A WORLD BANK Population OPERATIONS EVALUATION XEVAUAO and the World Bank IMPLICATIONS FROM EIGHT CASE STUDIES 0514 Xri ' X 2~~ Operations Evaluation DepartmentFiEcP Population and the World Bank IMPLICATIONS FROM EIGHT CASE STUDIES Other Titles in the Series PREPARED BY THE WORLD BANK OPERATIONS EVALUATION DEPARTMENT Forestry: The World Ban]k Experieince (1992; contains summaries in French and Spanish Evaluation Results for 1989 (1991; contains summaries in French and Spanish) The Aga Khan Rural Support Program in Pakistan: Second Interim Evaluation (1990; contains sunmaries in French and Spanish) Evaluation Results for 1988: Issues in World Bank Lending Over Two Decades (1990; also available in French) Agricultural Marketing: The World Bank's Experience 1974-85 (1990; contains summaries in French and Spanish Project Performance Results for 1987 (1989; also available in French) Renewable Resource Management in Agriculture (1989) Educational Development in Thailand: The Role of World Bank Lending (1989) Rural Development: World Bank Experience, 1965-86 (1988; also available in French and Spanish) Project Performance Results for 1986 (1988; also available in French) Cotton Development Programs in Burkina Faso, C6te d'Ivoire, and Togo (1988; also available in French) Sri Lanka and the World Bank A Review of a Relationship (1987) The Jengka Triangle Projiects in Malaysia Impact Evaluation Report (1987) The Twelfth Annual Review of Performance Results (1987; also available in French) The Aga Khan Rural Support Program in Pakistan An Interim Evaluation (1987) Population and the World Bank IMPLICATIONS FROM EIGHT CASE STUDIES Operations Evaluation Department The World Bank Washington, D.C. Copyright X 1992 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing June 1991 The opinions expresse!d 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. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. To present the results of evaluation with the least possible delay, the typescript has not been pre- pared in accordance with the procedures appropriate to formal printed texts, and the World Bank ac- cepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy por- tions for classroom use is granted through the Copyright Clearance Center, 27 Congress Street, Salem, Massachusetts 01970, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Publica- tions, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from Distribution Unit, Office of the Publisher, Department F, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116 Paris, France. ISSN: 1011-0984 Library of Congress Cataloging-in-Publication Data Population and the World Bank: implications from eight case studies. p. cm.-(A World Bank operations evaluation study, ISSN 1011-0984) Includes bibliographical references. ISBN 0-8213-2081-5 1. Developing countries-Population policy-Case studies. 2. World Bank-Developing countries-Case studies. I. International Bank for Reconstruction and Development. Operations Evaluation Dept. II. Series. HB884.P663 1992 363.9'09172'4-dc2O 92-13541 CIP Foreword This is the first assessment that the Operations and Eval- that should make the Bank's future population activities uation Department has undertaken of the Bank's role in the more effective. population field. It differs from evaluations undertaken by The study was written by Ronald Ridker of the OED others over the past 25 years in that it is based on a series of with substantial contributions from Ronald Freedman. It is detailed case studies of experiences in individual countries. based on a set of detailed case studies prepared by consult- This allows the authors more scope to investigate reasons ants working under Mr. Ridker's direction: Ronald Freed- for decisions and outcomes, and it provides a more solid man (Indonesia), Jay Satia (India). James Phillips foundation on which to build a picture of what can be done (Bangladesh), Warren Robinson (Kenya), Sara Seims (Sene- to improve performance in this area. gal), and Thomas Merrick (Colombia, Brazil, and Mexico). The study begins by reviewing and assessing develop- John Kantner provided valuable comments, especially on ments in eight countries, in each case asking how demo- the India case study. The major sources for these studies graphic and related social and economic developments were Bank documents, both published and internal, non- unfolded during the last 20-25 years and what role the gov- Bank studies and reports, field visits, and interviews with ermnent, other donors, and the Bank played during this pe- government officials and the staffs of both the Bank and riod. It tries to bring the picture up to date, asking whether other donor organizations. Governments of all countries in- recent changes in Bank approach adequately take the les- cluded in this study have had an opportunity to comment. sons of earlier experience into account. Using these materi- This study, plus two of the case studies, have been distrib- als as a starting point, the last section discusses major issues uted to and discussed by the Bank's Executive Directors; all of concern to the Bank that have implications for its future three are presented in this document. efforts in this field. The picture that emerges is one of considerable diversity of activities and experiences, from which it is not easy to Yves Rovani generalize. The presentation tries to give some feel for this Director General diversity without overburdening the reader. Nevertheless, Operations Evaluation the study offers concrete suggestions for a shift in emphasis February 1992 v Prologo Esta es la primera evaluaci6n que realiza el Departa- poblaci6n, lo que redundara en una mayor eficacia de di- mento de Evaluaci6n de Operaciones sobre la funci6n del chas actividades. Banco en el sector de poblaci6n. Difiere de otras evalua- El estudio fue preparado por Ronald Ridker del Depar- ciones efectuadas en los Ciltimos 25 afios en que se sustenta tamento de Evaluaci6n de Operaciones, con la importante en una serie de estudios detallados de casos practicos en colaboraci6n de Ronald Freedman. Se basa en un conjunto determinados paises. Esto permite a los autores un mayor de estudios detallados de casos prActicos preparados por campo de acci6n para investigar las razones en cuanto a de- consultores bajo la direcci6n del Sr. Ridker: Ronald Freed- cisiones y resultados, y proporciona una base mAs s6lida man (Indonesia), Jay Satia (India), James Phillips para establecer una visi6n de conjunto de lo que puede hac- (Bangladesh), Warren Robinson (Kenya), Sara Seims erse a fin de mejorar el desempenlo en esta esfera. (Senegal) y Thomas Merrick (Colombia, Brasil y Mexico). El estudio comienza por examinar y evaluar los cambios John Kantner proporcion6 valiosos comentarios, en espe- demograficos y socioecon6micos conexos que se han pro- cial sobre el estudio de casos de la India. La principal fuente ducido en ocho paises en los iiltimos 20 a 25 afios y la fun- de informaci6n de estos estudios fueron los documentos ci6n que han desempefiado los gobiernos, otros donantes y del Banco, tanto publicados como internos, estudios e in- el Banco en este periodo. Se intenta actualizar esta visi6n formes ajenos al Banco, visitas en el terreno y entrevistas general, y se plantea el interrogante de si los Cltimos cam- con funcionarios de gobiemo y del Banco y de otros organ- bios en el enfoque empleado por el Banco toman en cuenta ismos donantes. Los gobiernos de todos los paises inclui- en forma adecuada las enseiainzas derivadas de experien- dos en este estudio han tenido la oportunidad de formular cias anteriores. Utilizando esta informaci6n como punto de comentarios. Este estudio, mas dos de los estudios de casos partida, en la Ciltima secci6n se analizan los principales te- prActicos, se han distribuido a los Directores Ejecutivos del mas que revisten interes para el Banco y que tienen reper- Banco quienes los han examinado; los tres se presentan en cusiones en sus programas futuros en este Ambito. este documento. El panorama resultante es de una considerable diver- sidad de actividades y experiencias que hace dificil llegar a una generalizaci6n. La presentaci6n trata de dar una idea Yves Rovani de esta diversidad sin abrumar al lector. Sin embargo, el Director General estudio oftrece sugerencias concretas para un cambio de ori- Departamento de Evaluaci6n de Operaciones entaci6n en las actividades futuras del Banco en materia de Febrero de 1992 vi Avant-Propos C'est la premiere fois que le Departement de l1'valua- toutefois des suggestions concretes en vue d'inflechir l'ac- tion retrospective des operations (OED) entreprend d'etud- tion de la Banque et d'accroitre son efficacite future. ier le r6le joue par la Banque dans le domaine de la L'etude a et6 tablie par Ronald Ridker, de l'OED, qui a population. Cette etude differe de celles qui ont ete effec- beneficie d'un apport substantiel de Ronald Freedman. Les tuees ailleurs au cours des 25 demieres annees en ce sens monographies detaillees sur lesquelles elle s'appuie ont et qu'elle se fonde sur une serie de monographies detaillees r6digees par des consultants diriges par M. Ridker. Ce sont: consacr6es A divers pays. Les auteurs ont eu ainsi plus de Ronald Freedman (Indonesie), Jay Satia (Inde), James latitude pour examiner les raisons qui sont A l'origine des Phillips (Bangladesh), Warren Robinson (Kenya), Sara decisions et de leurs consequences. Le resultat obtenu con- Seims (Senegal) et Thomas Merrick (Colombie, Bresil et stitue une base de depart plus solide pour la recherche des Mexique). John Kantner a presente des observations moyens d'ameliorer les programmes de population. precieuses, particulierement sur le cas de l'Inde. Les princi- L'etude commence par un tour d'horizon et une analyse pales sources utilisees sont des documents intemes et des de la maniere dont la situation a evolue dans huit pays. Les publications de la Banque, des etudes et des rapports ex- auteurs ont suivi le deroulement des 6venements dans les ternes, des observations sur le terrain et des entretiens avec domainies demographique, social et economique au cours des membres du gouvernement interesse et des agents de la des 20 A 25 dernieres ann6es, et le r6le joue par les autorites Banque et d'autres organisations donatrices. Les nationales, les autres bailleurs de fonds et la Banque pen- gouvernements interesses ont tous eu la possibilite de for- dant cette periode. Ils ont tente de donner un apercu de la muler leurs observations. La presente 6tude, ainsi que deux situation actuelle en se demandant si les changements ap- des monographies, qui sont incluses dans ce document, ont portes recemment A l'approche de la Banque tiennent ete distribu6es aux Administrateurs de la Banque et compte comme il convient de l'experience acquise. A partir examinees par eux. de cela, la derniere partie analyse les principales questions qui se posent A la Banque dans l'optique de ses interven- tions futures. Yves Rovani 11 apparait que les activit6s et les r6sultats presentent Directeur gen6ral une diversite considerable, si bien qu'il est malaise de tirer Evaluation retrospective des conclusions generales. Les auteurs se sont efforces de des operations depeindre cette diversite sans lasser le lecteur. Ils proposent F6vrier 1992 vii Contents Forezvord v Prologo vi Avant-Propos vii Executive Summary I Resumen Ejecutivo 8 Resume Analytique 16 1. Introduction 25 2. Indonesia 27 Demographic Trends 27 Social and Economic Changes 27 Indonesia's Family Planning Program 28 World Bank Involvement 29 3. India 32 Demographic and Socioeconomic Trends 32 The Indian Family Welfare Program 33 T'he Bank's Role Prior to 1987-88 35 D)evelopments since 1987-88 37 4. Bangladesh 39 Demographic and Socioeconomic Trends 39 I'he Bangladesh Family Planning Program 40 The Foreign Assistance Community and the Bank 41 Implications and Lessons 43 5. Brazil, Colombia, Mexico 46 Demographic Trends, Development, and the Role of Government 46 External Assistance and the Role of the Bank 47 ix 6. Kenya 50 Demographic Developments and Principal Determinants 50 Foreign Assistance and the Role of the Bank 52 Implications and Lessons 55 7. Senegal 56 Demographic and Socioeconomic Trends 56 Population Policies and Family Planning Programs 57 Foreign Assistance and WVorld Bank Involvement 58 8. Major Issues for the World Bank 60 Neglect of Population Issues in Non-Population Sectors 60 Dialogue, Policy Promotion, and Sector Work 61 Project-related Issues 62 Donor Coordination and Involvement of Non-Governmental Agencies 65 Country Organization of Population and Family Planning Activities 65 Need for a Long-run Programmatic Approach 66 The Extent and Allocation of Bank Resources 66 Implications for Evaluating and Staffing Bank Population Activities 68 Figures 1.1 Case Study Countries, Birth, Death and Population Growth Rates, 1970-87 26 1.2 Case Study Countries, Total Fertility Rates, 1960-89 26 6.1 Family Planning Accessibility and Prevalence in Kenya: Travel Time, by Year Facility Started 52 8.1 Staff-weeks per Project, PHN vs. All Sectors 65 Statistical Appendix 1.1 Demographic and Socioeconomic Indicators, Selected Countries and Years 70 2.1 Donor Expenditures for Population Programs in Selected Countries, 1982-89 73 3.1 Summary Data for Population and Population-related Projects in Case Study Countries, FY7O-FY90 74 3.2 Project Costs by Component in Population and Population-related Projects for Case Study Countries 75 3.3 Appraisal Project Costs and Bank Financing, by Category of Expenditure, for Population Projects in Case Study Countries 76 4.1 Bank Lending by Sector, Case Study Countries 1970-90 77 5.1 Civil Works, Furniture, and Equipment in Bank Population Projects, Appraisal Estimates 79 6.1 Commitments for International Population Assistance by World Bank and Other Donors, 1952-89 80 References 81 Annex 1: The World Bank and Bangladesh's Population Program 85 Annex 2: The World Bank and Indonesia's Population Program 125 x Acronyms ADB Asia Development Bank ASCOFAME Association of Colombian Medical Schools BAPPENAS Indonesia National Development Planning Agency BEMFAM Brazil's IPPF affiliate BFS Bangladesh Fertility Survey BKKBN Indonesian National Family Planning Program BR Crude birth rate (total live births per 1,000 population) CONAPOP National Population Commission of Senegal CPR Contraceptive prevalence rate (percent of married couples in reproductive age group using some means of contraception) DHS Demographic and health surveys DR Crude death rate (total deaths excluding stillborn per 1,000 population) FP Family planning GDP Gross domestic product GOB Government of Bangladesh GOK Government of Kenya HRD Human resources development IDA International Development Association IEC Information, education, and communication IFS International financial statistics INPRES Indonesia National subsidy for local governments IPPF International Planned Parenthood Federation KAPS Knowledge aptitude and practice survey KDHS Kenya Demographic Health Survey LDC Less-developed country MCH Maternal and child health MIES Management information and evaluation system MIS Management information service MOH Ministry of Health MOHFW Bangladesh Ministry of Health and Family Welfare (formerly Ministry of Health and Family Planning) MOHPP Bangladesh Ministry of Health and Population Planning MSD Ministry of Social Development NCPD National Council on Population and Development of Kenya NICIPS National Indonesia Contraceptive Prevalence Survey NGO Non-governmental organization NPCC Bangladesh National Population Control Council NPFP National Program in Family Planning NRR Net reproductive rate OECD Organization for Economic Cooperation and Development OED Operations Evaluation Department ORG Operations Research Group, Baroda, India PAIP Population Action and Investment Plan PCR Project completion report PHC Primary health center PHN Population, health and nutrition PHR Population & human resources PPAR Project performance audit report PROFAMILIA Colombia's IPPF affiliate REPELITA Five-Year Development Plan RMB Resident Mission in Bangladesh SMP Social Marketing Project xi TFR Total fertility rate (number of live births per woman expected over a women's lifetime if her child-bearing experience at each age is the same as that of existing women at various ages) UNESCAP United Nations Economic and Social Commission for Asia and the Pacific UNFPA United Nations Fund for Population Activities UNICEF United Nations Children's Fund USAID United States Agency for International Development VCDC Village contraceptive distribution centers WHO World Health Organization WID Women in Development xii Executive Summary This study examines the Bank's activities in the popula- infant and child mortality in particular, have also declined tion sector by reviewing its role in eight countries. For each substantially-in the Latin countries, by about 50 percent country, it begins by asking what demographic and related since 1960. However, fertility, mortality, and population socioeconomic changes have occurred since 1968, when the growth rates remain high by most standards, especially in Bank announced that it would begin lending for this sector. Asia and Africa. Given the very young age structure of It also asks what government policies affecting these trends these countries, plus present rates of change in fertility and were instituted and what role was played by the Bank vis- mortality, their population growth rates are likely to remain A-vis other donors during this period. The conclusions near or above 2 percent per year for some time to come. reached are based on field visits, interviews, and a review Between two thirds and three fourths of this fertility de- of operations documents and literature. cline has been due to increased use of modem contracep- The picture that emerges is one of considerable diversity tive methods. Most of the remainder of the decline is of activities and experiences, from which it is not easy to accounted for by increasing age of marriage. These chang- generalize. Leaving this diversity and the qualifications es, in turn, have resulted from a combination of improve- aside for the moment, the study's overall conclusion is that, ments in social and economic characteristics favoring after slow, halting, and sometimes faulty starts, the Bank smaller families and later age of marriage. These include, has become progressively more effective in this field, for example, improvements in the educational, employ- though it could have done and achieved more. From this ment, and social status of women; declines in infant mortal- point onward-assuming forceful and consistent imple- ity; decreasing opportunities for child labor; and mentation of recent changes in the Bank's mode of operat- decreasing dependence on children for old age security. ing in this sector-most of what is required to effect the They also include family planning (FP) programs which en- desired demographic changes implicit in Bank sector policy deavor to make modem contraceptives readily available involves re-orientations and shifts in emphasis rather and provide information and education about them. than radical departures from recent practices and levels of In the three Latin countries, both factors-significant im- funding. provements in these social and economic indicators and sizeable and effective family planning services, in these cas- Demographic and Socioeconomic Trends es provided by non-governmental organizations (NGOs), quasi-public organizations, and commercial channels- Total fertility rates (TFRs) have declined in all case study have combined to bring about the observed demographic countries: marginally in Senegal; 20 percent of the distance changes. Indonesia has also experienced sizeable social and between the peak level of eight children per woman to a re- economic improvements, but remains a more rural, tradi- placement level of 2.1 in Kenya; 40 percent in Bargladesh; tional society with higher maternal and child mortality 50 percent in India; 60 percent in Indonesia; and mhore than rates. Its government-sponsored family planning program 70 percent in Brazil, Colombia, and Mexico. This news is has compensated for this difference in a variety of ways. particularly encouraging in the case of Bangladesh and Perhaps the most important of these is the large and effec- Kenya where, until very recently, there has been little to tive network of outreach workers extending down to the show for the efforts that have been made over the last smallest village. The Indian program has also compensated 20 years to reduce population growth rates. Mortality rates, for even more modest social and economic progress, but in I a different way: by focusing quite narrowly but effectively fashion; provided little in the way of technical assistance in on sterilization. Its program, however, has been set back at the preparation or implemention of projects in this field; times by political backlash resulting from periodic coercive and within projects, focused on providing civil works and campaigns. Because of its success in reaching a large frac- hardware. The Bank did help develop and support a major tion of couples desiring to terminate the reproductive population education program. In other software areas phase of their lives, the Indian program is now reaching a however, it tended to support implementation of projects point of diminishing returns. The Kenyan case is interest- developed by the Indonesian National Family Planning ing because the substantial social and economic progress of Program (BKKBN) with the technical assistance of other the last 25 years seems to have built up a pool of latent donors. For most of the period covered, there was no resi- demand for limiting family size which has not become op- dent technical staff member responsible for this sector and erative until very recently. A long period of program devel- sector work was limited. opment which has finally reached a critical threshold of Except perhaps for the first few years of its involvement effectiveness, plus active encouragement from the with the program, the Bank's approach was perfectly ap- President and other major leaders, finally appears to be propriate. The Indonesian authorities knew what they making a difference. Senegal provides an example of a wanted and proved to be good at implementation. Indeed, country with too little social. and economic development the BKKBN is commonly considered one of the most effec- and too little program development to have a significant tive family planning organizations in the world. Other do- impact on fertility. nors-in particular, the United States Agency for The Bangladesh case is particularly interesting because International Development (USAID)-were very active in there has been substantial prDgram development but very developing operational policies and implementation tac- little social and economic progress-an almost classic case tics. This proved to be a good de facto division of labor in which a supply side approach was utilized in a situation which was well-appreciated by the Indonesian authorities. in which one would not expect to find much demand. The report discusses five areas in which the Bank might Sufficient latent demand mus t have been present even here, have done more or operated differently in Indonesia. (1) A however, to lead to the contraceptive prevalence and fertil- Bank specialist resident in the country for a longer period ity decline that have occurred so far. Just how much longer of time and with a mandate to become actively involved in progress can continue in the absence of significant program development and implementation would have structural changes that affect demand is an open question.1 been useful. (2) Cooperation and collaboration with other The paper considers several possible explanations and calls donors could have been better during the first 15 years of for an investigation. Such research is important for the Bank participation. Since then, this aspect of the program future of the Bangladesh program and also for what it has improved substantially (3) The Bank could have en- might say about appropriate programs for countries with couraged earlier and stronger initiatives in the health sec- similar circumstances. tor, especially by more forcefully promoting maternal and child health (MCH) and its integration with family plan- The Bank's Role in Individual Countries ning. It is just now beginning to do this. (4) The Bank could have encouraged the development of non-family planning Indonesia. While the Bank has provided quite substantial strategies to enhance demand for smaller families, for ex- financial assistance to Indonesia's overall development ef- ample, by helping develop projects aimed at improving fort, lending for population has amounted to just 1 percent women's educational and employment opportunities. of the total. This amount has contributed about 10 percent Substantially more effort is still needed in this area. (5) to the total expenditures of the Government on its popula- Research and evaluation continues to need strengthening tion program. Other donors, including NGOs, have con- for at least two reasons: to extract lessons from this success- tributed another 20 percent. In providing these funds, the ful case for use elsewhere2 and to build research and Bank played a fairly traditional role, in the sense that it op- evaluation capacity into BKKBN where this function has erated more as a bank than as a development institution. been neglected. That is, by and large, it reacted to project proposals submit- Improvements in these areas were obviously not re- ted to it by the Indonesian authorities in an "arms-length"- quired, however, for the substantial progress that has been made so far. Moreover, Bank efforts in these spheres might 1. This way of stating things implicitly classifies information, education, have been resisted by BKKBN and other donors who pre- and communications (IEC) programs as being a part of a family planning program; that is, a supply-side measure. The idea behind this is that while IEC can disseminate information and encourage demand, there must be 2. For example, what factors explain this country's successful community- some prochvity already present-some latent demand created by more fun- based operations? Are they unique to Indonesia, or can they be applied to as- damental cultural, social, or economic forces-for these efforts to work on. sist other countries? 2 ferred the Bank to focus mainly on provision of resources accountability structure to give more weight to non-termi- for buildings and equipment. The Bank responded specifi- nal methods. Nor were there sufficient staff inputs-not in cally to the client's desires, and in this case, it was the right residence and also not in Washington-to do much more response, because of the considerable capacity of the client than pursue an arms-length approach. and the assistance of other donors. Since 1987, several significant changes have occurred. In Indonesia, the future may require different approach- Both the Government and the Bank appear to have become es from those of the past in three important respects. First, disappointed with results of the area projects and are will- there is much less need for massive infrastructure pro- ing to try a different approach. The sixth and seventh grams. Second, USAID has signaled its intention to scale projects, started in 1989 and 1990, reflect this through their back its inputs, in part because it believes Indonesia now focus on priority program components at the state level. needs less technical assistance and grant funds. BKKBN, They also include support for NGOs and social marketing. with USAIt! assistance, has responded by establishing pro- Second, since 1987, several good pieces of sector work have grams to promote private marketing and service delivery. been produced which appear to be having an effect on In- Third, the formula that has worked to bring fertility down dian thinking and the nature of the dialogue. The proposed from 5.5 to 3.5 children per woman may not work to lower Child Survival and Safe Motherhood Project for the first it further to a replacement level of 2.1. The Bank could ad- time focuses major attention on health issues of importance just to thlese changes by scaling back its activities as USAID for influencing fertility decisions. It also includes more de- is doing or by trying to fill the gap left by USAID. The re- tailed attention to field-level operations, which in.the past port suggests that, depending on the issue, both approach- have proven a major bottleneck to implementing policy es are needed. changes. Finally, Bank efforts in the education field-pay- India. As in Indonesia, until recently, the Bank has per- ing particular attention to women-have increased sub- formed more like a traditional bank in India than a devel- stantially. opment institution. In this case, however, this role was While these are all promising initiatives, much will de- much less appropriate because of strategic problems with pend on how forcefully and consistently they are pursued. the program. These problems include: excessive focus on The Bank will have to continue producing sector work and sterilization, which ignores younger couples' needs for research aimed at developing practical ways to resolve spacing and (as noted above) is running into diminishing structural problems; continue arguing-forcefully when returns; insufficient attention to program operations and necessary-for the application of the results of such work; quality (as opposed to program expansion); excessive cen- put more effort into enlisting the help of other donors in tralization and reliance on a single delivery system; and ne- this process; and increase the amount of time and effort de- glect of factors that can influence demand. These problems voted to supervision of field-level operations. All this is have persisted and been recognized for at least as long as likely to require more inputs by staff conversant with the the Bank has been involved in the Indian program. substantive issues, both at headquarters and in the field. The Bank's failure to have a significant influence on the Bangladesh. The Bangladesh program suffers from many Indian program is attributable to a variety of factors. of the same weaknesses as the India program. While there (1) The Bank's first population project in India started in has been less focus on sterilizations, the quality of family 1972, long after the country's approach was firmly en- planning services is poor, the MCH program is weak and trenched. (2) While the amounts lent constituted 28 percent not well- integrated with family planning, and scarce man- of the Bank's entire population portfolio up to 1988, these agerial talent plus ingrained administrative pattems make sums represented a small fraction-3.6 percent between improvements difficult. In addition, of course, the program 1980 and 1988-of total expenditures on India's Family operates in an extremely difficult physical environment in Welfare Program. All donors together contributed 12-14 which periodic crises which require priority constantly percent of government expenditures but have never threaten to wipe out progress. formed themselves into an effective coalition. (3) The na- Nevertheless, most of the progress achieved in increas- ture of the first five projects, called area projects, inhibited ing contraceptive prevalence and reducing fertility is attrib- the Bank from becoming involved in policy issues lying outside these (geographical) areas and discouraged experi- mental innovations even within those areas.3 (4) Both sec- 3. This is a disputable point. The original argument for this approach was tor wo:rk and staff inputs into the program have been that the Bank would have more influence within these areas than it could have in the overall program. It is more likely that it did not happen this way, inadequate. Prior to 1988, little sector work was done-cer- not because the concept was wrong, but because it was never accepted on the tainly not enough to provide the Bank with a basis for pro- Indian side that Bank-supported areas could be different from others, and posing tactical changes, for example, suggesting just how because the Bank never took the initiative to develop detailed designs for these projects that would have spelled this point out prior to implements. the program could change its incentive, monitoring, and tion. 3 utable to program inputs. The outreach program, which is economic improvements, further declines would be more more extensive and active than that in India despite its lim- certain and more rapid if structural changes could be itations, is probably the feature that makes the difference. achieved that enhance the demand for smaller families. In addition, donors have played a more intensive and ac- Brazil, Colombia, and Mexico. In contrast to other coun- tive role in both planning and operations; in effect, expand- tries, these countries received only a small part of their ex- ing the Government's implermentation capacity many-fold. penditures on family planning from external sources, and The Bank, which in this case has played an active, leading most of that was provided by NGOs rather than multilater- role, deserves considerable credit. al or bilateral agencies.4 Bank inputs were minimal. There The report lists seven factors that appear to have been have been no Bank- financed population projects in these critical. (1) The Government is clearly committed to solving countries. (One was prepared for Mexico, but was with- its population problem; has devoted a sizeable percentage drawn when a new government came to power.) Little was of its limited budget and managerial talent to the program; done to implement the few family planning subcompo- and has actively requested and utilized foreign assistance nents included with MCH in Bank-funded health projects, and advice. (2) The magnitude of foreign financial and tech- and policy dialogue seldom raised the topic. At least up to nical assistance has been substantial. (3) A considerable 1989, there has been very little sector work focused directly amount of the financial assistance, even within the five on population or family planning. In addition, lending to Bank-assisted population projects so far agreed to, has been other sectors, while it undoubtedly has had some indirect in the form of grants provided by co-financiers. This has al- impacts, was never undertaken with demographic consid- lowed more flexibility and greater focus on software com- erations in mind. ponents and has helped pay for additional Bank staff Three factors appear to account for this apparent neglect. inputs, both at headquarters aind in the field. (4) With these (1) Political sensitivities about "population control"policies additional staff inputs and with the encouragement of the of foreigners made it difficult for Bank staff to broach the Government, the Bank has plazyed an active role in all phas- topic with governments. These governments viewed Bank es of these projects-design, fund-raising, donor coordina- management as holding an unacceptable view and Bank tion, implementation, and evaluation-far more active than staff, caught in the middle, reacted by doing little. (2) By the is typical in Bank- assisted projects. (5) Collaboration time the Bank began lending for health services, which amongst donors has been substantial and co-financiers, might have provided a more acceptable approach to FP, who have typically provided more than half the funds in contraceptive prevalence was already comparatively high these projects, have played an increasingly active role. and the structure of service delivery well-established. Their involvement has now reached the stage where the (3) The lead agencies in these countries were NGOs or fourth project can be described more accurately as a semi-public agencies. Given government attitudes, the consortium project than a Bank project. (6) With the help of Bank lacked a way (unlike USAID) to work directly with the grant funds, flexibility has been built into the program these agencies or even to work indirectly with them, as it by, among other things, establishing an "Innovative has found in Kenya and elsewhere. Projects Fund"and a "Project Finance Cell."