires et définir de agencies and organizational cuentas. El diseio de los proyectos bons indicateurs de cette volonté frameworks. futuros deberà procurar mejorar la d'agir. Procéder autrement serait Borrower Ownership as the rendiciôn de cuentas de quienes son contraire aux intérêts du peuple Acid Test. More effort should be responsables de prestar un servicio o bangladais et aurait des retombées made to strengthen the political completar un proyecto. En el diseñ-o négatives sur l'ensemble du commitment to necessary institu- de proyectos deben participar todos programme d'assistance. tional change, and solid indica- los interesados, a fin de definir los Se focaliser sur les objectifs clés et tors should be sought that parámetros de la rendición de cuentas faire preuve de plus de sélectivité. demonstrate this commitment. To y establecer un marco para su Dans les pays où les projets ne cessent proceed without this is a disser- cumplimiento. de poser des problèmes d'exécution, il vice to the people of Bangladesh Necesidad de prestar más aten- pourrait être utile de se fixer un plus and bas negative externalities that ciôn a cuestiones de sostenibilidad. petit nombre d'objectifs, qu'il s'agisse affect the entire assistance No existe un solo enfoque adecuado des grands changements d'orientations program. para todos los sectores o proyectos, ou de l'impact sur le développement. Define Crucial Objectives, pero deberà hacerse mucho más Nécessité d'améliorer le suivi et Limit Goals, and Practice Greater hincapié en la necesidad de gastos l'évaluation des projets. Des progrès Selectivity. In countries that have ordinarios para operaciones y ont été accomplis dans ce domaine, manifested continual problems of mantenimiento, en la recuperación de mais on peut encore améliorer project implementation, it may be costos (aunque sôlo sea parcial en un l'évaluation durant l'exécution des useful to limit project objectives to principio) y en las estructuras projets. L'accent est souvent mis sur a smaller number of crucial policy institucionales u organismos de l'achèvement des travaux de génie changes and development impact ejecución y estructuras orgánicas. civil et autres réalisations mesurables, objectives. Identificación de los prestatarios alors que l'on devrait s'attacher plus Need to Improve Project con los objetivos del proyecto y su sérieusement à définir des indicateurs Monitoring and Evaluation. There grado de compromiso. Deberà spécifiques, qui soient intégrés au is further scope to evaluate devel- procurarse fortalecer el compromiso projet dès la conception. xiv Executive Summary opment throughout a pro- político con el cambio institucio- ject's implementation. Often nal necesario, y deberán bus- the focus of attention is on carse indicadores concretos que the physical completion of demuestren este compromiso. civil works or other such No se puede seguir adelante sin quantifiable targets, but estos requisitos, porque eso sería explicit outcome indicators perjudicial para el pueblo de should be included as part of the Bangladesh y tiene externalidades project design. negativas que afectan a todo el programa de asistencia. Definir objetivos cruciales, limitar metas e imponer una mayor selectividad. En los países que han manifestado problemas continuos de ejecución de los proyectos, puede ser útil limitar los objetivos a un número más pequeño de cambios cruciales de políticas y de repercusiones sobre el desarrollo. Necesidad de mejorar el seguimiento y la evaluación de los proyectos. Si bien estos aspectos han mejorado, sigue siendo necesario evaluar el desarrollo durante la ejecución de un proyecto. A menudo, la atención se centra en la conclusión física de las obras civiles o en otras metas cuantificables similares, pero deberá darse más consideración a indicadores explícitos de los resultados, que deberán contemplarse en el diseño de los proyectos. xv  ABBREVIATIONS AND ACRONYMS ADB - Asian Development Bank IEPS - Initial Executive Project Summary ADP - Annual Development Plan IFC - International Finance Corporation ARPP - Annual Review of the Project IIFC - Infrastructure Investment Facilitation Portfolio Center BOO/BOT - Build-Own-Operate/Build-Operate- IMED - Implementation Monitoring and Transfer Evaluation Division, Planning BPDB - Bangladesh Power Development Ministry Board IPC - Import Program Credit CAR - Country Assistance Review LCG - Local Consultations Group CAS - Country Assistance Strategy LDCs - Least Developed Countries CEM - Country Economic Memorandum LIB - Limited International Bidding CEO - Chief Executive Officer MIGA - Multilateral Investment Guarantee CODE - Committee on Development Agency Effectiveness MOF - Ministry of Finance CPP - Country Program Paper NGO - Nongovernmental Organization CPR - Contraceptive Prevalence Rate OED - Operations Evaluation Department CSP - Country Strategy Paper OD - Operational Directive DCA - Development Credit Agreement ORT - Oral Rehydration Therapy DFI - Development Finance Institution PBD - Planning and Budgeting Department, DSE - Dhaka Stock Exchange World Bank DWASA - Dhaka Water and Sanitation PER - Public Expenditure Review Authority PSIDF - Private Sector Infrastructure EIP - Expanded Immunization Program Development Fund ERD - Economic Relations Division PSIDP - Private Sector Infrastructure ERR - Economic Rates of Return Development Project ESF - Environmental and Social SDRs - Special Drawing Rights Framework SECAL - Sector Adjustment Loan ESW - Economic and Sector Work SOE - State-Owned Enterprise FAP - Flood Action Program SY - Staffyear FIAS - Foreign Investment Advisory Service TA - Technical Assistance GDP - Gross Domestic Product TORs - Terms of Reference GOB - Government of Bangladesh UNICEF - United Nations Children's Fund ICB - International Competitive Bidding UNDP - United Nations Development ICR - Implementation Completion Report Program IDA - International Development WHO - World Health Organization Association IDCOL - Infrastructure Development Company Ltd. Xvii  Achievements Despite Serious Constraints angladesh evokes profound, moving impressions. The country is the most densely populated state in the world. It has limited natural resources and is highly vulnerable to natural disasters. Famine is an ever-present danger: well over half the population of over 120 million have experienced extreme food shortages. Per capita income is about US$250 a year, and half the population, 80 percent of which is rural, live below a minimum poverty line. Emerging from a bloody war in 1971, the country's capacity to survive was widely doubted. But Bangladesh has survived, and has made progress in economic and social development. Since Independence, the economy has averaged most of the industrial, financial, and larger commer- about 4 percent annual growth. It now approaches cial sectors, and espoused a strong, direct public sec- self-sufficiency in rice production, and a host of social tor presence in the development process. For much of indicators (fertility rates, infant mortality, life the 1980s and 1990s, successive governments sought expectancy, primary school enrollment, and adult lit- to reverse the influence of the state in direct economic eracy) show steady improvement. Bangladesh has activities, but this was not a sustained effort: periods been a pioneer in initiatives adopted successfully by of gradual liberalization faltered in the face of opposi- other developing countries, such as village banking tion from political or vested interests. While a full dis- and oral rehydration therapy for children. course on the socio-political milieu in Bangladesh, and Can these positive achievements be sustained? its evolution over time, is beyond the scope of this Questions remain as to food security, aid dependency, Country Assistance Review (CAR), it must be recog- and the fragility of the ecosystem. nized that political, military, trade union, and bureau- The country has been undergoing a transforma- cratic issues combine to form a complex governance tion into a multiparty democratic system, encompass- framework. The interactions among these various ing also the establishment of a civil administration groups are not always transparent and have certainly structure. The history of political development in the complicated the development process and made pro- country has had a highly significant impact on the ject design and implementation more difficult. development path chosen and on economic outcomes. The World Bank has been involved in Bangladesh's The government of Sheikh Mujibur Rahman economic development since Independence. However the (1972-75), the country's first president, nationalized early years up to 1973 were colored by fundamental dif- 1 Bangladesh: Progress Through Partnership ferences regarding the country's post-Independence policy have expressed special interest. This may be most framework and the proper place of market systems. Since evident in programs in population, health care, and 1974-75 the World Bank and the Government of rural development, where donors have had a major Bangladesh (GOB) have been in development partnership, influence in shaping development policy. encompassing lending, policy advice, and donor coordi- One thing is certain: administration of the over- nation. Over the past two decades, the International all aid program in Bangladesh has been demanding. Development Association (IDA) has financed about a Implementation of the ADP has always been diffi- quarter of all foreign aid commitments to the country, cult. Twenty percent of a given year's program often covering all sectors of the economy, including import and remains unimplemented. As a result, the gap between balance of payments support. A massive amount of eco- commitments and disbursements has widened, and nomic and sector work, which has been made available to the pipeline backlog has grown. The backlog was the government and the donor community, has con- over US$6 billion by the end of 1995. There are var- tributed to policy reform and aid coordination. ious reasons for this. The already centralized bureau- cracy has become even more centralized (and hence The Development Challenge overburdened). Political involvement in the project A fundamental weakness of the Bangladesh economy administration processes makes civil servants even has been the low rate of investment and domestic more reluctant to make decisions. And Bangladeshi saving. Dependency on external concessional flows officials complain that donor conditions and admin- has been high. In the country's first 3 years, istrative requirements have become more demanding. Bangladesh received more aid than in all its 24 years A study undertaken by United Nations Development as East Pakistan; this aid reached over US$1.6 billion Program (UNDP), in a survey of 33 projects, found annually by 1993, when total aid disbursements that project implementation took, on average, 71 amounted to about US$15 per person, or about percent longer than expected. Combining time and US$90 for the average family.1 Assistance has cost overruns, this study concluded that such losses encompassed the full array of project aid, adjustment caused a 37 percent increase in total project cost.2 assistance, and food aid. Foreign assistance covered the bulk of the Gov- What Difference Did IDA Make? ernment of Bangladesh's Annual Development Plan The objective of this report is to evaluate the rele- (ADP) during the 1980s; only in the 1990s has pub- vance, efficacy, and efficiency of IDAs assistance to lic domestic resource mobilization improved suffi- Bangladesh over the past 16 years. This will encom- ciently to contribute more significantly to this ADP. pass an evaluation of lending and nonlending, includ- Given this, it is not surprising that foreign aid is itself ing aid coordination. Analysis will cover all sectors of a contentious issue. Many observers, domestic and the economy, as well as the key themes of public sec- foreign, hold that the country's use of foreign aid has tor management, institutional development, and pri- infringed on sovereignty issues or undermined self- vate sector development. Special attention will be reliance in the development process (Sobhan 1995). focused on poverty alleviation and the efforts that This so-called aid dependence is sometimes concen- have been made to reduce the number of poor living trated in particular areas or sectors in which donors below a poverty datum line. 2 I Relevance of IDA's Country Assistance Strategy DA articulated a formal assistance and development strategy in the form of Country Program Papers, Country Strategy Papers, Policy Framework Papers, and Country Assistance Strategy documents throughout the 1980s and 1990s. In the context of annual import credits up to 1986, there was a regular, short-term review of progress in the implementation of this strat- egy and of the adjustments that might be needed. This represents, with annual Country Economic Memorandums, a continuing, close overview of development progress, and of IDA's work. The need for a regular and relatively frequent review IDA's support also assisted other members of the of development strategy in Bangladesh has been a donor community to coordinate efforts that would function of intense donor interest in the country and create a critical mass of assistance in high-priority of the country's institutional weaknesses, particularly areas. in the late 1970s and early 1980s, which inhibited creation of a detailed multisector development IDA Strategy in the Early 1980s: Getting a Handle agenda consistent with macroeconomic stability and on the Country's Problems probable foreign aid flows. In the FY80 Country Program Papers, IDA explicitly A transition from a martial law regime to the recognized that government targets and expectations civilian government of President Ziaur Rahman was were unrealistic (for example, to reduce fertility to accomplished, and an elected Parliament was con- replacement levels by 1985 and eliminate food vened in March 1979. Development of an economic imports by the same year). IDA faced the delicate policy focus and investment planning process, how- task of reducing expectations while maintaining sup- ever, was tardy, and the need to placate the poor in port and enthusiasm over a medium-framework rural and urban areas, as well as the military, with- period. To a large extent this was achieved by focus- out alienating the urban middle-class and rural elite ing on key stated government policies or objectives (which provided a large measure of the president's that were uncontroversial and by clearly articulating political support) meant that many of the govern- constraints in absorptive capacity and domestic insti- ment's stated development goals were wildly unre- tutional capabilities. alistic. Fortunately, IDA's assistance strategy In the early 1980s, IDA strategy targeted four pri- encouraged the government to establish priorities. ority areas: 3 Bangladesh: Progress Through Partnership * Expand the production of food-grains so that gram Paper (CPP) in 1980, IDA recognized significant Bangladesh could secure a self-sufficient food supply. problems with the state-owned enterprises (SOEs), * Improve domestic resource mobilization so that particularly jute and textiles, and proposed projects to an increasing share of the country's ADP could help restructure inefficient public enterprises in these be financed with domestic resources. sectors. However, IDA did not then know the extent of * Persuade the government to undertake the institu- weakness throughout the public enterprise sector, nor tional and structural reforms necessary to improve was there a clear articulation of what the solution project execution in key economic sectors-with should be. This was equally true for the financial sec- both stabilization and project lending. tor; there was little appreciation of the burgeoning * Continue and strengthen the momentum of default culture and bad debt problems of the publicly improvements in the country's social indicators. owned banking system.1 During the first half of the 1980s, IDA envisaged Expanding Food-Grain Production. Severe food a continuation of program lending in the form of shortages in Bangladesh demanded a massive inflow Import Program Credits, accounting for about 25 per- of food imports, straining the country's balance of cent of the lending program, to help finance the struc- payments and reducing its resources for productive tural imbalance in the external accounts and provide activity. Periodic droughts and floods exacerbated the budgetary resources. These Import Program Credits country's severe food shortages by straining the were intended to be the vehicles to bring about general already poor distribution systems available to bring macroeconomic and structural reform. imported food and commodities to those in need. IDA Improving Project Execution. IDA explicitly rec- and the government recognized that expanding both ognized that previous technical assistance efforts to food-grain yields and the productivity of agricultural improve project execution through the use of foreign labor would be key to increasing income levels among consultants and advisers had not been entirely suc- the vast numbers of the poor in rural areas and bring- cessful. The strategy was to correct institutional weak- ing a measure of food security to the country as a nesses and lessen political involvement in all aspects of whole. And as food aid was progressively replaced civil administration, public enterprises, and banking with local production, other donor resources would be through continuing policy dialogue in conjunction freed for capital-goods imports that would ultimately with appropriately targeted ESW (Economic and Sec- improve productivity and employment opportunities tor Work). in other sectors. Sustaining Momentum on Social Progress. IDA The assistance strategy at this time proposed that would support a family planning and health care net- 40 percent of lending would be directed to the agri- work that had been launched with strong government cultural sector to improve the flow of inputs into the initiative. It planned a third population project, as well as sector. Strong emphasis was directed at increasing the credits for programs targeted at primary education and use of low-cost, minor irrigation; increasing fertilizer public school administration. The strategy recognized availability; and the use and accelerating introduction that labor-intensive export production offered good of high-yielding varieties of grain seeds. Attention was potential employment-generation. Accordingly, the ESW also to be given to rural infrastructure and to civil program and the policy dialogue sought to improve the works implemented by the Bangladesh Water Devel- trade regime, reduce the burdensome regulatory environ- opment Board in small-scale flood control and ment, and cultivate a more private sector-friendly envi- drainage. ronment. Improving Domestic Resource Mobilization. Even at this early stage in the assistance effort, the Bank rec- Assessment of the Relevance of IDA Strategy from 1980 ognized that external assistance could not finance the to 1985: A Slow Start on the Development Agenda ADP indefinitely. Tax mobilization efforts were poor, IDA's assistance strategy in 1980, while perhaps not as and there were serious weaknesses in public invest- ambitious as the government's development targets, ment programming. IDA proposed to assist the gov- was optimistic in its assessment of the development ernment through ongoing policy dialogue and sector constraints facing the country. Growth was projected work to improve this. At the time of the Country Pro- at about 6.5 percent annually during the first half of 4 Relevance of IDA's Country Assistance Strategy the 1980s, national savings would rise to about 10 ment needs and persistent balance of payments prob- percent of gross domestic product (GDP), and invest- lems was counterproductive, given the severe absorptive ment would reach about 18 percent of GDP by mid- capacity constraints and political unwillingness to decade. The reality was different. Growth averaged address major reform issues. Hence, efforts to improve about 3.5 percent annually, national savings reached project execution were to be intensified. Under the XIII only 4.5 percent of GDP, and investment rose to only Import Program Credit JPC (Credit 1655, for US$200 13 percent. No progress was made in improving the million), an array of measures was agreed with the gov- efficiency of the public enterprises, and the already ement on multiyear budget planning, project and con- poor performance of the financial system worsened. tract approval procedures, and improved monitoring The fundamental weakness with IDA assistance and evaluation. At the same time, it was realized that strategy in the early 1 980s was the belief that the gov- these IPCs were not effective instruments for bringing ernment would commit itself fully to removing the about sectoral or macroeconomic policy change. structural and institutional bottlenecks to economic Accordingly, IDA strategy shifted toward specific sec- growth. IDA did not yet fully appreciate the depths of tor-type adjustment operations, with more clearly tar- constraints and the limits of absorptive capacity that geted policy change parameters. were deep-seated in the country's administrative sys- A further shift in IDA strategy in the mid-1980s tem. The government made only tentative, dilatory increased the emphasis on energy sector projects, par- progress toward enhancing domestic resource mobi- ticularly in the oil and gas sector, to reduce the coun- lization and improving budgetary planning. Only as try's dependence on imported energy and accelerate the 1980s unfolded did the Bank gradually realize that development of the country's own gas reserves. The a core constraint in the economic development process earlier emphasis on agriculture waned, partly because was the public sector itself-one that was being sup- the already large project portfolio had delays in dis- ported with large transfers of donor funds, with weak bursing, and absorptive capacity was limited. From commitment of the political leadership to serious experience, it was recognized that seeking to restruc- reform. ture inefficient public enterprises without strong gov- While from 1982 there was large-scale denational- ement commitment was futile. As a result, no fur- ization of public enterprises, the financial state of many ther lending was envisaged to deal with these of these enterprises was precarious, with large debts problems in the industrial sector. Instead, IDA empha- accumulated during public ownership. Responsibility sized a speedier liberalization of the trade and invest- for these debt obligations was never resolved. Public ment regulation regime and proposed to support this resource mobilization efforts were dilatory, and the through sector adjustment lending and ESW designed financial results of the remaining publicly owned enter- to illustrate constraints and the actions needed to prises worsened. IDA's early assistance strategy to expand the role of the private sector in the economy. restructure SOEs in the textile and jute sector was not Population and family programs remained a prior- successful. ity, as did primary health care. However, the increas- ingly complex operations, with multiple donors, IDA Strategy in the Late 1980s and Early 1990s: Trying required supervision efforts, so the project implemen- to Adjust to Absorptive Constraints tation unit in the resident mission was to be expanded Some elements of IDA's assistance strategy were highly with bilaterally funded technical personnel. A change relevant and contributed to development progress. in education strategy put a much stronger focus on pri- The focus on increasing food-grain production was mary education and on accelerating access for females. correct, and while it took almost a decade to fully lib- By 1986 it was apparent that the financial and eralize agricultural input trade, the priority given to commercial banking sector was in distress, and urgent this task was appropriate. Continued support of pop- action was needed. ESW in the form of a comprehen- ulation and family planning programs was correct, sive sector study, together with a sector adjustment despite problems of implementation in the field. operation, was the planned approach to bring about By the mid-1980s it was becoming apparent that the necessary reform. The poor recovery performance seeking to continually expand the size of external of the Development Finance Institutions (DFIs) finally resource transfers to Bangladesh to address develop- convinced IDA that further line of credit-type projects 5 Bangladesh: Progress Through Partnership for both agriculture and industry could serve no useful and economic projections is clearly noticeable in the purpose without substantial institutional and system- strategy documents during this period. GDP growth wide reform. projections were still optimistic, as were projections of Domestic resource mobilization was still a major the improvements in national savings. But taking into problem, and IDA proposed an expanded program of account the massive flooding in 1988-89, which ser- ESW to cover public expenditure and tax reform. Con- ously affected agricultural output, these projections tinued efforts in the policy dialogue were to be directed were close to actual outcomes, suggesting an improved to reducing public sector subsidies, particularly for fer- understanding of economic dynamics in the country. tilizer and the food distribution programs. Since the The continued emphasis on population and family existing infrastructure was not being properly main- welfare programs, with greatly enhanced project activ- tained because of inadequate budget provisions, ity in this area, was relevant, and has been borne out attempts were to be made through specific elements in by results to 1996. The shift and focus on primary edu- the project lending and in the general policy dialogue cation in assistance to the education sector was also to increase allocations for operations and maintenance. relevant. Sector work undertaken in the early 1980s clearly showed that the economic returns to investment Assessment of the Relevance of IDA Strategy from 1986 in primary education were high. Moving away from to 1990: Recognizing Some Lessons but Not Others assistance to vocational training, higher education, and Some important lessons learned in the early 1980s administrative