(7) Advantage The Bank's current approach in Latin America is to focus has been taken of findings from pilot projects to improve on reproductive health and safe motherhood as the ratio- program performance. nale for family planning. This seems to be appropriate. A In this case, there are reasons to believe that program more aggressive approach is unwarranted and politically efforts should continue along present lines for the next few unwise. However, important issues of equity, service qual- years. Surveys indicate the existence of a pool of potential ity, and choice remain and are, in any case, part of a broader users who have not yet been reached and evidence that set of issues the Bank is working on in the health sector.5 prevalence is higher where density of service points is These Latin American cases are important because they greater, where frequency of contacts with family planning bring to the fore questions about the rationale of Bank lend- workers is greater, and where better-quality services are ing for population. If the Bank wants to work in countries provided. However, even if prevalence is increased from its that do not accept population control as the rationale, it current level of 35 to the 50 percent level obtained in the must base its population program on a broader and more Matlab pilot area-very unlilcely on a national scale-the flexible set of principles. This could start from a recognition total fertility rate would be aibout 3.7, still well above re- placement. To achieve lower levels in a reasonable period, a change in program directions to include more than the 4. Some international NGOs, however, have been supported from bilateral supply-oriented approach currently in use is likely to be and multilateral sources; in particular, USAID and United Nations Fund for needed. While desired familLy size appears to have de- Population Activites (UNFPA). 5. One dimension of such issues is the very high numbers of illegally-in- creased in Bangladesh even without significant social and duced abortions in these countries. See Singh and Wolf, 1991. 4 that the overall objective is promotion of sustainable im- fairly strongly on many of these issues and seems to have provements in living standards, and that family planning taken advantage of opportunities as they emerged. In this prograrms assist at the micro level-by improving family case, the absence of resident staff assigned to the sector health and choice-as well as at the macro level. In this does not appear to have been a serious barrier to progress, context, these programs are appropriate topics for policy given the liberal budgets for supervision and travel and in- dialogue and sector analysis to determine how, if at all, they puts from other donors with technical staff in the field. should be developed and used in specific circumstances. In The program appears to be more or less on the right track addition, these cases raise questions about the advisability for the time being and should continue to expand, though in some circumstances of lending for free-standing with substantially more emphasis placed on quality of ser- population projects, as opposed to health projects that in- vices, outreach, and IEC. However, even if the gap between clude family planning components. Had this approach desired and actual family size were totally closed, the TFR been used earlier and more consistently, the Bank may not would decline to only the 4-5 child range, which implies a have faced so much opposition and suspicion of its motives population growth rate that is still above 2 percent per year. in these countries. Between 1984 and 1987, the mean desired number of chil- Kenya. This case study provides a good illustration of the dren decreased from 5.8 to 4.4. To accelerate such change, virtue of-and need for-patience and perseverance in the something more than supply-side changes are needed. The population field. While donor efforts to persuade the report suggests some possibilities, but also indicates that Government to establish a major program have been research and pilot studies are extremely important in this substantial since the late 1960s, it was only in 1988-89 that case because of our limited knowledge about how to make clear evidence of the beginnings of a fertility decline began significant progress in reducing desired family size in the to emerge. face of a strong, pro-natalist family system. Besides supporting general economic development that Senegal. Apart from a rural health project designed in included substantial education funding, the Bank's inputs 1982 that focused on the provision of buildings and equip- have proceeded along two general lines. Starting in 1974, ment for expansion of basic health services, the Bank, four projects have been funded. The first two provided until recently, concentrated its efforts on helping the Gov- funds to the Ministry of Health (MOH) to establish a net- emient develop a comprehensive population policy. A work of rural health facilities and training schools; family 1985-86 sector review, which concluded that USAID and planning components in these projects were weak and UNFPA were doing as much as could usefully be done to largely ignored during implementation. During this peri- develop FP services, recommended this focus. It was imple- od, the population growth rate actually increased. The third mented, among other ways, by making development of and fourth projects have had more substantial components such a policy statement a condition of a structural adjust- related to population and family planning, but were initiat- ment loan. While some government officials may have ed too late (1988 and 1990) to have influenced the change in gone along with this request because it appeared innocuous fertility trends. The second strand of Bank inputs has in- enough, the policy statement and the process of developing volved policy dialogue, sector work, and efforts to influ- it have taken on a life of their own and resulted in substan- ence the evolution of the program in a variety of ways. tive progress that would have occurred more slowly in its Examples of the latter included successful efforts to per- absence, given stagnant social or economic development in suade the Government to establish an inter-ministerial co- recent years. This strategy may be especially important in ordinating -agency for population outside the MOH Francophone Africa, because of its pro-natalist colonial her- (included as a condition in the second structural adjust- itage that had to be explicitly and publicly broken with to ment loan), and then to provide it with more responsibili- legitimize family planning activities. ties; and efforts to persuade the MOH to liberalize The decision by the Bank to focus on policy development guidelines for providing contraceptives, to integrate FP was probably correct at the time because the Bank, unlike into the mainstream of MOH activities, and to offer steril- other major donors, lacked field staff to help implement ization services. Both strands of work have been important more complex operations, but did have access to senior pol- components of an overall package that included technical icy circles because of its much broader policy agenda. More assistance and commodity support from other donors and, recently, the Bank has decided to become more pro-active perhaps most important, clear signals from Kenya's presi- by developing a project that, among other things, aims to dent that family planning was to be taken seriously. extend family planning and other services to the lowest tier Could the Bank have done anything to speed up this pro- in Senegal's health delivery system. This is a risky ap- cess? While far more effort could have been made to devel- proach. USAID, for example, has restricted itself to operat- op effective IEC, outreach, monitoring, evaluation, and ing at the next level up because the Government's capacity research components, the Bank pressed the Government to administer and deliver services is so limited. Given the 5 urgency of the problem, these risks are probably worth tak- is well established, for example, that educating women-or ing and can be kept within acceptable bounds by starting simply keeping them in school for a few more years-en- on a small scale and escalating slowly; but they require courages later marriage and lower fertility rates within much more intensive and more intimate supervision than marriage. Other promising interventions include programs has been typical of Bank projects in Senegal. specifically directed to raise the probability that children will survive to adulthood, opening up job opportunities for Major Issues for the Banlc women in occupations that are competitive with child rear- ing (e.g., factory or office jobs rather than farm jobs), and The experience of the past 20 years clearly demonstrates improving social security programs so that adults feel less that even in poor countries thLat have not experienced much need of large families to provide such security. social and economic progress, a process of fertility decline Previous reviews of the Bank population program have can be initiated by a typical family planning program that called for more emphasis on such demand-oriented factors, focuses on the provision of contraceptive supplies, services, but very little has been done to turn this call into operation- and associated information. ][t will proceed more rapidly if al programs. Reasons include skepticism about the impor- that program includes a strong outreach component and tance of the argument for the necessity of demand-side provides high-quality services based on clients' percep- interventions; doubts about what can be accomplished; and tions of needs. This requires good field supervision, good staff compartmentalization and inertia. Underlying these training, and good motivation-all features that are diffi- factors may be the fact that population is typically treated cult to develop in poor, rural areas. as a sector with responsibility assigned to a specific admin- However, there is little evidence that even the best run of istrative unit, implicitly leaving other sectoral units free such programs can do much more than meet the needs of from this responsibility. Population is not a discrete sector those who are already predisposed to accept family plan- any more than economic development or poverty allevia- ning services.6 In all the cases studied, even if all such per- tion. All three are more in the nature of strategic objectives sons were enrolled in a family planning program-or even that should be the responsibility of all sectors. if the highest contraceptive p;revalence rates achieved in the Since 1987, several changes have occurred that are hav- best-run pilot programs were achieved throughout the ing promising impacts on this situation within the Bank. country-there is no country in the group of eight, except The administrative integration of education, population, perhaps the three Latin American countries, in which the health, and nutrition has allowed and encouraged thoughts total fertility rate would drop to acceptable levels.7 Other about how each sector can help achieve the others' goals. kinds of interventions will be required to go farther. Sup- The Women in Development and Child Survival initiatives are ply-oriented strategies may be sufficient to reduce the total forcing a focus on two of the most important possibilities.8 fertility rate from six to four, but in the absence of strong so- A few promising components are beginning to show up in cial and economic change, a clifferent strategy is likely to be education projects. Much more can and should be done. needed to reduce it from four to two. The Bank is uniquely qualified to do it. The Bank has been remiss in failing to develop programs Besides pointing up the need to broaden the scope of that focus on this issue. The most important thing it has population activities, the case studies suggest several im- done in this direction has been to promote general social provements that can be made within population projects and economic development. Above all else, this is probably and more traditional lines of activity. Early projects tended the strongest force we know cf to reduce desired family size to emphasize expansion of the physical infrastructure at the and encourage spacing. However, these efforts were not expense of "software"elements, the implicit assumption be- undertaken with their potential demographic effect in ing that the latter would be provided by government or mind. Had that been done-had the Bank searched for se- other donors. This assumption is now recognized as wrong lective interventions into the development process that had in more instances than it is correct and steps have been tak- the possibility of changing implicit benefits and costs of en to right the balance within projects. The case studies sug- large families-far more mig:ht have been accomplished. It gest, however, that there is still a way to go in some 6. Study of the Bangladesh case may require modification of this state- 8. Another recent initiative-the effort to develop integrated human re- ment. It is plausible that in this case, the successful IEC efforts and the dem- source projects-is worth trying, but in the end may not bear much fruit. The onstration effect of increasing prevalence played a role in decreasing desired danger is that such projects may end up being "Christmas trees"of indepen- family size, despite the absence of social and economic change. dent components that are administrative nightmares to implement. A more 7. There are, however, regions with.in some of these countries where fertil- practical approach may be to develop separate projects which incorporate ity has approached replacement: Yogyakarta (2.1), Bali (2.5), East Java (2.6), population concerns-for example, education projects that include concen- Kerala (2.2), and Tamil Nadu (2.6). In all cases, social and economic develop- trated efforts on enrolling and keeping girls in school for a longer period of ment and/or administrative capacity is above the national average. time, industrial projects that set aside specific jobs for women, etc. 6 countries. The tendency for population projects to grow in loan and grant funds and ways that it can be done. All these size andl fund increasing quantities of recurrent costs needs points suggest that the population "sector" is more staff- careful monitoring to ensure that absorptive capacity and intensive than most others and that greater allowance sustainability limits are not being overwhelmed. For asso- needs to be made for this difference. ciated reasons, more concern about cost containment and Finally, what can be said about the overall level of Bank intemal. efficiency within family planning projects and pro- effort in this field, an issue that has resulted in substantial grams its needed. Before becoming heavily committed to criticism of the Bank in recent years? In judging this issue, funding contraceptive supplies, more thought should be it must be remembered that financial input alone is a poor given to funding contraceptive production facilities in larg- indicator of level of effort, especially in the population er countries. Bank and country ability to assess project and field, and that the Bank's efforts cannot be assessed inde- program effectiveness is poor and will continue to be poor pendently of what other donors, with grant funds at their as long as project components associated with monitoring disposal, are doing. and evaluation and the establishment of research capacity Within the case study countries we find no evidence that are neglected. additional financial resources for the type of projects the The case studies also suggest several broader lessons. Bank is currently funding in these countries would have Efforts in the population field require patience but eventu- made much difference. Some additional funds could be ally pay off. This fact should influence the way targets and usefully used in specific instances: to allow more staff in- criteria for assessment of population activities are estab- puts for supervision and technical assistance functions; to lished. Non-project activities-dialogue; sector work; ef- undertake more non-project activities (more sector work, forts to establish policies, improve organizational more collaboration and coordination with other donors, arrangements, and change rules and regulations-are very and a more pro-active role in some countries); to expand important. At some stages, they are more important than population-related activities in non-population sectors. development of a project. Such activities need to be specif- Additional resources will also be necessary to develop ac- ically rewarded and encouraged. An impressive example is tivities in countries in which the Bank does not now have the Bar.k effort of the past year to establish a policy consen- programs. But such changes would represent a modest re- sus in Africa, which incidentally is now bearing fruit in pro- orientation and expansion of a program which, on the posals for useful projects. The Bangladesh case is especially whole, appears to be on the right track. important in suggesting the value of joint programming of 7 Resumen Ejecutivo En este estudio se analizan las actividades que ha venido ses en que, hasta hace muy poco, no se habian logrado cumpliendo el Banco en el sector de la poblaci6n de ocho muchos resultados en relaci6n con los esfuerzos desplega- paises. Se comienza por indagar, en el caso de cada uno de dos durante los uiltimos 20 afios a fin de reducir las tasas de ellos, acerca de los cambios demograficos y socioecon6mi- crecimiento de la poblaci6n. Las tasas de mortalidad, espe- cos conexos que se han producido desde 1968, mome?nto en cialmente en la infancia y en la nifiez, tambien han dismi- que el Banco anunci6 que comenzaria a otorgar prestamos nuido considerablemente, en un 50% en los paises de para dicho sector. Tambi6n se identifican las politicas oficia- America Latina a partir de 1960. Sin embargo, cualesquiera les establecidas que influyeron en estas tendencias y la fun- sean las pautas que se utilicen, las tasas de fecundidad, ci6n que cumpli6 el Banco frente a otros donantes durante mortalidad y crecimiento de la poblaci6n siguen siendo al- el periodo mencionado. Las conclusiones a que se ha llega- tas, especialmente en Asia y Africa. Habida cuenta de la es- do se basan en viajes de observaci6n, entrevistas y un ana- tructura demografica de estos paises en la que predomina lisis de documentos y trabajos en materia de operaciones. la poblaci6n joven, ademas de las tasas actuales de varia- El panorama resultante es de una considerable diversi- ci6n de la fecundiad y la mortalidad, es probable que en dad de actividades y experiencias que dificultan hacer una esas naciones las tasas de crecirniento de la poblaci6n se generalizaci6n. La conclusi6n general del estudio, dejando mantengan en un nivel anual cercano al 2% o aun superior de lado esta diversidad y sus distinciones por el momento, durante algdn tiempo. es que, despues de unos primeros pasos lentos, vacilantes y Entre las dos terceras y las tres cuartas partes de esta dis- a veces equivocados, el Banco se ha vuelto cada vez mas efi- minuci6n de la fecundidad ha obedecido al mayor uso de caz en este ambito, pero su Labor y logros podrian haber los metodos anticonceptivos modernos, en tanto que casi sido mayores. De ahora en adelante -en el supuesto de todo el resto de esta reducci6n se debe a que las personas se una aplicaci6n firme y homogenea de las ultimas innova- casan a una edad cada vez mayor. A su vez, estas modifica- ciones en materia de operaciones del Banco en este sector- ciones obedecen a la combinaci6n de mejoras registradas en la mayor parte de lo que hace falta para lograr las reformas las caracteristicas sociales y econ6micas que propician la re- demograficas pretendidas implicitas en la politica sectorial ducci6n del tamafio de la familia y una edad mayor para del Banco comprende cambios de orientaci6n y del enfasis contraer matrimonio. Se trata, por ejemplo, de mejoras de puesto, en lugar de apartamientos radicales de las practicas las condiciones de la mujer en cuanto a educaci6n, empleo y niveles de financiamiento recientes. y situaci6n social; la disminuci6n de la mortalidad infantil; menos oportunidades de trabajo para los nifios, y una me- Tendencias demograficas y socioecon6micas nor dependencia de estos para la seguridad en la vejez. Dichas mejoras tambien comprenden programas de plani- Las tasas de fecundidad total han bajado en todos los ficaci6n familiar con los que se procura que los anticoncep- paises objeto del estudio de casos: marginalmente en Sene- tivos modemos sean mas accesibles y se suministre gal; 20% de la distancia entre el nivel maximo de ocho nifios informaci6n y educaci6n a su respecto. por mujer a un nivel de reemplazo de 2,1 en Kenya; 40% en En la producci6n de los cambios demograficos observa- Bangladesh; 50% en India; 60% en Indonesia, y mas del 70% dos en los tres paises de Am6rica Latina se han combinado en Brasil, Colombia y Mexico. E]stas noticias son particular- ambos factores -es decir, mejoras importantes en estos in- mente alentadoras en los casos de Bangladesh y Kenya, pai- dicadores sociales y econ6micos y servicios de planificaci6n 8 familiar apreciables y eficaces suministrados, en estos ca- consideran varias posibilidades y se propugna una investi- sos, por organizaciones no gubernamentales (ONG), orga- gaci6n. Dicha investigaci6n reviste importancia para el fu- nizaciones semipiublicas e intermediarios comerciales. Si turo del programa de Bangladesh y tambien por lo que bien er Indonesia tambien se han registrado mejoras socia- podria revelar acerca de cuales serian los programas ade- les y econ6micas importantes, la sociedad en ese pais sigue cuados para paises en circunstancias similares. siendo mAs rural y tradicional con tasas mas altas de mor- talidad derivada de la maternidad y en la nifiez. Su progra- La funci6n del Banco en los distintos paises ma de planificaci6n familiar auspiciado por el Gobierno ha compensado esta diferencia de diversas maneras. La mAs Indonesia. Si bien el Banco ha prestado considerable asis- importante de ellas tal vez sea la red importante y eficaz de tencia financiera a las medidas de desarrollo generales de agentes de planificaci6n familiar que llega hasta las aldeas Indonesia, las operaciones crediticias para el sector de la mas pequefias. Con el programa de la India tambi6n ha poblaci6n han ascendido a tan s6lo el 1% del total. Esta compensado un progreso social y econ6mico aun mas mo- suma ha aportado los fondos necesarios para sufragar derado, pero de distinta manera: centrindose estrictamente aproximadamente el 10% del total de gastos puiblicos en el pero de modo eficaz en la esterilizaci6n. Sin embargo, dicho programa referido a la poblaci6n, en tanto que otros donan- programa se ha visto frenado a veces por la reacci6n politi- tes, incluidas las organizaciones no gubernamentales ca que han provocado las campafias coercitivas peri6dicas. (ONG) han aportado otro 20%. Al suministrar estos fondos, Debido al exito que ha tenido en llegar a una gran propor- el Banco desempef6 un papel bastante tradicional, en el ci6n de parejas que querian poner fin a la etapa reproducti- sentido de que se trataba de operaciones de un banco mas va de sus vidas, el programa indio esta ahora alcanzando que de una instituci6n de desarrollo. Es decir, en general re- su punto de rendimientos decrecientes. El caso de Kenya es accion6 con cautela ante las propuestas de proyectos pre- interesante puesto que el considerable progreso social y sentadas por las autoridades de Indonesia; fue poco lo que econ6imnico de los uiltimos 25 anios al parecer ha dado lugar proporcion6 a modo de asistencia tecnica en la preparaci6n a que se hiciera latente la necesidad de limitar el tamafio de o ejecuci6n de los proyectos en este ambito, y en los proyec- la familia, la que recien se manifest6 hace muy poco. El lar- tos mismos se centr6 en la construcci6n de obras civiles y el go periodo de *preparaci6n del programa que finalmente suministro de componentes fisicos. Sin embargo, el Banco ha llegado al umbral critico de eficacia, sumado al estimulo si contribuy6 a la formulaci6n y el respaldo de un progra- que le infundieron el Presidente y otros dirigentes impor- ma importante de educaci6n de la poblaci6n. Empero, en tantes, al final parece estar dando resultados. Senegal es un otras esferas de componentes 16gicos tendi6 a apoyar la eje- ejemplo de pais que tiene muy poco desarrollo social y eco- cuci6n de proyectos preparados por el Programa Nacional n6mico y donde se preparan muy pocos programas como de Planificaci6n Familiar de Indonesia con la asistencia tec- para que se pueda ejercer una gran influencia en la tasa de nica de otros donantes. Durante la mayor parte del periodo fecundidad. comprendido, no hubo ningon miembro del personal teni- Bangladesh constituye un caso de especial interes debido co residente que fuera responsable de este sector y los estu- a que se trata de una naci6n en la que ha habido una formu- dios sectoriales fueron limitados. Iaci6n importante de programas pero muy poco adelanto Salvo, tal vez, en los primeros anos de su participaci6n social y econ6mico, y es casi un caso cIasico de aplicaci6n en el programa, el enfoque del Banco era perfectamente del enfoque de la oferta en condiciones en que no se preveia adecuado. Las autoridades de Indonesia sabian lo que que- encontrar mucha demanda. Sin embargo, aun asi debe ha- rian y demostraron ser capaces de una buena ejecuci6n. De ber existido suficiente demanda latente como para redun- hecho, el mencionado programa de planificaci6n familiar dar en la prevalencia del uso de anticonceptivos y la normalmente es considerado como uno de los mas eficaces disminuci6n de la tasa de fecundidad que se han registrado en el mundo. Otros donantes -en especial, la Agencia para hasta ahora. Queda sin contestar el interrogante de cuanto el Desarrollo Internacional EUA (USAID)- participaron mas se puede avanzar sin contar con cambios estructurales activamente en la formulaci6n de politicas operacionales y importantes que influyan en la demandal. En el estudio se tacticas de ejecuci6n. Esta result6 ser una buena divisi6n de hecho de las tareas que mereci6 el elogio de las autoridades de Indonesia. 1. Esta modalidad de presentaci6n entrafia una clasificaci6n implicita de En el informe se analizan cinco esferas en las que el Ban- los programas de informaci6n, educaci6n y comunicaci6n como partes del programa de planificaci6n familiar; es decir, una medici6n desde el punto de CO podna haber hecho mas o actuado de manera diferente vista de la oferta. Ello se basa en la idea de que si bien se puede difundir la en dicha naci6n. 1) Habria sido uitil contar con un especia- informacid6n y alentar la demanda a traves de las actividades de informaci6n, lista del Banco residente en el pais por un periodo mas lar- educad6n y comunicad6n, ya debe existir derta proclividad -cierta deman- da latente generada por fuerzas culturales, sociales o econ6micas mAs esen- go y con la funci6n de participar activamente en la ciales- para que estas medidas produzcan algun efecto. preparaci6n y ejecuci6n del programa. 2) La coordinaci6n y 9 colaboraci6n con otros donarntes podria haber sido mejor vidades como lo esta haciendo la mencionada USAID, o durante los primeros 15 afos de participaci6n del Banco. bien tratando de Ilenar el vacio dejado por esta. En el infor- Desde entonces, este aspecto del programa ha mejorado me se indica que, segiin el problema de que se trate, es pre- considerablemente. 3) El Banco podria haber propugnado ciso aplicar ambos metodos. en forma mas temprana y decidida proyectos en el sector India. Al igual que en el caso de Indonesia, hasta hace de la salud, en especial promoviendo mas energicamente la muy poco el Banco ha desempefiado mas las funciones tra- salud maternoinfantil y su integraci6n en la planificaci6n dicionales de un banco que las de una instituci6n de desa- familiar, lo que reci6n esta comenzando a hacer ahora. 4) El rrollo en la India. Sin embargo, en este caso, el papel que Banco podria haber estimulado la formulaci6n de estrate- cumpli6 fue mucho menos adecuado en raz6n de los pro- gias de planificaci6n no familiar a fin de incrementar la ne- blemas estrategicos del programa. Estas dificultades con- cesidad de reducir el tamafici de la familia, por ejemplo, sisten en: excesiva atenci6n prestada a la esterilizaci6n, con contribuyendo a preparar proyectos destinados a mejorar lo que se ignora la necesidad de espaciar los nacimientos las oportunidades de educaci6n y empleo de la mujer. Aun que tienen las parejas j6venes y (como se sefnal6 antes) se es preciso realizar esfuerzos considerables en este ambito. estan observando rendimientos decrecientes; falta de aten- 5) Sigue siendo necesario el fortalecimiento de las activida- ci6n a las operaciones y la calidad de los programas (por des de investigaci6n y evaluaci6n, al menos por dos razo- oposici6n a la expansi6n de estos); excesiva centralizaci6n y nes, a saber: para extraer ensenianzas de este caso en el que dependencia de un Cnico sistema de prestaci6n, y la omi- se obtuvieron buenos resultados a fin de aplicarlas en otros si6n de considerar los factores que pueden influir en la de- paises2 y para fortalecer la capacidad de investigaci6n y manda. La persistencia e identificaci6n de estos problemas evaluaci6n dentro del Prograrna Nacional de Planificaci6n data, por lo menos, desde que se inici6 la participaci6n del Familiar de Indonesia, en el que esta funci6n ha sido des- Banco en el programa de la India. atendida. El hecho de que el Banco no haya podido ejercer una Sin embargo, evidentemente el mejoramiento de estos gran influencia en el programa de la India puede obedecer Ambitos no era necesario para lograr los considerables pro- a diversos factores. 1) El primer proyecto del Banco en ma- gresos que se han registrado hasta el momento. Asimismo, teria de poblaci6n en la India se inici6 en 1972, mucho des- la labor del Banco en estas esferas podria haber topado con pues de haberse afianzado firmemente el plan del pais. 2) Si resistencia por parte del Programa Nacional de Planifica- bien los montos entregados en prestamo constituian el 28% ci6n Familiar de Indonesia y otros donantes que preferian de la cartera total del Banco para el sector de la poblaci6n que el Banco se centrara principalmente en el suministro de hasta 1988, estas sumas representaban una pequefia pro- recursos para edificios y equipo. El Banco dio respuesta es- porci6n-3,6% entre 1980 y 1988- de los gastos totales en pecifica a los deseos del cliente, y en este caso fue la res- el Programa de Bienestar Familiar de la India. Si bien el to- puesta adecuada en raz6n de la gran capacidad del cliente tal de donantes aport6 entre el 12% y el 14% de los fondos y de la asistencia de otros donantes. necesarios para sufragar los gastos publicos, nunca se cons- En Indonesia, en el futuro tal vez hagan falta enfoques tituyeron en una coalici6n eficaz. 3) La naturaleza de los distintos de los aplicados en el pasado en lo que hace a tres cinco primeros proyectos, denominados proyectos zonales, aspectos importantes. En primer lugar, hay mucha menos le impidieron al Banco participar en cuestiones de politicas necesidad de programas masivos de obras de infraestructu- que excedieran del marco de estas zonas (geograficas) y ra. En segundo lugar, la USAID ha mostrado su intenci6n desalentaron las innovaciones experimentales incluso den- de disminuir progresivamente sus aportes, en parte porque tro de dichas zonas3. 4) Tanto los estudios sectoriales como considera que Indonesia ahora necesita menos asistencia los aportes de personal para el programa han sido insufi- t6cnica y fondos a titulo de dcnaci6n. El Programa Nacio- cientes. Fueron pocos los estudios sectoriales realizados an- nal de Planificaci6n Familiar de Indonesia, con la asistencia tes de 1988-sin duda no fueron suficientes como para de la USAID, ha reaccionado estableciendo programas en- proporcionar al Banco un fundamento para proponer cam- caminados a promover la comercializaci6n y la prestaci6n bios tacticos como, por ejemplo, recomendar siquiera c6mo de servicios por parte del sector privado. En tercer lugar, la f6rmula que ha servido para bajar la tasa de fecundidad de 5,5 a 3,5 nifhos por mujer puede no servir ya para reducirla 3. Se trata de un aspecto discutible. En un principio el fundamento de este aun mas, a una de reemplazo de 2,1. El Banco podria adap- sistema era que la influencia que ejerceria el Banco dentro de estas zonas se- tarse a estos cambios reduciendo progresivamente sus acti- ria mayor que la que habria tenido en el programa general. Lo mfhs probable es que no haya sucedido asi, y ello no se debe a que el concepto estuviera equivocado sino a que la India nunca acept6 que las zonas respaldadas por el Banco pudieran ser diferentes de otras y porque el Banco nunca tom6 la 2. Por ejemplo, ZcuAles son los factores responsables del exito de estas ope- iniciativa de elaborar estudios tenicos detallados para estos proyectos, en raciones del pais basadas en la comunidad? jSon exclusivos de Indonesia, o los que se hubiera explidtado minuciosamnente esta cuesti6n antes de la eje- pueden servir de ayuda a otros paises? cuci6n. 10 podria rnodificarse la estructura de incentivos, seguimiento dicas que exigen prioridad ponen constantemente en peli- y rendici6n de cuentas del programa a fin de dar mayor im- gro el provecho de los adelantos alcanzados. portancia a metodos menos drasticos que la esterilizaci6n. Sin embargo, la mayor parte del progreso logrado en ma- Tampoco hubo suficiente aporte de personal -en condi- teria de aumento de la prevalencia del uso de anticoncepti- ci6n de :residente ni en Washington- como para hacer mu- vos y reducci6n de la tasa de fecundidad puede atribuirse cho mas que emplear un enfoque distante. a los aportes del programa. Es probable que la diferencia ra- A partir de 1987 se produjeron varios cambios importan- dique en el programa de divulgaci6n que, pese a sus limi- tes. Al parecer el Gobierno y el Banco se sintieron decepcio- taciones, es mas amplio y activo que en la India. Ademas, nados cDn los resultados de los proyectos zonales y estan los donantes han desempefiado una funci6n mas intensa y dispuestos a probar un metodo diferente. Esto se refleja en activa en la planificaci6n y las operaciones, ampliando en la el sexto y el septimo proyectos, iniciados en 1989 y 1990, realidad la capacidad de ejecuci6n del Gobierno. Debe atri- por la atenci6n que se presta en ellos a los componentes buirse un gran merito al Banco, que en este caso ha cumpli- priorita:rios del programa a nivel de los estados. En ellos do un papel activo y de liderazgo. tambien se incluye el apoyo a las ventas subvencionadas y En el informe se enumeran siete factores que al parecer las organizaciones no gubernamentales. En segundo lugar, han tenido importancia trascendental. 1) Es evidente que el a partir de 1987, se han hecho varios estudios sectoriales Gobierno esta empefnado en resolver el problema relaciona- buenos que al parecer estan influyendo en el modo de pen- do con la poblaci6n; ha dedicado un porcentaje apreciable sar de la India y la naturaleza del dialogo. El Proyecto de de sus escasos presupuesto y talento directivo al programa, Supervivencia Infantil y Maternidad sin Riesgos que ha y ha procurado con ahinco y utilizado asistencia y asesora- sido propuesto presta por primera vez gran atenci6n a los miento externos. 2) La asistencia financiera y tecnica exter- problemas de la salud que revisten importancia como para na ha sido de considerable magnitud. 3) Gran parte de la influir en las decisiones en materia de fecundidad. En dicho asistencia financiera, incluso dentro de los cinco proyectos proyecto tambien se presta atenci6n minuciosa a las opera- para la poblaci6n respaldados por el Banco que han sido ciones za nivel de poblado, que en el pasado han demostra- aceptados hasta ahora, consiste en donaciones proporcio- do ser Ln punto de estrangulamiento importante para la nadas por los cofinanciadores. Ello ha permitido una mayor aplicaci6n de los cambios de politicas. Por uiltimo, han au- flexibilidad y una mayor concentraci6n de la atenci6n en mentado considerablemente los esfuerzos realizados por el los componentes 16gicos y ha contribuido a sufragar los Banco en el ambito de la educaci6n, prestandose atenci6n gastos por concepto de aportes adicionales de personal del especial a la mujer. Banco, tanto en la sede como en el terreno. 4) Gracias a estos Si bien las mencionadas son todas propuestas que pro- aportes adicionales de personal y con el estimulo del Go- meten dar buenos resultados, mucho dependera de la deci- bierno, el Banco ha desempefiado un activo papel en todas si6n y homogeneidad con que se las aplique. El Banco las etapas de estos proyectos -disefno, captaci6n de fon- debera continuar realizando estudios sectoriales y de in- dos, coordinaci6n de la ayuda de los donantes, ejecuci6n y vestigaci6n con miras a encontrar maneras practicas de re- evaluaci6n ex post-, mucho mas intenso de lo que suele solver los problemas estructurales; debera seguir serlo en los proyectos que respalda. 5) La labor de coordi- defendiendo -energicamente cuando sea necesario- la naci6n entre los donantes ha sido importante y los cofinan- aplicaci,6n de los resultados de dicha labor; debera poner ciadores, que por lo general han suministrado mas de la mas empefno en captar la ayuda de otros donantes en este mitad de los fondos para estos proyectos, han desempefia- proceso, y dedicar mAs tiempo y esfuerzos a la supervisi6n do un papel cada vez mAs activo. Su participaci6n en estos de las operaciones a nivel de poblado. Es probable que momentos ha llegado a la etapa que permite describir con todo ello requiera un mayor aporte por parte de funciona- mayor precisi6n al cuarto proyecto como de un consorcio y rios farailiarizados con los problemas de fondo, tanto en la no del Banco. 6) Con la ayuda de los fondos de donaci6n se sede ccmo en el terreno. ha dotado de flexibilidad al programa gracias a que, entre Bangladesh. El programa de Bangladesh adolece de mu- otras cosas, se ha establecido un "Fondo para proyectos no- chas de las deficiencias del programa de la India. Si bien se vedosos" y una "Unidad de financiaci6n de proyectos". ha centrado menos la atenci6n en las esterilizaciones, la ca- 7) Se han aprovechado los resultados de proyectos experi- lidad de los servicios de planificaci6n familiar no es buena, mentales para mejorar el desempefio del programa. el programa de salud matemoinfantil es endeble y no esta En este caso, hay motivos para considerar que las medi- bien integrado en la planificaci6n familiar; ademas, la falta das del programa deben seguir la misma orientaci6n actual de talenito en la direcci6n sumada a estructuras administra- durante los pr6ximos afios. Segun las encuestas, hay un nui- tivas profundamente arraigadas dificultan las mejoras. Por cleo de usuarios potenciales a los que aun no se ha llegado anladidura, desde luego que el programa se desarrolla en y hay pruebas de que la prevalencia es mayor cuanto ma- un medio fisico sumamente dificil en el que las crisis peri6- yor es la densidad de los puntos de servicio, cuanto mAs 11 frecuentes son los contactos con los agentes de planifica- prevalencia del uso de anticonceptivos ya era comparativa- ci6n familiar y cuanto mejor es la calidad de los servicios mente elevada y la estructura de prestaci6n de servicios es- suministrados. Sin embargo, incluso si aumentara el nivel taba bien afianzada. 3) Los principales organismos de estos de la prevalencia al 50% que se obtuvo en la zona experi- paises eran organizaciones no gubernamentales o institu- mental de Matlab, frente al actual nivel de 35% -lo que es ciones mixtas. En vista de la actitud de los gobiernos, el poco probable a escala nacional- la tasa de fecundidad Banco (a diferencia de la USAID) no disponia de una mane- total seria de un 3,7, aun muy superior al nivel de reempla- ra de trabajar en forma directa, ni siquiera indirecta, con di- zo. A fin de obtener niveles mas bajos en un periodo razo- chos organismos, como lo ha logrado hacer en Kenya y en nable, es probable que haga falta un cambio de orientaci6n otras naciones. del programa a fin de abarcar mas que el enfoque orientado El enfoque actual del Banco en America Latina consiste hacia la oferta que se utiliza a ctualmente. Si bien al parecer en centrarse en la salud para la procreaci6n y la maternidad en Bangladesh el tamafio aconsejable de familia se ha sin riesgos como fundamento de la planificaci6n familiar, y reducido aun sin que existieran mejoras sociales y econ6mi- ello parece lo correcto. Un metodo mAs agresivo no se justi- cas importantes, seria mas seguro y mas rapido lograr una fica y no es aconsejable desde el punto de vista politico. Sin mayor reducci6n si se pudieran concretar cambios embargo, siguen existiendo problemas importantes de estructurales que aumentaran la necesidad de reducir el equidad, calidad de los servicios y opciones que, de cual- tamafio de la familia. quier manera, son parte de un conjunto mas amplio de pro- Brasil, Colombia y Mexico. A diferencia de otros paises, blemas que el Banco esta tratando en el sector de la salud5. s6lo una pequefia parte de los gastos de estas naciones en La importancia de estos casos en America Latina radica planificaci6n familiar provinieron de fuentes externas, y en en que ponen en evidencia cuestiones relativas al funda- su mayor parte la ayuda recibida provino de organizacio- mento de las operaciones crediticias del Banco para el sec- nes no gubernamentales en lugar de organismos multilate- tor de la poblaci6n. Si el Banco quiere trabajar en paises que rales o bilaterales4. Los insumos aportados por el Banco no aceptan el control del crecimiento de la poblaci6n como fueron minimos. En estos paises no hubo proyectos para la fundamento, debe basar su programa para la poblaci6n en poblaci6n financiados por el Banco. (Se prepar6 uno para un conjunto de principios mas amplio y flexible. Esto po- Mexico, pero fue retirado al asumir un nuevo Gobierno.) dria partir del reconocimiento de que el objetivo general es No fue mucho lo que se hizo por aplicar los pocos subcom- la promoci6n de mejoras sostenibles en el nivel de vida, y ponentes de planificaci6n familiar comprendidos en el que los programas de planificaci6n familiar son una ayuda componente de salud maternoinfantil de los proyectos rela- tanto a nivel microecon6mico -mejorando la salud de la tivos a la salud financiados por el Banco, y en los diAlogos familia y las opciones- como macroecon6mico. En este sobre politicas rara vez se trat6 el tema. Al menos hasta contexto, estos programas son temas adecuados para un 1989 fueron muy pocos los estudios sectoriales centrados diAlogo sobre politicas y un analisis sectorial a fin de deter- directamente en la poblaci6n o en la planificaci6n familiar. minar de que manera deberian formularse y utilizarse en Ademas, si bien sin lugar a dudas las operaciones crediti- situaciones especfficas, si es que debe hacerselo. AdemAs, cias destinadas a otros sectores han producido algunos estos casos plantean interrogantes acerca de la convenien- efectos indirectos, nunca se emprendieron teniendo en cia en determinadas circunstancias de otorgar prestamos cuenta consideraciones de orden demogrAfico. para proyectos de poblaci6n independientes, frente a los Al parecer son tres los factores responsables de este apa- proyectos de salud que contienen componentes de planifi- rente descuido. 1) La sensibilidad de orden politico en tor- caci6n familiar. Si este enfoque se hubiera utilizado con an- no a las politicas de "control del crecimiento de la terioridad y de manera mas arm6nica, es posible que el poblaci6n" propuestas por extranjeros dificultaba al perso- Banco no se hubiera topado con tanta oposici6n y descon- nal del Banco empezar a hablar del tema con los gobiernos fianza acerca de sus motivaciones en estos paises. de estos paises, que consideraban inaceptable la posici6n Kenya. El estudio de este caso es un buen ejemplo de la adoptada por la administraci6n del Banco y el personal de virtud -y la necesidad- de tener paciencia y perseveran- 6ste, que se encontraba en el medio, reaccionaba haciendo cia en la esfera de la poblaci6n. Si bien desde los ultimos muy poco. 2) Para cuando el Banco comenz6 a otorgar pres- afios de la decada de 1960 han sido considerables los es- tamos para los servicios de salud, lo que habria hecho mAs fuerzos desplegados por los donantes para persuadir al Go- aceptable el enfoque relativo a la planificaci6n familiar, la bierno de que estableciera un programa importante, no fue 4. Sin embargo, algunas organizacicnes no gubernamentales intemaciona- les han recibido ayuda de fuentes bilaterales y multilaterales; especialnente, 5. Uno de los aspectos de dichos problemas es el elevado nu6mero de abor- la Agencia para el Desarrollo Intemacional EUA (USAID) y el Fondo de las tos provocados ilegales que existe en estos paises. Vease el trabajo de Singh Naciones Unidas para Actividades en Materia de Poblaci6n (FNUAP). y Wolf, 1991. 12 sino hasta 1988-89 que se vieron pruebas claras de que la caci6n y comunicaci6n. Sin embargo, aun cuando se cerrara tasa de fecundidad habia comenzado a bajar. totalmente la brecha existente entre el tamafio aconsejable Ademas de respaldar el desarrollo econ6mico general, de familia y el real, la tasa de fecundidad total disminuiria apoyo que comprendia sustanciales fondos para la educa- a la gama de tan s6lo 4 a 5 nifios, lo cual entrania una tasa ci6n, les aportes del Banco han seguido dos lineas genera- de crecimiento de la poblaci6n aiu superior al 2% al afio. les. A partir de 1974 se han financiado cuatro proyectos. Entre 1984 y 1987, el n(imero medio deseado de nifnos baj6 Con los dos primeros se suministraron fondos al Ministerio de 5,8 a 4,4. Para acelerar dicho cambio se necesita algo mas de Salud para establecer una red de instalaciones rurales de que modificaciones desde el punto de vista de la oferta. Si salud y hospitales clinicos; los componentes de planifica- bien en el informe se recomiendan algunas posibilidades, ci6n familiar de estos proyectos eran deficientes y se igno- tambien se sefiala la suma importancia que revisten las in- raron en gran medida durante la ejecuci6n. En este periodo, vestigaciones y los estudios experimentales en este caso, la tasa de crecimiento de la poblaci6n aument6 en realidad. toda vez que no se sabe bien de que manera lograr avances Si bien el tercero y cuarto proyectos han tenido componen- importantes en materia de reducci6n del tamafio apetecido tes mas importantes relativos a la poblaci6n y a la planifica- de familia dentro de un sistema familiar muy favorable a la ci6n familiar, se iniciaron demasiado tarde (1988 y 1990) natalidad. como para influir en un cambio de tendencia de la tasa de Senegal. Aparte de un proyecto de salud rural disefiado fecundidad. La segunda etapa de aportes del Banco ha en 1982 que se centraba en el suministro de edificios y equi- comprendido el dialogo sobre politicas, estudios sectoriales po para la ampliaci6n de los servicios basicos de salud, el y medidas encaminadas a influir en la evoluci6n del pro- Banco hasta hace muy poco habia concentrado sus esfuer- grama de diversas maneras. Entre los ejemplos de esto uilti- zos en ayudar al Gobierno a formular una politica integral mo se encuentran los esfuerzos desplegados que lograron en materia de poblaci6n. Este tema central se recomend6 en persuadir al Gobierno de que creara un organismo intermi- un anAlisis sectorial realizado en 1985-86, en el que se lleg6 nisterial de coordinaci6n en materia de poblaci6n fuera del a la conclusi6n de que la USAID y el Fondo de las Naciones Ministerio de Salud (que se incluy6 como condici6n en el Unidas para Actividades en Materia de Poblaci6n estaban segundo prestamo para fines de ajuste estructural), y de haciendo todo lo que podia hacerse para promover servi- que luego pusiera a su cargo mas responsabilidades, asi cios de planificaci6n familiar. Una de las maneras de poner- como medidas encaminadas a persuadir al Ministerio de lo en practica fue establecer la formulaci6n de dicha Salud de que liberalizara las pautas para el suministro de declaraci6n de politicas como condici6n de un prestamo anticonceptivos, integrara la planificaci6n familiar en la para fines de ajuste estructural. En tanto que algunos fun- principal corriente de actividades del Ministerio de Salud y cionarios del Gobierno pudieron haber aceptado este pedi- ofreciera servicios de esterilizaci6n. Ambas etapas del tra- do en raz6n de que parecia bastante inocuo, la declaraci6n bajo han constituido componentes importantes de un con- de politicas y el proceso de formularla han cobrado vida junto general que comprendia asistencia tecnica y propia y redundado en un avance importante que de otro financiamiento en especie de otros donantes y, lo que tal modo se hubiera producido en forma mas lenta, teniendo vez sea mas importante, indicaciones claras del Presidente en cuenta el estancamiento de la evoluci6n social o econ6- de Kenya en el sentido de que la planificaci6n familiar iba mica de los ultimos afios. Esta estrategia puede revestir es- a tomarse en serio. pecial importancia en los paises de Africa de habla ZEI Banco podria haber hecho algo por acelerar este pro- francesa, debido a su herencia colonial favorable a la nata- ceso? Si bien se podria haber puesto mucho mas empefio en lidad con la que hubo que romper en forma explicita y pu- formular componentes eficaces de informaci6n, educaci6n blica a fin de legitimizar las actividades de planificaci6n y coniunicaci6n, divulgaci6n, seguimiento, evaluaci6n ex familiar. post e investigaci6n, el Banco presion6 con bastante firmeza Es probable que en ese entonces la decisi6n del Banco de al Gobiemo respecto de muchos de estos temas y al parecer centrarse en la formulaci6n de politicas haya sido la correc- ha aprovechado las oportunidades a medida que estas iban ta puesto que si bien dicha instituci6n, a diferencia de otros surgiendo. En este caso, la falta de personal residente asig- donantes importantes, carecia de personal en el terreno nado al sector no parece haber sido un obstaculo grave para para ayudar en la ejecuci6n de las operaciones mAs comple- los avances, en vista de los presupuestos liberales para ta- jas, si tenia acceso a los circulos superiores de formulaci6n reas de supervisi6n y los viajes y aportes suministrados por de las politicas debido a sus actividades mucho mAs am- otros donantes que tenian personal tecnico en el terreno. plias en esta esfera. Hace muy poco, el Banco ha decidido Al parecer el programa por el momento esta en la buena volverse mas activo preparando un proyecto que, entre senda y habria de seguir ampliAndose, aunque poniendose otras cosas, esta orientado a extender los servicios de plani- considerablemente mayor enfasis en la calidad de los servi- ficaci6n familiar y de otra naturaleza al estrato mas bajo del cios, la divulgaci6n y las actividades de informaci6n, edu- sistema de prestaci6n de servicios de salud en Senegal. Se 13 trata de un enfoque peligroso. Por ejemplo, la USAID se ha co, es probable que se necesite una estrategia distinta para restringido a desarrollar sus actividades en el nivel inme- reducirla de cuatro a dos. diatamente superior en raz6n de que el Gobierno tiene una El. Banco se ha mostrado remiso a formular programas capacidad muy limitada de administraci6n y prestaci6n de que no se centraran en este tema. Lo mas importante que ha servicios. En virtud del caracter acuciante del problema, es hecho en este sentido ha sido promover el desarrollo social probable que valga la pena asumir estos riesgos, que pue- y econ6mico general. Quiza sea este, por sobre todas las co- den mantenerse dentro de lIfmites aceptables si se empieza sas, el impulso mas fuerte que se conozca para reducir el ta- a pequefia escala para ir avanzando lentamente; sin embar- manio de familia apetecido y alentar el espaciamiento entre go, se requiere una supervisi6n mucho mas intensa y mas los nacimientos. Sin embargo, estas medidas no se empren- estrecha de la que normalmente ha existido en los proyec- dieron teniendo en cuenta sus potenciales efectos demogra- tos del Banco para Senegal. ficos. Si se hubiera hecho asi -si el Banco hubiera procurado incorporar medidas selectivas en el proceso de Principales problemas para el Banco desarrollo que tuvieran la posibilidad de modificar los be- neficios y costos implicitos de las familias grandes- se hu- La experiencia de los ultimnos 20 afios demuestra clara- bieran alcanzado muchos mas logros. Es bien sabido que, mente que aun en los paises pobres en los que no se ha re- por ejemplo, la educaci6n de la mujer -o simplemente su gistrado mucho progreso social y econ6mico, el proceso de asistencia a la escuela durante unos pocos ainos mas- pro- disminuci6n de la tasa de fecundidad puede iniciarse con mueve su casamiento a una mayor edad y tasas mas bajas un programa de planificaci6n. familiar tipico que se centre de fecundidad dentro del matrimonio. Otras medidas que en el suministro de materiales anticonceptivos y la presta- prometen dar buenos resultados comprenden programas ci6n de servicios e informaci6n conexos. Los avances seran especificamente destinados a aumentar la probabilidad de mas rapidos si en dicho programa se incluye un componen- que los ninios lleguen a la edad adulta, crear oportunidades te de divulgaci6n intensa y se prestan servicios de alta cali- de empleo para las mujeres ajenas a la crianza de los hijos dad basados en las percepciones de los clientes acerca de (por ejemplo, en fabricas u oficinas en lugar de tareas agri- las necesidades. Para ello se requiere una excelente super- colas), y mejorar los programas de seguridad social para visi6n en el terreno, buena capacitaci6n y s6lidas motiva- que los adultos no sientan tanta necesidad de tener familias ciones, aspectos que son muy dificiles de generar en las grandes que proporcionen dicha seguridad. zonas rurales pobres. Examenes anteriores del programa del Banco en materia Sin embargo, hay muy pocas pruebas de que incluso el de poblaci6n han exigido que se prestara mas atenci6n a di- mejor administrado de dichos programas pueda hacer algo chos factores orientados hacia la demanda, aunque se ha mas que satisfacer las necesidades de aquellos que ya estan hecho muy poco para plasmar esta exigencia en programas predispuestos a aceptar los servicios de planificaci6n fami- operacionales. Los motivos para ello son el escepticismo liar6. En todos los casos estud:iados, aun si todas las perso- reinante acerca de la importancia del argumento que fun- nas participaran en un programa de planificaci6n familiar damenta la necesidad de medidas orientadas hacia la de- -o aun cuando se alcanzaran en todo el pais las tasas mas manda; dudas acerca de los logros que pueden alcanzarse, altas de prevalencia del uso c[e anticonceptivos obtenidas y la compartimentaci6n de funciones e inercia del personal. con los programas experimentales mejor administrados-, Estos factores se sustentan en el hecho de que por lo general no existe ningun pals en el grupo de los ocho mencionados, se ha tratado a la poblaci6n como un sector que es de res- salvo quizas los tres de America Latina, en el que la tasa de ponsabilidad de una unidad administrativa determinada, fecundidad total pudiera disrninuir a niveles aceptables7. liberando asi tacitamente a otras unidades sectoriales de Hara falta otro tipo de medideas para lograr un avance ma- esta funci6n. El sector de la poblaci6n no es mas indepen- yor. Si bien tal vez las estrategias orientadas hacia la oferta diente que el desarrollo econ6mico o el alivio de la pobreza. resulten suficientes para reducir la tasa de fecundidad total Los tres conceptos se encuadran mas en la naturaleza de de sies a cuatro, a falta de un gran cambio social y econ6mi- objetivos estrategicos que deberian ser responsabilidad de todos los sectores. 6. El estudio del caso de Bangladesh puede exigir la modificaci6n de esta A partir de 1987 se han producido varios cambios cuyos afirmaci6n. Es posible que en este caso, el 6xito de las tareas de informaci6n, efectos estan dando buenos resultados en esta situaci6n educaci6n y comunicaci6n y el efecto de demostraci6n del aumento de la dentro del Banco. La integraci6n administrativa de educa- prevalencia desempefiaran su funci6n en la reducci6n del tamafio apetecido * . ., de familia, pese a no existir modificaciones sociales y econ6micas. cion, poblaci6n, salud y nutrici6n ha promovido y ha dado 7. Existen, sin embargo, regiones dentro de algunos de estos paises en las lugar a ideas acerca de la manera en que cada sector puede que la tasa de fecundidad se ha acercado ala de reemplazo: Yogyakarta (2,1), ayudar a alcanzar las metas de los demas. Las propuestas Bali (2,5), Java Oriental (2,6), Kerala (2o2) y Tamil Nadu (2,6). En todos los ca- acerca de La f *nci6n de la m en el proceso de desarrollo la sos, el desarrollo social y econ6mico yio la capacidad administrativa son su- muler y periores al nivel medio nacional. Supervivencia infantil estan obligando a centrar la atenci6n 14 en dos de las posibilidades mas importantes8. Si bien estan Esto deberia influir en la manera en que se fijan las metas y comenzando a salir a la luz algunos componentes que pro- las pautas para la evaluaci6n de las actividades en materia meten dar buenos resultados en los proyectos de educa- de poblaci6n. Revisten mucha importancia las actividades ci6n, aim puede y debe hacerse mucho mas y el Banco esta que no son especificas de los proyectos, a saber: el dialogo, singularmente capacitado para ello. los estudios sectoriales, las medidas encaminadas a fijar las Ademas de sefialar la necesidad de ampliar el ambito de politicas, mejorar las estructuras organicas y modificar las las actividades relacionadas con la poblaci6n, los estudios normas y reglamentaciones. En algunas etapas, ellas son de casos recomiendan varias mejoras que pueden introdu- mas importantes que la preparaci6n de los proyectos. Es cirse en los proyectos relativos a la poblaci6n y actividades preciso recompensar especificamente y fomentar dichas ac- mas traclicionales. En los primeros proyectos se tendi6 a po- tividades. Excelente ejemplo de ello es el intento realizado ner enfasis en la ampliaci6n de la infraestructura fisica a por el Banco el anio pasado por lograr un consenso en ma- costa de los "componentes 16gicos", con lo que implicita- teria de politicas en Africa, que dicho sea de paso ahora esta mente se suponia que estos ultimos serian proporcionados dando resultados al surgir propuestas para proyectos pro- por el gobierno u otros donantes. Sin embargo, fueron mas vechosos. El caso de Bangladesh reviste especial importan- los casos en que este supuesto demostr6 ser err6neo y se cia al sefialar el valor que tiene la programaci6n conjunta de han tomado medidas para lograr un equilibrio dentro de los prestamos y las donaciones y la manera de hacerlo. To- los proyectos. No obstante, los estudios de casos indican dos estos aspectos indican que el "sector" de la poblacion que auin queda mucho por hacer en algunos paises. Es pre- hace un uso mas intensivo de personal que la mayoria ciso seguir con detenimiento la tendencia de los proyectos de los demas y que se debe tener mas en cuenta esta dife- para la poblaci6n a crecer en tama-no y en necesidad de fi- rencia. nanciamiento de los costos ordinarios, a fin de asegurar que Por ultimo, Zque puede decirse acerca del nivel general no se sobrepasen la capacidad de absorci6n y los LHmites de de la labor del Banco en esta esfera?, cuesti6n esta que ha sostenibilidad. Por razones de la misma indole es preciso provocado importantes criticas hacia el Banco en los ulti- prestar mas atenci6n a la restricci6n de los costos y a la efi- mos afios. Al analizar este tema debe recordarse que los ciencia interna de los proyectos y programas de planifica- aportes financieros por si solos son un indicador deficiente ci6n familiar. Antes de redoblar el empefno en el del nivel de empefno puesto de manifiesto, especialmente financiamiento de elementos anticonceptivos, se debe pen- en el ambito de la poblaci6n, y que la labor del Banco nci sar mas en el financiamiento de las instalaciones de produc- puede evaluarse sin considerar lo que estan haciendo otros ci6n de anticonceptivos en paises mas grandes. La donantes que tienen a su disposici6n fondos para otorgar capacidad del Banco y los paises para evaluar la eficacia de en calidad de donaci6n. los proyectos y programas es deficiente y seguirA siendolo En los paises que fueron objeto del estudio de casos no se en la medida en que sigan descuidandose los componentes han encontrado pruebas de que los resultados hubieran de los proyectos vinculados con el seguimiento y la evalua- sido mejores si se hubiera contado con recursos financieros ci6n ex post y el fortalecimiento de la capacidad de investi- adicionales para el tipo de proyectos que el Banco financia gaci6n. en estas naciones. En casos especificos se podrian utilizar Los estudios de casos tambien revelan varias ensefianzas algunos fondos adicionales a fin de permitir un mayor de mayor alcance. La labor desarrollada en el ambito de la aporte de personal para las tareas de supervisi6n y asisten- poblaci6n exige paciencia pero en definitiva da resultado. cia tecnica, emprender mas actividades no relacionadas con los proyectos (mAs estudios sectoriales, mas colaboraci6n y 8. Si bien vale la pena intentar otra propuesta reciente -la labor encami- coordinaci6n con otros donantes, y un papel mas activo en nada a preparar proyectos de recursos humanos integrados-, es posible que algunos paises), y a fin de ampliar las actividades relacio- en definitiva no sea de mucho provecho. El riesgo que se corre es que dichos nadas con la poblaci6n en otros sectores. Tambien se nece- proyedtos pueden terminar siendo "arboles de Navidad" de componentes independientes que constituyen pesadillas administrativas en el momento sitaran recursos adicionales para desarrollar actividades en de su ejecuci6n. Tal vez sea mas prictico el metodo de preparar proyectos in- paises en los que el Banco actualmente no tiene programas. dependientes que incluyan aspectos de interes concernientes a la poblad6n, Sin embargo, dichos cambios representarian una reorienta- por ejemplo, proyectos de educaci6n que abarquen la concentraci6n de es- fuerzos en la matriculaci6n de las nifias y en su asistencia a la escuela por un cion y ampliaci6n moderadas de un programa que, en periodo mnas largo, y proyectos industriales dentro de los que se reserven de- general, parece estar bien encaminado. terminadas tareas para las mujeres. 15 Resume analytique La pr6sente etude analyse les activites menees par la au Kenya; 40 % au Bangladesh, 50 % en Inde; 60 % en Banque dans le secteur de la population en examinant le Indonesie et plus de 70 % au Bresil, en Colombie et au r6le qu'elle a joue dans huit pays. Pour chaque pays, elle- Mexique. Cette evolution est particulierement encoura- commence par identifier les changements demographiques geante dans le cas du Bangladesh et du Kenya oii les resul- et les transformations socioeconomiques connexes surve- tats, tout r6cemment encore, etaient tres modestes malgre nus depuis 1968, date A laquelle la Banque a annonce qu'el- les efforts deployes depuis 20 ans pour reduire les taux de le commencerait A accorder des prets pour ce secteur. croissance de la population. Les taux de mortalite, en parti- I'etude cherche egalement A identifier les mesures gouver- culier la mortalite infantile et juvenile, ont egalement forte- nementales qui ont pu influer sur ces tendances et le r6le ment baisse dans les pays d'Amerique latine, soit d'environ que la Banque a joue A l'6gard des autres bailleurs de fonds 50 % depuis 1960. Les taux de fecondite, de mortalite et de pendant cette p6riode. Les conclusions de l'etude reposent croissance demographique n'en restent pas moins eleves, sur des visites sur le terrain, des entretiens et l'examen selon la plupart des normes, surtout en Asie et en Afrique. d'un certain nombre de documents et publications opera- Compte tenu de l'extreme jeunesse de la population de ces tionnelles. pays et des taux actuels de variation de la fecondite et de la Le tableau qui se degage de cette etude est un kaleidos- mortalite, il est probable que leurs taux de croissance de- cope d'activites et d'experiences qui ne se prete guere aux mographique resteront pendant encore un certain temps generalisations. Si on laisse de c6te pour le moment cette aux alentours ou au dessus de 2 % par an. diversite et ces reserves, l'etude arrive A la conclusion gen& Cette baisse de la fecondite est due, pour les deux tiers rale que la Banque, apres un demarrage lent, hesitant et ou les trois-quarts, a un recours accru aux methodes contra- parfois mauvais, est devenue de plus en plus efficace dans ceptives modernes. Pour le reste, elle est essentiellement ce domaine, meme si elle aurait pu faire et accomplir da- imputable au recul de l'Age au mariage. Ces changements vantage. Dorenavant - A supposer que les changements resultent A leur tour d'une serie d'ameliorations d'ordre so- recents apportes aux modalites d'action de la Banque dans cioeconomique qui tendent A encourager des familles ce secteur soient appliques avec vigueur et coherence - ce moins nombreuses et un recul de l'Age au mariage. Citons, qu'il faut, essentiellement, peur op6rer les changements de- par exemple, les progres constates dans le niveau d'instruc- mographiques souhaites qu'implique la politique sectoriel- tion, les possibilites d'emploi et la condition des femmes; la le de la Banque, c'est une reorientation et une refocalisation baisse de la mortalite infantile, la diminution des possibili- des efforts, et non une modification radicale des pratiques tes d'emploi pour la main d'oeuvre enfantine et une moin- et des niveaux de financement recents. dre dependance A l'6gard des enfants en tant qu'assurance- vieillesse. II convient 6galement de mentionner les pro- Tendances d6mographiques et grammes de planning familial (PF) dont le but est de diffu- sociecornomiques ser largement des contraceptifs modernes et de fournir des services d'information et d'education dans ce domaine. Les indices synthetiques de fecondit6 ont diminue dans Dans les trois pays d'Amrnique latine, ce sont les effets tous les pays ayant fait l'objet d'etudes de cas: marginale- conjugues de ces deux facteurs une amelioration sensible ment au Senegal; 20 % de l'&cart entre le maximum de huit des indicateurs sociaux et economiques et l'existence de enfants par femme et un niveau de remplacement de 2,1 services de planning familial bien developp6s et efficaces, 16 en l'occurrence foumis par des organisations non gouver- Bangladesh, mais pour les enseignements qu'on pourra en nementales (ONG) et quasi-publiques, et par des voies tirer pour des pays dont la situation est analogue. commerciales qui ont engendre les transformations demo- graphiques observees. L'lndonesie a elle aussi connu des Le role de la Banque dans chaque pays progres socioeconomiques appreciables mais, dans ce pays, la societe est plus traditionnelle et rurale, et les taux de Indonesie. La Banque a beaucoup contribue, sur le plan fi- mortalite matemelle et juvenile y sont plus eleves. Le pm- nancier, A l'effort global de developpement de l'Indonesie, gramme de planning familial lance sous l'egide du Gouver- mais ses prets au secteur de la population n'ont represente nement a compense cette difference de diverses fasons, que I % de son aide totale. Ce montant a contribue pour en- dont la p]lus importante peut-ttre est le vaste reseau efficace viron 10 % aux depenses totales consacrees par l'Etat A son d'animateurs qui s'etend jusqu'au plus petit village. Face A programme de population, la part des autres bailleurs de des progres socioeconomiques encore plus modestes, le fonds, y compris les ONG, s'elevant A 20 %. En foumissant programme de l'Inde A egalement compense ce manque ces fonds, la Banque a joue un role assez traditionnel en ce d'amelio:ration mais de facon differente, en focalisant l'ef- sens qu'elle a agi plus comme une banque que comme une fort, de faqon tres etroite mais efficace, sur la sterilisation. institution de developpement: dans l'ensemble, c'est com- Ce programme a cependant essuye des revers A cause des me une banque commerciale qu'elle a reagi aux proposi- r6percussions politiques qu'ont suscite des campagnes p& tions de projets qui lui etaient soumises par les autorites riodiques de coercition. Ayant reussi A toucher une grande indon6siennes; elle n'a fourni qu'une assistance technique proportion de couples desirant cesser de procreer, le pro- limitee pour la preparation et l'execution des projets dans gramme de l'Inde a maintenant atteint le stade des rende- ce secteur et, pour ces projets, elle s'est contentee de finan- ments decroissants. Le cas du Kenya est interessant car les cer les travaux de genie civil et le materiel. La Banque a ce- progres socio&onomiques importants des 25 dernires an- pendant elabore et finance un vaste programme nees semblent avoir engendre une forte demande latente de d'education en matiere de population, mais, pour les autres limitation des naissances qui n'a commence A se manifester aspects non materiels, elle a generalement soutenu l'ex6cu- que tres recemment. Une longue periode de gestation tion des projets mis au point par le BKKBN (Programme aboutissant A un programme efficace, ajoutee aux encoura- national de planning familial de l'Indonesie) avec l'assis- gements energiques du President et d'autres dirigeants, tance technique d'autres bailleurs de fonds. Pendant pres- semblent finalement avoir fait une difference. Au Senegal, que toute la periode A l'tude, la Banque n'avait aucun par contre, le d6veloppement socioeconomique et l1'tat d'a- specialiste resident responsable de ce secteur et le travail vancement du programme ont et insuffisants pour avoir sectoriel a e limite. un impact sensible sur la fecondite. Sauf peut-etre pendant ses premieres annees de partici- Le cas du Bangladesh est particulierement interessant pation au programme, l'approche suivie par la Banque etait dans la mesure oui il y a un programme bien developpe parfaitement justifiee. Les autorit6s indonesiennes savaient mais des progres socioeconomiques tres modestes - cas ce qu'elles voulaient, et elles s'etaient montr6es competen- quasi classique d'un pays qui a decide d'agir sur l'offre tes au niveau de l'execution. Le BKKBN n'est-il pas genera- dans une situation ofi l'on ne s'attendait pas A trouver beau- lement considere comme l'une des organisations de coup de demande. Or, meme dans ce cas, il devait exister planning familial les plus efficaces dans le monde. D'autres une demande latente suffisante pour aboutir aux taux de bailleurs de fonds - en particulier l'Agence des Etats-Unis prevalence de la contraception et A la baisse de la fecondite pour le developpement intemational (USAID) - jouaient enregistres jusqu'A present. Reste A savoir jusqu'A quand alors un r6le tres actif dans l'elaboration de principes ope- ces progres pourront se poursuivre en l'absence de modifi- rationnels et de tactiques d'execution. C'etait-lA, de fait, une cations structurelles profondes modifiant la demandel. Le division du travail rationnelle, qui etait bien appreciee par rapport envisage plusieurs explications possibles et preco- les autorites indonesiennes. nise une action de recherche. Des travaux de recherche sont Le rapport examine cinq domaines dans lesquels la Ban- importants non seulement pour l'avenir du programme du que aurait peut-etre pu faire davantage, ou agir differem- ment, en Indonesie. 