training was appropriate, given the very became manifest in a revised assistance strategy. There poor development outcomes up to 1986. was an urgent need to address sectoral and macroeco- nomic constraints, with much stronger and more IDA Assistance Strategy from 1991 to 1996: sharply focused policy conditionality than had previ- Toughening Conditionalities ously been achieved under the IPCs. Trying to deal with It is readily apparent from a review of the Country inefficient public enterprises in the absence of strong Strategy Paper (CSP) of 1992 and the Country Assis- government commitment and political will was not tance Strategy (CAS) of 1995 that IDA's assistance productive. IDA's assistance strategy during this period strategy for Bangladesh became more focused on sought to create a better environment for private sector implementation and on pursuing new operations only development and to deal with the distortions in trade, in areas where there was manifestly strong govern- pricing, credit allocation, and interest rates. ment commitment. Clear performance benchmarks But certain key lessons were not incorporated, and were established to mark progress on the development some mistakes and problems recurred. IDA continued agenda and to determine the pace and areas of new to act as if relatively minor administrative adjustments IDA commitments. Specific government actions were by the government would hasten and improve project identified, and quantifiable targets set. The CAS design and execution. The real nature of the gover- emphasized a strong and sustained improvement in nance problems within the central administration the environment for private sector development; an were not fully appreciated. The emphasis placed on enhanced program of support for population and pri- the energy sector, most notably oil and gas and indus- mary health care; a focus on overcoming the deficien- trial energy efficiency, which involved all state enter- cies in infrastructure, particularly power (but only fol- prises, was inappropriate. These governance problems lowing a demonstrated commitment to improve cost involved widespread corruption and poor account- recovery and reduce system losses); a continuation of ability within the political framework. In these areas support for primary education, with an enhanced the Bank had limited expertise. effort to improve access of females to both primary Hindsight is always helpful in evaluating a strat- and secondary education; and a continued focus on egy: the painfully slow progress made in liberalizing rehabilitation of infrastructure such as rural roads and agricultural input prices and distribution systems was flood-control and drainage systems. disappointing, given the high priority assigned to grain Adjustment lending was curtailed to about 15 per- self-sufficiency. The continued emphasis given to has- cent of the proposed lending program, and the sector tening this liberalization was appropriate and relevant. operations were undertaken in a manner reflecting Greater realism in expected development outcomes IDA's assessment of the government's willingness and 6 Relevance of IDA's Country Assistance Strategy capacity to proceed with a broad reform agenda in a approach to delays in policy reform or project execu- timely manner. Earlier IDA efforts to encourage an tion and implementation. The CAS has become much improvement in public sector resource mobilization more realistic about outcomes and expectations. The finally bore fruit in the 1990s, mainly because of a CSPs in both 1990 and 1992 underestimated GDP much improved revenue collection performance, but growth, investment to GDP ratios, and the improve- IDA recognized that fundamental reform of the civil ment in national savings. The risk assessments of the service structure was unlikely. As a result, IDA proposed strategies were much more conservative. adopted a strategy of redefining the task of the public During this period the CASs displayed a greater sector, encouraging more private sector participation appreciation of governance issues and their effects on in activities that were normally the sole preserve of the project implementation and outcomes. It was realized public sector. Through its ESW program, IDA sought that while certain reform measures might be politically to develop a comprehensive reform agenda for public difficult, there was no alternative for the government administration that included increasing the responsi- but to tackle these issues. Sector adjustment lending in bilities of other levels of government. the 1990s has not been entirely successful, precisely The 1995 CAS emphasized approaches to bring because IDA did not follow its own strategy of about speedier progress in absolute poverty reduction, addressing root causes and witnessing a demonstrated and plans to work more closely with nongovernmen- government commitment to deal with issues known to tal organizations (NGOs) in developing the design of be controversial. The financial sector provides an projects. In a number of areas, the government has obvious example. Nevertheless, it is clear that IDA has sought to involve NGOs in project implementation. In learned the lessons of the 1980s and is becoming more 1996 the Board approved the first example of such an determined in seeking clearly quantified performance approach with the Poverty Alleviation Project (Credit benchmarks before proceeding with new operations. 2922-BD), which will channel funds through an The continued emphasis on population and pri- autonomous nonprofit organization to NGOs provid- mary education is entirely relevant, and seeking a ing micro-credit to the very poor. This is a highly inno- speedier reduction in absolute poverty by using more vative project and suggests that the general approach innovative methods of project implementation is could be broadened to encompass other sectors. IDA appropriate. One criticism of the CAS could be that is also seeking to support the development of funds to IDA has not attempted to go farther in seeking alter- support private sector involvement in the power sector native implementation mechanisms that increase and other infrastructure areas. accountability and a more immediate assessment of Environmental issues started to take on greater outcome. For example, there is still scope for rural prominence in IDA's CAS from about the mid-1980s. local authorities to be more closely involved in main- Attention to the environment was strengthened in the tenance of rural roads and minor flood-control and 1990s with a more comprehensive program of ESW drainage schemes. and a more active dialogue on natural resource man- agement issues. Completion of an Environment Strat- Refining Country Strategy Relevance: Lessons egy Paper provided a direct input into the government's for the Future own National Environmental Action Plan, and IDA This assessment focuses on the relevance and appro- has sought to strengthen the capacity of local institu- priateness of the CAS for Bangladesh at the time it was tions charged with protecting the environment and put forward. Bankwide, strategies have since become monitoring environmental degradation. The 1995 CAS more sharply focused, with a far greater emphasis on advocates future investments in flood control and short- to medium-term impact. This is appropriate drainage only in the context of a strategic framework given the experiences in many IDA countries over the of national land and water planning. past two decades, and it allows a more realistic appre- ciation of outcomes and development progress. Assessment of the Relevance of IDA Strategy from 1991 Table 2.1 illustrates the OED evaluation of the rel- to 1996: A More Rational Approach evance of the Bank's assistance strategy for IDA's assistance strategy in the 1990s has become Bangladesh. The 10 criteria show that the Bangladesh more relevant because of a less compromising CAS has become more relevant and more realistic in 7 Bangladesh: Progress Through Partnership TABLE 2.1: RELEVANCE OF IDA'S COUNTRY ASSISTANCE STRATEGY Period Issue 1980-85 1986-90 1991-96 Sensitivity ito po-1inic3l vconom%. 3nd goi,ernance isus1 24 Awareness and ippre,iiaIon ot inS[itational con-irairnt' I I 3 Identification ift dc%clopnenr Iusues 3 - Identification of solurions io consrraini ' Prioritization ot asistance in accord wnh constraints : 3 Appropriate neo of instrumemt propc1d I .3 Accuracy ot sirareg reflecina experienLe 2 - 3 Realism of ot1.1comne proleclon ] I. Realism of ri;k a;se;sneni I I 3 Adequacy ol monitorable progress indictors 16 I 3 Average score 1.6 1.N 3.1 Rating system: 4 - Highly satisfactory 2 - Marginally satisfactory 3 - Fully satisfactory 1 - Deficient. Source: Bank data. its expectations. It might be argued that the more * Risk Assessments Should Be Realistic. The secure balance of payments and food supply situation validity and relevance of CASs are greatly in 1995 allowed IDA to take a less compromising enhanced by pragmatic assessment of the diffi- position on policy reform and lending commitment culties of achieving progress in certain areas. than in 1980. Given the much more precarious food This makes risk assessments more meaningful and political situation in 1980, there is validity to this and adds a needed cautionary tenor to develop- argument. However, it is clear that IDA now has a ment impact expectations. more sophisticated appreciation of institutional limi- * Experience Needs an Honest Assessment. Before tations in the public sector in Bangladesh, and is seek- defining a future CAS, more attention should be ing to counter them by more targeted interventions directed to determining the effectiveness of the and alternative project implementation approaches. last strategy. If objectives were not achieved, the Assessment of IDA's evolving strategy in causes for this need to be documented, as do the Bangladesh suggests or reinforces five key generic changes in approach that will result in their ulti- lessons for all CASs: mate achievement. * Institutional Country Assistance Strategies * Performance Benchmarks Should Be Moni- Should Be Specific. CASs invariably underesti- torable. Country strategies become much more mate the complexity and time necessary to bring relevant and operationally useful if there are about institutional change or capacity enhance- clearly defined or quantified way points in both ment. This may involve complex governance, policy reform and outcomes. These benchmarks and perhaps sociological, issues that are not need to be specific to gauge government owner- amenable to speedy change. A greater apprecia- ship and commitment, and so serve to define the tion and articulation of these governance issues most effective lending program. would both reduce unrealistic expectations and * The Lending Program Requires Flexibility. encourage the development of ways to avoid or CASs should allow considerable flexibility in lessen these institutional constraints. adjusting the lending program in response to progress in identified key reform areas. 8 Strategy Implementation: Emphasis, Instruments, and Efficiency DA commitments to Bangladesh during the period under review (from FY80 to FY96) amounted to US$5.965 billion and involved approval of 93 projects. For much of the 1980s, commitments averaged about US$10 per capita; this fell in the 1990s to about US$4.8. These levels of commitment are low compared with Bankwide averages. Disburse- ments during the period totaled US$4.65 billion. From 1980 to 1985 disbursements to Bangladesh amounted to 9.1 percent of total IDA disbursements. From 1986 to 1990 this proportion rose to 9.3 percent; however, in the 1990s in all CAS periods, it is notable that commitments in Bangladesh's share declined to 5.7 percent. This this sector accounted for 4 percent of the total. Before reflects absorptive capacity limits within the country, concluding that there is a major "disconnect" between the increasingly difficult institutional constraints that articulated strategy and implementation, some caution impeded project execution, and holdups in key sec- is necessary. This sector attracts many donors, and tors (power) or policy-based loans (Jute Sector IDA projects in this sector have involved many cofi- Adjustment), attributable to the slow pace of key nanciers and high donor participation. The catalytic reforms or tardy implementation of agreed reform role of IDA in the population and primary health care actions (see table 3.1). sectors is highly influential, and inferences drawn For much of the 1980s, adjustment lending con- from commitments alone as to the priority accorded stituted about a third of total commitments, with IPCs the sector by IDA will be misleading. every year from 1980 to 1986. The scale of adjustment lending declined in the 1990s to about 15 percent, and Emphasis will undoubtedly decline further without key reform The prominence given to agriculture in the lending pro- initiatives in certain sectors (such as finance). By over- gram was appropriate, given the stated high priority all sector allocations, the recipient of the most assis- attached to an increase in food-grain production and tance has been the agricultural sector, with 20 percent the desire to improve food-grain security and self-suffi- of total commitments, followed by the energy and oil ciency. The focus on balance of payments support lend- and gas sectors (15.3 percent), transportation (14 per- ing up to 1986 was necessary, given the large food and cent), and education (7 percent). Given the continued energy import requirements and the initially low export priority ceded to population and primary health care base. The relatively high proportion of commitments 9 Bangladesh: Progress Through Partnership TABLE 3.1: COMMITMENTS (BY SECTOR), 1980-95 Millions of US$ Percent of 1otal Sector All loans Adjusrment All loans Adjusrment Agriculture 1,I88.6 20 - Education 411.6 - Power 636.3 1.3 11 4 Industry and finance 67. -438.4 11 Multisector 1,237.0 1141.. 21 C) Oil and gas 453.u - Population, health, and nutrition 239.8 - 4 - Telecommunications 35.0- 1 - Transportation 815. 14 - Urban development -7.6 -1 - Water supply and sanitation 50.0 -1 - Other 1 3.0 - 3 - Total 5,9h5.4 1.56.- 100 30 Source: Financial database. made in the power and oil and gas sectors is notable. FY96 and, while still below the Bankwide average of There were serious shortages in power availability, and 18.8 percent, the gap had been dramatically narrowed. access to power was deficient for most of the popula- In FY97 the disbursement ratio rose to 23.7 percent tion. Rapid development of the country's natural gas overall, and 27 percent for project assistance. resources was crucial to reduce dependence on As can be seen from table 3.2, the World Bank has imported energy, and thus improve balance of payments had to devote levels of supervision effort consistently sustainability. This sector was an ideal candidate for above Bankwide averages or averages for IDA coun- private sector involvement, particularly from foreign tries. These efforts, which had to be progressively investment sources, but government policies inhibited increased from the late 1980s to deal with an ever- this. Perhaps IDA's efforts should have focused on a increasing buildup of undisbursed funds, together more intensive dialogue to improve the policy environ- with project cancellations, brought positive results, as ment surrounding the oil and gas sector, and reinforced shown by a decline in the number of projects rated as this stance with a more restricted lending program. unsatisfactory in the Annual Review of the Project The relatively high proportion of lending in the Portfolio (ARPP). In FY80, 12.5 percent of the ongo- transportation sector is appropriate, given the poor ing portfolio was given a problem rating. This climbed quality of roads, particularly rural roads, the impor- to 29.5 percent by FY86, and stayed at this level for tance of waterborne transport, and the inefficiencies the remainder of the 1980s. The proportion of prob- of the country's main ports. This sector includes the lem projects gradually declined during the 1990s, to Jamuna Bridge Project, an infrastructure project that 21.4 percent in FY96 (still above Bankwide averages). will have a profound effect on the links between large Even if the ARPP ratings were not entirely consistent, portions of the country. it is clear that there are severe capacity and institu- tional constraints in Bangladesh, and these may have Implementation gotten worse rather than b In Bangladesh, project implementation has been a per- One problem is the sheer size of the country's sistent problem, negatively affecting the progress of ADP In 1996 there were 1,100 projects in this devel- IDA's development assistance strategy. Implementation opment program, and the agency responsible for problems have undermined the effectiveness of IDA's monitoring progress, the Implementation Monitoring assistance efforts. Throughout the 1980s the disburse- and Evaluation Division (IMED) in the Planning Mi- ment ratios for Bangladesh (averaging about 11 per- istry, is overwhelmed. The government has sought to cent) were well below Bankwide and regional averages. introduce systems for prioritizing the monitoring A concerted effort made in the early 1990s has borne function, but the only solution may be more rigorous fruit. The disbursement ratio reached 17.3 percent in selection of projects for the ADP. The onerous 10 Strategy Implementation: Emphasis, Instruments, and Efficiency TABLE 3.2: PORTFOLIO RATINGS AND SUPERVISION EFFORT Portfolio ratings Supervision effort (Average, FY86-961 (Staffweeks/ppoject, Projects rated i Average # projecti less than under supervision sati%factorn looP FYS6-89 FY90-92 FY93-96 Bank%%ide 1,706 16.2 15.80 1. 1 .1.4 Bolivia 16 13.1 23.10 -.6 12. Tunisia 29 6. 10.30 11.1 15 S C6te d'hor l 13.4 14.20 16.1 26.2 Chile 14 5.2 11.30 11.8 13.5 Mala%ia 1 I8 2.0 8.40 2i.1 9. Moro. 32 10.4 12.20 14.5 1 5.8 Colomhia 28 21.0 15.90 1.3 42.x Poland 1.5 18. 1. 22.9 Bangladesh 38 23.0 26.17 29.9 32.9 a. Implementation progress rating. Source: Planning and Budgeting Department, World Bank (PBD). bureaucratic and administrative functions required at 3.3). The share of ESW in the total assistance effort all stages of the project cycle, from conception to for Bangladesh has been consistently higher, however, completion, impede implementation. These proce- compared with the South Asia Region. The sector dures have been resistant to change. work undertaken has been generally consistent with From 1980 to 1996, 63 projects were reviewed the stated GAS. An economic report has been cor- and rated by OED. Of this number, 24 projects, or 38 pleted every year of the period under review, and this percent, were rated unsatisfactory. This compares with has provided a foundation document for the annual a Bankwide OED rating of 33 percent. In the same Consultative Group meetings. Agriculture and flood period, 35 percent of rated Bangladesh projects had a control and drainage, with 11 sector reports, has led likely sustainability rating, compared with 45 percent the sector effort, followed by industry (7), energy (6), Bankwide, and 36 percent of rated Bangladesh projects and education (5). From the mid-1980s, sector work had a negligible institutional development rating, com- in the financial sector (4) and public sector manage- pared with 24 percent Bankwide. ment (4) has accelerated, which is in keeping with the In sectoral performance, the industry and finance sectors did poorly. Of 13 operations, only 3 (23 per- TABLE 3.3: INTENSITY OF ESW cent) had a satisfactory rating. The success rates for other sectors are as follows: agriculture, 79 percent; a energy, 60 percent; education and health, 66 percent; Bolivia I.h 18.3 and technical assistance, 50 percent. Tunisia 4.2 Is. The outcome performance of IDA projects is C6te d'lvoire 4.3 1S.8 Cie3.0 23.- clearly a cause for concern. Chapters 4 and 5 discuss Chi these outcomes and the efficacy of the Bank's CAS. Morocco 63 19.9 These disappointing results certainly reflect inherent Colombia 5. 21.1 difficulties in the overall policy environment in Poland 6.9 2S.8 Philippines X. 3 213.5 Bangladesh, the undefined role of the public sector, Nigeria 11.2 21.1 and its low institutional capabilities. But they also Bangladesh 1 127 21.8 reflect poor project design driven by overoptimistic Pakistan 11).4 V.9 All countries 4-6.7 22.0 outcome expectations. Note: Countries are ranked according to population size. Economic and Sector Work ESW SYs (economic and sector work staffyears) are yearly averages, based on data for FY86-96. SY data include The intensity of ESW on Bangladesh has been at about borrower country-identified staffyears only. the Bankwide average over the past decade (see table Source: PBD. for angades ha bee cosisentl hiher,howver Bangladesh: Progress Through Partnership identified problem areas and higher-priority develop- part of the GAS. The scope and range of this coordi- ment constraints. A comprehensive environmental nation effort now encompasses groupings of donors review was completed in 1991, and a detailed report that support broad sectoral investment programs on food policy in 1992. Two public expenditure under IDA leadership. This has been achieved for pop- reviews have been completed, the last in 1996, and a ulation and health, flood control, and primary educa- tax reform study was completed in 1989, again in tion. Efforts are currently under way to expand this keeping with the perceived need to improve public sec- sectoral coverage. tor resource mobilization. In the context of the 1994 Because of the large permanent donor presence in Country Economic Memorandum (CEM), a private the field, IDA, in conjunction with the government, sector assessment was undertaken. has organized a Local Gonsultations Group (with 18 The relevance, timing, and quality of this sector subgroups), which meets regularly in Dhaka to deal work have been good. The tax reform study and the with common implementation issues. The resident environmental review both fed directly into govern- mission has expanded over the past six years, taking ment studies and strategy formulation shortly after on a larger share of supervision and becoming more completion. The financial sector review, completed in deeply involved in project preparation. The mission 1987, was the foundation for the coming sector organizes regular sectoral meetings in Dhaka with adjustment operation and provided a solid basis for other resident donors to discuss policy issues, sector the policy dialogue. In recognition of the need to cre- work programs, and project implementation. ate two million new jobs every year (and a more inten- These donor coordination efforts have improved sive drive to reduce poverty at the lowest income lev- cross-dissemination of sector work and knowledge els), there has been a perceptible shift toward sector and minimized "donor competition" at the sectoral work that addresses employment, particularly in the level. Given each donor's unique administrative sys- rural areas, and issues related to micro and small tems and development assistance preferences, this has enterprises. Since poor public sector administration is been no easy task. Other large donors, such as the a key obstacle to the effectiveness of the overall devel- Asian Development Bank (ADB), have taken the lead opment effort, IDA completed a comprehensive public in reform and assistance in certain sectors (railways administration study in 1996, which in form and con- and ports) and have supported IDA in different parts tent is on the cutting edge of Bank work in this sector. of the country (rural electrification). Some donors There are anomalies, however, in the sector work with more modest assistance means and programs program. The relatively strong emphasis given to the maintain that their own assistance efforts should be industrial sector in the ESW program is misplaced. more explicitly and readily identifiable to those to The work intensity directed toward this sector up to whom they are accountable (for example, parliaments the mid-1980s has not produced commensurate and boards). IDA has been sensitive to this issue, rewards in fundamental reform of SOEs or in a greatly which on occasion bears on project design. enhanced share of industrial output. Another anomaly has been the limited amount of sector work in popu- Participation, Collaboration, and Awareness lation and primary health care. Merely counting sector Traditionally IDA and the World Bank have interacted reports (2) can be misleading, because a considerable mostly with government departments and interested amount of analysis is undertaken in the context of donors, and as a result have often been perceived as project preparation. There is scope for more sector distant by observers and other agents in development. work in population and primary health care. This has often led to misunderstandings counterpro- ductive to the reform process or to achieving the goals Aid Coordination set out in IDA's GAS. IDA has sought to remedy this Over 50 official donors provide assistance to Bangla- situation in Bangladesh. The country has a large, desh, and about 26 participate as members or active, and vocal NGO community, and IDA has made observers of the Aid Consultative Group, which meets efforts to engage these NGOs in the development every year in Paris. This group is chaired by IDA, and, process and in assisting IDA in the design and given that official aid covers about 80 percent of the appraisal of projects. The resident mission has been ADP, the donor coordination function of IDA is a vital active in outreach to the NGO community and has 12 Strategy Implementation: Emphasis, Instruments, and Efficiency sponsored a study on strengthening the relationship of The Efficiency of IDA's Assistance Program NGOs with the government and enhancing their con- Table 3.4 provides an indication of the overall effi- tribution to the development process. In rural devel- ciency of IDAs assistance program for Bangladesh. The opment, primary health care, family planning, and primary measure is staffyears (SYs) spent for each education, IDA has used local NGOs working at the approved project. Bangladesh does not compare favor- village level to obtain beneficiary inputs into project ably with the outcomes for the Bank as a whole. Even design and effectiveness. NGOs are becoming part of when we look at lending SYs by approved project, project implementation, as shown by the recently Bangladesh does not meet Bankwide norms, and is approved Poverty Alleviation Microfinance Project almost 40 percent higher in staff time spent. From the (August 1996). mid-1980s, lending completion staff time has been con- Led by the resident mission, an external relations sistently higher than Bankwide averages, as well as strategy is being developed to bring about greater averages for the South Asia Region. Adjustment lend- awareness of IDA's role in the country and IDA's CAS. ing cannot explain this difference. For example, in This is to be targeted at all segments of society, includ- FY93, for multisector lending, Bangladesh lending ing the illiterate poor in rural and urban areas. Semi- completion SYs were over twice the South Asia Region nars and workshops have been undertaken with mem- average, and 4.3 times the Bankwide average. bers of the local press and the academic community, This less efficient performance indicates the diffi- and attempts have been made to make IDA documents culties inherent in dealing with Bangladesh's assistance more accessible (summarized versions of project activ- program and reflects the weak institutional capabili- ity, some translated into Bangla). This positive devel- ties in the country. Project supervision is more opment seems to be bearing fruit. During the field mis- demanding, as seen in the high average supervision sion for this CAR, many commentators remarked on a effort shown in table 3.2. Table 3.5 shows the diffi- greater openness on the part of IDA about its activities culties in implementing the assistance strategy. in the country. They also noted the greater ease of Bangladesh requires over 50 percent more calendar access to documentary information about these activ- time than the Bankwide average to take a project from ities. Such components of implementing IDA's CAS concept to Board approval. Lending completion SYs could enhance the quality of policy dialogue with the were much better in the early 1980s. Various reasons government. may account for this. First, the whole ADP was less TABLE 3.4: COMPARATIVE DATA ON THE EFFICIENCY OF BANK ASSISTANCE PROGRAMS priar meaur is) staffyear (S) spntfo ec ConapprovedsSY pojc project Banladsh oe no opr faor eenwdilgo a No. of Commit- Counin~~anlaes doe nota mee Bakwd norms an isicimn 1/3)101 ()11o Bol~ i %. 1.1 21 ..s - ,_h 20.3 Tunsist enl h t1 a verage as wla Care d hoire 24f.or of. 1 t . i . A Chile 1.39.3 2 ~ 25B5.22. S 13 1.52l. .2 Mort cC 351).0 15..'-6 4A9,'9 -.6 3 40 Colombia 1~3 1I.4 34 3 s 8.3 3. 6 Poland 262.o 11.1 1 23 h.is df 1f1.4 4. Fo 6.5 P-.i iIi p pi n es 388.9 16 45 ;:s 8.63.-2 Niazeria 584.6 191.1.2 1 1I 4.4-1 14.3 4.6 1. BangFadeYh 638.2 226.3 i 46 394 fo 13.9 4.9 16.2 Pakiscan m638p6 l3.9 Y3 wre 1 ot.1 t. 9.4 average, an 9.3 me th1akid0vrg.4 All coliniri e 23801.- ,,611.3 2, i50 2>' 1 1[34 I 1. Note: Countries are ranked according to population size. ESW SYs (economic and sector work staffyears) are yearly aver- ages, based on data for FY86-96. SY data include borrower country-identified staffyears only. Commitments are in US$ mil- lions. Lending SYs denote project development from identification to Board presentation; the other two main categories in the total SYs are supervision and ESW Column (5) is total SYs per project; column (6) is lending SYs per project. Source: PBD, financial database. 13 Bangladesh: Progress Through Partnership TABLE 3.5: AVERAGE ELAPSED TIME IN MONTHS FOR FY90-96 APPROVALS (IEPS TO BOARD PRESENTATION) Number of A%crage. Average. A%erage, Countr approak IEPS-appraisal appraisal-board IEPS-board Bol ii3 25 16i 9.6 2b.1 Tunisij 19 2.'6.5 C6[e Ji oire 23 2. I 12.5 411.5 Chile 15 4.9 .5 MaL [a 1. 4.Q 12.0 Mo ro-o 2- 1. 10. i Colombia 21 1w. 14.3 33.9 Poland 2.3 1-1 12.8 2m. Philippine; 31 14.- 9.6 2-4.3 N'gurwa 21 1 Q. 14.j)33 X Bangladeh 26 29.8 13.8 -13.6 Paki;an 36 18.1 13.5 31.6 Total -76 2(j 13i. lh-.43m Note: IEPS = Initial executive project summary. Source: PBD. onerous for the government, and the institutional or longer. This reflects the administrative logjams in capacities were not so stretched. Second, in program the country, perhaps a less than full commitment to lending (the IPCs), staff time was relatively low, the project, and deficient project preparation and reflecting the weak policy conditionality attached to design. This also suggests an operational remedy for these credits. Third, the quality of lending, particularly the World Bank-that effectiveness conditions should in the industrial and financial sectors, was not good. he kept minimal, and most of the actions required As part of this CAR, a review was undertaken of should be taken before Board presentation. This may 83 completed projects (some of which were approved lengthen the time between negotiations and Board pre- before the start of the CAR review period) to deter- sentation, but it would create greater urgency to com- mine the time taken between Board approval and plete requested actions, given the "aging" of the Staff effectiveness of the credit. Of the 83 projects, 59 (over Appraisal Report or the President's Report and the rel- 70 percent) took longer than three months to be evance of the indicators and information shown in declared effective, 45 (54 percent) took six months or those documents. longer, and 24 projects (29 percent) took nine months 14 N1 Development Outcomes t_ I he growth performance of Bangladesh, with the exception of flood or drought- related years, has been in the range of 3.5 to 4.5 percent a year. Investment, both public and private, has been deficient, especially when compared with countries in the South Asia Region, as well as countries with similar per capita incomes. This investment performance has shown some progress in the 1990s, with a consequent modest improvement in the trend GDP growth rate. Viewed over a long trend period, two indicators suggest that profound changes have occurred in the framework of tized savings. High real administered interest rates on the Bangladesh economy. First has been the remark- deposits have been a positive inducement. An increas- able decline in the relative share of agriculture in GDP ing number of people can now meet basic survival from over 40 percent in the early 1980s to 25 percent needs for food and shelter resulting in a rising mar- by the late 1990s. While the general trend is consistent ginal propensity to save as incomes rise. Workers with the growth of low-income countries (as industry remittances from abroad have risen steeply in the lat- and services emerge from a low initial base), the mag- ter part of the 1980s and 1990s, adding to the pool of nitude of the change is puzzling (see figure 4.1). There private savings. The public sector has improved its appear to be serious anomalies in the agricultural and savings performance in the past five years. An average food-grain output data in the 1990s, inconsistent with annual current surplus of about 1.1 percent of GDP in fertilizer use and small-scale irrigation prevalence. As the latter part of the 1980s evolved into a surplus a result, the perceived relative decline in the impor- equivalent to 3.3 percent of GDP by FY95. This was tance of agriculture must be viewed with some caution achieved entirely on the revenue side, with an (table 4.1 displays performance data). improved tax collection performance and introduction A much more encouraging and positive trend over of a value-added tax system. the past 15 years has been the steady and strong The balance of payments has always been a weak- growth in domestic savings performance, notably pri- ness of the Bangladesh economy. Throughout much of vate savings. Several factors have contributed to this. the 1980s, reserves were at precarious levels, given the The increasing monetarization of the economy, country's susceptibility to the shocks caused by nat- together with a broad extension of banking services ural disasters. Very high current account deficits aver- into rural areas, has expanded the capacity for mone- aging over 9.0 percent of GDP in the early 1980s were 15 Bangladesh: Progress Through Partnership FIGURE 4.1: ECONOMIC GROWTH COMPARISONS earning opportunities. Food imports alone accounted Real GDP Growth (%)for about 20 percent of all imports during much of the 14 -1980s. Only with increasing liberalization in input supplies to agriculture in the late 1980s could food- 1980-90 grain output grow sufficiently to significantly reduce 12 - [] 1991-95 food-grain imports. 1996 The severe flooding in 1988 brought this problem 10 -into much sharper focus, and a comprehensive stabi- lization and liberalization initiative supported by the 8 IMF, IDA, and other donors was launched. From 1989 -7 to 1994 there was a sustained improvement in overall amacroeconomic management, and a sustained period Sof structural adjustment and liberalization. The bal- ance of payments performance improved dramatically, 4 foreign reserves reached historically high levels, infla- tion was at an historic low, and there was a strong 2 improvement in national and domestic savings. Growth in exports has been strong in the 1990s, 0 a led by ready-made garments and other nontraditional China India Pakistan products. Export values in 1995 were equivalent to over 17 percent of GDP, rising from just over 8 percent Soury Ordieveomnsndctr.97,WrdBn of GDP in 1988 (see figure 4.2). In part, this has been the result of a decline in the real exchange rate in the mid-1980s, which continued through the 1990s. The only sustainable through the flow of concessional for- progressive liberalization in the trade and investment eign assistance. The causes for this persistent external regimes, which, again, accelerated in the early 1990s, imbalance were inherently structural; little was done has been significant. For example, before 1990, up to about 1984 to change relative prices and reduce unweighted average nominal tariff protection was the distortions affecting the development of export- 88.6 percent. By 1996 this had declined to 24.6 per- TABLE 4.1: MACROECONOMIC PERFORMANCE 1981-85 IAa 6note 1991-95 GDP gro-virh ipercentl 1 3.81) 3.9 14.4I GDP comp19thri9tn i parn i Agricumrro 42.mne t a au do1.4 Industro 13.40 1b z .T Services 44.30 4 .toi h le i Consumption t 98.7o w a ht ltr Private 91.00S. I ~ Public Thh)1.9 I141.11 Investment 13.2i v 1 . 1av.6 Private -.1')0 -.9 Public 6.2') o26.- Gross dornes ic ;i ing-1.0 . Gross naiinnel sr hings 4.3been s 1 9 Current accounb raemnca ot pan rnonthe n9o20ntadti Exports rod13u i Imports whIh 19.9 Foreign relierr alizato ni ithh ott r anpod r n s t Consumer pricew aigaiunl. perioe alreragt 11.2e 9r 3.4 Source: World Bank data. 16 Development Outcomes FIGURE 4.2: GDP PER CAPITA (1980-96) US$ 250 200 - 150 100 50 0 I l l l l l l l l l lllll| 1980 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 Note: Data for 1996 are estimated. Source: World Development Indicators 1997. cent. The growth in ready-made garment exports from While Bangladesh has gone through periods of almost zero in 1982 to over US$2 billion by 1995 is privatization, government and policymakers have indicative of what can be expected of Bangladesh in been slow to fully accept that the private sector is cru- labor-intensive, light manufacturing activities. To cial to the development of the country's infrastructure date, this potential has not been fully realized, and sig- or primary resources base (such as natural gas). nificant investments in consumer electronics, con- Despite official announcements, limited or no progress sumer durables, and other consumer products have has been made in private sector investment in power, not occurred as in other parts of the South Asia energy, telecommunications, and transportation. At Region and China. This reflects the country's infra- the time of Independence in 1971, the development structure constraints (particularly power) and still path chosen borrowed much from the path selected onerous bureaucratic and administrative require- earlier by China. Ironically, in the 1980s and 1990s, ments, as well as the political uncertainties perceived China pioneered foreign private investment in power by foreign investors (see figure 4.3). and other sectors, and encouraged foreign private One area of macroeconomic stabilization and inflows into all aspects of industry and manufacturing. adjustment that requires further attention is fiscal per- Bangladesh needs to follow this example. formance. Progress has been made in improving central government savings performance, but the overall bud- Poverty Alleviation and Human Development get deficit remains high (6.8 percent of GDP in 1995). While GDP growth performance during the 1980s and A major problem during the 1980s and early 1990s has 1990s has not been spectacular, good progress has been the poor performance of the SOEs. By 1991 the been made in improving the general well-being of the total losses of the loss-making firms reached 2.1 percent population. With aggressive immunization programs, of GDP, and while progress has been made in reducing particularly from the mid-1980s, infant mortality has this, the state enterprise sector, including public utilities, declined and life expectancy has improved. Population continues to drain the economy (see table 4.2). programs, again from the mid-198s, with strong gov- 17 Bangladesh: Progress Through Partnership FIGURE 4.3: EXPORTS (US$ MILLIONS), 1982-95 $ Value Compound annual Share Share Share growth 13.2% 3,000 - 25.6% 2,500 -100% 2,000 - Total exports ' 00 48.23% 1,500- 100% # 71.7% 1,000 - 100% 53.9 w 53.4% Nontraditional exports 00 Re ayMmad 500 - ME:=C1 Ready-made 23.7% 35.5% garments 0 I I I I I I 1982-83 83-84 84-85 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 Traditional exports: Raw jute, jute manufactures, jute specialty products, tea, and leather products. Nontraditional exports (major groups): ready-made garments (RMG), fish products, urea, naptha. Source: World Bank Operation Information Systems database, internal documents. ernment commitment, have reduced fertility and the country. External assistance has been accompanied curbed population growth rates, although most people by (or has facilitated) rising total government expen- still live a precarious existence. But as seen in table ditures on health and family planning. 4.3, the proportion of the population with a daily food Family planning services are now widely avail- intake below 1,805 calories has declined. able. Field workers, both government and nongovern- ment, are the main agents. The contribution of physi- Family Planning, Health, and Nutrition cians, private clinics, and traditional practitioners By supporting expansion of health infrastructure and remains relatively small. There has been a rapid trained staff, as well as policy and institutional increase in contraceptive prevalence under socioeco- improvements, IDA operations have facilitated signif- nomic conditions not normally conducive to accep- icant positive outcomes in family planning and health. tance. Fertility has declined rapidly over the past two Over the past 20 years, Bangladesh has built an exten- decades. Many people express willingness to have sive health infrastructure; the country has 3,775 fam- small families and to delay first childbirth. The con- ily welfare clinics or health subcenters (rural dispen- traceptive prevalence rate (CPR) has reached 45 per- saries) at the union level and 349 Thana (subdistrict) cent and knowledge of contraceptive methods is wide- health complexes. These are at least nominally well- spread, although varying by group. The CPR has risen staffed with medical and health personnel. A network from 8 percent in 1975 to 45 percent in 1993, total of 5,000 salaried family welfare assistants and health fertility has declined from around 7 to 3.3, and the assistants carries outreach services to the villages, with rate of population growth has edged downward from domiciliary visits and at monthly fixed sites. In addi- over 3 percent to around 2.2 percent (WHO 1996). tion, NGOs run health facilities in different parts of Infant mortality fell from over 140 in 1971 to 88 18 Development Outcomes TABLE 4.2: FISCAL PERFORMANCE (PERCENT OF GDP) 1986-90 1991 1992 1993 1994 1995 Over.ill budCt dcticit -'.4 -.2 -5.9 -5.. -6.0 -h.S Re%c nue 8.8 9.M 11.l) 1.1.. 12.1 Expenditure 16.3 1h.. 10.8 18.1 1s.C Currenr 1.9 <. , 8.3 . .9 .8 Annual de%elopmenr proirm . b.2 6.3 -.1 8.4 8.6 Gro4s 1-,e ol publL enlerprles 1.1 2.1 1.9 1.1 0.8 n.j. Ne: lo;ses 01 public enicrpri;es ).8 1.s 1.6 Q.- ..6 n.J. Foreign tinhncin. 1 7.. 4.9 4.9 4 n.a. Not available. Source: World Bank data. by 1994, primarily as a result of the government's lack of integration with the health services, interrup- expanded immunization program (EIP). Between 1985 tions of contraceptive supplies, insufficient travel and and 1994 the proportion of children immunized daily allowances for field staff, and needless con- against six major childhood diseases increased from 2 straints on the operations of NGOs. The quality of to 62 percent. The EIP, with strong support from the health care is perceived to be poor, drugs are often in United Nations Children's Fund (UNICEF) and the short supply, and medical staff are thought to be indif- World Health Organization (WHO), has relied on a ferent to patient concerns. A 1994 study of 40 sub- large-scale program of social mobilization involving centers found that patients were adequately examined government workers, NGOs, and the media. The in only 37 percent of cases: the average consultation rapid spread of oral rehydration therapy (ORT) to lasted less than a minute (Zahan 1994). treat diarrhea, developed by the International Center for Diarrhea Disease Research, Bangladesh, has also Education contributed to the decline in child deaths. In addition, The increased targeting of vulnerable, disadvantaged the incidences of malaria, cholera, and typhoid have groups and primary education from the mid-1980s been reduced, and smallpox has been eradicated. has had positive results. After primary school enroll- There has been a modest decline in maternal mortality ment had stagnated for almost 30 years, gross enroll- over the past five years, probably reflecting better fam- ment rose from 55 percent in 1985 to 63 percent in ily planning services and wider use of tetanus toxoid. 1990. Attendance of girls, particularly in the rural The family planning and maternal and health ser- areas, has increased, and female enrollment in sec- vices still have serious weaknesses. Internal efficiency ondary schools grew from 24,068 in 1992 to 44,569 of the system is low because of management, staffing, in 1994. The enrollment rate for girls in secondary and logistical problems. Some facilities are underused; schools increased from 28 percent in 1980 to 42 per- others are overcrowded. Contraceptive use is increas- cent in 1993, and is now about the same as for boys. ingly skewed toward more expensive, less efficient Literacy and nonformal education programs cover temporary methods. Family planning is hampered by about a million students in all age groups, and 75 per- TABLE 4.3: SOCIAL INDICATORS ______________________________ 119191 19S6 1990 1944 (per lsitr i ts o t e op r biration o N G. IOI s Th 94.e q Lihea cirr.i i4.8 pecee 56o be p1 Ferrfeir t pr t trn.in e2 2.9 2u.4 Percent popnl.iliyn 3pcith ofii% cc eolorco inikA- Motn 2A 1 2 3.2 S.41 .i1 Percent p0pUlation %% i[h diil% Edlori intake bqJu%% 1,S11i 50.1 21 ~ I28.A fl i.. n.a. Not available. Source: World Bank data. 19 Bangladesh: Progress Through Partnership cent of these services are delivered by NGOs. riculum lacks relevance, textbooks are outdated, Government commitment to education has teachers are poorly trained and supervised, logistical increased. Education outlays have risen from 7.9 per- support is weak, and buildings are in poor condition. cent of the budget in 1980 to 10.3 percent in 1990 Instruction time is 2.5 hours each day in grades 1 and (education represented 9.6 percent of the budget in 2, and 4 hours each day in grades 3 through 5, the 1997). Infrastructure and teaching staff have lowest in South Asia. Actual instruction time is much expanded. There are now about 45,000 primary less, according to observers in the field. Access to edu- schools and 11,000 secondary schools in Bangladesh. cation remains inequitable, especially for the rural According to government statistics, approximately landless, urban poor, and girls. This is partly because 190,000 primary-school teachers serve 12.3 million the real cost of "free" education to consumers is high. students. The increase in school capacity and number The time costs of having children at school may be of teachers has made it possible to accommodate a considerable, and the direct private costs of education rapid increase in enrollment and gross enrollment are high. rates. From 1980 to 1990 the primary-school-age pop- ulation increased by 19 percent, and school enroll- Industry and Finance ment increased by 50 percent. Because of changes in government industrial sector Despite the achievements in general education, policy in 1982, a foundation was established to sup- access to primary education remains a problem for the port growth in the manufacturing sector (particularly poorest groups, and the quality of education delivered the nontraditional sectors) and the growth and diver- is less than satisfactory. Even now, the educational sification of exports. Performance of the sector during attainment of women in Bangladesh remains one of the 1980s was disappointing. Real growth averaged the lowest in the world. According to the 1991 Cen- only 2.6 percent a year between 1981 and 1990, sus, the literacy rate for females over the age of five below GDP growth of 3.2 percent annually. New pri- was 20 percent, compared with 35 percent for males vate investment was stagnant throughout much of the (see table 4.4). The 1991 Education Expenditure period, and the SOEs continued to perform poorly. Review reported the completion rate for the five-year Growth was affected by floods (1988), the sharp fall primary education program at 43 percent. in jute prices (1986), and political and labor distur- Internal efficiency of the school system is low, bances throughout the period. However, other policy reflecting high dropout and repetition rates. It takes an constraints combined to thwart the growth of the pri- average of 8.7 years of teaching to produce a single vate manufacturing sector. These included: graduate of the five-year primary cycle. This ineffi- ciency reflects the low number of classroom hours, * The high cost of capital and access to credit for poor quality and absenteeism of teachers, and lack of small manufacturing enterprises. system accountability. Retention rates in both formal * The government's policy of supporting sick private and government nonformal primary education com- and public industrial enterprises has weakened the pare unfavorably with programs sponsored by NGOs. banking system, and lack of exit for insolvent firms The quality of education is still deficient; the cur- has weakened stronger competitors. TABLE 4.4: EDUCATION COMPARATORS ___________________ IGross enropbi4wt ratc Aduu ipioeracr Priiware Scondar Trtiar i BaIrgnidesh i 19 4 2,d13 r 44 Nepal S4 owest 3) 6 Paicii65 46A 21 Sri Ln t onb 12 t fs Loci rincone i itounarbe,, 4forth 41 3 World l u33 r o n l it52 1i Source: World Bank data. 20 Development Outcomes * Despite some trade policy reform, throughout loan classification norms well below accepted interna- the 1980s the manufacturing sector was highly tional prudential standards. Financial deepening protected and oriented toward the domestic (Money/GDP) occurred throughout the 1980s, with market. This inward focus resulted in poor qual- the M2 ratio climbing from 19.6 percent in 1980 to ity and reliance on a small domestic market. about 28 percent in 1989-still low by the standards * Despite some denationalization, the high level of in neighboring countries. continuing public sector ownership in several large manufacturing subsectors (such as textiles Energy and Infrastructure and jute) has inhibited investment and techno- Bangladesh is one of the lowest per capita energy users logical progress. in the world, despite a threefold increase in per capita electricity consumption over the past 15 years. Despite the disappointing overall performance of Between 1980 and 1993 the installed generating the industrial sector, some promising structural capacity tripled, from 822 to 2,608 megawatts, or changes occured, most notably in the nontraditional about three times the rate of GDP growth. However, export industries of ready-made garments, frozen fish massive investments are needed to keep pace with pre- and shrimp, leather products, and natural gas-based sent demand growth. At present, only about 14 per- fertilizers. These sectors registered an annual average cent of the population is served with electricity. Esti- growth rate of 11 percent from 1982 to 1990 (see mates suggest that generating capacity needs to be table 4.5). Throughout the decade, good progress was increased by almost 1,200 megawatts by the turn of made in the diversification of exports from a declining the century. The provision of adequate electricity is world jute market, the country's main export product. hampered by having over 25 percent of capacity out of Ready-made garments, for example, grew from almost service, awaiting rehabilitation, and having one of the zero in 1982 to over US$700 million by the end of the highest numbers of employees per consumer in the decade. world. There is frequent loadshedding, and a massive Bangladesh still has a small and relatively underde- system loss of over 35 percent, caused mainly by ille- veloped financial system. The range and diversity of gal connections or corrupt billing practices. institutions is narrow and the financial products avail- Access to basic infrastructure, such as power sup- able are limited. The sector contributed 2.1 percent of plies, is unavailable to nearly 80 percent of the popula- GDP in 1989, which declined to 1.8 percent in 1992. tion. Only about 50 percent of the population in urban Development of the capital market has been slow, and areas has access to safe water, and telecommunication institutional and regulatory frameworks are weak. penetration (at 2 lines per 10,000 people) is still among Domestic savings rates are low compared with the coun- the lowest in the world (see table 4.6). Over the past try's neighbors (Bangladesh, 7 percent; Pakistan, 17 per- two decades the total road network increased by over cent; Sri Lanka, 19 percent; and India, 21 percent). 