1) 11 aurait et utile d'avoir dans le pays un specialiste resident qui reste plus longtemps et prenne une part active a l' elaboration et l' execution des pro gram- 1. Cette facon de presenter les choses implique que les activites d'informa- mes. 2) La collaboration avec d'autres bailleurs de fonds tion, education et communications (IEC) font partie integrante des program- mes de planning familial, cest-A-dire qu'il s'agit de mesures ax6es sur l'offre. aurait pu etre meilleure durant les 15 premieres annees de On suppose en effet que si les services d'IEC peuvent diffuser l'information participation de la Banque. Cet aspect du programme s'est et encourager la demande, 11 doit y avoir d'ores et djA une certaine predis- fortement ameliore depuis. 3) La Banque aurait pu encou- position-- une demande latente cr6ee par des forces culturelles, sociales ou economiques fondamentales - pour que ces efforts puissent aboutir. rager les autorites A agir plus t6t et plus resolument dans le 17 secteur de la sante, surtout en contribuant vigoureusement les naissances et (comme il est note plus haut) aboutissant A promouvoir la sante maternelle et infantile (SMI) et son maintenant A des rendements decroissants; un manque integration avec le planning familial. Elle vient seulement d'attention accordee au fonctionnement et A la qualite du de commencer A le faire. 4) Pour stimuler la demande de fa- programme (par opposition A l'expansion); une centralisa- milles moins nombreuses, la Banque aurait pu encourager tion excessive et le recours A un systeme unique de presta- 1'elaboration de strategies autres que le planning familial, tion; et un desinteret pour les facteurs pouvant influer sur par exemple, en aidant A mettre sur pied des projets visant la demande. Ces problemes ont persiste, et sont connus de- a ameliorer les possibilites d`6ducation et d'emploi offertes puis au moins aussi longtemps que la Banque participe au aux femmes. II reste encore beaucoup A faire dans ce domai- programme de l'Inde. ne. 5) 11 est toujours necessaire de renforcer le travail de re- Le manque d'influence de la Banque sur le programme cherche et d'evaluation, et ce pour au moins deux raisons: de l'Inde s'explique par divers facteurs. 1) Lorsqu'elle a Ian- pour tirer de cette experience concluante des enseigne- c6 son premier projet de population en Inde en 1972, la stra- ments utilisables ailleurs2 et pour bien integrer les moyens tegie du pays etait deja bien etablie. 2) Si les montants de recherche et d'evaluation dans le BKKBN, cette fonction pretes ont constitu6 28 % de la totalite de son portefeuille ayant et jusqu'ici negligee. au secteur de la population jusqu'en 1988, ces sommes ne Cependant, des ameliorations n'6taient manifestement representaient qu'une petite fraction - 3,6 % entre 1980 et pas necessaires dans ces domaines etant donne les progres 1988 - des depenses totales consacrees au Programme de notables r6alises jusqu'a present. En outre, les efforts de la protection de la famille de l'Inde. Les bailleurs de fonds, Banque auraient pu se heurter a l'opposition du BKKBN et pris ensemble, ont contribue pour 12 a 14 % aux depenses des autres bailleurs de fonds qui preferaient que celle-ci se publiques, mais ils n'ont jamais forme une coalition effica- borne essentiellement A financer les constructions et le ma- ce. 3) La nature etroitement regionale des cinq premiers t.riel. La Banque s'est conformee aux voeux de son client, projets a dissuade la Banque d'intervenir dans des ques- ce en quoi elle a eu raison car l'Indonesie avait des moyens tions de politique g6nerale qui depassaient ce cadre regio- considerables et recevait en plus I'assistance d'autres nal (geographique) et a decourage l'innovation, meme dans bailleurs de fonds. ces regions3. 4) Tant les etudes sectorielles que le temps de En Indonesie, une approche differente pourrait s'av6rer travail consacre par le personnel de la Banque au program- necessaire a l'avenir, et ce pour trois raisons importantes. me ont 6te insuffisants. Avant 1988, peu de travail sectoriel Premi2rement, le pays a beaucoup moins besoin de pro- a 6te fait - certainement pas assez pour que la Banque soit grammes massifs d'infrastructure. Deuxiemement, l'USAID en mesure de proposer des changements tactiques, par a indique son intention de reduire sa contribution, en partie exemple, de suggerer comment on pourrait modifier les parce qu'elle estime que l'Indonesie a maintenant moins be- structures du programme - incitations, suivi et responsa- soin d'assistance technique et de dons. Avec l'assistance de bilits - pour donner plus de poids aux methodes non ir- l'USAID, le BKKBN a dejA reagi en prenant des dispositions reversibles. Le temps de travail consacre par le personnel pour encourager le secteur prive A promouvoir et A distri- de la Banque - sur place comme A Washington n'6tait pas buer les services. Troisiemement, la formule qui a permis de non plus suffisant pour permettre A la Banque d'intervenir ramener le taux de fecondite de 5,5 A 3,5 enfants par femme de plus pres. ne permettra peut-etre pas d'atteindre le niveau de rempla- Depuis 1987, plusieurs changements importants se sont cement de 2,1. Face A ces changements, la Banque peut faire produits. Le Gouvernement comme la Banque semblent deux choses: reduire le champ de ses activites, A l'instar de avoir et6 d6eus par les resultats des projets regionaux et l'USAID, ou essayer d'occuper le vide laisse par cet organis- sont prets A adopter une approche differente. Les sixieme et me. Selon le rapport, ces deux approches sont necessaires, septieme projets, amorces en 1989 et 1990, temoignent de en fonction des problemes en. jeu. cette evolution puisqu'ils portent sur des volets prioritaires Inde. En Inde, comme en Indonesie, la Banque jusqu'A re- du programme au niveau de l'Etat. Ils prevoient egalement cemment s'est comportee davantage cornme une banque un appui pour les ONG et le programme de ventes subven- traditionnelle que comme une institution de developpe- tionn6es. Deuxiemement, depuis 1987, la Banque a produit ment. Dans ce cas, ce r6le 6tait cependant beaucoup moins justifie A cause des probl6mes strategiques lies au program- me, A savoir: une concentration excessive sur la sterilisa- 3. C'est 1a un point contestable. Uargument au depart etait que la Banque tion, ignorant la necessite pour les couples jeunes d'espacer aurait dans ces domaines plus d'influence qu'elle n'en pouvait avoir dans le programme global. 11 est probable que cela ne s'est pas passe ainsi, non pas parce que le concept etait mauvais, mais parce que les interlocuteurs indiens n'ont jamais accepte que les domaines d'action appuyes par la Banque puis- 2. Par exemple, quels sont les facteuirs qui expliquent le succes des opera- sent etre differents des autres, et parce que la Banque n'a jamais pris l'initia- tions communautaires de ce pays? Ces facteurs sont-ils uniques a l'lndonesie tive d'e1aborer, pour ses projets, des plans detailles qui auraient explicite ce ou peuvent-ils etre appliqu6s a d'autres pays? point avant l'ex6cution. 18 plusieurs etudes sectorielles de qualite qui semblent avoir ment l'assistance et les conseils de l'ext6rieur. 2) Le volume un effel: sur les attitudes en Inde et sur la nature du dialo- de l'assistance financiere et technique exterieure a ete subs- gue. Le Projet survie de l'enfant et maternit6 sans risque en- tantiel. 3) Une part considerable de cette assistance finan- visage pas la Banque focalise pour la premiere fois ciere, meme dans le cadre des cinq projets de population l'attention sur des questions de sant6 qui ont une grande in- appuyes par la Banque, a ete fournie sous forme de dons fluence sur les decisions en matiene de fecondite. Ce projet par des cofinanciers, ce qui a permis d'agir avec plus de s'interesse egalement de plus pres aux operations sur le ter- souplesse et d'axer les efforts sur les elements non mat& rain qu:i, dans le passe, se sont revelees un gros obtacle A la riels, et en meme temps de financer un un apport supple- mise en, oeuvre des reformes. Enfin, la Banque a beaucoup mentaire de personnel de la Banque, tant au siege que sur intensifie les efforts qu'elle deploie dans le secteur de l'edu- le terrain. 4) Avec ces ressources additionnelles en person- cation, en particulier en faveur des femmes. nel et les encouragements du Gouvernement, la Banque est Toutes ces initiatives sont encourageantes, mais le resul- intervenue activement dans toutes les phases de ces projets tat d6pendra pour beaucoup de la vigueur et de la constan- - conception, mobilisation des fonds, coordination avec ce avec lesquelles elles sont menees. La Banque, pour sa les bailleurs de fonds, execution et evaluation - beaucoup part, devra continuer, par ses etudes sectorielles et ses tra- plus activement qu'elle ne le fait normalement dans les pro- vaux de recherche, de chercher des moyens pratiques de re- jets qu'elle soutient. 5) La collaboration entre les bailleurs soudre les problemes structurels; continuer de militer - de fonds a e tres etroite, et les cofinanciers, qui ont gene- energiquement lorsque c'est necessaire - pour que les re- ralement fourni plus de la moitie du financement de ces sultats de ces travaux soient appliques; faire plus d'efforts projets, ont jou6 un role de plus en plus actif. Leur degre de pour mobiliser l'appui d'autres bailleurs de fonds dans cet- participation est maintenant devenu si etroit qu'il serait te entreprise; et accroitre le temps et les efforts consacres a plus juste de voir le quatrieme projet comme un projet con- la supervision des operations sur le terrain. Pour cela, la sortial que comme un projet de la Banque. 6) Grace aux Banque aura probablement besoin d'utiliser tant au siege dons qui ont e foumis, on a pu integrer une certaine sou- que sur le terrain, plus de personnel connaissant bien les plesse dans le programme, notamment en creant un Fonds problernes en jeu. pour les projets novateurs et une Cellule de financement Bangladesh. Le programme du Bangladesh souffre A peu des projets. 7) Les enseignements tires des projets pilotes pres des memes faiblesses que celles qui caracterisent le ont ete mis A profit pour ameliorer la performance du pro- programme indien. Si le pays fait une place moins impor- gramme. tante A la sterilisation, la qualite des services de planning Dans ce cas, il y a lieu de penser que le programme de- familiat est mediocre, le programme de SMI est deficient et vrait etre poursuivi dans le sens des orientations actuelles mal integre au planning familial, et la penurie de gestion- pendant les annees a venir. D'apres certaines enquetes, il naires et de traditions administratives enracinees rendent existe une masse d'utilisateurs potentiels qui n'ont pas en- toute amelioration difficile. En outre, il ne faut pas oublier core ete touches par les services, et il est prouve que le taux que le programme opere dans un environnement physique de prevalence est plus eleve lA oii la densite des points de extremement difficile oii des crises periodiques exigeant services est grande, oii les contacts avec les agents du plan- une attention prioritaire des autorites menacent constam- ning familial sont frequents et oii les services sont de quali- ment les progres accomplis. t6. Toutefois, meme si l'on portait le taux de pr6valence N6anmoins, les progres realises pour accroitre la preva- actuel de 35 % au niveau de 50 % obtenu dans la zone pilote lence de la contraception et reduire la fecondite sont, pour de Matlab - ce qui est peu probable A l'chelle nationale - la majeure partie, imputables au programme. Les activites l'indice synthetique de fecondite serait d'environ 3,7, ce qui de vulgarisation, qui sont plus etendues et plus dynami- est encore bien superieur au niveau de remplacement. Pour ques qu'en Inde malgre leurs limitations, sont probable- abaisser encore ce taux en un laps de temps raisonnable, il ment l1'l1ment qui fait la difference. En outre, les bailleurs faudra probablement reorienter le programme de facon A de fonds ont pris une part plus active a la planification et completer l'approche actuelle axee sur l'offre par d'autres aux op6rations, ce qui a en fait demultiplie les moyens types d'action. Si le nombre d'enfants idal par famille sem- d'execution de l'administration. Ii convient A cet egard de ble avoir diminue au Bangladesh, meme en l'absence de louer la Banque qui, en l'occurrence, a joue un role actif et progres socioeconomiques sensibles, une nouvelle diminu- influent. tion serait plus sure et plus rapide si l'on pouvait operer des Le rapport enumere sept facteurs qui semblent avoir ete changements structurels tendant A accroitre la demande de essentiels. 1) Le Gouvernement est manifestement resolu A familles moins nombreuses. resoud re son probleme de population: il a consacre une Bresil, Colombie et Mexique. A la difference des autres part appreciable de ses ressources budgetaires et gestion- pays, ces pays n'ont recu de l'exterieur qu'une faible part nelles limitees au programme, et a sollicite et utilise active- des fonds consacres au planning familial, et ce sont des 19 ONG, et non des organismes multilateraux ou bilateraux, nalite de son action, elle doit alors fonder son programme qui ont fourni la majeure partie de ces fonds4. La contribu- de population sur des principes plus generaux et plus sou- tion de la Banque a e minimale. Elle n'a finance aucun ples. On pourrait, par exemple, commencer par reconnaitre projet de population dans ces pays. (Un projet a e prepare que l'objectif global est de promouvoir une am6lioration pour le Mexique, mais il a et arrete lorsqu'un nouveau durable du niveau de vie et que les programmes de plan- gouvernement est arrive au pouvoir.) Peu d'efforts ont e ning familial sont utiles au niveau microeconomique - en faits pour mettre a execution lies quelques 6lements de plan- ameliorant la sante et les choix de la famille -ainsi qu'au ning familial inclus avec la SMI dans les projets de sant6 fi- niveau macroeconomique. Dans ce contexte, il y a lieu d'in- nanc6s par la Banque, et ce sujet a rarement ete aborde dans clure ces programmes dans les themes des discussions de le dialogue avec les pouvoirs publics. Au moins jusqu'en politique generale et de l'analyse sectorielle, afin de deter- 1989, tres peu d'etudes sectorielles ont porte directement miner comment, le cas echeant, ils devraient etre concus et sur les questions de population ou de planning familial. De utilises dans des circonstances specifiques. En outre, ces surcroit, si les prets aux autres secteurs ont surement eu des etudes de cas posent la question de savoir s'il est souhaita- effets indirects, ils n'ont jamais pris en compte la dimension ble, dans certaines circonstances, de preter pour des projets demographique. purement de population, plut6t qu'a des projets de sant6 Trois facteurs semblent expliquer ce desint&ret apparent. ayant des volets de planning familial. Si elle avait adopte 1) A cause des resonnances politiques des politiques de re- cette approche plus t6t et plus systematiquement, la Ban- gulation des naissances preconisees par l'etranger il etait que n'aurait pas rencontre autant d'opposition et de me- difficile pour le personnel de la Banque d'aborder cette fiance dans ces pays. question avec les gouvernements. Aux yeux de ces der- Kenya. Cette etude de cas montre bien combien il est utile niers, les positions de la direction de la Banque etaient inac- - et necessaire d'etre patient et perseverant dans le domai- ceptables; pris entre deux feux, le personnel de la Banque a ne de la population. Malgre les efforts intenses deployes r6agi par l'inaction. 2) A l'epoque oui la Banque a commen- par les bailleurs de fonds, depuis la fin des annees 60, pour ce a preter pour les services de sante - ce qui aurait pu etre convaincre les autorites de lancer un programme de grande un vehicule plus acceptable pour promouvoir la PF - la envergure, ce n'est qu'en 1988-89 qu'on a commence a voir prevalence de la contraception etait dej3 relativement 6le- baisser la fecondite. vee et la structure de prestaticn des services bien etablie. 3) En plus d'un appui a l'effort global de developpement Dans ces pays, les organismes chefs de file etaient des ONG economique, qui comportait des fonds importants pour l'e- ou des organismes semi-publics. Compte tenu de l'attitude ducation, la contribution de la Banque a pris essentielle- des gouvernements, la Banque (a la difference de l'USAID) ment deux formes. A partir de 1974, quatre projets ont e n'avait pas encore trouve le moyen de travailler directe- finances. Les deux premiers ont mis des fonds a la disposi- ment, ou meme indirectement, avec ces organismes, com- tion du Ministere de la sante pour la cr6ation d'un reseau me elle a su le faire au Kenya et ailleurs. de centres ruraux de formation et de soins de sante; les vo- L'approche actuelle de la Banque en Amerique latine lets de planning familial de ces projets laissaient a desirer et consiste a mettre l'accent sur lVhygiene de la reproduction ont e largement ignor6s pendant l'execution. Durant cette et sur la maternite sans risque comme justifications des ac- periode, le taux de croissance de la population a en fait aug- tivites de planning familial. Cette approche semble appro- mente. Les troisieme et quatrieme projets faisaient une plus priee, et une attitude plus agressive ne serait ni justifiee, ni grande place aux activit6s de population et de planning fa- judicieuse sur le plan politique. II n'en reste pas moins milial, mais ils ont e lanc6s trop tard (1988 et 1990) pour d'importants aspects - equit6, qualite des services et choix avoir influence les tendances de la fecondit6. La contribu- - a resoudre, et ces aspects s'inscrivent de toute facon dans tion de la Banque a prix une deuxieme forme, a savoir son le contexte plus large des questions sur lesquelles la Banque dialogue avec les pouvoirs publics, ses travaux sectoriels et travaille actuellement dans le secteur de la sante5. les efforts qu'elle a faits pour influencer de diverses facons Ces cas d'Amerique latine sont importants car ils mettent l1&volution du programme. C'est ainsi qu'elle a reussi a en cause le bien-fond6 des prets de la Banque au secteur de convaincre le Gouvernement de creer, en dehors du Minis- la population. Si la Banque veut travailler dans des pays tere de la sante, un organisme de coordination interminis- qui n'acceptent pas la limitation des naissances comme fi- teriel pour les questions de population (condition a laquelle etait assujetti le deuxieme pret d'ajustement structurel), puis de lui confier davantage de responsabilites; et elle s'est 4. Certaines ONG internationales ont cependant beneficie d'appuis bilat& efforcee de persuader le Ministere de la sante de liberaliser raux et multilateraux, en particulier cle l'USAID et du Fonds des Nations les directives de distribution des contraceptifs, d'integrer la Unies pour les activites en matiere de population (FNUAP). PF aux activit& traditionnollos du Minist&e de la sant6 et 5. 1l convient de signaler a cet eganl l'existence d'un tres grand nombre d'avortements illicites dans ces pays. voir Singh et Wolf, 1991. d'offrir des services de sterilisation. Ces deux lignes d'ac- 20 tion ont et6 des elements importants d'une strategie globale observee ces derni&es annees. Cette strategie pourrait etre qui comportait une assistance technique et une aide en na- particulierement importante en Afrique francophone A cau- ture d'autres bailleurs de fonds, et, plus important encore se de son heritage colonial nataliste avec lequel il faut rom- peut-etre, des signaux explicites du President du Kenya in- pre explicitement et publiquement si l'on veut legitimiser diquant qu'il fallait prendre au serieux le planning familial. les activites de planning familial. La Banque aurait-elle pu faire quelque chose pour acce- La decision de la Banque de se concentrer sur l'elabora- lerer ce processus? Si elle aurait pu faire beaucoup plus tion d'une politique d'ensemble etait probablement justi- d'efforts pour mettre sur pied des composantes efficaces fiee A l'6poque car, contrairement aux autres bailleurs de d'IEC, de vulgarisation, de suivi, d'evaluation et de recher- fonds, elle n'avait pas le personnel de terrain voulu pour che, la Banque s'est montr6e assez resolue, face au Gouver- mettre a execution des operations plus complexes, mais elle nement, sur un grand nombre de ces questions et semble avait acces aux plus hauts responsables du fait que son pro- avoir tire parti des occasions qui s'offraient A mesure qu'el- gramme d'action etait beaucoup plus vaste. Plus recem- les se presentaient. Dans ce cas, l'absence d'agents resi- ment, la Banque a deide de jouer un r6le beaucoup plus dents affectes A ce secteur ne semble pas avoir et une dynamique en lancant un projet qui, entre autres, vise a of- grosse entrave vu l'existence d'amples ressources pour la frir des services de planning familial et autres au plus bas supervision et les deplacements, et la contribution apportee echelon du systeme de soins de sante senegalais. C'est IA par d'autres bailleurs de fonds qui disposaient eux, d'un une entreprise risquee. L'USAID, par exemple, se contente personnel technique sur le terrain. d'agir au niveau juste au-dessus, car l'Etat a des moyens si Le programme semble etre plus ou moins sur la bonne limites pour administrer et fournir les services. Etant donne voie pour le moment, et il faudrait continuer de le develop- la nature urgente du probleme, il est probable que ces ris- per, tout en mettant beaucoup plus l'accent sur la qualite ques valent la peine d'etre pris et qu'on peut les contenir des services, la vulgarisation et les services d'IEC. Toute- dans des limites acceptables en commencant A petite echel- fois, mene si l'on arrivait a combler totalement l'ecart entre le et en elargissant les operations lentement; ce type de pro- le nombre d'enfants ideal et le nombre d'enfants effectif par jet requiert cependant une supervision beaucoup plus famille, l'indice synthetique de fecondite serait encore de intense et plus etroite que celle qui caracterise normalement 4 A 5 enfants, ce qui implique un taux de croissance de la les projets de la Banque au Senegal. population encore superieur A 2 % par an. Entre 1984 et 1987, le nombre ideal d'enfants est tomb6 en moyenne de Problemes cdes a resoudre pour la Banque 5,8 a 4,4. Pour accelerer cette evolution, une strategie ax6e sur l'offre ne suffit plus. Le rapport propose un certain L'experience acquise ces 20 dernires annees montre clai- nombre d'options, mais fait valoir que la recherche et les rement que, meme dans les pays pauvres qui n'ont pas en- 6tudes pilotes sont extremement importantes dans ce cas, core fait beaucoup de progres sur le plan social et car on ne sait pas encore tres bien comment on peut aider A economique, on peut amorcer une baisse de la fecondite reduire le nombre d'enfants ideal dans un systeme forte- avec un programme type de planning familial centre sur la ment nataliste. distribution de moyens de contraception, de services et in- Senegal. A part un projet de sante rurale, concu en 1982, formation connexes. Cette baisse interviendra plus rapide- qui consistait a financer des constructions et du mat6riel ment si le programme fait une grande place A la pour etendre les services de sante de base, la Banque, jus- vulgarisation et fournit des services de qualite adaptes aux qu'A une date encore recente, s'est efforcee avant tout d'ai- besoins des clients. Cette strategie exige un bon encadre- der le Gouvernement A 6laborer une politique generale en ment sur le terrain, une bonne formation et une bonne mo- matiere de population. Cette strategie avait et recomman- tivation - autant de conditions qui sont difficiles A reunir dee par une etude sectorielle de 1985-86, qui avait conclu dans les zones rurales pauvres. que l'USAID et le FNUAP faisaient dejA tout ce qui pouvait Toutefois, il ne semble pas, meme avec le programme le etre fait utilement pour developper les services de PF. Pour mieux administre, qu'on puisse faire beaucoup plus que re- mettre en oeuvre cette strategie, la Banque a decide, entre pondre aux besoins de ceux qui sont dejA predisposes A ac- autres, de faire de l'6laboration de cette declaration de cepter des services de planning familial6. Dans tous les cas politique de population l'une des conditions d'un pret A etudies, meme si toutes ces personnes participaient A un l'ajustement structurel. Certains responsables gouveme- mentaux ont peut-etre accede A cette demande parce qu'elle paraissait assez anodine, mais cette declaration et son pro- 6. L'6tude du cas du Bangladesh peut nous amener a modifier cette affir- cessus d'elaboration ont fait boule de neige et entrame de mation. II est plausible que dans ce casle succ6s des efforts d'IEC et l'effet de grands pDrogris, progres qui auraient sinon e beaucoup d6monstration d'une pr6valence cmoissante de la contraception aient con- grands progr~~~~~s, progr~~~s qui ~tribu6 A r&luire le nombre d'enfants id6al, malgr6 I'absence de progri!s so- plus lents, A cause de la stagnation sociale et economique cioconomiques. 21 programme de planning familial - ou meme si l'on pou- ment due au fait que la population est generalement trait6e vait etendre a tous le pays les, taux de prevalence de la con- comme un secteur A part qui releve administrativement traception les plus 6leves enregistres dans les meilleurs d'un service donn6, ce qui libere implicitement les autres programmes pilotes - il n'y a pas un seul pays parmi les services sectoriels de cette responsabilite. Or la population huit, sauf peut-tre les trois d'Amerique latine, oii l'indice n'est pas un secteur d'activite distinct, pas plus que le de- synthetique de fecondite toimberait A un niveau accepta- veloppement economique ou la reduction de la pauvrete. ble7. D'autres types d'interventions seront necessaires si Dans ces trois cas, il s'agit. plus d'objectifs strategiques dont l'on veut aller plus loin. Des strategies axees sur l'offre peu- la responsabilite incombe A tous les secteurs. vent Wtre suffisantes pour rainener de 6 A 4 l'indice synth& Depuis 1987, on a assiste a plusieurs changements qui tique de fecondite, mais, en I'absence de transformations ont des repercussions positives sur la situation au sein de la socio6conomiques profondes, une strategie differente sera Banque. L'integration administrative de l'education, de la probablement necessaire pour le ramener de 4 A 2. population, de la sante et de la nutrition a permis et encou- La Banque n'a pas su elaborer des programmes centres rage une reflexion sur la facon dont chaque secteur pouvait sur cette question. Les efforts les plus importants qu'elle a concourir a la realisation des objectifs des autres secteurs. faits dans ce sens ont consiste A promouvoir le developpe- Certaines initiatives, Le rdle de la femme dans le dkveloppement ment social et economique general. Plus que tout autre fac- et La survie de l'enfant, focalisent actuellement I'attention sur teur, c'est la probablement le moteur le plus puissant que deux des possibilites les plus importantes8. Plusieurs com- nous connaissons pour reduire le nombre ideal d'enfants et posantes prometteuses commencent A apparaitre dans les encourager l'espacement des naissances. Mais ces efforts projets d'6ducation. Cependant, on pourrait et on devrait n'ont pas ete entrepris dans la perspective des effets d6mo- faire bien davantage. La Banque est exceptionnellement graphiques qu'ils pouvaient avoir. Si cela avait ete fait - si bien placee pour cela. la Banque avait cherch6, en favorisant le developpement, En plus d'indiquer la n&essite d'elargir le champ des ac- des moyens d'intervention selectifs susceptibles de modi- tivites de population, les etudes de cas proposent plusieurs fier les avantages et couits implicites des familles nombreu- ameliorations qui peuvent etre apportees dans les projets ses - elle aurait sans doute pu accomplir bien davantage. de population et les secteurs d'activit6 plus traditionnels. II est bien connu, par exemple, que le fait d'instruire les Les premiers projets tendaient A privilegier l'expansion des femmes - ou le simple fait de les garder A l'ecole quelques infrastructures materielles aux depens des elements "non annees de plus - tend A reculer l'Age au mariage et A abais- materiels", ces derniers etant censes etre pris en charge par ser les taux de fecondite des couples en question. II y a l'Etat ou par d'autres bailleurs de fonds. Cette hypothese d'autres moyens d'intervention prometteurs, et notamment s'est averee plus souvent fausse que correcte, et des mesu- les programmes qui visent sp6cifiquement a accroitre les res ont ete prises pour retablir l1'quilibre dans chaque pro- chances de survie des enfants jusqu'A l'Age adulte, la crea- jet. D'apres les etudes de cas, on serait encore loin du tion, pour les femmes, de possibilites d'emploi, entrant en compte dans certains pays. Il convient de suivre de pres la concurrence avec le metier de mere (par exemple, un travail tendance des projets de population A prendre de l'expan- d'usine ou de bureau plutot qu'un travail A la ferme) et l'a- sion et A financer un volume croissant de charges r6curren- melioration des programmes de securite sociale pour que tes, afin de ne pas depasser la capacite d'absorption ni les les adultes ressentent moins ia necessite d'une famille nom- limites de viabilite A long terme. Pour des raisons connexes, breuse pour avoir ce type de securite. il est necessaire de se preoccuper davantage de la maitrise Des etudes anterieures du programme de la Banque dans des cofits et de l'efficience interne au sein meme des projets le secteur de la population avait recommande de mettre da- et programmes de planning familial. Avant de s'engager vantage I'accent sur ces facteurs lies A la demande, mais massivement A financer la distribution des contraceptifs, il peu d'efforts ont ete faits pour que cette recommandation faudrait envisager de plus pres la possibilite de financer, debouche sur des programmes operationnels. Plusieurs dans les grands pays, des moyens de production de contra- facteurs expliquent cette inaction: scepticisme quant A la ceptifs. L'aptitude de la Banque et des pays a evaluer l'effi- necessite d'interventions axees sur la demande; doutes sur ce qui pouvait etre accompli; et compartimentalisation et 8. Une autre initiative r6cente - 1'6laboration de projets int6gr6s de res- inertie du personnel. Cette situation est peut-etre egale- sources humaines - vaut la peine d'etre tent6e, mais pourrait en fin de compte ne pas donner beaucoup de r6sultats. Le danger est que ces projets finissent par etre une collection de composantes ind6pendantes qui sont ad- ministrativement tres difficiles a administrer. 