200 percent, to nearly 14,000 kilometers. The financial sector has been dominated by the This development tended to focus more on government since nationalization of the commercial national and regional highways than on rural feeder banks in 1971. The quality of financial intermediation roads connecting rural markets. The rail network has has been poor, and loan recovery performances declined in importance as a mover of freight and pas- steadily deteriorated throughout the 1980s and early sengers. This is a result of the expansion in road traf- 1990s. By 1985 nearly 30 percent.of the NCB's loan fic, but also a function of poor management. In the portfolio was impaired by overdue repayments 1980s, rail freight traffic declined by 20 percent. (Watanagase 1996). This percentage was based on Agriculture, Food Security, and Flood Control TABLE 4.5: MANUFACTURING PERFORMANCE Agricultural growth (from 1980 to 1996) averaged (PERCENT) about 2.1 percent a year, or the same pace as popula- :FY73-80o F N sI-90 FY91 F')92 FY93 tion growth. This is curious, given the large increases Growth race -.8 2 0 2.-4 ) in fertilizer use, introduction of high-yielding seed Share of GDP 12.10 111.1 9.8 1li 1 10.-1 varieties, and more widespread irrigation.1 Other data Source: World Bank data. suggest that agricultural production has grown faster 21 Bangladesh: Progress Through Partnership TABLE 4.6: COMPARATIVE OVERVIEW OF INFRASTRUCTURE PROVISION Indicator Bangladesh India Pakistan Sri Lanka Philippines Thailand I Mala5 sia Percent of households with electricity 12 54 .3,131 33 46.0 -1 64.0 System losses as percent of output 35 1 24 Is 19.0 11 16.0 Population with access to safe drinking water (%) (nationwide) 8 "i 81.0 -- . Telephone density, lines/1,000 2 16 12 10.3 31 111.3 Paved road density, km/million persons 59 CP3 29 16 242.0 i13 n. Percent paved roads in good condition 15 IS I' 11.0 ill n.a. Note: Data generally pertain to the years 1990-94. Source: 1994 World Development Report, A Survey of Asia's Energy Prices, and International Telecommunications Union (ITU) reports. than suggested by the GDP calculation. For example, has benefited from the so-called Green Revolution, from 1990 to 1995, fertilizer use increased by 36 per- but more modestly than other countries. This is par- cent, minor irrigation (as measured by shallow tube- tially the result of the inefficient distribution of inputs wells and low-lift pumps in operation) increased by (fertilizer and irrigation), but with the introduction of 101 percent; yet the stated food-grain production low-lift pumps and shallow tube-wells, this situation increased from 18.7 million tons to 19.1 million tons started to change. By the end of the 1980s, with cor- (2 percent) (Narbye 1989). Data from the Household plete liberalization of the market for irrigation equip- Expenditure Survey conducted in 1991-92 show that ment and fertilizer which increased availability and per capita intake of calories from cereals increased by lowered prices, the sector enjoyed periods of robust 8 percent from 1984, and that the share of the food growth despite severe flooding in 1988 and 1989. budget going to cereals declined from 58 to 50 per- One cannot discuss agriculture in Bangladesh with- cent. From some of these proxy data, unofficial esti- out discussing flood control and drainage. During the mates suggest that agricultural growth has been aver- annual cycle of flooding, the three main rivers engulf aging 1 to 2 percent higher than population growth. It almost two-thirds of the agricultural land, not always in is clear that by the early 1990s, considerable progress a completely predictable manner. This annual flooding had been made toward achieving rice self-sufficiency. and retreating of the rivers (barsha) is a central fact of Rice self-sufficiency has been attained in years of rural life and of the agricultural system, providing water normal weather. However, without export markets, and soil nutrients and recharging the aquifer. The gov- domestic rice prices fall sharply each time self-suffi- ement and donors have sought to make this flooding ciency is reached, undermining incentives for further less lethal, and at the same time improve agricultural growth. Attention must be given to the development productivity. Progress has been mixed. of rice exports as well as crop diversification into Early emphasis was on the construction of mas- higher-value items. Although crop diversification is sive systems of barrages, embankments, and canals demand-driven, alternative crop possibilities should with support from the donor community. However, still be explored. This will require efforts in extension technical problems soon became manifest. Contro- and technology transfer, using governmental and non- versy as to how to "manage" this flood water persists. governmental agencies. For example, silting raises the river bed, necessitating In 1980 input supplies and prices were controlled higher embankments; if these are breached, the lethal- by the government, and output prices, particularly for ity of the flooding is extended. There is the problem of food-grains, were controlled in part by procurement drainage. Common structures are "polders," which policies and food distribution programs. The country use embankents to completely seal-off a compart- 22 Development Outcomes ment of land from flooding. However, heavy monsoon coordinated by the World Bank. Progress toward con- rains leave excess water in the polder that needs to be sensus has been slow. drained, requiring sluice gates (which require perpet- ual maintenance or the land becomes waterlogged). To Sum Up And the silting problem persists outside the polder, The 1980s and 1990s saw progress in economic and further aggravating the drainage problem. social development, albeit not at the pace one might Many of these externalities with regard to flood prefer. Reform has accelerated from the late 1980s, control programs were known in the early 1980s. and this has resulted in an improved investment and However, a consensus view was never achieved and commercial environment. But in contrast to other large-scale civil works proceeded. But the massive countries in the region and in Asia, Bangladesh has floods in 1988, with the loss of over 2,000 lives, been tardy in limiting the role of central government brought the issue into much sharper focus. A host of in all aspects of economic development and focusing conflicting proposals were put forward, ranging from on creating a more dynamic investment environment. massive embankments to more modest riverbank ero- Employment creation remains a central challenge. sion protection. Following a G-7 meeting in London in Although there is no shortage of entrepreneurship or 1989, a Flood Action Program (FAP) was designed to individual ingenuity in the country, government, both bring together a series of studies and activities that at the local and the central level, has yet to define the would eventually culminate in a coherent set of pro- proper balance of support and intervention and create posals. The bulk of FAP products are studies, presided the enabling environment that will reduce transaction over by the government's Flood Plan Coordination costs and allow entrepreneurship to flourish. Organization, which is advised by a panel of experts 23 u Efficacy of IDAs Assistance Strategy: Overall Assessment ecause it is one of the poorest countries in the world, any assessment of development progress and achievement in Bangladesh will be colored by the desire to reduce absolute poverty and human hardship as speedily as possible. As a result, expectations will always risk being overoptimistic and outstripping what realistically might be expected. Complex political, social, and economic factors all constrain the development process, and even when identified, are not always easy to overcome. This produces impatience and frustration for all those involved in the country's development and can lead to hasty gen- At the start of the 1980s, IDNs task was to convince eralizations and impracticable suggestions. policymakers that while the public sector was important In assessing the effectiveness of a single agency's par- in the development process, the main dynamic engines for ticipation in this process, no matter how large, it is vital to economic development were markets and private initia- remember that other agencies are involved. Establishing a tive. This dialogue continues, and IDA can claim only par- counterfactual scenario with no IDA involvement is not tial success in this endeavor. Progress has been made, most technically possible. And measuring IDA's performance notably from the nid-1980s, in liberalizing the trade without assessing the performance of others involved in regime, regulating investment, distributing agricultural the development effort may ascribe to IDA a capability inputs, and gaining access to foreign exchange. But and independence of action that it does not enjoy. progress has been tortuously slow, and successive govern- Over the past decade and a half there has been ments have never fully endorsed the notion that the pri- progress in both economic and social development. While vate sector needs to be encouraged rather than con- economic performance has not been remarkable, the strained. In many respects, authority and decisionmaking macroeconomic situation in 1996 was more robust than are more centralized today than a decade ago. Deficient in 1980, and the more rapid growth performance of the service delivery in rural clinics and hospitals, inadequate 1990s suggests what can be achieved with even more or completely absent maintenance of critical infrastructure determined efforts to overcome constraints and create an such as flood controls and drainage systems, massive Sys- optimal policy environment. Many social indicators show tem losses in power generation and distribution, and steady improvement. This should encourage a redoubling highly inefficient ports are but a few examples of the cor- of efforts to bring about further progress. IDA has con- responding diminution of accountability (see box 5.1). tributed much to this progress and has strongly supported In the context of Macroeconomic Stabilization and important changes in the Bangladesh economy. Adjustment, IDA has taken a secondary role to the IMF, 25 Bangladesh: Progress Through Partnership although a prodigious amount of good-quality ESW has tax reform in the late 1980s guided the GOB to imple- contributed to the overall policy dialogue. IDA has taken ment reform programs in the early 1990s. An Environ- a strong stance on public domestic resource mobilization. mental Review fed directly into the development of the Progress on this throughout the 1980s was slow. But with government's own environmental action plan, as did the appropriate sector work (Public Expenditure Reviews (1996) Public Expenditure Review. and a tax reform study), good progress has been made in There were anomalies in the ESW work program, the 1990s, and macroeconomic management has however. The relatively strong emphasis given to work in improved (see box 5.2). The government is now con- the industrial sector was misplaced; this has not resulted tributing to the ADP from domestic fiscal resources, but in a commensurate reward such as fundamental reform more needs to be done. In this context, it can be said that of SOEs or a greatly enhanced share of industrial output. IDA has contributed to reducing the country's aid depen- A further anomaly has been the relatively small amount dency. The IPCs, which had fully disbursed by mid-1988, of sector work in population and primary health care. were not full, effective mechanisms for bringing about Implementation of projects in Bangladesh has been a fundamental policy change, and their discontinuation persistent problem, affecting the results of IDAs Country was appropriate, if late. In the early 1980s, however, Assistance Strategy. Implementation problems have been given the country's very low reserves and limited ability difficult to overcome, and disbursement ratios (averaging to import essential food and oil, it is debatable whether about 11 percent) throughout the 1980s were below IDA's structural policy reform stance should have been Bank norms. Efforts to improve this have been successful, less compromising, with much stronger conditionality but the supervision intensity effort per project in contained in the IPCs (see box 5.1). Bangladesh is still about 50 percent higher than the IDA's Country Assistance Strategy became more Bankwide average. There are many reasons for this rela- sharply focused, more relevant, and more realistic in tively poor implementation performance. The limited the 1990s. The improving macroeconomic position institutional capacities in the country are under strain, has provided more opportunity to suspend activity in and this has worsened as the ADP has expanded. There sectors of strong policy disagreement. Nevertheless, are too many projects in the ADP, which makes monitor- IDA now manifests a much more sophisticated appre- ing problematic. The government's own procedures and ciation of institutional limitations in the public sector. administrative arrangements are cumbersome and time- And IDA is seeking to neutralize these limitations by consuming. Project design has often been overoptimistic more targeted interventions and alternative project in estimating project completion times; some project implementation approaches. designs are too complex for local institutional capabili- IDA's strategy throughout the period under ties. Attempts at institutional reform have met with little review was efficacious. The early emphasis on success, and the only way to improve this has been to cir- expanding food-grain production and improving cumvent poorly functioning institutions and find alterna- birth control prevalence and primary health care was tive implementing agencies. highly appropriate, and sustained efforts in these In Project Outcomes from 1980 to 1996, 63 proj- areas brought positive results. IDA's tenacity has not ects were reviewed and rated by OED. Of this number, brought equally strong dividends elsewhere. IDA 24 projects, or 38 percent, were rated as unsatisfac- assistance to restructure SOEs and the financial sector tory. This compares with a Bankwide OED rating of (particularly in the 1980s) was not successful. Only 33 percent of projects as unsatisfactory. With regard gradually did IDA realize that the core constraint in to sustainability, 35 percent of rated Bangladesh pro- the development process in Bangladesh was the pub- jects had a likely sustainability rating, compared with lic sector itself. Political involvement in all aspects of 45 percent Bankwide. With regard to institutional civil administration, in public enterprises, and in com- development performance, 36 percent of rated mercial banks and corruption within the civil service Bangladesh projects had a negligible institutional and public utilities all created strong vested interests development rating, compared with 24 percent determined to maintain the status quo. Bankwide. The intensity of ESW was consistent with Bankwide Sector performance has been variable; the industry norms, and some notable successes in assisting govern- and finance sectors performed extremely poorly. Out of ment policy formulation were gained. Studies in trade and 13 operations, only 3 (23 percent) had a satisfactory rat- 26 Efficacy of IDA's Assistance Strategy: Overall Assessment BOX 5.1: EVALUATION OF IDA'S IMPORT PROGRAM CREDITS FROM 1980 TO 1988 uring the period ciency. Initially the opera- patterns and circum- created a desensitizing 1980-85, the tosfcsdo oiy sacso h aiu fetb leitn Board approved issues related to the prime socioeconomic groups. pressure on the political five IPCs, which dis- objective of increasing The government main- authorities to be more bursed $619.1 million in food-gram production. tamed that IDA did not resolute in addressing balance of payments As the 1980s progressed, display a full understand- pressing economic support to mid-1988. howevei; the content of ing of social and political problems. This program followed action programs changed issues, or of the institu- But reform progress eight similar operations to encompass a wider tional frameworks was made in some areas, in the 1970s. These range of sector and surrounding these issues. and while the pace may operations were macroeconomic issues. This criticism was not have been slow, by the intended to provide Although ostensibly justified, end of the 1980s solid assistance to sustain a performance-oriented, It was clear that positive results were minimum level of criti- conditionality for second progress up to about emerging. Improvements cally essential imports tranche releases tended to 1987 was uneven, and were made in fertilizer (food and oil supplies) be general and permitted conditionality, while distribution and market- and to alleviate recurring an elastic assessment of comprehensive in its ing, and sustained IDA balance of payments dif- performance: tranching intent, was not fully attempts finally bore ficulties. While some of these operations did complied with, nor were fruit in liberalizing the aspects of these difficul- not provide an effective reforms energetically environment for low-cost ties could be traced to enforcement mechanism. followed up. Through- minor irrigation inputs. inappropriate macroeco- This general lack of out the period, political Progress was made in nomic management, timing and specificity commitment to a broad- improving overall man- much of the difficulty reflected IDA!s delicate based adjustment agement of the country's was caused by external task of balancing the program was fragile: no food system and the shocks (oil price rises), country's acute short- institutional framework public food-grain distrib- adverse terms of trade term need for foreign for integrated manage- ution programs. This movements (declining exchange and budgetary ment of an adjustment was particularly notice- world jute prices), and a resources against fulfill- process was created. able after the massive poorly developed capac- ment of a wide range of During the 1980s, the floods of 1988. Gradual ity to generate foreign needed structural flow of external assis- progress was made in the exchange earnings. reforms. tance exceeded US$12 liberalization of the trade Given that much of the From the ESW billion, and, while this regime and the environ- population was living at undertaken, IDA was was successful in gradu- ment surrounding invest- the edge of survival, dra- able to identify key ally advancing macro- ment. This established a matically reducing constraints and obstacles, economic stabillzation foundation and accep- aggregate demand to and brought these to the and financing a large tance for the more reduce balance of government's attention. share of public invest- aggressive reforms in payments difficulties in For much of the 1980s, ment, it may have these two areas. Between the very short term was the government argued undermined a more 1982 and 1986, nearly not humanely feasible. that while this ESW was determined reform 600 nationalized firms These operations of high quality, it did not effort. This assistance, were denationalized, sought to identify and pay sufficient attention to particularly balance of against strong domestic , remedy critical issues equity considerations and payments support, may opposition. related to production effi- the differential behavioral have inadvertently (continued on next page) 27 Bangladesh: Progress Through Partnership BOX 5.1: EVALUATION OF IDA'S IMPORT PROGRAM CREDITS FROM 1980 TO 1988 (CONTINUED) Reform progress in resource mobilization manifest in the early tions. Because the coun- the financial sector was efforts or in improving 1980s, by the mid-1980s try's foreign reserve levels less successful. Through- budgetary planning and it was becoming apparent had climbed to about out the 1980s, few con- management. Project that a more detailed and 3.75 months of import crete steps were taken to implementation delays holistic sector approach equivalent, from only 0.5 improve the country's continued, and perhaps was needed, with opera- months in 1980, it was commercial banking sys- worsened, during the tional mechanisms, such feasible to drop the IPCs tem, and there was little decade, reflecting lack of as Sector Adjustment in favor of more conven- or no attempt to rehabil- progress in overcoming Loans (SECALs), that tional adjustment itate the variety of spe- institutional and admin- would have a more operations. cialized DFIs. Little istrative constraints, specific reform agenda progress was made in While the need for and more demanding improving domestic these IPCs was clearly tranche-release condi- Source: This pr6cis of the effectiveness of IDA!s IPCs in the 1980s was drawn from World Bank data. ing. The success rate for other sectors is as follows: agri- tality has begun to drop markedly, primarily as a culture, 79 percent; energy, 60 percent; education and result of the government's expanded immunization health, 66 percent; and technical assistance, 50 percent. program, supported by IDA and other donors. The outcome performance of IDA projects in some sectors Despite these achievements, the family planning is clearly a cause for concern, and health services have serious weaknesses. Internal The Aid Coordination function of IDA in Bangladesh efficiency of the system is low because of management, is crucial, given the large number of donors assisting the staffing, and logistical problems. The quality of deliv- country. The scope and range of this coordination effort ery of health care in public facilities, particularly in the has expanded, most notably from the late 1980s, and now rural areas, is perceived to be poor, and drugs are encompasses groupings of donors who support broad sec- often in short supply. IDA and other members of the tor investment programs under IDA leadership. This has donor community and the government are aware of been achieved for population and health, flood control, these problems, but attempts to deal with them have and primary education. Because there is a large donor failed. The rising demand for family planning and presence in the field, IDA, in conjunction with the gov- maternal and child health services is placing an erment, has organized a Local Consultations Group increasing organizational and financial burden on the (LCG), with 18 subgroups. The LCG meets about 6 times system. Meeting the goal of replacement fertility will a year; the subgroups meet periodically. These coordina- require a considerable increase in efficiency, greater tion efforts have been effective, reliance on NGOs and the private sector, and greater By supporting expansion of health infrastructure cost recovery. A revitalized service delivery system will and trained staff, as well as policy and institutional call for greater community involvement, considerable improvements, IDA operations have facilitated signif- administrative decentralization, and a more limited icant positive outcomes in Family Planning and role for government at the national level. Health. This assistance, together with that of other IDAs assistance efforts in Education and Human donors, has been accompanied by rising total govern- Resource development have been effective. The ment expenditures on health and family planning. The increased targeting of vulnerable, disadvantaged groups contraceptive prevalence rate has increased to about and primary education has had positive results. After 45 percent from 8 percent in 1975, and total fertility primary school enrollment had been stagnant for almost has declined from around 7 to 3.3 percent. Infant mor- 30 years, gross enrollment rose from 55 percent in 1985 28 Efficacy of IDA's Assistance Strategy: Overall Assessment to 63 percent in 1990, and to over 100 percent by 1995. centralized control, and a complex array of policy dis- Attendance of girls, particularly in rural areas, has tortions and regulations. Throughout much of the increased, and NGOs are active in nonformal literacy 1980s, there was little genuine government commit- programs. Government commitment to education has ment to creating an industrial sector driven by market increased: education outlays have grown from 7.9 per- forces. As a result, government ownership of projects cent of the budget in 1980 to 10.3 percent in 1990. The was deficient, and IDA consistently miscalculated gov- large increase in school capacity and number of teachers ement resolve to effect real change.1 enabled primary school enrollment to increase by 50 The situation improved in the 1990s. IDA!s move to percent in the 1980s. IDA has had much less success in SECALs with broad sector reform agendas has been vocational training and specialized education, with neg- appropriate, and while not fully successful, has gener- ative rates of return for assistance to public vocational ated a momentum for more liberalization throughout and technical training centers. the economy and created a growing constituency for As with health care, however, the internal effi- wider reform. Where success has been attained (the ciency of the public primary school system is low, and Export Development Project), the objectives were clear the quality of education is deficient. Teachers are and specific and the project was modest in scope and poorly trained and supervised and logistical support is expectations. Good progress has been made in trade poor. Dropout rates are high; it takes an average of 8.7 reform and investment deregulation, and this has been years to produce a single graduate of the five-year pri- sustained. But the regulatory environment is still oppres- mary cycle. System accountability is inadequate and sive, and the extensive public sector involvement in retention rates for formal government primary educa- industry has yet to be reduced or removed. tion compare unfavorably with programs sponsored Attempts by government, with the assistance of by NGOs. Increasing the quality of education requires IDA and other donors, to create a strong and healthy improved incentives and greater accountability for Financial sector in Bangladesh have failed. All OED- teacher performance, improved teacher training, more rated projects during the review period were unsatis- parental involvement in resource mobilization and factory and not sustainable. In many respects, the school