11 serait peut-etre plus pragma- 7. Dans certains de ces pays, il y a cependant des r6gions oft la f6condite tique d'elaborer des projets s6par6s qui int6grent les pr6occupations est proche du niveau de remplacement: Yogyakarta (2,1), Bali (2,5), Java-Est d6mographiques - par exemple, des projets d'education qui visent sp6cifi- (2,6), Kerala, (2,2) et Tamil Nadu (2,6). Dans tous les cas, le d6veloppement quement a scolariser les filles et a les garder a l'6cole plus longtemps, social et economique et/ou la capacit6 administrative sont superieurs a la des projets industriels qui r6servent des emplois sp6cifiques pour les moyenne nationale. femmes, etc. 22 cacite des projets et programmes est limitee, et elle le a et bien critiquee ces dernires annees? Pour se faire une restera tant qu'ils negligeront d'inclure des activites de sui- opinion sur ce point, il convient de se rappeler que la con- vi et d'evaluation et d'etablir des capacites de recherche. tribution financiere est A elle seule un pietre indicateur du Ces etudes de cas offrent plusieurs enseignements plus niveau d'effort, surtout dans le secteur de la population, et generaux. Toute intervention dans le secteur de la popula- qu'on ne peut pas evaluer l'action de la Banque indepen- tion exige de la patience, mais cette patience finit par etre damment de ce que font d'autres donateurs, qui eux dispo- recompensee. Cette constatation doit influencer la facon sent de fonds sous forme de dons. dont les objectifs et les criteres d'evaluation des activites de Dans les pays retenus pour les etudes de cas, rien n'indi- population sont fixes. Les activites hors projet - dialogue, que qu'un surcroit de ressources financieres, pour le type etudes sectorielles, efforts tendant A definir des orienta- de projets que la Banque finance actuellement dans ces tions, A ameliorer l'organisation et A modifier la reglemen- pays, aurait fait une grande difference. Dans certains cas, tation - sont tres importantes. A certains stades, elles sont des fonds supplementaires auraient WS utiles: pour accroi- plus importantes que l'elaboration d'un projet. Ce type tre le temps de travail consacre aux fonctions de supervi- d'activites doit etre appreie A sa juste valeur et encourage. sion et d'assistance technique; pour lancer un plus grand Citons, par exemple, les efforts remarquables deployes par nombre d'activites hors projet (plus de d'etudes sectoriel- la Banque l'an passe pour instaurer un consensus en Afri- les, un plus gros effort de collaboration et de coordination que, leq[uel, soit dit en passant, a maintenant debouche sur avec les autres bailleurs de fonds, et un r6le plus dynami- des propositions de projets utiles. Le cas du Bangladesh est que dans certains pays); pour elargir les activites de popu- particulierement important car il montre le merite d'une lation en dehors du secteur de la population. Des prograrnmation conjointe des prets et des dons, et la facon ressources supplementaires seront egalement necessaires d'operer cette coordination. Toute ces remarques donnent A pour developper les activites dans les pays oii la Banque penser que le "secteur" de la population fait une utilisation n'a pas actuellement de programme. Mais ces changements plus intensive de personnel que la plupart des autres sec- ne representeraient qu'une reorientation et une expansion teurs, ce dont il faut tenir compte au plan des ressources. modestes d'un programme qui, dans l'ensemble, parait etre Enfin, que peut-on dire du niveau global des efforts faits sur la bonne voie. par la Banque dans ce domaine, point sur lequel la Banque 23 1 . Introduction This study assesses the role of the Bank in the population which will lag far behind unless assisted by special sector, by reviewing the specific experience in eight coun- programs. tries. In each case, the study poses the following three ques- In other respects, the countries selected for this study are tions: very dissimilar. They are at very different stages of social, - What demographic and related socioeconomic changes economic, and demographic development. As Table 1.1 have occurred since 1968, when the President of the (see statistical appendix) indicates, their values for a variety Ban.k first announced that it would begin lending for of development indicators span the range experienced by Bopulankfir announced that it would begin lending for developing countries as a whole. The range of values spe- popWhat govemment polides affecting these trends were cifically for demographic indicators is more clearly seen in * What government policies affecting these trends were Figures 1.1 and 1.2 and warrants special comment. No i What role was played by the Bank vis-a-vis other popu- country in this group of eight is at a stage prior to the be- lWhation dooro s playedur b this period? Bki-soepu ginning of a demographic transition, the point at which lation donors during this period? population growth rate is low because of high and roughly From these case studies, which for the most part take a his- equal birth and death rates. Senegal and Kenya are clearly toric approach to events, a number of cross-cutting issues in the first stages of demographic transition, in which death are extracted for further discussion and suggestions are rates are falling without commensurate declines in birth made for future activities. rates. The result, predictably, is high and rising population The countries examined in this study-three in Asia, growth rates. In the other six countries, population growth three in Latin America, and two in Africa-were selected rates are declining at different speeds, almost impercepti- for two reasons. First, all are countries that would benefit bly in the case of Bangladesh and at significant rates in the from programs to alter demographic trends. Second, they case of the three Latin American nations. All eight, howev- span the range of experience covered by most developing er, remain far from the third stage reached by many OECD countries. In all these countries, maternal, infant, and child countries, in which birth and death rates are roughly equal mortality rates are far above desirable levels for the major- at low levels of population growth. ity of their populations. So too are population growth Second, their population programs differ greatly. rates--even in Latin America, where they are still above Senegals program is the least developed; Kenya's the next two percent per annum. In all of them, high fertility rates least-developed. The Asian countries all have large, well- are an important factor contributing to these demographic established family planning programs in place. In Brazil features. Table 1.1 in the statistical appendix provides and Colombia, these matters are more or less left to NGOs comparative data on these and other critical indicators of and para-public organizations, which operate quite size- general socioeconomic well-being and advancement for the able and effective programs. In Mexico, family planning study countries. Social and economic development may services are offered through both the Government social se- eventually ameliorate the problems of rapid population curity system and its Ministry of Health. Third, these coun- grow*,h. However, only in Latin America is development tries cover the range of Bank experience with population likely to occur fast enough to achieve acceptable rates of programs, from very heavy involvement in the case of improvement in the current demographic trends; and even Bangladesh, to almost no involvement in the three Latin here, there are large pockets of relative backwardness American cases. 25 Figure 1.1: Case Study Countries, Birth, Death, Figure 1.2: Case Study Countries, Total Fertility Rates, and Population Growth Rates, 1970-87 1960-89 9 60 60 8 _ j_ Kenya 50 -Growth K Bangladesh 40Rate Mexiro f ZZZSnea 7 .,,, >-.. ee,! - 4% Colombia - ~~~~~~~~~~~~~~~~~~~~~~Senegal /40 0 :e 30 India 2 es-~~ndnsi o20 2 1 0 1%3 C173 0 2 ., . . . . . . . . . . . . . . . E 0 5 10 15 20 25 60 70 80 89 Crude Death Rate Years Note: Growth rates are natural rates ol increase, i.e., excluding migration. Source: World Bank data. Source: World Bank data. Thus, these eight cases provide a good sample of experi- one to observe and speak to a variety of other issues, for ex- ence from which to make generalizations. The following ample, what program features commonly work. However, sections review these experiences one by one. This exercise such issues are not discussed in as much depth as the topic is then used to answer more general questions about the warrants, because of the particular focus of this report. Bank's role in this field. The case study approach allows 26 2 . Indonesia Demographic Trends num. This is partly because mortality has fallen along with fertility and partly because it has a very young age distribu- Since 1970, Indonesia's fertility has fallen more rapidly tion, a product of higher fertility rates in previous decades. than aryone expected at the time. This was because of its relatively low level of development, its cultural diversity, Social and Economic Changes large areas of Muslim fundamentalism, and a legacy from before the "New Order" period of a bloated bureaucracy, Among the three Asian countries reviewed, Indonesia is political instability, and a poor health structure. most clearly characterized by the kinds of developmental The total fertility rate, estimated at 5.5 for 1967-70, had changes generally believed to be favorable to lower fertility fallen to 3.3 by 1984-87. This 40 percent decline, while not rates. A major program for massive increases in education- as large or as rapid as that in a few other Asian countries, is al levels has made primary education virtually universal, substantial in terms of the initial expectations. The decline greatly increased both secondary and higher education, is even more impressive when seen as 65 percent of the dis- and produced greater equity for boys and girls. Education- tance between the initial high-fertility level and the replace- al achievement is related in Indonesia to wanting and hav- ment level of 2.1. ing fewer children, greater knowledge of and greater use of The most important immediate cause of the fertility de- contraception, and later marriage. There have also been cline was an increase in the use of contraception by married major increases in mass media communications and gov- couples to limit the number of children and to increase the emient-organized information and education programs at spaced interval between births. In the 1960s, the contracep- the village level. Both the younger and older generations tive prevalence rate was probably less than 10 percent. By are increasingly linked to the worldwide communication 1976, it had increased to 19 percent, and by 1987, it was 48 system. The development of a denser and more extensive percent. Between 20 and 25 percent of the fertility decline transportation network has increased the circulation of was due to an increasing age at marriage. Most of the re- people, goods, and ideas. There has been an associated in- mainder is accounted for by increased use of contraception. crease in the ownership and availability of consumer The significant increase in contraceptive prevalence re- goods. There have also been improvements in the health flected an increasing demand for small, planned families. and longevity of both children and adults. Undergirding all Between 1976 and 1982, the mean desired number of chil- these changes is broad improvement in the economy and dren decreased from 4.6 to 3.2, and the proportion of wom- considerable decrease in poverty. Over the period 1965 to en wanting no more children increased from 37 to 51 1988, real GNP per capita is estimated have increased by a percent. By 1987, 42 percent of mothers with two children very healthy 4.3 percent per year. and 6,7 percent of those with three said that they didn't Associated with social and economic changes have been want ainy more children. Forty percent of all currently-mar- fundamental changes in the outlook of the young adult ried women of childbearing age in 1987 were not yet using population, making them more independent of familial di- contraception and either wanted no more children or want- rection. This is evident, for example, in later marriages, a ed to postpone the next pregnancy for at least two years. sharp decline in arranged marriages, and increased influ- While Indonesia's fertility decline is impressive, its pop- ence of peer groups. All of these changes are in directions ulation is still growing rapidly-1.8-1.9 percent per an- generally believed to be conducive to lower fertility. 27 One other aspect of Indonesian society pertinent to the have been in leadership positions for more than success of its family planning program is its unusual capac- 15 years. ity for communication to and from the local community 4. Significant social and economic development which has level and for mobilizing local community involvement. affected attitudes about marriage, family life, and repro- This is a critical element for a successful family planning duction; increased openness to new ideas; and created a program. transportation and communication infrastructure These substantial social and economic changes contrib- through which the program could function as it expand- uted to the fertility decline in two important ways; by de- ed throughout the Indonesian islands.2 creasing the demand for children and increasing the 5. Considerable financial and technical support from demand for contraception, and by greatly improving the international and bilateral donors. This support has not general national infrastructure which facilitated the work subverted indigenous Indonesian direction of the pro- of the national family planning program. In this develop- gram, in part because the government has paid an in- ment context, the National Family Planning Program creasing share of program cost, reaching 70 percent by (BKKBN) helped to legitimize small planned families and 1980. The BKKBN has been generally successful in gain- provided the contraceptive supplies and services to make ing and coordinating donor assistance for its policies that possible. There is little doubt that both the favorable and it has worked closely with donors in developing development trends and the effective family planning pro- new initiatives. gram contributed to the fertility decline, although these in- 6. Considerable success in working with religious leaders. fluences cannot be separated and quantified. This has been facilitated by a tradition of working out problems through frequent conferences and discussions Indonesia's Family Planning Program in which confrontation is avoided and accommodation and consensus are stressed. Muslim leaders were asked The BKKBN has created a program that operates effec- for advice and co-opted to a common enterprise. The ef- tively at every level-the nation, the 27 provinces, the 301 fectiveness of this approach is evident in the relatively districts, the sub-districts, and eventually, at the village and high acceptance rates, even in areas of Java known to be hamlet level. Through this hierarchy, it has succeeded in more orthodox in their Muslim identification. reaching the mass of married couples in the several hun- dred thousand hamlets below the village level. While activ- The Indonesian program has been given a high ranking ities outside Java are somewhat less intense, the Indonesian planninn proM rams. Its performance was also very favor- program comes closer than any other to achieving the goal planning as I performanew a lso ve favo of complete, country-wide, community-Board coverage hably assessed as a model for other agencies in the Bank's BKKBN, formally a coordinating rather than a line agency, v prelpensive and in angemento desing le- works through such ministries as those for health, educa- velopment program, and in a book devoted to deriving les- tion, home affairs, information, religion, and the armed ser- vices. However, BKKBN also has important implementation capacities w,hich enable it to carry out some important activities directly and to test expand and 1. Three features help explain this effectiveness. First, there is considerable some important aciiisietysocial pressure to make decisions by consensus and for individuals to abide hasten new initiatives. by that consensus once reached. Second, everyone in the administrative hi- erarchy, including the village chieftain, is rated by how well he (and his con- In summary, then, the success of the family planning stituency) achieve agreed-to targets. BKKBN has been successful in getting program is attributable to: family planning targets accepted by top levels of the Government as one of the most important to achieve. Third, BKKBN has been successful in placing 1 .The continuing strong government support from the their representatives throughout the country and providing them with a de- President and through him, the whole administrative gree of importance in the eyes of the villagers. Thus, after national targets are structure. Because of this, plus its effective leadership, agreed to and disaggregated by region, the local BKKBN representative visits BKKBN has successfully engaged support from many community leaders to discuss the extent to which the community can help in their achievement. Once fully discussed and agreed to, social pressure takes ministries and agencies, including those for health, edu- over to make individuals feel that it is unfair and socially irresponsible to cation, religion, home, and defense. have too large a family. These features, plus some program activities like the 2. An administrative structur-e that facilitates communica- safaris"-more intensive family planning campaigns that are held periodi- cally-have led some observers to describe the program as coercive. To some tion and mobilization of arction at the grass-roots level extent, one's view of the situation depends on his or her cultural background. for a wide range of development activities, including No hard evidence has been presented on this matter and BKKBN has never family planning. This system is unusually effective.1 condoned coerdve practices. 2. Some of this development resulted from special programs that have 3. Stable and effective leadership. The charismatic chair- supplemented the impact of general economic development. Especially man of the BKKBN and rrmost of his principal deputies noteworthy is Indonesia'sadult literacy and non-formal education program, which is assisted by the Bank. It is among the largest of its kind in the world. 28 sons from successful management development programs small relative to the size of the development budget. How- (Samuel Paul, 1982). ever, the symbolic value of the Bank's lending and support Of course, the program is not without its problems. has been of great importance in legitimizing the program BKKBN has become a large, complex bureaucracy with for diverse political and religious groups in Indonesia, par- many of the problems typical of such organizations, includ- ticularly in the program's formative years. ing increasing difficulties in maintaining good internal Nearly all the Bank lending was provided to BKKBN5, communications and in reacting rapidly and innovatively and two-thirds of this to expand its physical plant and to charging events. Its strong target-achievement orienta- equipment. Approximately 40 percent was used to con- tion conflicts with equally strong desires to insist on accu- struct headquarters, warehouses, and training facilities for rate reporting, focus on quality, and promote voluntarism BKKBN, not only in Jakarta, but in the 27 provincial and at the local level. Research and evaluation capacity remains 301 district capitals. Another 26 percent went for transport, inadequaate. Uniformity of coverage could be improved. IEC, and other equipment. The remaining third was used More immediate problems include the need to introduce for production of materials and curricula for various pro- cost recovery and increase private sector participation in grams, including sizeable population education and IEC order to compensate for a planned decline in USAID fund- programs, plus some support for research and evaluation, ing of contraceptive supplies (see below) and to improve salaries, and consultants. coordination with the Ministry of Health as the demand for The buildings put in place are impressive-significantly more clinical methods like NORPLANT increases. better than other nearby government buildings, particular- However, none of these problems is of a character or de- ly in the countryside. Had Bank financing not been avail- gree of severity that threatens future progress. Given the able, it is probable that the Government would have program's strengths and momentum, plus prospects for provided BKKBN with similar buildings, but they would continued social and economic development, it is highly have been built more slowly and been fewer, smaller, and probable that the contraceptive prevalence rate will contin- more spartan. The impact of the Bank's inputs, therefore, ue to rise and the fertility rate will continue to decline dur- was to speed up the expansion of the program and to help ing the next few years without any radical change in it achieve status in the eyes of the public and its own staff. program directions. Whether the TFR will reach a replace- Similar comments can be made about the transport and ment level of 2.1 in the near future without such a change other equipment provided by the Bank loans. They allowed is an open question, however. a more rapid deployment of more and better equipment than could otherwise have been afforded. These are not in- World Bank Involvement consequential achievements, particularly for a new pro- gram that must change social mores and overcome doubts After the announcement of a national family planning to acquire a clientele. program in 1968, the Bank initiated a series of discussions The high-level dialogue that preceded the first of these on population policy with government leaders, at least one projects, while not critical for the initiation of this long pe- of which involved a meeting between the President of In- riod of Bank involvement, was certainly useful. Given the donesia and the President of the Bank. In 1972, the first high regard in which the Bank was held at the time, this di- population project was approved. Three other projects fol- alogue may have assisted the Government in consolidating lowed, in 1977, 1980, and 1985. Together, these first four support for a forceful program in its early days. It certainly have r esulted in Bank lending to this sector of US$122.4 sent a strong message to the staff of both the Bank and the millio]n1.3 During the 1980s, these projects contributed about Government about the extent of support the Government 10 percent of total program expenditures. The Government could expect from the Bank, and this may have encouraged financed 70 percent-up from 40 percent in the early 1970s. the country to proceed more rapidly and aggressively than Other donors, including NGOs, financed the remaining it might otherwise have done. 20 percent.4 A fifth project has just begun. This financial Dialogue on a more technical and operational level, support was probably not of crucial budgetary significance while it could have been very helpful-especially in the for the program. Not only was the Government strongly early years when the program was being designed-ap- committed to the program, but the amounts involved were pears to have played only a minor role until lately. One way to judge this is in terms of sector and analytical work un- 3. Includes actual disbursements for the first three projects and the loan dertaken by the Bank to underpin such dialogue. The first amouni for the fourth. major sector report on population and family planning ap- 4. These estimates are based on a comparison of figures from Repelita III, loan amounts for Population III and IV, and donor contributions for the pe- riod 1982-88 taken from Global Population Assistance Report 1982-88, UNFPA, 5. The fourth project provided some support to a newly-created Ministry New York, 1989. of Population and Environment. 29 peared only in 1980, when the third project was about to be- rent costs (which often also have a low foreign exchange gin. The second, a review of trends in fertility and component) when it believed this was necessary for the contraceptive prevalence, was issued in 1986, one year after success of the project. In contrast, the Bank has been willing the fourth project was initiated. Research and studies in- to provide a larger fraction of local costs when there is a cluded within the population projects do not appear to savings or a balance of payments gap which it believes a have contributed much either. In at least two instances in government is incapable of filling itself. However, the Bank which potentially important operations research studies has been reluctant to finance recurrent costs (unless they were included to test out new ideas, the results were not are incremental, and then only on a declining basis) on very useful, in part because of inadequate collaboration grounds that the project is not sustainable unless the with Indonesian counterparts in the initial stages. Howev- government is willing and able to take over recurrent cost er, two quite recent reports--a study of needs and pros- requirements by the time the project ends. The two agencies pects for the family planning program in the 1990s and a seem to have different views about what constitutes much-needed financial and economic analysis of the popu- discipline and a show of commitment. A consequence of lation and health sector-are highly relevant to program this difference is that USAID has become involved in needs and have influenced the design of the fifth popula- program operations to a greater extent than has the Bank, tion project, now under implementation. and that may be the point to USAID's willingness to fund In addition, of course, the Bank has contributed indirect- recurrent costs. ly to the population program through its other loans to In- This situation has also resulted in the Bank being per- donesia. Population lending per se has constituted only 0.8 ceived as not providing intellectual leadership to the Indo- percent of US$14.8 billion lent to date. However, 10 percent nesian program. This is in sharp contrast to its has gone to the education sector, 1 percent to health, and acknowledged, important intellectual leadership role in the remainder to the major economic sectors-industry, ag- macroeconomic policy and some other sectors in Indonesia. riculture, transport and communications, etc. While none While the Bank has population staff capable of playing this of these loans was designed or justified in terms of its pos- role, the implicit or explicit choice was to let other donors sible effects on fertility, they must have had a substantial with competent resident staff lead the way in this sector. impact on desired family size and the willingness to do There are three explanations for the emergence of this something about it. They could, however, have had a larger pattern. First, the GOI has preferred grants for software impact if thought had been given to possible linkages with components and program operations and loans for capital fertility in their design. This is especially true insofar as ed- components. Second, the operating style of various partici- ucation and health projects are concerned. The same might pating agencies has played a role. In contrast to USAID and also be true for projects involving communications and UNFPA (the largest donors excluding the World Bank), the transportation. For example, more effort in the education Bank typically delegates little decision-making authority to sector might have been put into female education, and in the field and has limited field representation. In this case, the health sector, into maternal and child health care. there was one senior specialist located in Jakarta for about In contrast to the BanKs focus on plant and equipment, six years during the 1970s. Since it operates at arm's length, other donors have concentrated more on providing com- Bank staff feel more comfortable with components like con- modities and technical expertise that helped, for example, struction, which require few changes after the design stage in designing training programs, and improving work rou- has been completed, as opposed to components like train- tines and managerial system. These donors also financed ing or IEC, in which learning by doing and adaptation dur- pilot projects to test new initiatives aimed particularly at in- ing implementation are more important. Also, it is much volving NGOs and local communitites. Often the Bank easier to develop large projects by focusing on construction. came along later to help scale up the more successful of Third, given the intellectual leadership displayed by these innovations to full operations. Thus, a de facto divi- BKKBN and several major donors, there has been less need sion of labor has developed in which the Bank has focused for the Bank to become involved in technical policy issues. on program hardware and other donors, on software ele- In the early years, Bank staff sometimes raised questions ments. The net result for the program has been a fairly well- about the managerial capacity of BKKBN and its frequent balanced package of foreign assistance. reorganizations. Over time, however, this attitude has been It is interesting to observe how these different approach- replaced with statements of praise and confidence. es have been justified. USAID, for example, has become After a long period of poor Bank relations with other do- increasingly reluctant to finance construction in most coun- nors, the situation has greatly improved in recent years, as tries on grounds that it requires mostly local-cost finance evident in cooperative funding of specific projects, regular which governments should be able to finance themselves. exchanges of information, and cooperative evaluation of On the other hand, USAID has been willing to fund recur- program components of mutual interest. Such cooperation 30 and an improved division of labor among donors would be cial requirements will increase. The numbers are not large fostered if reports on various donor activities, at least as and certainly smaller than the amounts that have been pro- they relate to the Bank's activities, were made a regular part vided for construction and equipment in recent years, ac- of supervision, project completion, and audit reports. tivities which can now be scaled back; but they involve What might the Bank have done better or differently to recurrent costs rather than capital costs, for which-as indi- have increased its contribution in the population field in cated above-both the GOI and the Bank have trouble jus- Indonesia? First, a representative for the Bank in the popu- tifying the use of loan funds. lation field (and possibly also the health and education Third, the Bank might usefully have considered both a fields) might have been located in Jakarta for more of the more integrated approach to population and health and period under review and given some latitude to make ad- more substantial strategic assistance to upgrade the health justments in project components as the need arose. system as a whole. This would have served both health and A case can still be made for locating someone in the field population objectives. The Fifth Population Project, which on a long-term basis. In the future, programmatic attention includes substantial maternity and child health compo- will increasingly focus on improving software components nents and provides over a quarter of its funds to the Minis- like outreach and IEC, cost recovery, and other financial try of Health, is a promising beginning to correct this bias. issues--components not easily handled with an arms- Fourth, the Bank could have done more to assist and en- length approach. However, it must be recognized that such courage BKKBN in upgrading its research and evaluation a move would have been more productive if undertaken work. BKKBN has been quite successful at innovating earlier. Now that BKKBN and its programs are well estab- "from the driver's seat," even without many sound re- lished, the need for more intimate and continuous contact search studies that investigate the relationships between in- by the Bank is less. Given BKKBN's present capacity to puts and outputs to determine what works and what does manage and innovate, if additional external financing is re- not. However, this success may have been due in large part quired, the Bank should consider program or sector loans. to the rapid social and economic development that took Especially in the case of Indonesia, these might require few- place during the last 20 years, plus the fact that fertility lev- er rather than more inputs from Bank staff. els were coming down from very high levels. It may be Secoind, the Bank might have considered funding recur- much harder to bring the fertility rate down from its cur- rent costs at an earlier stage. Had it done so-and if the GOI rent level of 3.4 to a replacement level of 2.1, particularlv if been willing to accept loans for this purpose-its involve- there is a slowdown in economic development during the ment in program and policy activities might have been next decade. The Bank could make a substantial contribu- more substantial. This issue also continues to have rele- tion if it could assist the BKKBN in finding ways to upgrade vance for the future. On the one side, the need for funding its capacity to undertake research that might assist it as it contraceptive supplies and services is growing, in part be- enters these uncharted waters. Also, the very success of the cause of successful expansion of the program in the past, Indonesian program deserves careful study to determine and in part because of the need to continue increasing the which of its elements might usefully be employed else- contraceptive prevalence rate if the target of a replacement where.6 In the process, the Bank would find itself involved fertility rate of 2.1 is to be reached early next century. On in very challenging intellectual issues, a by-product of the other, USAID and other donors have indicated their in- which would be a positive change in its image. tention to scale down participation by 1995. BKKBN is try- ing to respond to these trends by promoting privatization of supply and greater cost recovery in public programs. The need for external funding will depend in large part on how 6. For example, much of the success of this program has been attributed to successful these efforts are. Since they are not likely to be the role of outreach and community institutions in achieving social and be- havioral change. How this was achieved and how these community institu- that successful in rural and more remote areas, where the tions were mobilized for this task is not weU understood. Nor is it clear how bulk of non-users reside, there can be no doubt that finan- they might be mobilized to tackle related problems like high maternal and perinatal mortality which require more clinically-based delivery of services. 31 3 . India Demographic and Socioeconomic Trends women wanting no more children. This is a notable achievement, but it would be incorrect to assume that there Between 1970 and 1987, India's total fertility rate fell is little remaining unmet need. Indonesia had a comparable from about 6.0 to 4.1, a 32 percent decline. This represents 18 percent figure in 1976 in the midst of a continuing in- roughly half the distance between the initial rate of 6.0 and crease in contraceptive prevalence, with growing numbers the replacement level of 2.11. The current level is roughly wanting no more children and using contraception. A fur- the same as the median for developing countries with pop- ther indication of latent demand is that by 1988, five out of ulations of at least 45 million. Its decline has been consider- six women in India, with at least three children wanted no ably less than that of Indonesia, China, or Thailand, but more, and many of those already had more than they want- somewhat greater than that of the Philippines, Bangladesh, ed. In addition, these figures do not reflect the latent de- and Pakistan. This rate of decline has been approximately mand for contraception for spacing purposes. matched by the decline in mortality rates. The result is that India has made steady but slow progress in improving the population growth rate has remained at or slightly general socioeconomic conditions. Real GNP per capita above 2 percent per year during the last three decades. grew 1.8 percent per year during the 1965-88 period. The About a third of this decline can be attributed to an in- gross primary school enrollment rate for females increased crease in the age of marriage, which is quite low in India from 56 to 67 percent during the 1970-85 period. Between compared to other Asian cotntries. Most of the remainder 1971 and 1989, the infant mortality rate declined from 139 is attributable to a decline in marital fertility resulting from to 91 per 1,000 and literacy among wives increased from 17 an increase in contraceptive use from 10 to 39 percent over to 39 percent. It is probably for this reason that the average this period. Nearly 90 percent of the supplies and services desired family size has declined slowly, from 3.65 to 3.4 be- involved are provided by the Govemment's program. tween 1980 and 1988. This is not much higher than esti- Latent demand for contraception appears to have been mates at about the same time for Indonesia (3.2), Mexico substantial, even in 1970, before this major decline in fertil- (3.0), and Colombia (3.1). However, further progress is like- ity. This is reflected in national surveys of that time indicat- ly to be limited without much greater social and economic ing that half the women of reproductive age wanted no progress. In particular, improvements in the education and more children and that 80 percent of this group were not status of women and in child health and longevity are ur- using contraception. In other words, 40 percent of all wom- gently required, especially in India's northem states. en of reproductive age could be considered "in need" of These national figures can be misleading, however, be- contraception at that time. By 1980, the increase in contra- cause of the extreme diversity of these indicators within the ceptive use had reduced this proportion in need to 18 per- country. Fertility (and contraceptive prevalence) rates vary cent. Between 1980 and 1988, this figure has remained greatly among Indian states, from approximately 2.2 in Lourghlv conlstant, suggesting that further increases in con- Kerala and 2.6 in Tamil Nadu to between 4.7 to 5.3 in the traceptive use were offset by increases in the proportion of four northem states of Uttar Pradesh, Madhya Pradesh, Rajasthan, and Bihar, which together constitute 40 percent - ~~~~~~~~~~~~~of India's population. Anrudh Jain (1985) has explained 1. These estimates come from Satia (1991). Using the UN estimate of a TFR of 5.7 for 1965-70 and the 4.2 figure for 1988 from the ORG study yields a 25 these differences by differences in social development, as percent decline. indexed by female illiteracy and infant mortality. These 32 indices are substantially higher and development program many knowledgeable observers judge the program's suc- performance has been generally weaker in these northern cess to be modest. In part, this stems from the fact that the states. program set unrealistically ambitious goals for itself which it could not meet. In addition, many programs which start- The Indian Family Welfare Program ed later than the Indian program have accomplished more, which has contributed to the perception of a comparative The Indian family planning program-renamed the lack of achievement. However, these judgements also stem Family Welfare Program in 1977 when MCH was incorpo- from several major problems with the program's design rated into it-is the oldest and, with the possible exception and operation. of China's, the largest program in the world. Early years Excessive focus on sterilization. Since the early 1970s, the were characterized by large-scale IEC campaigns to spread Indian family planning program has focused on steriliza- contraceptive knowledge and gain acceptance for a small tion as the primary means to reduce fertility. Despite rhet- family as a norm. It was a time when red triangles (the pro- oric about a cafeteria approach, little has changed since that gram's symbol) and posters portraying a happy couple time other than a shift in focus from men to women. This with two healthy children became ubiquitous. Those years emphasis can be seen in the following figures on the per- of naive enthusiasm gave way to periods of frustration that centages of married couples of reproductive age using var- sometimes resulted in coercive campaigns which created a ious types of contraception:3 political backlash and set the program back. All the while, however, the principal channel for delivering services was Natural being expanded. Today, there is one primary health center Family (PHC) per 40,000 and one subcenter (SC) per 5,700 persons, Year Total Sterilization IUD Pill Condom Planning operated by state Ministries of Health and Family Welfare, with budgets and guidance from the center. This is an im- 1970 14 6 1 0 3 4 pressive achievement in a country with a population of 1980 34 22 * 1 5 6 close to 850 million. When fully staffed, the PHCs are capa- ble of providing comprehensive health and family plan- Note: = <0.5 ning (FP) services, including sterilizations and IUD insertions. They also support and supervise outreach ser- This focus results from several factors. (1) Targets are set vices operated from subcenters. Services at these sub- in terms of equivalents to sterilizations, for example, one centers are supposed to be provided by one male and one sterilization being equal to three IUD acceptors, nine pill female multi-purpose worker, who are responsible for users, or 10 condom users over a year. In practical terms, a rudimentary health, MCH, and FP outreach activities. focus on sterilizations means fewer people to contact and However, only about half of the male workers are in place motivate to reach a given target. Also, sterilizations require in the northern states. Other delivery mechanisms include no re-motivation and supply and are easier for supervisors intensive campaigns to enroll sterilization acceptors in ru- to monitor and verify. (2) Substantial 'compensation pay- ral areas twice per year and social marketing schemes to ments' are provided only for sterilizations. (Acceptors re- distribute condoms and pills. Funding for the program, ceive 100 rupees for a sterilization, but only 10 for an IUD amounting in 1987-88 to about 7.50 rupees (US$0.50) per insertion and nothing for other methods.) (3) Spacing meth- capita, coomes mainly from the center and constitutes 0.7 ods are not readily and reliably available. The program still percent of the central budget. In addition, a number of pub- relies heavily on campaigns and camps for service delivery lic and private agencies-including the postal service, the and neglects the type of regular contact, follow-up, and re- defense establishment, the railways, and a number of large supply arrangements required for spacing. Many multi- manufacturing enterprises-have quite active programs. purpose workers still have not been trained to insert IUDs. NGOs are not a major force. The Family Planning Associa- No regular outlet for condoms exists in a large number of tion of India, for example, the country's largest and oldest, villages-60 percent according to one study. Pills, intro- provides services to about one percent of the population, duced only recently, are widely considered unsafe and in- largely in urban areas. jectables and implants are not yet offered. (4) While In addition to its overall impact, the program's major ac- complishments include meeting the contraceptive needs of 2. In 1965 and 1966, an effort was made to promote the IUD, but it ceased about two thirds of couples not desiring additional chil- when medical complications began to be reported. In the view of many dren, and making knowledge and general acceptance of knowledgeable observers, this was a mistake which set the program back. family planning-that is, sterilization (see following para- 3. These figures are derived from ORG surveys. Government service sta- tistics have very similar numbers for sterilizations but higher numbers for graphs)-nearly universal. Despite these accomplishments, temporary methods, which knowledgeable observers believe to be inflated. 33 knowledge of sterilization is nearly universal-indeed, that who are visited or come into clinics often report that the is what the words "family planning" mean to many- quality of services is poor. In one study, a third of steriliza- almost half of women of childbearing age do not know tion acceptors complained of post-operative complications about modern spacing methods. and expressed dissatisfaction with follow-up services. Oth- The emphasis on sterilization has meant that a large frac- er studies report low utilization of available services and a tion of the program's potential audience-younger couples preference for private services where available and afford- who might be interested in spacing-are more or less ig- able. nored. In addition, because of its success in reaching high- The specification of targets-in terms of numbers of ster- parity couples interested in termination, the program is re- ilizations performed, IUDs inserted, and condoms distrib- alizing diminishing returns. Indian women on average uted-without regard to local conditions and individual have their third child around age 30, so even if childbearing needs exacerbates these problems. Program managers ai ter age 30 were completely eliminated by sterilization, the should be held responsible for making available high- TFR would not decline to replacement levels. India has quality services, tailored to couples' needs as those needs demonstrated that it is possible to rely on sterilization to change over their lifetimes. This requires, not rigid, bring the TFR down from six to four, but its target of a re- uniform target-setting from afar, but multiple methods, placement level of 2.1 will not be possible without placing multiple delivery channels, attention to delivering high- much more reliance on temporary methods.4 quality services, and some form of feedback and client par- Insufficient attention to program operations and quality in the ticipation. interests of program expansion. This is a general point that Excessive centralization and single-channel delivery system. may explain a number of program deficiencies, especially The near monopoly of the Departments of Family Welfare weak outreach and poor service quality The focus of the for delivery of services contributes to the above problems. program managers has been on increasing the inputs be- The private sector, which currently provides two thirds of lieved necessary to increase the contraception prevalence health services, should be enlisted to play a major role. rate. Thus, a great deal of attention has been given to Other government programs-for example the Integrated achieving specified norms for the distribution of clinics and Child Development Services program which operates in staff, but much less attention has been paid to how the sys- 40 percent of the country's villages-should be involved. tem operates. Health centers and subcenters have been con- Social marketing and community-based distribution sys- structed, but maintenance is a problem. Staff have been tems need substantial expansion and strengthening. hired and trained, but on-the-job training, supervision, and Neglect of factors that influence demand for contraception and the establishment of efficient work routines have been ne- small families. Despite rhetoric to the contrary, the Govern- glected. Also neglected are the reporting systems and oper- ment has focused narrowly on family planning-the provi- ations research studies necessary to monitor, test, and make sion of supplies, services, and related information-as corrections in program procedures, as well as the great re- almost the sole means of reducing the population growth gional disparities in culture, level of development, and ad- rate. No other agency outside the Ministry of Health and ministrative capacities. Family Welfare has been assigned any real responsibility Several surveys have indicated that a large majority of for this goal. In particular, the Ministry of Education and its married women have never been visited by a health or fam- state counterparts have done little to focus on the special ily planning worker. This is not surprising, given the large educational deficiencies of women, especially in the high- number of tasks and the caseload assigned to the multi- fertility states. Nor has the Government done much to in- purpose workers and the absence of rules for establishing volve NGOs and the private sector. The Planning Commis- priorities. An important suggestion made in a recent inter- sion and cabinet committees are supposed to encourage nal Bank report is that both FP and MCH outreach efforts and coordinate such agencies, but their efforts to date have should focus on women who are about to deliver or have been unimpressive. Even the Ministry of Health and Fami- recently delivered, since this is the group with the greatest ly Welfare has neglected programs in its own realm that are need for both family planing and MCH services. Those important for influencing effective demand, specifically the MCH program and spacing aspects of family planning. Both are needed to reduce infant and child mortality. It is 4. Anrudh Jain (1989) reported on three computer simulations that make difficult to believe that fertility rates can be reduced much the case very clearly. If sterilizations alone were used to achieve replacement further without resorting again to coercive measures, un- fertility of 2.1 by the year 2020, 55 percent of couples in the 20-24 age bracket would have to be sterilized. If the method mix of 1980 were maintained, the less the probability of children surviving is substantially figure would be reduced to 24 percent. If all methods were readily available, increased. the figure might be reduced to 7 percent, though it might still have to be 25 Defenders of the program argue that there were few ef- percent for the 25-29 years age group. The latter figures may still be unaccept- fee ers to steriliat in the erl wers Tey ably high, implying that the goal may not feasible. fective alternatives to sterilization in the early 1970s. They 34 also maintain that the urgency of the population problem In 1989 and 1990, the Bank provided two additional and the magnitude of the administrative problems in a credits, of $124.6 million and $96.7 million, respectively. country as large as India justified the use of special cam- If this new and much higher level of lending continues, paigns, an emphasis on numbers rather than quality, and the picture of Bank participation will look quite different the use of targets and incentives. They argue that the focus after a few years. As of now, however, even with these was on expanding supply rather than demand because of large credits added in, the Bank's contribution to the In- the presumption-which proved to be correct-that sub- dian program has not been greater than 4 percent of gov- stantial latent demand was already present. Moreover, this ernment expenditures. approach has worked; at the least, it seems to have worked Nature of the projects and their administration. The over- no worse than programs in other countries at similar socio- all goal of the first five of these projects was to accelerate economic levels and rates of progress. the expansion of the service delivery network for family The question is whether more could have been accom- planning and MCH in specific districts more rapidly plished had these problems been resolved and whether this than resources would permit the Government to do else- approach can continue to work in the future. In retrospect, where. By concentrating resources in a few districts, the the peaks in acceptor rates reached during special cam- Bank hoped to have more influence, at least in these dis- paign years were surely not worth the public backlash, and tricts, than it could on a broader level, and then to use the the program neglect and falloff in acceptors that occurred experience gained there to influence the national pro- in subsequent years. Surely, this approach, now fully insti- gram. The first project (1973-80) was considered experi- tutionalized and difficult to change, will find it more and mental, designed to complete and test the Government's more difficult to make progress in the future. Its modest service delivery pattern and identify ways to strengthen successes to date not withstanding, little has been done to it. It was implemented in six districts of Uttar Pradesh anticipate the fact that its methods cannot continue work- and five districts of Karnataka. The second (1980-88), ing this well in the future. third (1983- ) and fourth (1985- ) projects, termed area projects, were meant to implement a model plan or blue- The Bank's Role Prior to 1987-88 print in selected districts of various states. The third and fourth projects also included some support for MIS, IEC, While the Bank's role has evolved over time, it can most and training at the state level. The fifth project (1988- ) easily be characterized by two time periods; that is, before provided support for the metropolitan areas of Bombay 1987-88 and post-1988. During the first and longer of these and Madras and one district in Tamil Nadu. No signifi- (1972-88), the Bank provided a modest amount of financial cant policy conditionality was attached to these projects. support. For the most part, this assisted the Government in Results from the first two of these projects-the only carrying out its preconceived plans, and it had little influ- ones so far completed-suggest that they have had no ence on the program's directions. This judgement is based differential impact on the program. In comparing both on a review of the magnitude and timing of the funding, project and non-project districts, the overall character of the nature of the projects and the way they were adminis- the program was the same, contraceptive prevalence tered, the character of the dialogue and sector work in- rates increased at roughly the same slow rates, and most volved, the role that the Bank played in other sectors, and indices of MCH performance (e.g., use of clinics, immu- the Bank's relations with other donors during this period. nization rates) indicated no upward trend. Three factors Magnitudes and timing. Implementation of the first Bank- seem most important in explaining this outcome: financed population project was initiated in 1972, long after * Neglect of inputs other than infrastructure. Between 40 India's approach to its population problems was firmly en- and 60 percent of costs in the first four projects were trenched. Up to 1988, the Bank funded five projects involv- expended on civil works. (In the fifth project, civil ing US$245.7 million. While that constituted 28 percent of works constituted 11 percent of total costs.) If furni- the Bank's population portfolio to that date, it was a small ture and equipment are included, these figures rise to fraction-3.6 percent during the 1980-88 period-of total between 51 and 68 percent. While the emphasis on expenditures of India's Family Welfare Program. Donors as training, MIES, operations research, and IEC in- a whole contributed 12-14 percent of government expendi- creased over time in part as a result of Bank pressure, tures, and the Bank provided about a fourth of this amount. the quality and effectiveness of these components The absence of Bank resources, therefore, would have made were generally weak, depending more on the interest little difference to the program funding. However, the pres- of the state and project unit concerned than on the ence of these funds clearly indicated to both the Govem- Bank's presence. In general, issues of field-worker ment and other donors the Bank's support for the general training, motivation, work program, performance directions of this program. 35 measurement, and feedback were not well-ad- Dialogue and sector work. In the early 1970s, family plan- dressed.5 ning was still a politically sensitive topic in India, with * Insignificant differentiation between areas. In formulat- some arguing that it diverted attention and resources from ing the area projects, it was anticipated that locally development and the alleviation of inequalities-the fun- appropriate approaches would be developed and damental cause of large families. The Bank, quite appropri- would serve to guide national program development. ately, played a low-key role, but it continued to play this However, no conditionality was introduced into role throughout the remainder of the 1970s and well into project documents that would require this to happen, the 1980s, during a time when it would have been helpful and no plans were developed for this purpose.6 In to challenge the Government on issues being raised by pro- practice, the approach continued to be based on the gram critics. One reason for this is that, since the Bank his- uniform model plan7, and the difference in the speed torically has played a reactive role in India, it would have with which some districts achieved the model stan- been out of character to have played a different role in this dards was not great enough to significantly affect the sector. Another important reason, however, is that compre- outcome. hensive sector work did not exist until very recently. The * The fact that project as well ais non-project districts operated Bank sponsored a few studies (for example, some reviews under the constraints of the overall program. Many of these of results from research and pilot experiments). It also in- constraints were more bindling than mere lack of physi- cluded hortatory statements in Country Economic Memo- cal inputs. Examples include excessive focus on steril- randa (for example, about the negative consequences of ization, excessively large target populations and rapid population growth, the need to strengthen service de- numbers of tasks per worker, a weak MCH program, livery, and the need to emphasize spacing methods and continued neglect of other factors that could influence raise the age of marriage), and it periodically reviewed five- demand such as female education and employment op- year plan strategies for the sector. However, these efforts portunities, and weak institutional capacity (due, were either too piecemeal or too general to be of operation- among other reasons, to frequent changes in top man- al use. How, for example, should the accountability system agement, limited depth of managerial skills, and inade- be altered in order to place less emphasis on sterilization, or quate use of feedback, monitoring, evaluation and work rules changed to ensure more effective outreach ef- research). The first five Bank-funded projects did little to forts? The Bank had no answers to such questions. The first try to change these constraints. significant break in this pattern occurred in 1989, with the Preliminary results from the other projects are mixed. publication of two papers that began to develop a strategic Progress seems to be good in IKerala, West Bengal, and Ma- image of what was needed.8 To be sure, over this whole pe- dras, but disappointing in Karnataka and Bombay. riod, the Government did nothing to encourage such ef- forts, but had a continuous stream of materials as useful as these two studies been produced, it would have been diffi- cult to ignore them. 5. Training improved over time, with an increasing emphasis being placed Modest Bank staff inputs. Considering the size and impor- on in-service training. However, it is still largely oriented toward providing tance of this program, staff inputs were meager. From the knowledge and some technical skills, as opposed to improving work rou- early 1970s to the mid-1980s, the program operated with tines and quality of services. The MIES developed in the first project and sup- p ported in all subsequent projects has resulted in some improvements in field- one full-time person in Washington, a series of consultants, level reporting, but analysis and use of data gathered has been limited. While and PHN staff brought in for short-term assignments. Dur- funds were included for operations research in all projects, output has been in negligible. In the first project, these activities were derailed by the steriliza- g the first half of this period, there was one person in the tion campaigns during the 1975-77 eimergency period, and in subsequent resident mission in New Delhi with other responsibilities as projects, by the neglect of program managers who did not see their utility. Two population research and training centers were established by the first project, but took a long time to become operational. The fourth project sought to establish a State Institute of Health and Family Welfare, but plans have 7. By the end of the second project, a few differences had emerged: the been scaled back because of the recurrent cost burden. The first project paid number of sub-centers, plus staff and equipment for them, increased a little little attention to demand generation except for research to study the issue. more rapidly; the proportion of constructed versus rented facilities was Subsequent projects included significant IEC components which led to size- greater; and IEC and management training inputs were somewhat greater. able increases in the quantity and qua:.ity of such activities over time. How- But there were no differences in other inputs such as medical supplies, multi- ever, the overall impact of these activities was constrained because they were purpose worker training, and field supervision, or in the effective availabili- implemented top-down, more as publicity and propaganda campaigns, and ty of spacing methods. not backed by research on client groups; professional skiUs for effective IEC 8. "Improving Family Planning, Health and Nutrition Outreach in India: activities were limited; and program managers were not adequately con- Experiences from some World Bank Assisted Programs" (World Bank Dis- vinced of the need for IEC. cussion Paper, 1989) and "Family Welfare Strategy in India: Changing the 6. State governments resisted the few efforts that were made-for exam- Signals" (Sector Review Paper, 1989). A third study, "Strengthening the Role ple, to supply project areas with more transportation equipment-on of NGOs in the Health and Family Welfare Sector of India"(1989) was also grounds that this would involve unacceptable discrimination. helpful. 36 well. In the mid-1980s, a second senior staff member was new, non-population projects that have come on stream added. Since the end of the 1970s until 1987, there was no since 1987. Nevertheless, a number of these projects are rel- one in Delhi working on population. Given these numbers evant for population.9 In addition, as noted above, a num- and the level of Bank activity in other fields for which these ber of strategy papers and high-quality studies began to be staff we:re responsible, it is not surprising that little sector produced. Finally, the central Government-as well as the work was undertaken and that supervision was sorely ne- Bank-have become disappointed with progress and ap- glected at times. pear more willing to consider alternative approaches and to Assistance to other sectors and agencies. The Bank is the substantially increase expenditure levels.10 These last two largest donor to India for economic development pro- factors have resulted in a more substantial dialogue be- grams. While seventy percent of its lending has been pro- tween the Government and the Bank. This, in tum, has led vided to the energy and the agriculture and industry to some changes in the nature of subsequent projects. sectors, only 1.7 percent has been allocated to the popula- The sixth (1989- ) and seventh (1990- ) projects differ tion and. health and nutrition sectors combined, and only from their predecessors in important ways: they are consid- 1.6 percent to the education sector (see Table 4.1 in statisti- erably larger, both in terms of funds and the number of cal appendix). Except for a few urban development projects states covered; they operate at the state rather than the dis- which included some assistance to the family welfare pro- trict level; their stated goal is to improve priority program gram, and the Tamil Nadu Nutrition Project, none of the components such as development of human resource man- non-population projects tried to take into account the de- agement skills and training, rather than expansion of the mographic impact they might have had. This is especially delivery system; they include support for NGOs and for so- telling in the education sector, where the only assistance cial marketing; and they include significant operations re- provided prior to 1987 was for an agricultural education search components."' In addition, the third, fourth, and project that placed no special emphasis on women. Nor did fifth projects were restructured in 1990 to include some of the Bank attempt, until quite recently, to assist in the devel- these new features.12 opment of other channels for delivery of family planning While these are very hopeful signs which indicate a more services, supplies, and messages. This pattern was largely pro-active role on the part of the Bank, two notes of caution the result of the Indian Government's desires, but there is are in order. First, although some key deficiencies of the little evidence that the Bank tried to have it any other way present approach are now fairly well accepted by central Relations with other donors. While the Bank's inputs to the program managers,'3 they have not been translated into Indian population program were small, combined donor practical, "implementable" proposals for their correction. inputs were in the range of 12-14 percent of total outlays. For example, the appraisal documents for these new As the largest of the donors during the 1980s, the Bank projects recognize the need to emphasize spacing and might have been expected to take the lead in developing a methods other than sterilization, but present no concrete coordinated donor approach to program strategy issues. plans to attack key issues-such as work routines, target- This never happened, nor did any significant amount of co- financing develop. This is due in large part to the Indian Government, which served as the donor coordinating 9. The Integrated Child Development Services, a second Tamil Nadu nu- agency and discouraged jointly-funded projects. In fact, the trition project, a vocational training project, two technical education projects, and a proposed basic education project are all cases in point. The vocational area projects approach effectively kept donors apart, with training and technical education projects include components designed to in- each one working in a different part of the country. While crease access for women to modern industrial training, and a major emphasis the Bank's relations with other donors were cordial, consul- of the basic education project is to enroll female children and keep them in school for a longer period. tations were limited in large part to discussions by visiting 10. India's Draft Eighth Five-Year Plan (1990-95) allocates US$2.7 billion to Bank missions, because the Bank did not have a significant the family welfare program, a substantial increase over the Seventh Plan, and local presence for most of the period. the Bank has indicated its desire to increase its commitments to roughly 10 percent of this amount. An additional factor may be the deterioration in In- dia's macro-economic situation in recent years, which has resulted in signals Developments since 1987-88 to sector ministries to do what they can to maximize resource transfers. 11. In fact, both projects include specific agreements that project states will release designated blocks from existing family planning targets for research Since 1987, this picture has begun to change in promising purposes. ways. Fi.rst, staff inputs (for the combined PHN-plus-edu- 12. This restructuring occurred mainly because of slow disbursements re- cation sector) have increased. From 1987 through 1989, four lated (mainly) to exchange rate changes. However, advantage was taken of this situation to include additional areas and programs, including social mar- full-time senior staff worked on these sectors, one of whom keting of contraceptives, NGO activities, and strengthening of implementa- was locztted in Delhi. In 1990, two additional senior persons tion and monitoring capability at the ministry level. were added. This may prove to be of only marginal direct 13. Though perhaps not the local program managers, many of whom ap- benefit t the popuation setor becaue of the umber of pear to feel that the program is on the right track, but only suffers from short- benefit t:o the population sector because of the number of ages of resources and skilled personnel. 37 ing, monitoring, and accountability-that have kept the paid to requirements for implementation. For example, it program focused on sterilization. All of these remain to be includes functional guidelines for changes in work routines worked out during implementation. These projects provide for field workers, field-level coordination with the Integrat- the resources necessary to do things differently in a number ed Child Development Service Project, field-level report- of respects, but do not mandate specific changes. Accord- ing, and community participation-a first for this series of ingly, much will depend on how much advantage program projects. Moreover, it contains hardly any funds for civil managers take of the opportunities offered to them and works, the understanding being that adequate funds for how persuasive the Bank can be in arguing for the changes this purpose are included in the sixth and seventh projects. it deems necessary. On the other hand, it is weak on details about the family Second, as Table 3.3 indicates, the sixth and seventh planning components (included here under the safe moth- projects have allocated roughly the same proportions of erhood portion of the project). While this is understand- total project costs to construction and to equipment as was able, given the focus of this project, special care will have to the case in the first four projects. While this does not neces- be taken to ensure that this component is not overlooked in sarily mean that expansion rather than human resource the midst of these laudable efforts to improve the long-ne- development remains the priority, it does mean that special glected health aspects of the Family Welfare Program. care will have to be taken to ensure that these financial The next step is to translate these good intentions into allocations do not drive the time allocation of staff and field-level results. On the Bank's side, this will require more judgements about implementation progress as they have in intensive and continuous attention to supervision by staff the past. conversant in the substantive issues than has been the case The Child Survival and Saf e Motherhood Project (1991-) in the past. This will be especially important for the Child appears to have applied these lessons. More attention is Survival Project, which is national in scope. 