management, and further encouragement of malaise in the banking sector is as chronic today as it NGOs and private providers. Since 1992, IDA has was in 1980. Comprehensive ESW undertaken by the addressed these issues in a sustained and aggressive Bank identified the problems and the solutions. But manner, and this is bearing fruit. the complex governance issues involved in the sector, Over the past 15 years, government efforts, with the politicization of lending and government owner- IDA assistance, to create a dynamic Industrial sector ship, and the way in which these factors would under- have not been successful. Only two out of nine proj- mine the reform agenda were not acknowledged or ects have been rated as satisfactory: of the two ongo- addressed. Throughout the 1980s and 1990s, govern- ing projects, one had significant difficulties and was ment commitment to reform was deficient and project recently canceled. The pace of reform of the policy ownership was weak. IDA consistently overestimated environment has been slow, and the successes that government commitment, and had unrealistic imple- have emerged (such as ready-made garments) cannot mentation expectations as a result. be attributed to government or IDA (see box 5.2 for a It is unfair to suggest that no benefits accrued discussion of the IFC's activity in the country). The sit- from IDAs efforts. The sector reform adjustment oper- uation improved in the 1990s, with more aggressive ation brought about some liberalization and reduced trade policy reform following good IDA ESW and the volume of directed credit, and the operation high- more intense policy dialogue on trade liberalization lighted the nature and seriousness of the problems issues. IDA's attempts to deal with the highly ineffi- affecting the banking system and the urgent need for cient SOEs have largely failed. The quality and depth pervasive reform. But an autonomous and efficiently of the ESW undertaken by the Bank certainly diag- functioning central bank and legal loan recovery sys- nosed the problems and constraints. However, few tems have not been fully established. realized the complexity of unbundling the issues of IDA has been effective in its assistance to Agriculture. government ownership, strong and highly politicized Good progress has been made toward achieving rice self- labor movements, a bureaucracy seeking to maintain sufficiency, and while this may have taken longer than 29 Bangladesh: Progress Through Partnership BOX 5.2: IFC ACTIVITY IN BANGLADESH uring the period Between 1980 and was cumbersome. These credible clients, however, under review 1992, the IFC made four circumstances put this was difficult, given the 19x0-96), IFC investments in Bangla- firm at a disadvantage in endemic default culture. activity in Bangladesh has desh, totaling US$10.27 a fiercely competitive Client servicing and been limited. There were million. By the end of international market, persistent follow-up were various reasons for this. FY96, the IFCs portfolio The situation improved central features of IDLCs For many years the envi- exposure in the country in the early 1990s, with success. Today, folowing ronment for private sec- was down to US$3.7 mil- a relaxation in the IDLCs lead, there are tor activity in the country lion. ICs experience in import ban on raw nine leasing companies. was not encouraging. these projects was mixed, hides, lowering of import To augment IDLCs lend- There was a wide array Two projects, Bata Shoes tariffs (15-22.5 percent), ing resources, the IFC of regulatory and admin- Limited and IDLC, a and introduction of approved a U.S. istrative impediments to leasing company, were bonded warehousing. dollar guarantee in 1995 foreign investment and successful. The other two This was a successful to ANZ Grindlay's Bank the development of the projects, Dynamic Tex- IFC project, and the for a Taka loan to IDLC. indigenous private sector. nigenuprvtsco. tiles and IPDC (and an equity investment has IPDC, the IFC's third In addition, government earlier investment, made now been divested, with investment in the country, and private sector rela- in 1979) were problem a solid gain, had a difficult start. tions tended to be adver- projects. Even the suc- Another IFC success Several investments suf- sarial, reflecting the resid- cessful projects were was the investment (in fered repayment prob- ual effects of the hampered by 1985) in IDLC, the coun- lems because of the nationalization and cen- controlled trade policies try's first leasing com- absence of an effective trally planned economic and cumbersome govern- pany. This was in collab- legal system, which inhib- policies followed in the mental procedures. oration with a Korean ited enforcement of the 1970s. The business envi- For example, IFCs leasing company, KDLC. borrower's contractual ronment was adversely investment in the local At that time, medium to obligations. However, affected by the absence of expansion of a multina- long-term project finance IPDCs operations have an effective legal system tional shoe manufacturer, was scarce in Bangladesh, significantly improved and weak enforcement of while yielding satisfac- and the commercial now that the government existing regulations. This tory dividends, was ham- banks focused on short- has demonstrated its policy and legal environ- pered by government term trade financing, commitment to strength- ment (combined with a policy that banned the Although awareness of ening the legal frame- relatively underdeveloped import of raw hides and this additional source of work. IPDCs more recent domestic market and a imposed a high import medium-term financing investments have regis- shortage of entrepreneur- tariff (45-60 percent) on took some time to tered substantially better ial talent, managerial leather-finishing chemi- develop, leasing became repayment performance, expertise, and skilled cals. The procedure for popular with entrepre- and with the develop- labor) hampered private claiming the duty draw- neurs as they realized its ment of the Dhaka Stock sector development, back facility for export advantages. Selecting Exchange (DSE), PDCs desired, dDA was instrumental in liberalizing the agricul- was relevant; for much of the 1980s, howeve IDAs tural input trade. Easier farmer access to minor irrigation, efforts to strengthen the institutions delivering these ser- equipment, power tillers, and fertilizer has improved vices were not fully effective. IDA has not been successful smallholder productivity and incomes. IDA advice and in improving credit delivery systems for agriculture-they assistance for extension, research, and crop diversification are as bad today as they were in 1980. Much of the 30 Efficacy of IDA's Assistance Strategy: Overall Assessment equity holdings are show- first private sector invest- services have been successful outcome and ing healthy capital gains. ment since gas concession provided to the local Secu- a positive development blocks were awarded rities and Exchange Corn- impact are greatly Future Agenda through bidding in the mission, to the Govern- reduced. As noted in Chapter 4, early 1990s. A rural ment of Bangladesh on The role of the govern- the economic reform telecommunications privatization and private ment in ensuring a momentum was fitful project was approved for sector involvement in the facilitative and nurtur- throughout the 1980s. telephone connection to power sector, and to the ing environment for From late-1991, however, rural homesteads and Telecommunications private sector develop- economic liberalization rural business growth Regulatory Board. ment is crucial to the took on a more sustained centers in the northern success of any project character and foreign and half of Bangladesh. Other Lessons for the Future or undertaking. The domestic investor interest projects include a hotel, a IFC involvement in regulatory and adminis- expanded. As a result, cement plant, a housing Bangladesh highlights trative environent IFC activity accelerated finance company, and a important generic lessons should be user-friendly, from about mid-1992. A leasing company. There is applicable to IFC activity in and should be dynamic local presence was estab- now a much-expanded many developing countries, and flexible in over- lished at the resident pipeline, which includes coming unforeseen mission in 1993, and transactions in power, 9 All projects and yen- obstacles. This requires increased efforts were telecommunications, tures require strong both policy reform and made by Washington- cement, housing finance, operational manage- bold government lead- based staff and by the leasing, health care, ment and technical sup- ership to overcome New Delhi South Asia urban transport, the gas port. In environments entrenched bureaucratic Regional Mission to sector, and textiles, where the necessary practices and attitudes. expand the project In additon to its managerial and techni- 9 Strong collaboration is pipeline. Six projects were direct investment activi- cal expertise is in short needed between the IFC approved in the last two ties, the IFCs Foreign supply, it is essential and other parts of the years, for a total amount Investment Advisory Ser- that foreign partners World Bank Group to of US$55.77 million in vice (FIAS) has been active maintain a close identify and overcome IFC and participant bank throughout the 1980s; this almost day-to-day- constraints to private loans and equity. The IFC support has expanded managerial and techni- sector development. Board approved an equity significantly in the 1990s. cal support function. Already established IFC investment in an unincor- IFC has worked closely a No matter how viable a private sector contacts porated joint venture with IDA on aspects of venture may look in can be valuable sources with a U.S. oil and gas the ESW program, with financial and economic of information in this company for gas explo- completion of a Private terms, if the legal, regu- area and in determining ration and development Sector Assessment in 1994 latory, and financial the sequence of reform in the northeast region of that provided inputs to the systems are deficient, needed to overcome Bangladesh. This is the 1994 CEM. Advisory the chances for a these constraints. responsibility for this must rest with the government, expanding mangrove forestry: this has addressed environ- which has undermined credit delivery with periodic loan mental issues and protection of vulnerable shorelines. forgiveness schemes. The specialized public financing Overall, IDA has been effective in Flood Control institutions had no incentive structure to enforce loan and Drainage. This was an area of considerable techni- recovery. IDA has made a valuable contribution in cal controversy and, of course, high visibility. IDAs 31 Bangladesh: Progress Through Partnership strategy, based on 1970s sector work, was to avoid blocks have been leased, and offers for the remainder are large projects in deeply flooded areas with high civil under evaluation. work costs per hectare. IDA efforts were concentrated IDA support in the development of Bangladesh's on smaller sub-projects in shallow-flooded areas with Infrastructure has been effective in expanding the physi- low development and operating costs. Over time, many cal stock, and the development impact, particularly in policymakers came to recognize the worth of this the rural areas, has been positive. As with other sectors, approach. IDA sector work in this area was instrumen- however, the creation of sustainable institutions has not tal to developing the National Water Plan in 1986, been successful. For urban water there is a lack of com- which established a multisector framework for planning mercial orientation with very poor cost recovery; munic- water management; minor irrigation and shallow tube- ipal revenues are inadequate to cover the maintenance of wells were vital elements of this plan. IDA also took a existing systems, let alone new investments. Good suc- strong coordinating role in the development of a Flood cess has been achieved in IDA's assistance for rural Action Plan in 1989, which moderated demands by roads, using local materials. Adequate maintenance has both government and some donors for massive civil been a problem, but under the latest project, efforts are engineering works with uncertain technical and envi- being made to involve local councils in maintaining road ronmental outcomes and economic viability. drainage and culverts. This approach (strongly endorsed IDA was not so successful in improving the institu- by the Planning Commission) could lead to greater tional performance of the Bangladesh Water Develop- council participation in other areas, improving account- ment Board. Little progress was made in dealing with ability and sustainability. deficient operation and maintenance of flood control Although IDA's assistance strategy has been effi- and drainage schemes. From the early 1990s however, cacious and effective, there have been failures. There there has been a greater degree of beneficiary participa- has been a consistent failure to fully recognize institu- tion, with local councils becoming more involved in the tional constraints and compose effective means to deal operation and maintenance of smaller schemes. with them. The complex governance issues that sur- On balance, IDA has had a positive impact in the round the public sector's role in the economy have Energy sector. But, if one notes the resources directed to become more clearly understood, but resolution of the sector, and balances this against the sustained these issues has yet to be fully developed. That decen- achievements, the overall efficacy of IDA's assistance is tralization of authority in the implementation of modest. IDA has sought to encourage the government to development efforts improves accountability and allow a greater private sector role in energy, and at least involvement of beneficiaries has been slow to gain a greater degree of commercial autonomy for public util- acceptance. Despite over 15 years of policy dialogue, ities, but government has resisted. Little progress has the enabling environment for private sector develop- been made in reducing system losses in the main urban ment is still deficient. Much ESW has identified con- areas, despite separation of generation from distribution. straints and suggested solutions, but liberalization has The relative success of IDA's efforts in rural electrifica- proceeded fitfully. This has undermined development tion and distribution is the product of the decentralized progress and impeded a speedier reduction in overall administrative approach adopted with the creation of poverty. There is no doubt that the political environ- rural distribution cooperatives that are more account- ment has been difficult for much of the period under able to the communities they serve, and operate with a review, and this has limited the effectiveness of IDA's more commercially oriented administrative structure. policy dialogue. A consistent theme throughout the While five of seven projects in the oil and gas sector have period has been IDA recommendations for reform and been rated as satisfactory, the long-term sustainability of change that have been thwarted by government the public sector's total monopoly position in the sector unwillingness to act boldly. Nevertheless, persistence is questionable. IDA has encouraged the government to by IDA in this dialogue has brought about some allow a greater role for the private sector in gas devel- notable successes that have improved the human con- opment, with some success. Eight out of 23 exploration dition in Bangladesh. 32 Lessons for the Future o improve the future effectiveness of IDA assistance in Bangladesh and accelerate economic development and reduction of poverty, two questions need to be asked. What should the government do, and what should the private sector and NGOs do? How can the services provided by the public sector be made more responsive to stakeholders' needs? The answers to these questions should determine IDA's assistance strategy and the type of lending and ESW that should be done in each sector. The generic lessons for the future cut across all sectors, and should guide future CASs. Redefining the Role of the Public Sector. In all sec- unit performance has not been made accountable to the tors this role needs to be redefined. There is great community served. No penalty has been imposed for potential scope for the private sector, NGOs, and local deficient performance. Public employees in Bangladesh, governments to be involved in providing services. whether teachers, doctors, agricultural extension offi- Local and municipal governments should be more cers, bank loan officers, or electrical engineers, do not prominent in all aspects of infrastructure development operate under performance and incentive structures that and maintenance, as well as education and primary encourage accountability to the communities or clients health care. Government involvement in commercial they are expected to serve. activities must be reduced. Paying Greater Attention to Sustainability Issues. Increasing Emphasis on Accountability of Systems. While no single approach will work in all sectors or Future project design must improve the accountability projects, much more emphasis must be given to recur- of those responsible for providing a service or complet- rent expenditure requirements for operations and ing a project. This design must involve all stakeholders, maintenance, to cost recovery (if only partial in the to define the limit of this accountability and set a frame- beginning), and to institutional structures or imple- work for enforcing it. As a general principle, account- menting agencies and organizational frameworks. ability is enhanced by decentralized participation in Policy reforms should be difficult to reverse or under- decisionmaking, finance, and organization. Organiza- mine, and should produce tangible benefits quickly. tional reforms have been tried in all sectors, including By doing so, reforms will establish a constituency of the civil service and quasi-government authorities; by beneficiaries with a strong vested interest in the sus- and large, all have failed. This has happened because tainability of initiatives undertaken. 33 Bangladesh: Progress Through Partnership Improving Project Monitoring and Evaluation. ceptive prevalence increases, and phase out its retailing There is further scope to evaluate development of family planning services. The NGOs and private throughout project implementation. Often the focus providers (allopathic physicians, pharmacists, and tra- of attention is the physical completion of civil works ditional practitioners) should be encouraged to gradu- or other such quantifiable targets, but more thought ally fill the resulting gap. In family welfare and health, needs to be given to explicit outcome indicators; these a greater role for private providers will increase the should be part of the project design. Projects often run scope and variety of service alternatives available. Pub- over seven or eight years, and waiting for the Imple- lic funding of hospital-based curative procedures mentation Completion Report (ICR) to provide a crowds out cost-effective basic health care, and govern- measure of effectiveness takes too long. Better moni- ment programs should work with, rather than restrict, toring and evaluation would assist the government in indigenous practitioners and midwives (and upgrade administrative reform designed to streamline project their skills). The government should provide safe water design and implementation. and sanitation, control of communicable diseases, and Recognizing Ownership Issues. Many projects in health education. In education, priority must be given Bangladesh show symptoms of incomplete ownership to improving the quality of primary and secondary edu- and low commitment to project objectives. Long cation. The role of the public sector in technical and delays in effectiveness, delays in appointing key staff, vocational education should be limited, and programs or failure to take action on agreed policy measures are should be developed in collaboration with private examples. More effort should be made to strengthen employers, with costs primarily borne by the employer the political commitment to necessary institutional and trainees. Subsidies to universities should be phased change, and solid indicators should be used to indicate out and user charges raised. To increase accountability, this commitment. To proceed without this essential schools should eventually become the responsibility of foundation is a disservice to the people of Bangladesh reconstituted local governments, with teachers and and has negative externalities that affect the entire administrators being employed by (and responsible to) assistance program. these community-based administrations. Defining Crucial Objectives and Limiting Goals. Increasing Emphasis on Accountability of Systems. In countries that have manifested continual problems At present, administrators, managers, and service of project implementation, it may be useful to limit providers in the health and education sectors are not project objectives to a smaller number of crucial pol- held accountable to the communities they serve. Within icy changes and development impact objectives. Pro- the current framework of the government's programs jects often expand during the preparation and and projects, and Bank support, the scope for improv- appraisal cycle because of the well-intentioned desire ing this accountability is intrinsically limited. Govern- to generate maximum impact. However, as shown on ment has been very good at activities such as mobilizing many occasions in Bangladesh, this may result in dis- communities, increasing school enrollment, vaccina- appointments for the borrower and for IDA. For tion, and contraceptive acceptance. It has been less able adjustment-type operations there is a strong case for a to bring about the systemic improvements needed to sequence of single-tranche operations, with policy upgrade school quality or monitor performance of reform to be enacted before Board presentation. health care provision. Organizational reforms have Adaptable lending should be a key instrument in deal- been tried; so far all have failed because the major, ing with sequential policy reform. While this may politically powerful stakeholders (government physi- increase lending staff time, there will be a compen- cians, family welfare workers, and teachers) have satory saving on supervision effort; this may help the undermined them, believing the reform measures to be borrower to focus on key issues and expedite consen- against their interests. All stakeholders could support a sus building within the country. reform agenda in which all see benefit. As a general principle, greater accountability is enhanced by decen- Social Sector Development tralized participation in decisionmaking, finance, and Redefining the Role of the Public Sector. In family plan- organization. This means giving greater financial and ning, the government should progressively withdraw managerial autonomy to local units within the country's from the role of primary provider of services as contra- administrative systems. These units should have greater 34 Lessons for the Future responsibility for planning and budgeting, collecting ject cannot be deemed a success: many of the new gradu- fees, and determining how collected funds and govern- ates had great difficulty entering the labor market. ment transfers should be used. This would improve Intensifying Sector Work. Over the past 20 years, accountability, provide signals about the appropriate only S sector reports have been prepared on the edu- mix of services, encourage greater community partici- cation and population/health sectors. Other studies pation, and minimize administrative overhead. have dealt with these sectors only peripherally. This is Paying Greater Attention to Sustainability Issues. not enough to keep a sharp focus in the ongoing pol- Budgetary constraints will limit achievement of gov- icy dialogue in the sectors. The recently established ernment goals for family planning, health, and educa- Health Economics Unit in the Ministry of Health and tion. To relieve this constriction, an amalgam of Family Welfare has already produced reports on key approaches will be necessary. The goals may have to issues in the health sector, including a public expendi- be adjusted to more realistically reflect the budgetary ture review; an analysis of recurrent costs, user fees, envelope. Perhaps budgetary allocations will have to and the poor, costing the package of essential health be increased. Greater reliance must be placed on the services; and mobilizing resources through hospital private sector and the NGOs. More vigor should be user fees. There is much for the Bank to build on in applied to increasing the efficiency of service provi- health and education. sion, together with increasing the charges to benefi- Optimizing the Role of NGOs. NGOs are impor- ciaries of publicly provided services. Redefining the tant in health and education, both in piloting innova- role of government is desirable from an accountabil- tive approaches and in providing services. The GOB ity perspective, and charges to beneficiaries for cura- has become increasingly supportive of NGOs and has tive medical care and for higher education are justi- accepted them as partners in achieving national goals. fied on equity and efficiency grounds. However, the GOB has concerns about accountability Recognizing Ownership Issues. All projects and efficiency and maintains a strong regulatory role. undertaken in the social sectors in Bangladesh exhibit Bank relationships with the NGOs have been good, symptoms of low ownership of projects (delays in and Bank operations have benefited from the appointing key staff, delays in approving project pro- approaches taken by some NGOs. However, discus- formas, and failure to take action on agreed policy ini- sions in the field suggest that many NGOs are reluc- tiatives are a few examples). In future assistance, tant to approach the Bank because of the complexity stronger efforts must be made to strengthen the polit- of Bank procedures and procurement regulations and ical commitment to institutional change and to the need for government guarantees. There is scope for develop more effective procedures and control mecha- greater NGO-government collaboration and interac- nisms in the provision of public services. tion. The Bank-IDA would benefit considerably from Improving Monitoring and Evaluation. Greater this, as both teams seek to improve effectiveness in the thought must be given to choosing explicit outcome indi- social sectors in Bangladesh. cators of development impact that can be monitored dur- Reshaping Lending Operations. The time may ing project supervision, given the long gestation period of now be appropriate for a transition from a "project" to projects undertaken in the social sectors. In the past, a "program" form of lending for core operations in the greater attention was focused on physical implementa- population, health, and education sectors. This pro- tion. While ICRs assess project outcomes more deeply gram approach would require agreement with the gov- than supervision reports, the time lag of ICR completion ement on a wide range of policies, objectives, strate- averages about eight years from project approval. A strik- gies, and expenditures. Disbursements would be linked ing example is the Technical Education Project, which was to policy performance and satisfactory progress of the approved in May 1984. Closing was extended three expenditure program. This approach will require times, until June 1992: the ICR was issued in October agreement on a "new vision" for the role of govern- 1993. There were 11 supervision missions between Octo- ment and other stakeholders in the sector, as well as a ber 1984 and November 1991, and at no time was the review of the sector expenditure program and its project's rating less than satisfactory, largely because phys- financing. The existing project approach of large and ical rehabilitation and procurement of equipment were complex operations with many subcomponents is not relatively timely. Yet from an outcome perspective the pro- designed to carry the significant policy changes that 35 Bangladesh: Progress Through Partnership might be necessary to move to new frameworks for gradualist approach often enhances the power of the providing population, health, and education services. more permanent (relative to changing political leader- ship) centralized administrative bureaucracy, main- Industrial and Financial Sector Development rains the scope for rent-seeking behavior, and dilutes Public Sector Ownership of Commercial Banks and the confidence of domestic and foreign investors who Industrial Enterprises. Government ownership of have witnessed, or experienced, previous "liberaliza- commercial banks and industrial enterprises does not tion" efforts. For the success of 1DAs future assistance provide the appropriate incentive framework for to Bangladesh, IDA must seek fundamental reform in either internal efficiency and speedy response to bank supervision enforcement, legal debt recovery changing circumstances or for sound allocation of enforcement, and closure of unviable SOEs before credit within various parts of the economy. Reform of proceeding in other areas of industrial and financial dysfunctional financial and SOE sectors must include, sector restructuring. at the outset, an explicit statement of government Attention to Sustainability Issues. Experience sug- intent to privatize, and also a widely publicized and gests that multi-tranching of adjustment credits does monitorable action plan to implement privatization in not necessarily ensure sustained commitment to a parallel with increased efforts to strengthen central long-term reform agenda. It might be beneficial to bank supervision and enforcement of prudential regu- have a series of single-tranche adjustment or sector lations. During the 1980s, the GOB denationalized a operations, with each having more limited, but cru- large number of SOEs, but the momentum has been cial, objectives. This may bring a sharper focus to the variable, and a policy of protection for the inefficient policy dialogue and facilitate a more sensitive under- remains. Over the long term, this is not sustainable standing of short-term political constraints. It may and represents a tax on the more productive. also improve the process of consensus building neces- Redefining the Role of the Public Sector. For both sary to ensure that all stakeholders are committed to the industrial and the financial sectors to achieve a the reform effort. While this approach may increase more dynamic contribution to the country's develop- Bank staff time in project appraisal, there will be a ment, it is necessary that there be a clearer definition compensatory tradeoff against staff time spent in of the appropriate role of the public sector. Improving supervision. It will also help implementing agencies in the technical skills within the central bank and grant- Bangladesh because weak implementation capacities ing the necessary autonomy to enforce sound pruden- can be more sharply concentrated. tial regulation, improving the efficiency and impartial- Technical Assistance, Institutional Development, ity of the legal system, liberalizing further the trade and Sector Adjustment. IDA's assistance to these two regime and the overall regulatory environment, and sectors over the past decade and a half has shown that improving the quality of the supporting infrastructure institutional strengthening is an indispensable part of are all areas where the GOB should focus attention. sector reform. Attention must be given to the timing IDA can support this effort, but should do so only and focus of technical assistance designed to improve when there is a demonstrated willingness to assume a the capabilities involved in the sectors. For example, less interventionist public sector role. Stronger atten- good bank supervision skills and systems are critical tion must also be given to micro-credit facilities in to improving the overall financial system. Technical both rural and urban areas, which have been shown to assistance efforts in this area should precede more be viable and successful by NGOs. Greater links need comprehensive sectorwide reform efforts. to be progressively forged between NGOs engaged in Increasing Emphasis on Accountability. The micro-credit provision and the formal banking system. almost complete lack of project success in these sectors Pace and Sequencing of Reform. Gradualism-in over the past 15 years suggests that accountability on introducing international standards and good prac- the part of designated implementing agencies is defi- tices in the banking system or encouraging efficiency cient. The recently approved microenterprise line of in the industrial sector-has untoward side effects. credit project, however, which will disburse through a The sense of urgency and commitment to a sustained specially created foundation for onward transmittal to reform program is undermined, which allows coali- participating NGOs lending to very small enterprises, tions of vested interests to thwart the reform intent. A is a promising sign. 36 Lessons for the Future Energy and Infrastructure ity Supply Authority, illustrates what might be The Appropriate Role of the Public Sector. Much more achieved. Annual performance contracts that bind serious consideration must now be given to the appro- senior management to achieving organizational and priate role of the public sector in the provision and operational targets would certainly help. The reverse maintenance of infrastructure. This implies that the of this coin (of course) is the need to encourage broad private sector, foreign and domestic, has a critical part managerial autonomy and discretion. in investment and operations in power generation and Donor Assistance and Domestic Self-Reliance. distribution, the oil and gas sector, telecommunica- Many donors have provided assistance to all aspects of tions, water and sewerage operations, ports, and all infrastructure development in Bangladesh. Much has forms of transportation, and suggests that various lev- been achieved in the past two decades. In some cases, els of the public sector must became more actively however, most notably the power and oil and gas sec- involved and that a proper regulatory role for the state tors, the flow of concessional resources may have per- must be made explicit. Local authorities and munici- versely undermined the perceived urgency for reform palities should be given more autonomy and manager- and domestic resource mobilization. While there is ial discretion in investment and the maintenance of indeed a shortage of reliable generating capacity in the infrastructure. Where funding and institutional capa- country, much could be achieved by a determined pro- bilities are weak, these should be progressively gram to reduce system losses and improve collections. strengthened. In clearly commercial areas such as the These deficiencies represent a tax on the poor who are oil and gas sector, telecommunications, and power, without adequate electrical power. While the power sec- public ownership has not optimized exploration, pro- tor will need external resources, these should only duction, or distribution. There is clearly a place for pri- accompany a demonstrated commitment to improving vate sector partnership in these sectors in Bangladesh. institutional performance. This may suggest a more Where public corporations remain, they must be program-based approach by IDA in the power sector, allowed considerable managerial autonomy and discre- with solid agreement with the government on policies, tion to operate on purely commercial lines. objectives, strategies, and expenditures. Disbursements Attention to Sustainability Issues. Project design could then be linked to policy performance and clearly must pay much closer attention to sustainability issues monitorable indicators of system performance. in infrastructure projects. Expanding generation capacity or urban access to potable water serves little Agriculture and Flood Control purpose if nontechnical system losses are not Redefining the Role of the Public Sector. As with other addressed vigorously or if cost recovery remains dila- sectors, the public institutions supporting agriculture tory. Expanding the rural road network without main- have not changed speedily to the more dynamic nature taining the existing network undermines the develop- of agriculture in Bangladesh, and service delivery per- ment impact of the investments made. A recurrent formance has been poor. The positive achievements theme throughout projects in the energy and infra- made in the agricultural sector reflect the liberaliza- structure sectors is the substandard performance of tion of input trade and greater private sector involve- responsible institutions and a general approach to ment. This has brought about significant technology governance that reduces accountability rather than transfer and suggests that in research and extension, enhancing it. This is not the result of limited human public and private contributions are synergistic. capital and technical skills (the Local Government Greater beneficiary participation, with local councils Engineering Department in the Ministry of Local Gov- becoming more involved in the operation and mainte- ernment, Rural Development, and Cooperatives has nance of smaller flood control and drainage schemes, competent engineers and technicians, for example), has had positive results. This should be extended. but rather a function of existing bureaucratic and Patient Policy Dialogue Can Help to Define Cru- administrative systems. Tinkering with internal struc- cial Objectives. IDA!s coordinating efforts in develop- tures will achieve little unless the reward/sanction phi- ing a Flood Action Plan have been beneficial in a losophy within the service provider groups is changed. highly complex area with far-reaching environmental The much better collection performance of the Rural consequences. Perhaps a similar approach may be use- Electricity Boards, compared with the Dhaka Electric- ful in developing agricultural diversification. 37 1 ANNEXES ANNEX A: EXECUTIVE SUMMARY, REPORT NO. 17455-BD, FOR BOARD DISCUSSION Introduction The World Bank and IDA have been involved in Bangladesh is a country that evokes strong impres- the economic development of Bangladesh from the sions, and few represent such a testing challenge for very beginning. This partnership with the Government the development mandate of the World Bank and IDA. of Bangladesh (GOB) has encompassed lending, policy Famine has been an ever-present danger, and this advice, and donor coordination. Over the past two densely populated country has been highly vulnerable decades, IDA has financed about one-quarter of all to the natural disasters of flooding and cyclones. foreign aid commitments, covering all sectors of the Annual capita income is below US$250, and half the economy, including balance of payments support. population, which is 80 percent rural, still lives below Furthermore, a massive volume of sector work has a minimum poverty line. As Bangladesh emerged from been completed and made available to the government a bloody civil war in 1971, many doubted the coun- and the donor community. This has had a considerable try's capacity to survive and foresaw a precarious exis- influence on development initiatives and, while not tence supported by the largesse of the richer nations. always accepted by the GOB, it has certainly fueled They have been proved wrong. debate. Bangladesh has not only survived, but has made some good progress in economic and social develop- The Development Challenge ment. The country now approaches self-sufficiency in The need for foreign assistance is clear. The most fun- food-grains and a host of social indicators (fertility damental weakness of the Bangladesh economy has rates, infant mortality, life expectancy, primary school been low rates of investment and saving. Given that so enrollments, and adult literacy) have shown steady many people are living at a precarious margin, where improvement. Since Independence, growth has not consumption is necessary to survive, this may be read- been spectacular, but it has been steady and has accel- ily understandable. A further development challenge erated in the 1990s. has been the relatively low level of development of the Despite these achievements, doubts remain human resource endowment, which makes it difficult regarding whether they can been sustained. Questions to improve technology and labor productivity in the remain as to the further deceleration of population short term. But foreign aid has been forthcoming from growth, and whether food security can be further the start. In the country's first three years, Bangladesh enhanced, or if the country is caught in an "aid trap" received more aid than in all its 24 years as East Pak- of perpetual dependency on concessionary foreign istan, and annual commitments exceeded US$1.5 bil- financing flows. There is no doubt that despite the lion by the early 1990s. progress made, it has not been quite enough-the This foreign assistance itself has been a con- absolute number of poor has grown and the country's tentious issue. Many observers, both domestic and ecosystem is very fragile. foreign, have expressed concern that the country's use Throughout this period, the country has also been of foreign aid has both infringed on national sover- undergoing a political transformation and the devel- eignty issues and undermined self-reliance in the devel- opment of a multi-party democratic system. This opment process. It is not possible to definitively settle transformation has also included the establishment of this issue. As with many complex economic and social a civil administration structure that has regularly been propositions, there are elements that can support one reshaped and adjusted. This history of political devel- opinion or another. Given the fungibility of resources, opment has had a very significant influence on the aid indeed helped fund budget deficits and inefficient development path chosen by Bangladesh and on eco- state-owned enterprises (SOEs), and perhaps reduced nomic outcomes. The first government nationalized the pressure to improve efficiency and cost recovery in most of the industrial, financial, and larger commer- public utilities. However, all this assumes that public cial sectors, and espoused a strong, direct public sector sector institutions would rapidly or magically function role in the development process. Successive govern- efficiently in the absence of aid. This is obviously ments have sought to reverse the role of the State in uncertain. What is more certain is that many programs direct economic activities, but this effort has not been and projects would have proceeded more slowly, if at consistently sustained, faltering in the face of opposi- all, and this delay would have had very serious ramifi- tion by political or vested interests. cations and adverse social costs. Examples abound, 39 Bangladesh: Progress Through Partnership such as the population program, child immunization, with a stronger tax collection performance. minor irrigation development, and primary education. The balance of payments has always been a weak- One thing is certain: administration of the entire ness of the Bangladesh economy. Throughout much of aid program has been demanding for the government. the 1980s, reserves were at precarious levels, given the Implementation of the Annual Development Plan country's susceptibility to the shocks brought about (ADP) has always been difficult, and 20 percent of the by natural disasters. Current account deficits in the program for a given year is often not implemented. As early 1980s averaged over 9 percent of GDP, and were a result, throughout the 1980s and early 1990s, the gap only sustainable through the flow of concessional for- between commitments and disbursements has widened, eign assistance. The causes of this persistent external and the backlog in the pipeline has grown. This reflects imbalance were inherently structural, but little was the very centralized nature of administration in the done until the mid-1980s to bring about a fundamen- country and the weak institutional capabilities through- tal change in relative prices or a reduction in the dis- out the public sector. In many respects this represents tortions that affected the development of export earn- the prime development challenge. One of the key con- ing opportunities. By 1986 this had started to change, straints has been the public sector itself, and no govern- and following a sustained adjustment effort in the late ment of the 1980s or 1990s has maintained a clear 1980s, the balance of payments improved dramati- vision of an appropriate role for the public sector, and cally, and foreign reserves reached historically high the role that the private sector and NGOs play in the levels. Growth in nontraditional exports has been very development of the country. As will be seen, this is a strong in the 1990s, led by ready-made garments. consistent theme throughout this report. This reflects a decline in the real exchange rate from 1985, and the progressive liberalization of the trade Progress to Date and investment regimes. The growth of ready-made Growth performance has not been spectacular, and with garments-from almost zero in 1982 to over US$2 bil- the exception of flood- or drought-related years, it has lion in 1995-is a notable achievement. It is indicative been in the range of 3.5 to 4.0 percent annually. Invest- of what can be expected of Bangladesh in relatively ment, both public and private, has been deficient, labor-intensive, light manufacturing activities, pro- although investment performance has shown some vided the environment is facilitative and infrastructure improvement in the 1990s, with a consequent modest constraints (in power, for example) can be reduced. improvement in the trend GDP growth rate. But two While public savings has improved, the overall bud- indicators suggest that some profound changes have get deficit remains high, at about 6.8 percent of GDP in occurred in the basic framework of the Bangladesh 1995. A major difficulty has been the poor performance economy. First, the share of agriculture in total output of the SOEs-gross losses reached a staggering 2.1 per- has declined, despite an increase in food-grain produc- cent of GDP in 1991. Progress has been made in reduc- tion, which had approached self-sufficiency by the early ing this, but the SOEs and utilities remain a drain on the 1990s. This growth pattern is consistent with the record economy. Bangladesh has gone through periods of pri- of many other low-income countries and suggests a vatization, but momentum has been fitful. Government degree of economic diversification and policymakers have also been slow to acknowledge Another encouraging and positive trend has been the role that the private sector can play in helping to the steady and quite strong growth in domestic savings develop the country's infrastructure. performance, notably private savings. This suggests a While GDP growth may not have been spectacu- number of positive things. The economy has become lar good progress has been made in improving the increasingly monetized and a growing proportion of general well-being of the population. An aggressive the population has been able to meet basic survival immunization program, particularly from the mid- needs for food and shelter, resulting in a rising mar- 1980s, has brought about declines in infant mortality ginal propensity to save. Strong growth in worker and improved life expectancy. Population programs remittances from abroad has also added to the pool of have been successful in reducing fertility and curbing private savings. Finally, by the 1990s, the public sector population growth. In spite of these achievements, had improved its own savings performance, an complacency is unwarranted. Large segments of the improvement achieved entirely on the revenue side population still live a precarious existence-47.5 per- 40 Annexes cent of the people consumed fewer than 2,122 calories the reduction of poverty. Improving yields and labor daily in 1990. Although this percentage is down from productivity in the rural areas was viewed as a way to a massive 73.2 percent in 1981, the development chal- increase incomes for the bulk of the population. In lenge is very much alive. keeping with these views, the assistance strategy of the Overall, Bangladesh has seen progress in eco- time envisaged that 40 percent of lending would be nomic and social development in the 1980s and directed to the agricultural sector, with an immediate 1990s, albeit not at a pace one might prefer. Undoubt- goal of increasing the flow of inputs to farmers. Atten- edly the speed of reform accelerated in the late 1980s, tion was also to be given to rural infrastructure and and this has led to an improvement in the investment agricultural credit programs. and commercial environment. But the country must At the time, IDA recognized the problems of the generate 1 million new jobs a year, which will require SOEs, and proposed to assist in the restructuring of the a more dynamic investment environment. There is no jute and textile sectors. It was hoped that this would shortage of entrepreneurship or individual ingenuity improve domestic savings. Together with sector work in the country, but government, at both the local and to improve public investment programming, this strat- the national levels, has yet to define the proper balance egy would make it possible for increasing amounts of of support and intervention needed to create a suitable local resources to be channeled into the ADP. enabling environment that reduces transaction costs While perhaps not as optimistic as the govern- and allows entrepreneurship to flourish. ment, IDA was quite confident that a quick transition to a sustainable growth path could be realized. IDA'S COUNTRY ASSISTANCE STRATEGY Growth was projected to be about 6.5 percent annu- ally, and it was envisaged that investment would reach 1980-85 18 percent of GDP by 1985. The reality was different. At the beginning of the 1980s, a transition from a mar- Growth averaged 3.5 percent annually, and invest- tial law regime to the civilian government of President ment reached only 13 percent of GDP. No progress Ziaur Rahman was accomplished. Many of the early was made in improving the SOEs (although quite a development goals were very optimistic, such as few were denationalized), and the already poor per- replacement fertility levels and elimination of food formance of the financial system worsened. imports by 1985. In the 1980 CPP, IDA explicitly rec- For all the difficulties, however, there were ele- ognized that these targets and expectations were unre- ments of IDA's assistance strategy that were highly alistic. It then faced the quite delicate task of reducing relevant and contributed to development progress in a expectations, while maintaining government and donor number of areas. The focus on food-grain production support and enthusiasm over a medium-term frame- was correct, and while it took almost a decade to fully work period. To a large extent this was accomplished liberalize agricultural input trade, the priority given to by focusing on key stated government objectives and this task was appropriate. The early support given to clearly articulating constraints in both absorptive the population program was also correct. capacity and domestic institutional capabilities. If there was a fundamental weakness in IDA's IDA strategy in 1980 focused on three priority assistance strategy at this time, it was the belief that areas: government was fully committed to a reform of civil service institutions aimed at reducing excessive regula- * Expand food-grain production to achieve self- tion and administration-to real reform of the ineffi- sufficiency. cient public enterprises and a genuine drive to create a * Improve domestic resource mobilization to more conducive environment for the private sector. cover an increasing share of the ADP. Only gradually through the 1980s did IDA realize that * Improve project implementation and a core constraint in the development process in execution. Bangladesh was the public sector itself. Given the extreme hardship caused by periodic droughts and floods, together with the strain that food 1986-90 imports entailed for the balance of payments, expand- Some important lessons were learned in the early ing food self-sufficiency was seen as a key element in 1980s that did manifest themselves in a revised CAS. 41 Bangladesh: Progress Through Partnership By 1986, it was concluded that there was a much inappropriate. Finally, while