38 4. Bangladesh Demographic and Socioeconomic Trends and mortality may have changed the climate for fertility regulation. Third, social and economic changes unrelated to The decline of the total fertility rate in Bangladesh from the family welfare program might explain the increased use 7 to 4.9 and the increase in contraceptive prevalence from of contraceptives. 3 to 35 percent since 1970 was unexpected by most Can the fertility decline be explained by factors unrelat- observers. Bangladesh is one of the poorest countries in ed to contraception? Studies of changes in all the proximate the world. Illiteracy is very high, especially for females. It determinants of fertility (e.g., marriage age and duration, is a patriarchal society in which most women are in pur- postpartum infecundity, contraceptive use and abortion) dah, their status is low, and they are dependent on their suggest that the most important factor by a large margin is fathers, brothers, and sons. The density of population on increased use of contraception, accounting for 80 percent of available land is the highest in the world in this agricul- the decline. The only other factor of significance is the tural economy and environmental risks, especially flood- change in marriage pattems, which accounts for most of ing, are serious. Families and individuals face many risks the remainder.2 for which relatives, especially adult sons, are the only Can this fertility decline be attributed to improvements available insurance. in health conditions or declines in mortality? While there This is a complex of conditions generally considered to have been slow declines in mortality during the first half be classic determinants of high desired and actual fertility. of this century and somewhat more rapid declines in in- These conditions led many competent observers1 as recent- fant and child mortality in recent years, mortality remains ly as the last decade to believe that in Bangladesh there very high: infant mortality, about 12 percent; child mortal- could not be a significant demand for family planning ser- ity to age five, about 20 percent; maternal mortality, 70 vices and lower fertility before substantial structural chang- times Western levels; and a life expectancy of only 51 es occurred. Indeed, they remain a basis for the plausible years. The risks of losing children, husbands, and sons is idea that further gains will be difficult and that a plateau still very high. will be reached in fertility levels long before the recent de- Can the decline be attributed to social and economic cline brings fertility to replacement levels. changes unrelated to the program? Little evidence of this The significant increase in contraceptive practice and the can be found in published indicators of such change. Since fertility decline in Bangladesh have been associated with an 1970, GNP per capita has changed very little and remains extensive family planning program and considerable donor one of the lowest in the world, unemployment and land- commitment. However, before concluding that the family lessness remain high, and illiteracy among adult women, planning program accounts for these trends, three compet- while it has fallen modestly, remains above two-thirds. The ing interpretations must be considered. First, this trend in Government has not developed social security programs fertility might be explained by factors unrelated to in- creased use of modem contraception being promoted by 2. The change in marriage patterns is mainly an increase in the age of mar- the program. Second, improvements in health conditions riage. Duration of marriage has lengthened, partially offsetting the age effect. The only other change of significance has been a decline in postpartum infe- cundity, which has had a positive effect on fertility. Changes in the extent of 1. See, for example, Demeny (1975); and Arthur and McNicoll (1978). abortion are not known. 39 that would substitute for the family and children as gram, which provided information, legitimation, and safeguards against the many continuing hazards of life. contraceptive supplies and services. Although the family Most of the country's 66,000 villages remain inaccessible by planning program began with (and continues to have) road, and river transport is inadequate even during flood- many serious problems, it can nevertheless be credited ing season. with building a system which made possible these changes There is, of course, the possibility that unrecorded in contraceptive use and fertility changes have taken place. Some sociologists and anthro- pologists report that villagers appear more prosperous than The Bangladesh Family Planning Program 20 years ago; others, however, have reported greater impoverishment. The Food for Work program has opened This program had to operate in an extremely difficult en- up additional employment opportunities for women. Also, vironment. (1) As indicated above, the program has re- surveys report increasing nurnbers of persons who indicate ceived no assistance from improvements in social and that they cannot afford to maerry as early or have as many economic conditions, which still seem to favor large fami- children as they would like.3 However, these changes are lies. (2) The physical environment makes implementation small relative to the magnitude of changes that appear to of all programs in the countryside very difficult. The sea- have occurred in other countries with similar demographic sonal flooding that occurs over vast areas of the country- changes. side greatly complicates logistics and communications on a Apparently, then, it is possible for fertility to decline and regular basis, and natural disasters result in periodic shifts contraceptive use to increase without much change in those in priority away from longer-term programs like family social, economic, and health variables generally believed to planning and education. (3) The administrative structure be crucial pre-conditions for demographic change. This and operating style of the Government is highly centralized makes Bangladesh an especially important case to study in and hierarchical, better suited for developing plans, issuing view of the great interest in whether similar demographic orders, and dealing with foreign agencies like the World changes are possible in many other poor countries in which Bank than for obtaining feedback, working collaboratively there is no obvious sign of the onset of fertility decline. with field workers and clients, and obtaining the coopera- One interpretation is that there has been a significant tion of agencies over which it has no direct control-condi- latent demand for family limitation for some time despite tions that seem to be necessary for an effective family adverse socioeconomic conclitions. Cleland and Wilson planning program. (4) Managerial talent with an innova- (1987) argue that this is the case in many pre-transition tive, problem-solving bent-especially important in a pro- societies and that such demand becomes manifest when gram where the best practice is not well defined-is ideas about the legitimacy and efficacy of birth control are generally very scarce in Bangladesh, and especially scarce widely diffused. in the population program because of all the more immedi- In 1975, the mean desired number of children in Bang- ate competing demands elsewhere in the country. (5) This ladesh was about four-identical to that in Taiwan in 1965 limited managerial talent found itself inundated by de- when it was well launched cn its precipitous fertility de- mands from all quarters, from the central political authori- cline. This number decreased to 2.9 in 1989-the level for ties who insisted on quick results and from a multitude of Taiwan, China in 1976 when its fertility rate was 2.7. In suc- foreign aid donors-multilateral, bilateral, and private-all cessive surveys, a substantial number of Bangladeshi wom- eager to help but all with different motivations, procedures, en not using contraception have said that they wanted no and philosophies. more children-one indicator of demand. Until recently, In addition, the sense of urgency that surrounded the be- however, these data, indicating considerable demand, have ginning of the program in 1971 resulted in the premature been discounted as polite statements of what many respon- initiation of a large, complex program without a testing of dents thought the interviewers wanted to hear. strategies or a gradual scaling-up, thereby locking in place After the fact, it now seems plausible that pre-existing la- problems that persist to this day. The history of the first IDA-assisted population project in Bangladesh is indica- tive. Although it was designed by a small working group from the Planning Commission and the Bank that had little 3. Armindo Miranda from the Chr. Michelsen Institute in Norway has put knowled fiel Conditions and operantial radlities forward an interesting hypothesis to explain the increasing age of marriage. knowledge of field conditions and operational realities, po- Since the 1960s, the dowry required to marry off a daughter has increased litical and administrative pressures in the country resulted dramatically, a phenomenon he explains by a bulge in births after World War in an atmosphere in which anything short of a comprehen- II currently reaching marriage age, which temporarily increases the supply sive and ambitious program was unthinkable. The result of eligible women more than eligible men (because men marry at much later ages than women). It is not clear what effect this phenomenon has on marital was a large, complex, unwieldy project and administrative fertility rates. structure. At the outset, consideration was given to a vari- 40 ety of strategies, including an independent family planning of childbearing years have been visited by female village outreach program, an integrated program of basic health workers at least once quarterly. These workers also de- and family planning services, and a set of special develop- liver immunization and other simple health services to ment programs aimed at generating demand for family households, contributing to the credibility of the family planning services. In the end, it was decided to try all three planning messages. This is a two-edged sword, howev- strategies simultaneously A categorical family planning er, since these activities sometimes crowd out the time program was established in the population wing of the that is supposed to be devoted to family planning. Ministry of Health, and this operated fairly independently * Mass-media campaigns, along with the information ac- of the rest of the Ministry. A maternal and child health tivities of the clinics and outreach workers, has succeed- (MCH) program which incorporates family planning as ed in making knowledge about contraception almost one of its components was also established in the Ministry universal in this largely illiterate society. According to of Health. Programs for demand generation were estab- survey research, about 6 percent of all married women lished in other ministries in the fields of education, agricul- had heard about contraception in the mid-1960s. By ture, rural development, labor, women's affairs, and 1969, this had increased to 52 percent, although few communications. All told, eight ministries were involved. women knew where to obtain services. By 1983, 99 per- Funds were provided to purchase civil works, equipment, cent knew of at least one method. Today, most women technical assistance, fellowships and training, contracep- know about several modern methods, where to obtain tives, salaries, and other operating costs for these pro- them, and their costs. grams. An inter-ministerial coordinating body, established * Non-clinical distribution channels are beginning to be without a secretariat, never operated very well. The next effectively used. By 1989, almost 40 percent of couples two projects, each one double the size of the previous one, practicing contraception were using condoms and pills continued the rapid expansion of this elaborate system. supplied by pharmacies and other commercial outlets The result was a series of projects which continuously through a very successful social marketing project. threatened to overwhelm the carrying capacity of the Gov- * By 1990,120 NGOs were collaborating with the Govern- ermnent and a program with several serious, unresolved ment program in providing contraceptive services, problems. Perhaps the most important is the failure to inte- estimated to reach as many as 20 percent of the contra- grate the health and family planning programs at interme- ceptive users. They have now been authorized to receive diate levels between the center and the local community, a foreign aid funds for this activity. factor contributing to inadequate quality of care and a cause of serious friction between the separate cadres. Compound- The Foreign Assistance Community ing this problem is the fact that most health workers tend to and the Bank be men and most family planning workers women. Also im- portant, the program continues to be based on rigid, uniform How was it possible in this environment to accomplish standards, irrespective of local conditions, and unrealistical- this much? Seven factors seem to have been critical. ly high targets and workloads for outreach workers, which Government commitment and interest in assistance. First and also ac[versely affect service quality and morale. foremost, the determination of the government to solve the Despite these problems, the program has a number of country's population problem-to proceed publicly and important achievements to its credit: rapidly despite the political risks involved, to allocate a sizeable portion of its limited budget4 and managerial tal- * MCH clinics have been constructed, equipped, and ent to this program, and to request and utilize foreign assis- staffed in 2,716 of the 4,325 unions (the basic local unit of tance and advice-was crucial. This is probably the most Government). These union-level clinics provide non- surgical contraceptives and primary care for side effects plus some basic maternal and child health services. Many of these primary centers are in extremely poor 4. From fiscal year 1974/5 (when separate accounts for family planning and health began to be maintained) to 1986/7, almost 6 percent of the devel- shape, however, because of lack of maintenance. opment budget and 5 percent of the revenue budget were devoted to family * Tubectomy and vasectomy services are available in ev- planning programs and another 3.5-4.0 percent was devoted to health. The ery district and subdistrict hospital at no cost to the pa- figures for family planning are high by world standards. Indonesia, for ex- ample, has devoted about 1 percent of its budget to population and family tienit. Such services are also available from NGOs planning. Per capita expenditures have been rising; in 1987/8, the figure for operating in most district headquarters and large towns. both population and health was $1.60, an increase of perhaps 50 percent in * There are now 23,500 female outreach workers deployed real terms over what it was a decade earlier, but still quite modest by most to deliver family planning services to cou es in their standards. (Based on personal communication from Charles Griffin, Univer- to deliver family planning services to couples in their sity of Oregon, who has recently completed a study of population funding in homes. Surveys indicate that about 40 percent of women Bangladesh for The World Bank.) 41 important factor explaining why more progress has been project development, the development of Five-Year Plans, made in Bangladesh than in Plakistan. and cycles of organizational changes. The coincidence of The magnitude of the foreign assistance made available to these cycles placed the Bank in a unique position to collab- Bangladesh has also been critical. Between 1982 and 1988, orate in the policy development process. Bangladesh has received nearly US$43 million per year for It is significant that in this case the Bank found no diffi- its population programs, sufficient to finance about two- culty getting involved in broader policy issues without re- thirds of its expenditures on these programs. Two-thirds of sorting to sector or program loans. The most important this external assistance has come from bilateral donors, 18 reasons why this occurred are that the Government wanted percent from multilateral donors, and 15 percent from it that way and that, with donor assistance, the Bank was NGOs. The Bank's contribution has been made through able to respond with heavy staff inputs and dynamic lead- three IDA credits, approved in 1975, 1979, and 1986. The ership on the ground. The synchronization of project and first of these projects cost alrmost US$46 million, of which Five-Year Plan development cycles was also helpful. The IDA and co-financiers contributed more than US$40 form that the loans took was probably unimportant. million. Each subsequent project was at least double its pre- Collaboration amongst donors has been substantial. The decessor in size. The Fourth Project, scheduled to begin in first project included seven co-financiers who provided 1992, is estimated to cost $600 million, three-fourths of funds for a project designed in large part by the Bank and which is likely to be contributed by IDA and co-financiers. the Government. This arrangement continued in the second While considerable waste nmust have been involved in and third projects but with increasing donor participation. attempting to move as rapidly as these figures imply, these It has now reached the stage in which it is more accurate to projects helped to promote a degree of progress that describe the fourth loan as a consortium, rather than a Bank, otherwise would have bee:n impossible without such loan. Only two major donors, USAID and UNFPA, remain extensive aid. outside this consortium, but both attend major meetings The character of the foreign assistance. In contrast to the typ- sponsored by the consortium and have even participated in ical Bank-funded population projects, two-thirds of the ex- the design of the fourth project. This collaboration has been ternal funding for these projects came from co-financiers on important, not only for the additional resources that it mo- a grant basis. One consequence has been that a larger frac- bilized, but also because the consortium members have tion of foreign assistance has been provided for software proved to be an effective voice insisting on more attention components (technical assistance, fellowships and training, being paid to quality of services provided (as opposed to and operating costs) than is typical. Table 4 provides an in- simple expansion of the delivery system). dication of this, at least for the Bank projects. Another and Mechanisms for innovation and flexible problem solving. perhaps more important consequence has been the flexibil- Mechanisms have been established to encourage innova- ity that these grant funds have provided to project manag- tive activities and help solve unanticipated problems that ers to get the job done. Examples of how this worked and threaten progress. In the second project, an "Innovative the importance of this element are given below. Projects Fund" was established to cover research needs not The Bank's pro-active role. With the encouragement of the anticipated at the time of appraisal. Research on such issues Government, the Bank has played a very active role in all as contraceptive acceptability and compensation to steril- phases of these projects-design, fund-raising, donor coor- ization clients and tests of innovative approaches applied dination, implementation, and evaluation-far more active by NGOs have been funded in this way. The fund is fi- than is typical in Bank projects. This special role was made nanced by IDA and the cofinanciers, administered by IDA, possible by the assignment of more staff than is normally and supervised by an expert committee of Bangladeshi pol- allowed, both in the field and in headquarters, to work ex- icymakers. The third project established a "Management clusively on the Bangladesh population program. Supervi- Development Unit," funded by co-financiers, to resolve in- sion was not intermittent, as is typically the case. The ternal communications problems and serve as a manage- additional staff, in turn, was made possible by the availabil- ment consulting team reporting to senior MOH officers. ity of grant funds that paid for these inputs. Initially, one Much of their activity has involved the vetting of directives person was assigned to the field mission, but, as the before they are issued and the provision of feedback on projects have grown in size, t]his staff has grown until, to- their implementation. The fourth project envisions the de- day, there are four persons in Dhaka and four in Washing- velopment of a Health Economics and Financing Group in ton working on various aspects, including handling the the Ministry of Health and Family Welfare, responsible for complex accounting arrangements for the co-financiers and economic and financial analysis on which to base financial assisting the Government in the development of Five-Year planning and policy formulation. Plans for the sector. The impact of this involvement has Use of pilot project findings. Before the favorable national been enhanced by the synchronization of several cycles, the trends in Bangladesh were established, a pilot project in the 42 Matlab area demonstrated that the significant latent de- projects, could result in further increases in contraceptive mand measured there was real and that a high-quality ser- prevalence. These include the following: vice program adapted to the local cultural situation could have significant effects. This project (Phillips, 1987), mainly * Latent demand, reflected as unmet need in a variety of supported by USAID with technical support from the Pop- surveys, continues to be substantial. Such indicators ulation Council, succeeded in bringing contraceptive prev- have proved to have considerable validity in Bang- alence tip to more than 50 percent from initially very low ladesh. levels, in an area not substantially better off than the rest of * Recent national surveys also indicate that 60 percent of Bangladesh in social, economic, and health terms. It had at women of childbearing age have not been contacted by least two very important effects in Bangladesh. First, the a family planning worker in the previous quarter. Part of demonstration that such a success was possible greatly the reason for this is that more than one-third of unions strengthened the morale and commitment of Bangladeshis (local government units) still do not have clinics around and donors alike. Secondly, the programmatic lessons of which program activity is typically organized. These this carefully designed and monitored project influenced clinics are planned for construction and operation with- the development of the Bangladesh program in a number in the decade. of important ways. To cite only one important example, the * More limited surveys have indicated that there is an in- Matlab findings indicating that the program needed to in- verse correlation between distance to clinics and extent crease greatly the density of field-staff deployment influ- of outreach activity on the one hand, and the contracep- enced {the important decision of the GOB to increase tive prevalence rate on the other. Where density of ser- considerably the field staff a few years ago. The Matlab ex- vice points is high, as it is in special pilot project areas perience is being studied by scholars and program leaders like Matlab and some sites covered by NGO activities, all over the world for its possible relevance in other places, prevalence rates above 50 percent have been achieved. especially those with what appear to be unfavorable social, Simply increasing clinic density further would increase economic, and cultural circumstances. contraceptive prevalence. Thus, the foreign aid community has found a variety of * In addition to increasing the density of service points ways to expand the (initially) very limited absorptive ca- and frequency of contacts, additional increments of pacity of the Government for undertaking work in the pop- prevalence should be possible from improvements in ulation field. Within this community, the Bank played a the quality of services along lines that have already leading and catalytic role. No other agency-neither the proved successful in some locations: better training and Government nor any other donor-was in a position to increased mobility of workers, more effective supervi- play this role. Had the Bank not done so, the confusion, sion, and decentralization to achieve more flexible local waste, and bottlenecks involved would have resulted in programming. Substantial improvements in efficiency substantially fewer funds mobilized for this sector and sub- and outreach intensity could be realized by improving stantially less progress. coordination of NGO field programs with government Several Bank staff members provided dynamic and outreach efforts and better delineating their roles. Fur- skilled leadership, willing to take risks and to cope with in- ther improvements in efficiency will be realized when evitable resistance from Bangladesh and Bank bureaucratic male workers of the health wing retire in the 1990s and structures to achieve flexibility and unconventional proce- are replaced by multi-purpose female health and family dures. Key members of the Bank staff found innovative planning workers. Especially important would be re- ways to get things done and to work them through the duction in the workload of outreach workers by focus- Bank system. The Bank can be credited with eventually ac- ing on women who are pregnant or have recently commodating itself to such different procedures. delivered, the group most interested in-and in need of-MCH care at any point in time. Implications and Lessons It is difficult to predict how much increase in contracep- Questions have been raised about whether the progress tive prevalence is likely as a consequence of these quantita- made to date can continue much longer without significant tive and qualitative improvements. However, even if it social and economic change or a radical change in program increased from its current level of 35 to close to the 50 per- direction to compensate for lack of such change. There are cent level obtained in Matlab (very unlikely on a national several reasons to believe that diminishing returns from the scale) the total fertility rate would be about 3.7, still well present approach have not yet set in, and that further in- above the replacement level of 2.1. creases in infrastructure and outreach, plus improvements To achieve prevalence rates required for replacement- in program quality along lines already tested in pilot level fertility, new approaches that go beyond these supply- 43 side initiatives and increase the level of demand are likely Washington and in the field, seems essential. If such ap- to be needed. These probably will require basic structural proaches are not feasible for the Bank in other settings, the changes involving such matters as the education, status, Bank should consider allowing or encouraging donors with and employment of women, infant and child mortality, and grant funds and flexible procedures to take the lead in the improved economic conditions. Given the intractability of early phases of program development, only coming in with these problems and how little is known about the speed large, traditional projects when the program begins to take and extent to which demand might respond to these chang- shape. This may mean playing more of a supportive than a es, plus uncertainties about the limits of the supply-side ap- leadership role in the early stages. The alternative, of proach, prudence strongly suggests that both approaches course, is to find ways to change Bank procedures with re- be pursued vigorously and simultaneously To date, only spect to population projects. the supply-side approach has been forcefully pursued.5 The Bangladesh experience illuminates both the positive The most important lesson to glean from the Bangladesh and negative consequences of very large projects which experience is that a supply-oriented approach can increase press hard against the absorptive capacity limits of an ad- contraceptive practice and decrease fertility significantly, ministrative body. On the positive side, there can be no even in exceptionally unfavorable social, economic, and in- doubt that these limits gave way more rapidly and led to stitutional circumstances, providing there is sufficient la- more progress than had they not been breached. Such an tent demand. A key element seems to be trained and well- approach is most useful when one is confident of the direc- supervised workers who visit rural households on a regu- tions in which a program is to move. If this confidence is lar basis and offer a range of reliable family planning missing, it is more appropriate to scale up in a more orderly services. fashion from smaller projects and pilot studies. Failure to A second lesson of importance is that developing a suc- do so in the Bangladesh case led to considerable waste and cessful supply-side approach in such inhospitable settings the institutionalization of some prematurely-made deci- is a complex, difficult, and time-consuming process involv- sions which have proved difficult to change. Nevertheless, ing considerable risk of strategic error. This has a number of the Bangladesh experience has been a considerable success implications for new programs and first population in terms of what was expected in its early phases. Under projects. Such projects must be carefully tailored to specific such difficult circumstances and operating in uncharted local circumstances; they cannot be packaged and trans- waters, waste and false starts may have been inevitable. ferred to new settings, as a first power project might be. The Before drawing implications for other countries and for preparation period is therefore likely to be longer, perhaps the future of the Bangladesh program, it is important to try involving pilot studies and certainly involving more inten- to understand the origin of the demand which prior theory sive staff and technical assistance inputs than the Bank usu- and experience in such unfavorable situations suggested ally supplies. In such settings, there is much to be said for would not be present. One explanation, already noted, is "starting small" and scaling up only after learning from that a significant latent demand for smaller numbers of successes and failures. It helps considerably to have staff in children has always existed even in pretransition societies, the field with access to grant funds and authority to use and that it becomes manifest when ideas about the legiti- them in reasonably flexible ways. macy and efficacy of birth control are diffused (Cleland and Such situations are not ideal for application of the tradi- Wilson, 1987) and contraceptives made easily available and tional Bank approach to lending, involving large projects, affordable. loans, standardized procedures, arms-length involvement This latent demand involves more than a few isolated in- in preparation, and intermittent supervision. Ways around dividuals who deviate from the norm; it involves more these problems were found in the Bangladesh case in part than a third of the population. Where could it have come by using the flexibility provided by other donors. There from? Perhaps, for these persons, traditional values have may be ways other than those used in Bangladesh to been eroded not by progress but by its opposite: their in- achieve the flexibility and the intimate knowledge of the creasing economic deprivation, a result of three decades of situation required. However, a sufficient number of well- deterioration in the rural economy which may have made qualified personnel concentrating on the country, both in many parents believe that children can no longer be count- ed on for support in sickness or old age. Under conditions 5. Some recent encouraging efforts on the demand side should be noted, of extreme deprivation, they must know that they them- however. The General Education Project calls for efforts to recruit female selves cannot be counted on to support parents and rela- teachers and enroll more girls in primary school; a Female Secondary School tives. At the same time, modern communications are Assistance Project is being developed to provide stipends for girls from rural areas; a non-formal education project has been proposed; and an Intersecto- ral Population Activities Program to involve other ministries and agencies is families leading a good life with fewer children. This too being developed. may be altering fertility desires (Freedman, Khoo, and Su- 44 praptilah, 1981; and Freedman and Freedman, 1986). The In view of the importance of the Bangladesh experience presumption is that demand for contraceptives and fertility for other pre-transition countries, this case should be thor- reduction may increase because of the combination of oughly studied to develop a more definitive explanation of Malthusian pressures and new ideas affecting aspirations. what has happened so far and to monitor future events. Finally, to some extent, a strong supply-oriented pro- While it seems likely that the downward trend in fertility gram may itself generate new demand. This possibility is will continue for some time without structural changes, it is based on the view that structural factors are less important an open question when, if at all, a plateau will be reached. than group norms and behavior which slowly change as If and when it does occur, the program will have to change the program becomes more visible and as leaders and pace- in radically different ways to continue progress. Some con- setters themselves become acceptors. tingency planning to anticipate this day would be prudent. 