hindsight is always won- greater need to address sector and macroeconomic derful in evaluating a past strategy, the slow, almost tor- constraints, and to do so with much more sharply tuous, progress made in liberalizing agricultural input focused policy conditionality. It was also recognized prices and distribution systems is quite remarkable. that trying to deal with inefficient public enterprises in the absence of a very strong government commit- 1991-96 ment was unproductive. Instead, greater efforts were A review of IDA's strategy documents from the early made to create a more conducive environment for the 1990s clearly shows a hardening of IDA's strategy, private sector and to deal with the distortions in trade, which became more focused on implementation and pricing, credit allocation, and interest rates. the pursuit of new operations only in areas where Renewed efforts were made to improve project there was a manifestly strong government commit- execution, which included agreements with govern- ment. Strategy now reemphasized a strong and sus- ment on multi-year budget planning, project and con- tamed improvement in the environment for private tract approval procedures, and better monitoring and sector development, an enhanced program of support evaluation. Domestic resource mobilization was still a for population and primary health care, and continued major problem, and IDA proposed an expanded pro- support for primary education. Adjustment-type lend- gram of ESW to cover public expenditure and tax ing was reduced to about 15 percent of the proposed reform, as well as a more intense policy dialogue to lending program, and sector operations were under- reduce subsidies, particularly for fertilizer and food taken in a very selective manner in line with an assess- distribution programs. Population and primary health ment of the governments willingness to proceed with care remained a priority, but with increasingly com- a broad reform agenda in a timely manner. plex operations, multiple donors, and more compo- Earlier efforts to improve public sector resource nents. The education strategy changed, adopting a mobilization finally started to bear fruit in the early much stronger focus on primary education and accel- 1990s, but IDA recognized that a fundamental reform erating access for females. of the civil service was unlikely in the near term. As a The earlier emphasis on agriculture waned some- result, IDA adopted a strategy of redefining the role of what, in part because the already large project port- the public sector, encouraging greater private sector folio was delayed in disbursement, and absorptive participation in a wider array of activities that previ- capacity was limited. One major shift in strategy was ously had been the preserve of the public sector. The to increase the emphasis on the energy sector, partic- 1995 CAS emphasized approaches to bring about a ularly oil and gas, to reduce the country's depen- speedier reduction in absolute poverty, and worked dence on imported energy. Finally, the poor loan closely with NGOs to develop the design of projects to recovery performance of the DFIs convinced IDA achieve this aim. This, in turn, led to a search for inno- that additional line of credit-type projects in agricul- vative ways to implement IDA projects. Finally, in the ture or industry served no useful purpose in the late 1980s, environmental issues started to take on a absence of very substantial institutional, systemwide much greater prominence in IDA's country assistance reform. strategy, and IDA has sought to link investments in Although a number of key lessons were learned, flood control and drainage with a strategic framework some were not fully absorbed. Problems in implementa- of national land and water planning. tion and project execution worsened, and IDA dis- What was notable about the 1992 CSP and the played continuing faith that relatively minor adminis- 1995 CAS was the more clearly articulated perfor- trative adjustments by the government would solve the mance benchmarks that were put forward to mark problem. The real nature of the governance problems progress in the development agenda, but also to deter- within the central administration were not fully appre- mine the pace and areas of new IDA commitments. ciated. Also, it is recognized that while dealing with IDA's assistance strategy in the 1990s undoubt- public enterprises in Bangladesh was a high-risk edly has become more relevant simply because of a less endeavor, the emphasis placed on the energy sector- compromising approach to delays in policy reform or most notably oil and gas and industrial energy effi- project execution. Furthermore, this strategy led to a ciency, which involved solely state enterprises-was much more sanguine view of outcomes and more real- 42 Annexes istic expectations, with more conservative risk assess- STRATEGY IMPLEMENTATION AND EFFICIENCY ments. There was a much greater appreciation of gov- IDA commitments to Bangladesh during the period ernance issues and how they affected project objec- under review (FY80 to FY96) amounted to US$5.965 tives. The sector adjustment lending that was billion and involved approval of 93 projects. Dis- undertaken in the 1990s has not been entirely success- bursement during the period totaled US$4.65 billion, ful, in part because IDA did not follow its own strat- or about 9 percent of IDA disbursements in the 1980s. egy of addressing root causes and requiring a demon- This proportion declined quite dramatically in the strated government commitment to deal with issues 1990s, to 5.7 percent, reflecting the absorptive capac- that were known to be controversial. ity constraints within the country and the increasingly Concrete steps were taken to introduce more difficult institutional constraints that have impeded innovative methods of project implementation. If there project execution. Disbursement ratios during the is to be a modest criticism of the CAS in the 1990s, it 1980s were well below Bank norms, averaging about is that IDA did not attempt to go much farther in seek- 11 percent. A concerted effort was made to solve this ing alternative implementation mechanisms that problem in the early 1990s, and the disbursement increase accountability and a more immediate assess- ratio had reached 17.6 percent by 1996. As might be ment of outcome. Nevertheless, it is clear that IDA has expected given the implementation problems, supervi- a more sophisticated appreciation of the limitations of sion effort has consistently been above the Bank aver- the public sector in Bangladesh. ages, or averages for IDA countries. The intensity of ESW for Bangladesh has been at Generic Country Assistance Lessons about the Bank average over the past decade, although A review of IDA's country assistance strategy over the this intensity accelerated in FY94. The share of ESW past 15 years highlights and reinforces a number of in the total assistance effort for Bangladesh has been key generic lessons for all CASs. consistently high, however, when compared with the South Asia Region. The ESW undertaken has been * Monitorable Performance Benchmarks: consistent with the country strategy, and the timing Country strategies become much more rele- and quality have been good. For example, a tax vant and operationally useful if there are very reform study and an environmental review fed directly specific benchmarks in both policy reform into government studies and strategy formulation and outcomes. soon after completion. * Flexibility in the Lending Program: Strategies From 1980 to 1996, 63 projects were reviewed should allow considerable flexibility to adjust and rated by OED. Of this number, 24 projects (38 the lending program in response to progress percent) were rated as unsatisfactory. This compares in identified key reform areas. with a Bankwide OED average of 33 percent unsatis- * Realism of Risk Assessments: The validity and factory. With regard to sustainability, 35 percent had relevance of CASs is greatly enhanced by san- a likely rating, compared with 45 percent Bankwide. guine assessments of the difficulties of achiev- Of these rated projects, 36 percent had negligible insti- ing progress. Such assessments render the risk tutional development ratings, compared with a 24 per- assessments more meaningful, and expecta- cent Bank average during the same period. tions more realistic. The efficiency of assistance to Bangladesh does * Past Experience Needs an Honest Assess- not compare favorably with Bank averages. In lending ment: Before embarking on definition of a staffyears per project, Bangladesh is almost 40 percent future CAS, more attention should be given higher in staff time spent. This relatively less efficient to determining the effectiveness of the last performance is indicative of the difficulties inherent in strategy. dealing with Bangladesh's assistance program. * Realism in Institutional Capacity Assess- Bangladesh requires over 50 percent more calendar ments: CASs inevitably underestimate the time to take a project from project concept to the complexity and time necessary to bring about Board. There is another very interesting indicator of institutional change and capacity enhance- the difficulties of project assistance to the country. In ment. the context of this review, 83 completed projects were 43 Bangladesh: Progress Through Partnership analyzed to ascertain the time between Board munity involvement, considerable administrative approval and project effectiveness. It was found that decentralization, and a more limited role for govern- 29 percent took nine months or more to be declared ment at the national level. effective. This is strongly indicative of administrative logjams in the country, perhaps a less than full com- Education and Human Resource Development mitment to the project, and deficient project prepara- IDA was an early supporter of Bangladesh's education tion and design. efforts, although at the outset there was no coherent sectoral approach. Early assistance did not address SECTOR OUTCOMES primary education, which in retrospect can be seen as a mistake. This deficiency was remedied in 1980 with Health and Family Planning the first project in primary education, and emphasis In 1974 gross health statistics in Bangladesh reflected has remained on this area. These efforts have been a quite desperate situation. Life expectancy was only effective. After stagnant primary school enrollment for 45 years and infant mortality was 140 per 1,000 live almost 30 years, gross enrollment rose from 55 per- births. Total fertility rates were about 7, and the pop- cent in 1985 to 63 percent in 1990. Attendance of ulation was growing at over 2.5 percent a year. In girls, particularly in rural areas, has increased, and 1976 the government declared family planning as one NGOs are very active in nonformal literacy programs. of its top priorities. IDA became involved in this Government commitment to education has increased, effort, and gradually assumed leadership of a collabo- and education outlays grew from 7.9 percent of the rative effort of the international community, resulting budget in 1980 to 10.3 percent in 1990. The large in family planning and health projects of increasing increase in school capacity and teachers enabled pri- size and complexity. The core of these projects mary school enrollment to increase by 50 percent in financed rural health facilities, recruitment and train- the 1980s. IDA has had much less success in voca- ing of rural family planning and health workers, and tional training and specialized education; there have purchase of contraceptives and other medical supplies. been negative rates of return for assistance to public This effort has brought about significant positive vocational and technical training centers. outcomes. Fertility has declined to about 3.3, the con- As with health care, however, the internal effi- traceptive prevalence rate has reached 45 percent, and ciency of the public primary school system is low, and infant mortality has declined to 88 per 1,000 live the quality of education is deficient. Inadequately births. Between 1985 and 1994, the proportion of trained and supervised teachers receive poor logistical children immunized against 6 major childhood dis- support. Dropout rates are very high, and it takes an eases increased from 2 to 62 percent. average of 8.7 years to produce a single graduate of Despite these achievements, the family planning the 5-year primary cycle. There is a lack of system and health services have a number of serious weak- accountability, and retention rates for formal govern- nesses. The internal efficiency of the system is low ment primary education compare unfavorably with because of management, staffing, and logistical prob- programs sponsored by NGOs. To raise the quality of lems. The quality of health care delivery in public education, better incentives and accountability for facilities, particularly in the rural areas, is perceived to teacher performance, improved teacher training, more be poor, and drugs are often in short supply in these parental involvement in resource mobilization and areas. IDA, other members of the donor community, school management, and further encouragement of and the government are aware of these problems, but NGOs and private providers will be required. attempts to deal with them have failed. The rising demand for family planning and maternal and child Industry health services is placing an increasing organizational Over the past decade and a half, government efforts and financial burden on the system. To meet the goal and IDA assistance to create a dynamic industrial sec- of replacement fertility will require a considerable tor have not been successful. Only two of nine projects increase in efficiency, greater reliance on NGOs and have been rated as satisfactory, and of the two ongoing the private sector, and greater cost recovery. A revital- projects, one is experiencing significant difficulties. ized service delivery system will involve greater com- The pace of reform of the policy environment has been 44 Annexes slow, and the successes that have been achieved (ready- Agriculture made garments) cannot be attributed to government or IDA has been effective in its assistance to agriculture. IDA. Undoubtedly the situation improved in the Progress has been made toward achieving food-grain 1990s, with more aggressive trade policy reform. IDA self-sufficiency, and while it may have taken longer attempts to deal with the highly inefficient SOEs have than desired, IDA was instrumental in liberalizing the largely failed. The quality and depth of Bank ESW agricultural input trade. Easier farmer access to minor diagnosed the problems and constraints, but there was irrigation equipment, power tillers, and fertilizer has a continual underestimation of the complexity of brought about a fundamental change in smallholders unbending the issues of government ownership, strong productivity and incomes. IDA advice and assistance and highly politicized labor movements, a central for extension, research, and crop diversification was bureaucracy seeking to maintain centralized control, relevant, but for much of the 1980s, IDA's efforts to and a complex array of policy distortions and regula- strengthen the institutions delivering these services tions. Throughout much of the 1980s, genuine govern- were not fully effective. As in the financial sector, IDA ment commitment to, and a vision of, an industrial sec- has not been successful in improving credit delivery tor driven by market forces was absent. Government systems, which are as bad in agriculture today as they ownership of projects was thus deficient, and IDA has were in 1980. Much of the blame for this must rest consistently miscalculated government resolve to effect with the government, which has undermined credit real change. The situation has improved marginally in delivery with periodic loan forgiveness schemes. IDA the 1990s. IDA's move to SECALs with broad sector has made a valuable contribution to expanding man- reform agendas has been appropriate, and while not grove forestry, which has addressed environmental fully successful, their use has generated a momentum issues and the protection of vulnerable shorelines. for more economywide liberalization and has created a growing constituency for this wider reform. Flood Control and Drainage Overall, IDA has also been effective in flood control Finance and drainage. This was an area of considerable techni- Attempts by government, with the assistance of IDA cal controversy and, of course, high visibility, given the and other donors, to create a strong and healthy finan- periodic severe flooding in Bangladesh. IDA's strategy, cial sector in Bangladesh have also failed. All IDA pro- based on sector work in the 1970s, was to avoid large jects rated during the review period were unsatisfac- projects in deeply flooded areas with high civil work tory, and not sustainable, and in many respects the costs per hectare. Instead, IDA efforts were concen- malaise in the banking sector is as chronic today as it trated on smaller subprojects in shallowly flooded was in 1980. Comprehensive ESW was undertaken by areas with low development and operating costs per the Bank, particularly from the mid-1980s, which hectare. This was not a strategy the government fully identified the problems and the solutions. What was accepted, but many policyrakers gradually came to not fully understood were the complex governance recognize the worth of such an approach. IDA sector issues involved in the sector, the politicization of lend- work in this area was also instrumental in developing ing and government ownership, and how these factors the National Water Plan in 1986, which finally estab- would undermine the reform agenda. Throughout the lished a multi-sector framework for planning water 1980s and 1990s, government commitment to reform management and confirmed that minor irrigation and was deficient and project ownership was weak. It is shallow tube-wells were a vital element of the plan. unfair to suggest that no benefits accrued from IDA's IDA also took a strong coordinating role in the devel- efforts. The sector reform adjustment operation opment of a Flood Action Plan in 1989, which served (1990) brought about some liberalization and reduced to moderate demands by both government and some the volume of directed credit, and as it progressed, the donors for massive civil engineering works with uncer- operation highlighted the nature and seriousness of tam technical and environmental outcomes. the problems affecting the banking system and the IDA was not as successful in improving the insti- urgent need for pervasive reform. But an autonomous tutional performance of the Bangladesh Water Devel- and efficient central bank and legal loan recovery sys- opment Board, and little progress was made in dealing tem have not been fully established, with deficient operation and maintenance of existing 45 Bangladesh: Progress Through Partnership flood control and drainage schemes. Efforts have con- guiding vision for the next decade: tinued, however, and from the early 1990s there has (a) What should-and can-the government do, been greater beneficiary participation, with local and what should be done by the private sector councils becoming more involved in the operation and and NGOs? maintenance of smaller schemes. (b) How can the services provided by the public sector be made more responsive to the needs Energy of the stakeholders? On balance, IDA has had a positive impact in the energy sector. But if one takes note of the resources How these questions are answered in each sector that have been directed to the sector, and balances this and for each activity should determine IDA's assis- against the sustained achievements, the overall effec- tance strategy and the type of lending and ESW to be tiveness of IDA's assistance appears more modest. done in each chosen sector. The conclusions in each IDA has sought to encourage the government to allow sector chapter contain a variety of sector-specific rec- a greater private sector role in energy and a greater ommendations for the future. Outlined below are degree of commercial autonomy for public utilities, more generic lessons that cut across all sectors, and but government has resisted. Little progress has been would guide the future CAS. made in reducing system losses in the main urban Redefining the Role of the Public Sector. In all areas, despite separation of generation from distribu- sectors the role of the public sector needs to be rede- tion. A noteworthy contrast to these results has been fined. There is much more scope for the private sector, the relative success of IDA's efforts in rural electrifica- NGOs, and local governments to be involved in the tion and distribution. This is the product of the decen- provision of services. Much greater efforts must be tralized administrative approach adopted, which cre- made to encourage an expanded role for local and ated rural distribution cooperatives that are more municipal governments in all aspects of infrastructure accountable to the communities they serve, and oper- development and maintenance, as well as in education ate with a more commercially oriented administrative and primary health care. Government involvement in structure. commercial activities must be reduced. Increasing Emphasis on Accountability of Sys- Infrastructure tems. Future project design must seek to improve the IDA support in the development of Bangladesh's infra- accountability of those responsible for providing a ser- structure has been effective in expanding the physical vice or completing a project. Project design must stock. The development impact, particularly in the involve all stakeholders, simply to define the parame- rural areas, has been positive. As in other sectors, how- ters of accountability and to set a framework for ever, success has not been achieved in creating sustain- enforcing it. As a general principle, accountability is able institutions. For urban water, there is a lack of enhanced by decentralized participation in decision- commercial orientation with very poor cost recovery, making, finance, and organization. Organizational and municipal revenues are inadequate to cover main- reforms have been tried in all sectors, whether it be the tenance of existing systems, let alone new investments. civil service or quasi-government authorities, and they Good success has been achieved in IDA's assistance for have all largely failed. The simple reason is that unit rural roads using local materials. Adequate mainte- performance has not been made accountable to the nance has been a problem, but under the latest project, community served, and there has been no penalty for efforts are being made to involve local councils in deficient performance. Public employees in maintenance of road drainage systems and culverts. Bangladesh, whether they be teachers, doctors, agri- cultural extension officers, bank loan officers, or elec- LESSONS FOR THE FUTURE trical engineers, do not operate under performance To improve the effectiveness of future IDA assistance and incentive structures that encourage a greater in Bangladesh, and to accelerate the pace of economic accountability to their communities or clients. development and the reduction in poverty, two quite Need to Pay Greater Attention to Sustainability fundamental questions need to be asked to develop a Issues. More attention should be given to sustainabil- 46 Annexes ity issues in project design. There is no single approach in effectiveness and in the appointment of key staff or that will work in all sectors or projects, but much failure to take action on agreed policy measures are a more emphasis must be given to recurrent expenditure few examples. More of an effort should be made to requirements for operations and maintenance; to cost strengthen the political commitment to necessary insti- recovery, even if only partial in the beginning; and to tutional change, and solid indicators should be sought institutional structures or implementing agencies and to gauge this commitment. To proceed without this is organizational frameworks. Policy reforms should be a disservice to the people of Bangladesh, and often has sought that are difficult to reverse or undermine, that negative externalities that affect the entire assistance produce tangible benefits quickly, and thereby estab- program. lish a constituency of beneficiaries with a strong vested Define Crucial Objectives and Limit Goals. In interest in the sustainability of the initiatives. countries that have manifested continual problems of Need to Improve Project Monitoring and Evalua- project implementation, it may be useful to limit pro- tion. Project monitoring and evaluation have ject objectives to a smaller number of crucial policy improved, but there is further scope to evaluate devel- changes and development impact objecrives. Projects opment throughout a project's implementation. Often often tend to expand during the preparation and the focus of attention is on the physical completion of appraisal cycle simply because of the well-intentioned civil works or other such quantifiable targets, but desire to generate the maximum impacr. However, as more thought needs to be given to explicit outcome has been demonstrated on many occasions in indicators, and these should be included as part of the Bangladesh, this may be counterproductive, resulting project design. Projects often run over by seven or in disappointments for the borrower and IDA. For eight years, and the wait for the Implementation Com- adjustment-type operations, there is a strong case in pletion Report (ICR) to provide a measure of effec- Bangladesh for a sequence of single tranche opera- tiveness is too long. tions, with policy reform enacted before Board pre- Recognizing Ownership Issues. Many projects in sentation. While this may increase lending staff time, Bangladesh show symptoms of incomplete ownership there will be a compensatory saving in supervision and commitment to the project objectives. Long delays effort, and it may also serve to help the borrower 47 Bangladesh: Progress Through Partnership ANNEX B: REPORT FROM THE COMMITTEE ON DEVELOPMENT EFFECTIVENESS (CODE) Bangladesh: Country Assistance Review (CAR) tutional weaknesses and strong opposition by power- The Committee on Development Effectiveness ful defaulters have thwarted progress. It noted OED's (CODE) considered the Bangladesh Country Assis- conclusion that attempts by Government, with the tance Review (CAR) (IDA/SecM98-102), prepared by assistance of IDA and other donors, to create a strong the Operations Evaluation Department (OED), on and healthy financial sector have failed, and that in March 25, 1998. It commended OED for a particu- many respects, the malaise in the banking sector is as larly rich and candid report, including useful insight chronic today as it was in 1980. OED recommends into the historical background of the country. The that IDA accord the highest priority to bank supervi- Committee welcomed management's assertion that the sion enforcement, legal debt recovery enforcement, CAR's findings and conclusions had contributed sig- and closure of unviable SOEs. The Committee wel- nificantly to the preparation of the current CAS, comed management's response that banking reform is which is scheduled for Board discussion on March 31, a top priority for IDA and the IFC and that the pro- 1998. The lessons from experience and management's posed assistance strategy will help the Government to responses to the report's recommendations, which are rebuild three institutional pillars of sound banking: enumerated in the CAS, reflect a broad confluence of strong regulatory systems, well-managed banks, and views with OED's main concerns. an effective court system. The Committee welcomed the CAR's principal conclusion that IDA's assistance to Bangladesh over Client Ownership and Commitment the past fifteen years has been effective, and that the The issues of client ownership and political commit- partnership forged with the Government has led to ment were prominent in the Committee's discussion. progress in both economic and social development. The Committee took note of the finding in the CAR IDA has made a positive contribution to fundamental that a fundamental weakness in IDA's assistance strat- changes in the Bangladesh economy. Nevertheless, the egy has been the implicit assumption that Government Committee is aware that progress has not been uni- was fully committed to a reform of the civil service formly good in all areas and raised several issues that institutions, to real reform of the inefficient public are pertinent to the upcoming CAS discussion. These enterprises, and to a genuine drive to create a more are highlighted below. conducive environment for the private sector. OED rec- ommends that more effort be made to strengthen the Redefining the Role of the Public Sector political commitment to necessary institutional change The Committee noted OED's finding that the disap- and that solid indicators of commitment be sought. pointing progress in some areas is due, in large part, to Management acknowledged that the critical impor- deeply entrenched governance issues, including weak tance of client ownership and commitment is one of the or missing institutions and powerful vested interests lessons it has learned from experience in Bangladesh. that have sought to maintain the status quo. OED Indeed, OED found that in the early 1990s, there was identifies the public sector as one of the key con- a hardening of IDA's strategy as it became more straints to Bangladesh's development, and concludes focused on implementation and pursuing new opera- that the role of the public sector in each sector must be tions only in areas where there was a manifestly strong redefined. Management asserts that the fundamental Government commitment. The Committee welcomed challenge in pursuing the Bank's mission is to work the current strategy's plan to exercise selectivity by with Bangladesh to overcome these constraints by matching the Bank Group's response to the Govern- assisting in building or strengthening institutions that ment's willingness to initiate and complete essential guide or improve the management of development, reforms. that perform essential "governance" functions or The Committee endorsed OED's generic recoi- improve the environment for private sector develop- mendation that country strategies, to be more relevant ment, and that improve the delivery of key public and operationally useful, should include clearly social and infrastructure services. defined and monitorable benchmarks in both policy The Committee was particularly concerned that reforms and outcomes. These benchmarks serve to the financial sector is among those areas where insti- gauge Government ownership and commitment, and 48 Annexes thereby help to define the most effective assistance and is a vital part of the country assistance strategy. It program. noted that, for the most part, this donor coordination effort has been effective, and has served to improve Partnership, Collaboration, and Aid Coordination cross-dissemination of sector work and knowledge The Committee stressed the importance of partnership and to minimize "donor competition" at the sectoral with donors, NGOs, and other elements of civil soci- level. The Committee was particularly pleased that the ety within a framework of understanding with the report underscores that IDA can play a leadership role Government, and noted OED's finding that there is in providing development assistance to Bangladesh scope in Bangladesh for greater collaboration among without having to do everything itself. It also NGOs, Government, and IDA, particularly in the applauded the active involvement of the country direc- social sectors. It also emphasized that donor-driven tor in the field in the coordination of the Bank Group's institutional change cannot be sustained and that work with that of local donor partners, thus helping to change must come from within the society and culture select activities that best fit the Bank's comparative in which the institutions are to operate. Building civil advantage. With respect to IDAs relationship with the service and public support for key institutional Asian Development Bank, the Committee noted that reforms can help to alleviate problems of weak politi- regional management had visited Manila in February cal will. The Committee welcomed management's 1998 to start a process of frequent, regular reviews to statement that the current assistance strategy is ensure coordination of assistance to Bangladesh. premised on support from civil society as a con- stituency for institutional change and on the inclusion of domestic development partners in the work of the Bank Group as a way to increase its quality and sus- tainability. The Committee agreed with the OED conclusion that the donor coordination function of IDA is crucial Jan Piercy Chairman, CODE 49 Bangladesh: Progress Through Partnership ANNEX C: SUMMARY OF BANK-GOB DISCUSSIONS OF THE CAR, JUNE 2, 1999 Colleagues: recovery than with credit delivery. Dr. Rahman pointed out that this was a result of palming off bad The Country Assistance Review (CAR) report was loans (except in the case of natural disasters), which discussed with the Government of Bangladesh today at was again a government failure. Once again it seemed the NEC Auditorium, Sher-e-Banglanagar, Dhaka. Dr. that GOB officials did not strongly disagree, but with Masihur Rahman, Secretary, ERD, presided over the the language (para 40, last line), ie.: the blame should meeting, and representatives from different ministries rest with the Government. The same message could be and relevant GOB agencies made specific comments on conveyed more diplomatically. sections of the report relevant to them. Mohsin The DG, IMED, suggested a correction that Alikhan led the WB team, which included Arun Baner- IMED is under Planning Ministry, not under Planning jee, Kapil Kapoor, Robert Epworth, Syed Nizamuddin, Commission (p. 36, para 3.7). As regards monitoring Muhammad Iqbal Karim, and Subrata Dhar. of project benefits (p. 16, para 52), he observed that Dr. Masihur Rahman started the meeting with the these should be built into each project, and IMED observation that the report should be reviewed from does not have the capacity to do it. Dr. Rahman the perspective of lessons for the future that would help observed that project benefit should be monitored by better formulation of country assistance strategy and someone who is not part of project management. project designs. Mohsin Alikhan discussed briefly the IMED is the best institution to do it. However, mech- process of a Country Assistance Review, which is con- anisms needs to be devised to enhance IMED capacity ducted by OED, an independent arm of the World in this regard. Bank, and reiterated the emphasis on lessons for the A representative from the Power Division of the future. Energy Ministry contested the observation (Chapter 8, As Dr. Rahman opened the floor, a representative p. 6, para 8.27) about government resistance to from the Planning Division noted that the summary change. Dr. Rahman interjected this discussion with a note of the report was positive, but observations in the question: whether the Government failed or resisted. main text were not consistent with this summary note. He observed that this should be rephrased as the For example: he pointed out (para 13, page 7) the last Government tried hard, but met with limited success." line "Only gradually through the 1980s did IDA real- In conclusion, Dr. Masihur Rahman requested the ize that a core constraint in the development process Bank to look into the language of the document. He in Bangladesh was the government itself and its agreed with the substance, but observed that a little unwillingness to enforce the needed public sector change in semantics might make the document. reforms." GOB officials did not object strongly, but The Bank team that attended the meeting feels they thought that the language should be changed to strongly that the wording change should be made convey the same message. Dr. Rahman observed that without changing the substance. The present wording the report should be more focused on quality of pro- is not conducive to building client relationships. I ject implementation, employment creation, balanced would strongly urge that the report should be edited and equitable development, strengthening of multi- for choice of words. After all, the ERD/GOB showed year budgeting (three-year rolling investment), and alot of understanding in agreeing to go through the export growth. report at this late stage and are not asking for changes The Ministry of Agriculture said that the sentence in substance, but in presentations, which would be "Since 1995, there has been. . . ." on page 54, para mutually beneficial. 4.26, should be deleted, which the representative Fred, since you are in Washington, I leave the deci- claimed was not correct. He also said that the obser- sion to you. vation about resistance from the public institutions on page 73, para 6.25, was too harsh. He also contested Best Regards, the observation that government failure had under- Mohsin [Alikhan] mined credit delivery (p. 13, para 40) in rural areas. The problem, according to him, was more with credit 50 Annexes ANNEX D: GOVERNMENT RESPONSE Government of the People's Republic of Bangladesh Ministry of Finance Economic Relations Division No. ERD/IDA8/1/98/45 June 17, 1999 Dear Mr. Alikhan, Please refer to the discussions in the meeting held on 02.06.99 in ERD on the Bangladesh Country Assistance Review (CAR, proposed for publication) prepared by the World Bank's Operations Evaluation Department (OED). As discussed in the meeting, the Bangladesh Country Assistance Review would need some changes/corrections regarding its language and information such as "have failed," "unwillingness," and "has resisted" in paragraph 39 and paragraph 13 of the executive summary of the CAR and paragraph 8.27 of chapter 8 of the CAR, respectively. However, the Government of Bangladesh (GOB) is pleased to provide clearance to publish the report in English, provided the report is changed before publication as per suggestions made by the GOB participants in the meeting. With regards. Yours sincerely, (Rokeya Sultana) Deputy Secretary 51  ENDNOTES Chapteri Chapter4 1. Average per family aid receipt is equivalent to about six 1. Bangladesh has three growing seasons, and hence three months' wages for the poorest adult workers. rice crops per year. "Aus" is rice planted at the start of the wet 2. For a discussion of the increasingly centralized civil ser- season and harvested while the water in paddy fields is still shal- vice administration and the burden this imposes, see World low. "Aman" rice grows in deep water and is harvested at the Bank 1996e. end of the wet season. "Boro" rice is planted and harvested dur- ing the dry season, and hence is dependent upon irrigation. The Chaper 2land area that is triple-cropped increased between 1985 and Chapter 21991 from 1.4 to 2.3 million hecares. 1. The Region notes (with some justification) that at this early stage of IDA's involvement with the SOEs and the coun- try's financial sector, it was unreasonable to expect a fuller understanding of both sectors' deficiencies. Certainly by the 1. The Region believes, however, that there was a genuine mid-1980s, IDA had a much stronger appreciation of the deep- attempt to encourage successive governments to make hard seated malaise in both sectors. political choices, and given the importance of the SOEs and their impact on the budget, IDA persistence in seeking change Chapterwas warranted. This is undoubtedly a valid point. Nevertheless, Chaper 3OED maintains that a stronger focus on key sector and subsec- 1. By the end of FY97, only two projects were classified as tor policy reforms and actions is a more effective way of using problem projects, and both were near closing. IDA resources in the sector. 53  BIBLIOGRAPHY Bangladesh Institute of Development Studies. 1996. and Infrastructure Operations Division. Dhaka. "Rural Poverty in Bangladesh 1987-94." Dhaka. . 1993d. "A Report on Power Sector Reforms Bhattacharya, Debapriya. 1996. "A Tale of Two Sis- in Bangladesh." Inter-Ministerial Working Group, ters: Two Decades of the World Bank and the Inter- Ministry of Energy and Mineral Resources. Dhaka. national Monetary Fund in Bangladesh." Contem- 1993e. "Summary of Financing in the FPHP porary South Asia. Health Sector Projects." Ministry of Health and BRAC. 1996. BRAC 1995. Dhaka. Family Welfare. Dhaka. 1995. "BRAC's Health Programmes." Dhaka. 1994a. "Progotir Pathey. Progress Towards Canadian International Development Agency. 1993a. the Achievement of Goals for the 1990s." Bangla- Small-Scale Water Control Structures Project. desh Bureau of Statistics and Ministry of Planning Operational Status Survey. Vol. I. Main Report. and UNICEF Dhaka. Bangladesh Water Development Board. . 1994b. "Flow of External Resources into . 1993b. Small-Scale Water Control Structures Bangladesh." Economic Relations Division, Mi- Project. Operational Status Survey. Vol. II. Subpro- istry of Finance. Dhaka. jected Briefs. Bangladesh Water Development Board. . 1995a. "National Institute of Population Carty, Winthrop P. 1993. "Success in a Challenging Research and Training (NIPORT), an Overview." Environment: Fertility Decline in Bangladesh." Health Economics Unit, Ministry of Health and Population Reference Bureau. Washington, D.C. Family Welfare-A.K.M. Rafiquz Zaman. Dhaka. Centre for Health and Population Research. 1995a. . 1995b. "National Energy Policy." Ministry of 1995 Annual Report. Energy and Mineral Resources. Dhaka. 1995b. Biennial Work Plan 1995-96. . 1996a. "Bangladesh-Directory of Education 1996. "The Role of Pharmacies in Providing Programmes of the NGOs." Dhaka. Family Planning and Health Services to Residents . 1996b. "User Fees, Self-Selection and the Poor of Dhaka, Bangladesh." Working Paper 21. in Bangladesh." Health Economics Unit, Ministry Choudhry, Yusuf, Shawkat Ali, and R. Faruqee. 1995. of Health and Family Welfare. Dhaka. Institutional Development in the Water Sector. 1996c. "National Minor Irrigation Develop- World Bank Discussion Paper. Washington, D.C. ment Project." Ministry of Agriculture. Dhaka. Dahl, Henrik, and Pradeep Mitra. 1990. "Applying 1996d. "A Status Report on Bangladesh Tax Policy Models in Country Economic Work: Fourth Population and Health Project for the World Bangladesh, China, and India." World Bank Coun- Bank and Co-Financers." Ministry of Health and try Economics Department Working Paper 504. Family Welfare. Dhaka. Washington, D.C. . 1996e. "Memorandum for the Bangladesh Dubey, Vinod S. 1995. Impact of Economic and Sec- Aid Group 1996-97." Ministry of Planning and tor Work in South Asia: A Client Perspective. Ministry of Finance. Dhaka. Washington, D.C.: World Bank. Halcrow, William, & Partners Ltd. 1996. "National Economist Intelligence Unit. 1996. Country Report- Minor Irrigation Development Project: The Cost of Bangladesh. Series 1st Quarter 1995-2nd Quarter Power for Minor Irrigation." Government of 1996. London. Bangladesh, Ministry of Agriculture. Dhaka. Gisselquist, David. 1996. "Liberalizing Agricultural Hasan, Abid, and Owaise Sadat. 1996. "Bangladesh- Inputs Trade: Impact and Opportunity." World Government that Works. Reforming the Public See- Bank, Washington, D.C. tor." World Bank Private Sector Development and Government of Bangladesh. 1992a. "Flood Action Finance Division. Washington, D.C. Plan-North West Regional Study." Dhaka. Hino, Toshiko. 1996. "NGO-World Bank Partner- - . 1992b. Bangladesh Issues and Options in the ships: A Tale of Two Projects." World Bank Gas Sector. Vol. 1: Executive Summary and Main Human Capital Development Working Paper 67. Text. 10484-BD. Energy and Infrastructure Opera- Washington, D.C. tions Division. Dhaka. Hossain, Nadim. 1995. "The World Bank's Opera- - . 1992c. Bangladesh Issues and Options in the tions and Role in Bangladesh." World Bank, Wash- Gas Sector. Vol. M: Annexes. 10484-BD. Energy ington, D.C. 55 Bangladesh: Progress Through Partnership Humphrey, Clare E. 1990. Privatization in The Daily Star. 1995. "25 Years of Partnership." Bangladesh: Economic Transition in a Poor Coun- Dhaka. try. Boulder, CO: Westview. TITAS Gas. 1994. "TITAS Gas Transmission and Distri- IME 1991. "Bangladesh: Economic Reform Measures bution Company Limited-Annual Report 1993-94." and the Poor." Working Paper 91/39, Asia Depart- UNICEF. 1996. The Progress of Nations. New York. ment. Washington, D.C. USAID. 1996a. "Privatizing Fertilizer Distribution: Jarrett, Stephen. 1994. "Visit to Bangladesh: Health Bangladesh Case Study." Report 54. Washington, Systems Review at the Perilphery, 25 February-8 D.C. March." Health Systems Development, UNICEF. . 1996b. "Rural Electrification in Bangladesh-Final New York. Report." Part I: "Assessment of Post-1996 Technical Khan, M.R. 1989. "Sicknesses, Diseases, Treatments Assistance Needs." Center for Environment, Office of and Medical Costs by Socioeconomic Variables in Energy, Environment, and Technology, Washington, Bangladesh." Report 15, Bangladesh Institute of D.C. Development Studies. Dhaka. . 1996c. "Rural Electrification in Bangladesh-Final Krishi Bank. 1994. "Krishi Bank Parikrama." Yearly Report." Part 11: "Assessment of Power Generation by Journal of Bangladesh Krishi Bank. the Rural Cooperatives." Center for Environment, Office Manor, James. 1996. The Political Economy of of Energy, Environment and Technology, Washington, Decentralization. Washington, D.C. : World Bank. D.C. Narbye, O.D.K. 1989. "Internal Inconsistencies in Watanagase, Tarisa. 1996. "Banks in Distress: The Agricultural Statistics." In Bangladesh Agriculture: Case of Bangladesh." IMF Working Paper 90/81. Performance and Policies. Dhaka: UNDP. Washington, D.C. Papanek, Gustav E 1987. Fiscal Policy and Develop- World Bank. 1978. Bangladesh: Current Economic ment Strategy in Southern Asia. World Bank Dis- Trends. Vol. 11: Population Growth and Planning. cussion Paper DRD278, Development Research Kantoor Pisharoti. Report 1931. Washington, D.C. Department. Washington, D.C. . 1987. Bangladesh: Promoting Higher Growth Rural Electrification Board. 1994. Rural Electrifica- and Human Development. Washington, D.C. tion Board-Annual Report 1993-94. Washington, . 1990a. Bangladesh: Strategies for Enhancing D.C.: National Electric Cooperative Assn. the Role of Women in Economic Development. Selman, Sanford J. 1996a. "Bangladesh Private Power Country Study P-7899. Washington, D.C. Program-Guide to the Negotiation Process." . 1990b. Bangladesh: Review of the Experience World Bank, Energy and Project Finance Division with Policy Reforms in the 1980s. Operations Evalua- of South Asia Region, Washington, D.C. ton Department Study 8874. Washington, D.C. - . 1996b. "Bangladesh: Review of Unsolicited * 1992. Population and the World Bank-Implica- Offers for Build-Own-Operate Power Plants." tions from Eight Case Studies. Operations Evaluation World Bank, Energy and Project Finance Division Department. Washington, D.C. of South Asia Region, Washington, D.C. . 1994. World Development Report 1994: Shamannay, Unnayan. 1996. "Rural Electrification Infrastructure for Development. New York: Board of Bangladesh." Nreca International Ltd.; Oxford University Press for the World Bank. Sheladia Associates, Inc. . 1996a. "Partnerships, Possibilities and Pit- Sobhan, Rehman. 1982. "The Role of the World Bank falls: Working with Local NGOs in Bangladesh." in Bangladesh." In The Crisis of External Depen- ASHR, Reidar Kvam. Washington, D.C. dence: The Political Economy of Foreign Aid to . 1996b. Bangladesh: An Agenda for Action. Bangladesh. London: Zed. Washington, D.C. - . 1995. "The Crisis of External Dependence. The . 1996c. "Economic and Sector Work and Political Economy of Foreign Aid to Bangladesh." In Results on the Ground. "Operations Policy Experiences with Economic Reforms. Dhaka: Center Department. Washington, D.C. for Policy Dialogue. . 1996d. "Performance Indicators for Technical Stalker, Peter. 1994. Bangladesh: Out of the Shadows. Assistance Operations." Operations Policy Depart- Washington, D.C.: World Bank. ment. Washington, D.C. 56 Bibliography 1996e. "Bangladesh: Government that Works." In Reforming the Public Sector. Washing- ton, D.C. WHO. 1996. "Reproductive Health Care in Bangladesh: A Review of Recent Data and Some Proposals for Future Strategies and Activities." WHO: Reproductive Health Unit, Dhaka. Zachan, S. 1994. "Services at Thana Health Com- plexes Unsatisfactory." Dhaka. 57  OPERATIONS EVALUATION DEPARTMENT PUBLICATIONS The Operations Evaluation Department (OED), an in- Documents listed with a stock number and price dependent evaluation unit reporting to the World code may be obtained through the World Bank's mail Bank's Executive Directors, rates the development order service or from its InfoShop in downtown Wash- impact and performance of all the Bank's completed ington, DC. For information on all other documents, lending operations. Results and recommendations are contact the World Bank InfoShop. reported to the Executive Directors and fed back into For more information about this study or OED's the design and implementation of new policies and other evaluation work, please contact Elizabeth projects. In addition to the individual operations and Campbell-Pag6 or the OLD Help Desk. country assistance programs, OED evaluates the Bank's policies and processes. Operations Evaluation Department Summaries of studies and the full text of the Precis Partnerships & Knowledge Programs (OEDPK) and Lessons & Practices can be read on the Internet at E-mail: ecampbellpage@worldbank.org http://www.worldbank.org/html/oed/index.htm E-mail: OED Help Desk@worldbank.org Telephone: (202) 473-4497 How to Order OED Publications Facsimile: (202) 522-3200 Operations evaluation studies, World Bank discussion papers, and all other documents are available from the World Bank InfoShop. Ordering World Bank Publications The World Bank InfoShop serves walk-in customers Customers in the United States and in territories not only. The InfoShop is located at: served by any of the Bank's publication distributors may send publication orders to: 701 18th Street, NW Washington, DC 20433, USA The World Bank P.O. Box 960 All other customers must place their orders through Herndon, VA 20172-0960 their local distributors. Fax: (703) 661-1501 Telephone: (703) 661-1580 Ordering by E-Mail The address for the World Bank publication database If you have an established account with the World on the Internet is: http://www.worldbank.org (select Bank, you may transmit your order by electronic mail publications/project info). on the Internet to: books@worldbank.org. Please E-mail: pic@worldbank.org include your account number, billing and shipping Fax number: (202) 522-1500 addresses, the title and order number, quantity, and Telephone number: (202) 458-5454 unit price for each item. 59  OED SERIES 1998 Annual Review of Development Effectiveness 1997 Annual Review of Development Effectiveness Agricultural Extension and Research: Achievements and Problems in National Systems Developing Towns and Cities: Lessons from Brazil and the Philippines Financial Sector Reform: A Review of World Bank Assistance Fiscal Management in Adjustment Lending India: The Dairy Revolution Mainstreaming Gender in World Bank Lending: An Update Nongovernmental Organizations in World Bank-Supported Projects: A Review Paddy Irrigation and Water Management in Southeast Asia Poland Country Assistance Review: Partnership in a Transition Economy Reforming Agriculture: The World Bank Goes to Market The World Bank's Experience with Post-Conflict Reconstruction Zambia Country Assistance Review: Turning an Economy Around Proceedings Lessons of Fiscal Adjustment Lessons from Urban Transport Evaluation and Development: The Institutional Dimension (Transaction Publishers) Monitoring & Evaluation Capacity Development in Africa Public Sector Performance-The Critical Role of Evaluation Multilingual Editions Assessing Development Effectiveness: Evaluation in the World Bank and the International Finance Corporation Appreciation de Pefficacite du developpement : L'evaluation a la Banque mondiale eta la Societe financiere internationale Determinar la eficacia de las actividades de desarrollo: La evaluaci6n en el Banco Mundial y la Corporaci6n Financiera Internacional C6te d'Ivoire : Revue de l'aide de la Banque mondiale au pays Philippines: From Crisis to Opportunity Filipinas: Crisis y oportunidades Rebuilding the Mozambique Economy: Assessment of a Development Partnership Reconstruir a Economia de Mo(ambique 0 THE WORLD BANK 1818 H Street. N W. Washington, D.C. 20433, U.S.A. Telephone: (202) 477-1234 Facsimile: (202) 477-6391 Telex: MCI 64145 WORLDBANK MCI 248423 WORLDBANK Web: http //www.worldbank.org E-mail : books@worldbank.org 14293 MC C3- j)3 copies12I 170821 342930