45 5. Brazil, Colombia, Mexico Demographic Trends, Devfelopment, conducive to small-family values and to new attitudes and the Role of Government about the use of contraception. Among these changes are increased female education and labor force participation, Brazil, Colombia, and Mexico have experienced substan- rapid urbanization, and the spread of consumerism and ex- tial fertility declines (29-45 percent) since 1970, comparable posure to the mass media. Also contributing was change in to or somewhat greater than the declines in India, Indone- the economic climate. During the period of rapid growth in sia, and Bangladesh (Table 1.1 in statistical appendix). the 1960s and 1970s, many urban consumers formed expec- When measured as a proportion of the distance between tations about improved living standards, only to confront the beginning fertility levels and the replacement level of much harsher economic realities during the 1980s. The pace 2.1, the fertility declines are an impressive 51-79 percent. of fertility decline may have accelerated because families Their contraceptive prevalence rates of 53-66 percent are adjusted to the economic squeeze by deciding to postpone higher than those of the three Asian countries previously or stop having more children. Another factor that could be discussed. As compared with the Asian countries, contra- significant is the effectiveness of the IEC approaches used ceptive prevalence in the Latin American countries ac- in Latin America, in which FP advocates have used radio counts for more and nuptiality for less of the fertility spots, rock videos, soap operas, and comic books to deliver decline. While data are not very reliable, it appears that il- their messages. legal abortion may count for 25 percent of overall fertility The three Latin American countries clearly rank higher control in Latin America. than the three Asian countries and Kenya on a broad range The significant increases in contraceptive prevalence in of development indicators (Table 1.1). These include much the three Latin American countries are behavioral evidence higher GNP per capita, higher life expectancy, lower infant of a demand for family limitation. There is further evidence mortality, and much lower illiteracy Nevertheless, fertility of this in the low reported desired number of children, with levels for Brazil and Mexico are somewhat higher than means of 2.8 to 3.1. There is a striking convergence of these those for Indonesia, and Colombia's is only marginally rates with those for the three Asian countries (2.9 to 3.4).1 lower. India's fertility rate is only marginally higher than This was unexpected in view of large differences between Mexico's. Thus, while the higher development levels in Lat- the two groups in socioeconomic levels and in the character in America are consistent with their falling fertility, the ac- of their family planning programs. However, the propor- tual fertility levels attained do not put them in a different tion of women wanting no more children in the three Latin class from the three Asian countries studied, despite their American countries (64-69 percent) is considerably higher much lower development levels. Whether the stronger and than that in the three Asian countries (50-58 percent). much earlier family planning programs in the three Asian These fertility declines have taken place against back- countries explain this anomaly is a question that is dis- grounds of substantial socioeconomic changes that were cussed later. While all three countries had increased contraceptive prevalence and declining fertility, these outcomes were 1. The 3.4 figure for India is obtained by adding the mean number of ad- reached via quite different policies and organization of ser- ditional children desired to the number of living children. This is biased up- vices. Until the mid-1970s, Brazil had a pro-natalist policy. ward, because some respondents already had more children than they A wanted. After that, its laissez-faire stance opened the way to consid- 46 erable activity by NGOs, particularly BEMFAM, the Brazil- spite the fact that social and economic progress in the 1960s ian IPP]F affiliate, which was active in public education to and 1970s was substantially ahead of these countries. UN popularize and legitimize family planning, and in the pro- estimates place the TFR in Mexico at 6.4 as late as 1970-75, vision of some services; in the latter, going so far as setting when Brazil was at 4.7 and Colombia at 4.8. Mexico's fertil- up community-based distribution programs at the invita- ity fell precipitously after 1970-1975 (to an estimated 4.9 in tion of some states. The contraceptive pill was mainly pro- 1975-80 and 4.2 by 1980-85). Contraceptive prevalence in- vided by private pharmacies whose prices were very low creased from 12 percent in 1973 to 38 percent in 1979, and to because of price controls on pharmaceuticals, and because 53 percent by 1987. The onset of this rapid rise in contracep- local producers paid no royalties and had to compete with tive use and the associated decline in fertility coincided free pills from NGOs. Sterilization, the most common with the inauguration of a substantial national family method, was provided about equally by private and Gov- planning program. emnment agencies. The Government played an important, After a long period of pro-natalist policy, the Mexican but passive, role in the large increase in female sterilization, Government moved in the early 1970s to adopt a policy to both in government and private facilities, because the So- slow population growth and make family planning avail- cial Security Health system paid for Caesarian deliveries able through its social security system and the Ministry of which provided an opportunity for a tubal ligation as a side Health. Most of the increase in contraceptive prevalence arrangement, for an additional payment. Thus, while Brazil since the 1970s is attributable to these programs. Thus, the has had no formal policy to increase access to family plan- most likely explanation for Mexico's different experience ning services, the combination of strong NGO activities, is that latent demand built up during the 1960s and 1970s low-cost commercial supplies, and indirect subsidies for but did not become effective until government policy sterilizations resulted in the highest prevalence rate among changed and availability of supplies and services and in- the three countries. It is plausible that this rate would have formation increased. been even higher and fertility lower if there had been a NGOs have played an important role in Mexico in such well-organized national family planning program. This activities as training, mobilization of support by physi- would have increased prevalence, especially in the poorer cians, and work with special groups such as teenagers and regions and strata of the population where prevalence rates underserved poor groups. However, in terms of direct ser- are lower. vice provision, they have been much less irnporian; i;har In Colombia, NGOs have been the principal provider of their counterparts in Colombia and Brazil, where the direct FP services and supplies. The dominant role has been government role is much less. played by PROFAMILIA2, that country's IPPF affiliate. It is likely that three sets of interacting factors jointly ex- PIROFAMILIA established a network of family planning plain the experience of these three countries: (1) moderniz- clinics, conducted information and education campaigns, ing forces associated with the socioeconomic changes and launched male and female sterilization programs dur- previously described; (2) the impact of improvements in ing the 1970s, and remains the main provider of steriliza- the supply of FP services and supplies on demand, either tion services. Though drugstores are the main source of by increasing consciousness about being able to exercise pills in Colombia, PROFAMILIA helped to keep pill prices fertility control or by transforming latent demand into down by buying them in bulk and re-selling them to phar- actual demand, as availability and access to low cost con- macies. It was also a pioneer in organizing outreach service traceptives increased; and (3) economic setbacks amid ris- programs for rural areas and in social marketing efforts. ing expectations. The timing of demographic change in In contrast, the Government until recently has played a Mexico-little change at first despite socioeconomic im- distinctly secondary role. While it announced a national provements, and rapid changes after the introduction of the policy to slow population growth as early as 1969, its ef- national program-suggests that factors on the supply side forts were sporadic during the 1970s. Since 1981, the Min- are especially salient. istry cf Health has developed a program to improve maternal and child health services, which includes family External Assistance and the Role planning. After initial outside donor support by UNFPA, it of the Bank has also had support from the World Bank since 1985 for services in poor areas. In contrast to other regions, external assistance pro+ id ed Family planning and fertility decline in Mexico lagged only a small part of the costs of family planning in Latin considerably behind the trends in Colombia and Brazil, de- America and, of that, close to three-fourths was provided by private (NGO) rather than multilateral or bilateral agen- 2. Plus its predecessor, ASCOFAME, which focused on training and public cies. Nevertheless, these inputs played a critical role in the education, paving the way for later provision of supplies and services. expansion of services by enabling providers to deliver ser- 47 vices at a lower cost, helping them to legitimize and popu- already comparatively high and the structure of service larize FP and providing them with training. USAID, for delivery well established. example, channeled most of its support to local NGOs, di- * Importance of private and para-public providers. The most rectly and indirectly (through its contributions to interna- effective actors in the population field-NGOs in Brazil tional NGOs) and played a major role in supplying them and Colombia and the social security system in with low-cost contraceptives. A large share of UNFPA Mexico-are not institutions with which the Bank typi- funds, which went to the governments involved, were used cally works directly. Given the attitude of the govern- for non-FP activities such as policy development, research, ments, it was not possible to work indirectly with these and training. organizations as the Bank did in Kenya and elsewhere. A review of the possible avenues through which the WorldieBank mighthave ontribnuted throufert hit dclne The Bank's current approach in Latin America is to focus suggest tatk mihts inlencsinimal,utedtofe on reproductive health and safe motherhood-not the ben- efits of reducing population growth rates-as rationales for * Policy dialogue. While the Bank made a limited contribu- family planning. While some Latin Americans believe the tion to the policy changes that occurred in Mexico (see ulterior motive remains population control, this approach below) population policy was either not significant or does in fact speak to the main needs for family planning, too sensitive to deal with in Brazil and Colombia. given the present demographic situation of these coun- * Sectoral research and analysis. Some useful contributions tries.3 Fertility rates have come down and contraceptive were made in the human resource area generally, but prevalence rates increased to the point at which one can very little focused directly on population or family plan- foresee significant declines in population growth rates as ning prior to 1989. An exception was a 1974 sector report the age structure evens itself out over time. To force this on population, health, and[ nutrition in Brazil. process to occur more rapidly would be unwarranted and * Projects. Family planning was included as components politically unwise. However, important issues of equity, of MCH in health projects in Colombia and Brazil, but quality, and effectiveness in the utilization of resources for little was done to implement these subcomponents. A family planning still remain. Many of these problems are loan for population activitjies was prepared for signature subsets of issues that the Bank is already dealing with in its irt Mexico, but withdrawn when a new government work on health services in general. It therefore seems ap- came to power and decided to use grant funds from propriate and opportune to bring family planning into NGOs (some supported by USAID) for the same pur- these discussions, and into project lending focused on the pose. In general, health len,ding in many Latin American health sector. countries has been limited by governments' reluctance The cases of these three countries demonstrate the im- to borrow hard currency for social sectors. portance for the Bank of supporting its population work * Other lending. Lending to other sectors which can have with a rationale acceptable to all parties: government, Bank indirect effects on fertility-for example, education, management, and Bank operating staff. In these three cases, transport, and communications-was undertaken, but Bank management was perceived by government officials not with demographic considerations in mind. as holding an unacceptable view, and Bank staff were Three factors appear to explain the Bank's limited caught in between. While the staff backed away from the impact: population control issue, it continued to be hampered by * Political sensitivity. Particularly during the 1970s, when the perception within the Bank as well as within these the direction of policies and programs was being countries that they were out of step with what the Bank is shaped, Bank management was perceived in Latin really about in its population work-in other words, that America to be outspoken about the need to reduce pop- population control is really the Bank's strategic objective. If ulation growth rates, a rationale for family planning the Bank wants to work in countries that do not accept this programs that was unacceptable in these countries. This rationale, as well as in countries that do, it will have to base made it difficult for Bank staff to broach the subject with its population program on a broader and more flexible set these governments. In contrast, USAID, which also of principles. started with a hard-sell approach, changed tactics and In establishing such a set of principles, one could begin began working with NGOs, thereby avoiding confronta- with the recognition that the Bank's overall objective is to tion with the governments while still channeling signif- promote sustainable improvements in living standards, icant support to these countries. * Timing. By the time Bank policy changed to permit lend- ing directly for health (which offers another avenue for 3. Graphic evidence of one of these needs was presented in a recent article the Bank to approach FP), contraceptive prevalence was indicating the high inddence of illegal induced abortions in some of these countries. See Singh and Wolf (1991). 48 and that family planning programs influence variables re- developed and used in specific contexts. It is likely that lated to this objective on two levels. At the micro-level, they most such discussions would conclude that FP programs can improve family health and enlarge the range of control are appropriate for their micro-impacts in nearly all individuals have over their lives. At the macro-level, they circumstances, and for their macro-impacts in cases where can have major demographic, and therefore economic, population growth rates are "too high". But judgements impacts. Such programs are therefore appropriate topics for can differ considerably about what constitutes "too high" a policy dialogue and sector analysis in order to develop growth rate. tactical judgements about how, if at all, they should be 49 6. Keny a Demographic Developments and staff. Primary health centers were scattered throughout the Principal Determinants countryside to provide MCH/FP services for those who de- sired them. It was essentially a passive, clinic-based system During the last 20 years, Kenya has had one of the high- operated by medical personnel. Efforts to create an out- est total fertility rates in the world. These high rates coin- reach program did not get very far. Logistics, IEC, research, cide with very large desired family size which in turn is and evaluation components of the program were weak and plausibly explained by a strongly entrenched and distinc- often neglected. In 1982, an umbrella organization, the Na- tively African family systemr. Most observers have there- tional Council on Population and Development (NCPD), fore held out little hope that family planning programs was created with responsibilities for promotion, informa- would succeed. This view is corroborated by observers. tion, education, and communication programs (IEC), rela- "There is little reason to expect the demand for chil- tions with foreign donor agencies, and coordination of the dren [in Kenya] to decline in the absence of any FP activities of NGOs, but this too got off to a very slow change in economic, social, or cultural conditions, start. Lapham and Mauldin, in their 1984 study which pro- since it is solidly founded on African institutions" vided comparative ratings of family planning programs, (Frank 1987, p. 197). ranked the Kenyan program in both 1972 and 1982 as "Marriage in Kenya was traditionally, and in consider- "weak," its lowest category (although its ranking was high- able measure remains, a form of contractual quid pro er than other sub-Saharan countries). As late as 1987, Frank quo between two lineages, not a direct commitment and McNicoll described the program as a "conspicuous between husbands and wives.... The lineage, as well failure" whose impact was "thus far negligible."' (Frank as the husbands and wives, have interests favoring and McNicoll 1987, p. 209). high fertility.... These obstacles to change may erode, Because of such views, the results of the 1989 Kenya De- but only as the family system itself is modified ...." mographic Health Survey (KDHS, 1989) were unexpected (Frank and McNicoll, 1987, pp. 209-243 ff). by most observers. According to this survey, the total fertil- Other scholars have written in a similar vein about other ity rate was 6.7 in 1989, down by 13 percent from the 1984 countries in sub-Saharan Africa, as well as Kenya (see, for estimate of 7.7 and by 16 percent from the 1977-78 figure of example, Dow and Werner, 1983, Kamuzora, 1989, and 7.9. This decline is probably real, given its remarkable con- most recently Caldwell and Caldwell, 1990). sistency with other indicators. During this 1984-87 period, In addition, until very recently, Kenya's family planning the contraceptive prevalence rate increased from 17 to program was considered to be quite weak. Although its population and family planning policy was established in 1967-the first in sub-Saharan Africa-availability of ser- 1. In faimess, it should be noted that they did call attention to the potential vices and supplies of modern contraceptives grew very for improvement. Following their detailed analysis of barriers to fertility de- slowly. This slow growth resulted from an early decision to cline, they wrote: "Notwithstanding these caveats, an increasing recourse to fertility regulation in Kenya is not an implausible finding.... Even within the provide FP services through the MCH program. This was more traditional family pattemr.. some changes in reproductive behavior the responsibility of the MOE, which at the time gave pri- can be expected from increasing exposure to urban... lifestyles and more in- ority to rural health, not FE The first 10-15 years were de- tense competition for higher education and modem-sector employment, es- ority to ~~~~~~~~~~~~~~~pecially given the difficult economic dircumstances of the last decade." voted to creating the basic MICH network of facilities and (p. 240). 50 27 percent, most of this because of increased use of modern lagers are typically responsible for constructing and methods; the proportion of women indicating in surveys maintaining the school building and must bear the cost that they want no more children increased from 41 to 50 of textbooks, supplies, and uniforms, which can be sub- percent; the mean desired number of children declined (for stantial, especially for poor families. These characteris- husbands as well as wives) from 5.8 to 4.4; and the percent tics-high demand, increasing costs, and significant of women pregnant at the time of the survey declined from improvements in educational attainment-appear to II to 8.9. Moreover, II percent of women with a birth in the have interacted in the way theory says they should, to last 12 months were willing to say that they did not want to change parents' attitudes about the benefits and costs of have the last child when it was conceived, an extraordinary children and make them want to substitute quality for finding lby African standards. All these indicators began to quantity. move in. the directions noted during the 1977/78-to-1984 2. Improvements in health. The Government, plus the NGO period, but so slowly-the TFR, for example, only declined community, has also invested heavily in health services from 7.9 to 7.7-that they were discounted by most observ- since independence-almost 4 percent of GDP during ers. By now, however, the magnitudes and rates of change the 1970s. The result has been that, between 1970 and to- are such that they can no longer be ignored. day, life expectancy has increased from 46 to 58 years Of the total decline in fertility, from 7.9 to 6.7, about half and infant mortality has decreased from over 100 to appears to have resulted from an increase in the age of mar- about 703. This factor also is likely to have had its riage. The other half was probably due to an increase in use expected effect of reducing the number of children of contraception. Both factors have had to be strong enough parents need to ensure the survival of the desired num- to offset decreases in breastfeeding and postpartum ber and reinforcing the growing concern with quality abstinence which by themselves would have increased fer- over quantity. tility. It is not known whether any changes occurred in 3. Increases in female labor force participation. While a large abortion rates. fraction of Kenyan women have typically been recorded What explains this turnaround? While no in-depth stud- as participating in the labor force-in our sample, only ies have yet been undertaken, six factors operating in com- Senegal has a higher fraction-the character of this em- bination appear to be the most important. ployment seems to be changing. An increasing number are becoming involved in new occupations or in more 1. Education attainment and costs. The demand for education moe versins of inal occupations sc as as a means to social and economic advancement has ming and tradinin al armore chmet been strong in Kenya since colonial days. After indepen- itive with childbearing. This change undoubtedly has dence in 1963, the Government quickly announced a been associlated with mproved female education levels policy of universal, free, primary education and em- an impoved sta generally. barked on a massive school expansion program. During 4. Declining size of landholdings. Concerns about the grow- most of the 1970s, education expenditures were about ing sity of latin oncaa s and-a sml por- 7 percent of GDP, a larger percentage on a sustained ba- ing densnty of populaston arable land-a small por- SiS than nearly any other country in the world, andtinoKey'vatpnsac-hebenocd sisthan nearlr tanythato other country es in theswoldy,a since before independence. During the last two decades, much larger than that of other countries in this study, it has reached a point where farm familie- hmsle which ranged from 2.5 to 3.5 percent. The result has has reache point were numbes thilves can see the connection between the number of children been a dramatic improvement in educational attainment they have and the size of plots they can pass on to them. from the very low base existing at the time of indepen- 5 therhvemandsthe sie of seyvcan Asshon totem. 5.Improvements in supplies and services. Although modern dence. Most significant, a substantial fraction of women contraceptives were relatively abundant in the 1970s, have now become literate and have completed six years their effective availability was limited because of restric- of schooling (which many studies indicate to be the tive guidelines for their use in goverment health cn- threshold for maximum effect on fertility).2 This ics, the small number of supply points, lack of training, achievement has come at increasing cost to individual and lack of interest on the part of the health staff. By the families. While there are no primary enrollment fees, vil- mid-to-late 1980s, restrictions on to whom and how con- traceptives could be offered were relaxed, the density of supply points had increased, and a wider variety of de- 2. Female enrollment in primary education, which was roughly in the livery chanesgvren,N s, and comme- same range as that in Senegal, Bangladesh, and India in 1970, is now substan- livery channels-governent, NGOs, and commer- tially higher than in those countries. Secondary enronllment of females, which was similar to that of Senegal and Bangladesh in 1970, is substantially above those countries today, though still below that of India; and the female literacy 3. This last figure is substantially lower than those observed in Senegal, rate was recorded at 50 percent in 1985, placing it substantially above that of Bangladesh, and India, comparable to that of Indonesia, and not much above Senegal, Bangladesh, and India in that year. that of Brazil. 51 cial-had developed. NGOs, for example, with USAID Figure 6.1: Family Planning Accessibility and Prevalence support, initiated large-scale outreach programs (social in Kenya: Travel Time, by Year Facility Started marketing and community-based distribution schemes) 100% during this period, which the previous health center- 90% / based program lacked. Perhaps most important, as indi- g.9 80% / cated below, signals were changed in ways that made t ' 70% / health center staff give higher priority to the provision 2 60% - of services. 50% , The dramatic increase in supply points and its correla- E 40% tion with increases in contraceptive prevalence is graph- E 0 ically illustrated in Figure 6.1. It shows that, despite an / co anti-natalist policy since Independence, contraceptive CD / - ._ - - services were not available to a large proportion of the _ 10% *.. rural population until the mid-1980s. In 1970, less than 1960 1965 190 1975 1980 1985 19 90 10 percent of this population was within one day's travel Year to a delivery site. By 1980, this figure had risen to 40 per- - Less than 1 day Clinical prevalence cent, and by 1985, to more than 75 percent. By the end of - ...... Less than 3 hours - - - - Non-clinical prevalence that decade, almost 90 percent of the rural population -- Total Prevalence was within three hours of such services. Most of this in- Source: Hammerslough, 1991. crease occurred because of growth in the number of gov- ernment clinics, with a sizeable minority receiving supplies from field educators and family planning dis- tributors. The increase in use of clinical and non-clinical foreign aid donors, private investors, and tourists which in ethods, as revealed by al series of sample surveys, is turn has exposed the population to a materialistic, Western methods,easirevealed byr a series of sam ple lifestyle at the same time as it has increased opportunities also depicted for three points in time. 6. National leadership. While President Kenyatta was not with which to pursue it. In the process, the perceived value very interested in the pcipulation program, President of children and large families may have declined more rap- Moi has gone out of his way to promote it. This has le- idly than in other African countries with similar traditional gitimized the program and a smaller family norm and values. Second, fertility control for spacing has been accept- has prodded government officials-for example, district ed in Africa for generations. This may help explain why ac- commissioners and village chiefs, who have the power ceptance of contraception for limiting family size to allocate funds for construction and to assist NGOS proceeded rapidly in Kenya once attitudes towards large with transport, petrol, meeting rooms, and the like-to families changed. give the program more priority The President's leader- ship also led to an open debate on the population prob- Foreign Assistance and the Role lem, the outcome of which was greater public support of the Bank and consensus.4 What role has the Bank played in this process? First, the At least two other factors may be playing significant Bank, along with other donors, has encouraged and finan- roles. First, the country has been especially open to the in- cially supported the Government's economic policies; in fluences of Western lifestyles and attitudes about family particular, its heavy emphasis on education and health. size. The Government has allowed markets to operate rela- Since 1970, the Bank has provided loans and credits to tively freely and has encouraged private entrepreneurship, Kenya of more than US$2.6 billion, of which 4 percent went both foreign and local. This policy-in direct contrast to to the education sector. This percentage is close to average that of most of its neighbors--plus a relatively stable gov- for Bank lending in Sub-Saharan Africa, but larger than that ernment and strong economy has resulted in an influx of found in Bank lending programs elsewhere. Financial assis- tance to other sectors-transport and communications, for 4. Starting in 1984, NCPD hosted two national and a series of district-level example, which has amounted to 20 percent-may have population conferences that have proved to be very useful. The first confer- helped the program by providing needed infrastructure. ence, for example, which was co-chaired by the Vice President and the Min- However, there has not been-at least up to about 1989- ister of Health and attended by all Cabinet members, led to a series of debates that were continued in the press. One such debate over the use of De- any conscious attempt to slant these lending programs in poprovera led to a technical defense of its use by the Director of Medical ways that would maximize their fertility impact. (An exam- Services, which was reported in the press and may have been associated with ple would be specifically encouraging enrollment of wom- a significant increase in demand that occurred shortly thereafter. 52 en or introducing population education programs into the er forthcoming. The National Family Welfare Center, estab- curricula.) lished by the project to plan and support the activities of the Second, the Bank, along with other donors, has provided new MCH/FP program, was never provided with the staff a significant portion-perhaps 70-80 percent-of the funds required for the purpose. As a consequence of this and spent by the Government on its population and MCH pro- other factors, little progress was made in implementing grams. According to the UNFPA (1989), during the 1982-89 other components such as IEC, research, and evaluation, period, Kenya received more than US$8 million per year for and the impetus for a new category of health worker-fam- its population program. In per capita terms (because of the ily health field educators, who might have served as the nu- small size of its population), this is one of the largest cleus for an outreach program-was dissipated and amounts in the world.5 Bilateral agencies provided 54 per- eventually died. In the end, there was really nothing in this cent; multilateral agencies (including the World Bank), 10 project directly related to family planning, other than the percent; and private organizations, 36 percent of this agreement that when MCH services eventually came on amount. The Bank's contribution, in the form of four credits stream, they would include a FP component. for projects initiated in 1974, 1982, 1988, and 1990, has Also, no progress was made toward achieving the stated amounted to US$82.2 million. goals of the project-goals which must have appeared na- Bank involvement started in the fall of 1969, shortly after ive and unrealistic to many observers, even at that time. establishment of the Population Projects Department with- These were: (1) to reduce the population growth rate from in the Bank's Central Projects Staff. This was a time when 3.3 to 3 percent and (2) improve the health of mothers and the new and relatively inexperienced staff was under sub- children. By the time the project ended, the population stantial pressure to produce a "bankable project," particu- growth rate had in fact increased to 3.9 percent. The implic- larly in Sub-Saharan Africa. On the basis of a five-year it presumption was that construction and training would health plan prepared largely by expatriate advisors, the quickly lead to effective operations and the latter would Governrment, after considerable internal debate and discus- quickly lead to reductions in fertility and improvements in sions with donors, decided that its primary interest was ru- health. Under the circumstances of the time, it was a con- ral health and that it would consider FP only as a part of the siderable achievement to have successfully completed the MCH program. Accordingly, it proposed a project focusing construction and training program. Had project goals been on construction of rural health facilities and training specified in terms of future expansion of the system, with schools. 'The rationale was that, since FP was to be deliv- the explicit understanding that this would be an important ered through the MCH program, the establishment of a net- first step on a long and difficult path, perhaps the disillu- work of delivery points and training for that program was sionment that followed the end of this project could have the first priority. While WHO supported that position, been avoided. UNFPA and USAID objected on grounds that this strategy Considerable analysis and soul-searching went on before would have little or no impact on the population growth the second project was sent to the Board. The general con- rate during its lifetime. The Bank decided to support the sensus was that four factors stood in the way of progress: Governnment's position and went forward with the apprais- lack of commitment at senior levels in Government; weak al without the involvement of these other agencies. The organizational support provided by MOH; lack of effort in consequence was a project that was 89 percent civil works, dealing with continuing high levels of demand for children; 9 percent: furniture and equipment, and 2 percent technical and excessive focus on investment in building health infra- assistance, including fellowships and training. Nearly all of structure, as opposed to providing FP services through ex- this was provided to the MOH, where commitment to fam- isting health centers and the substantial NGO service ily planning was weak. Little thought and fewer resources network. were devoted to the problem of what to do about demand The second project continued support for expansion of or what to do about the very small gap between actual and the MOH health delivery system. However, in order to cor- desired family size. This was essentially a supply-side, clin- rect these earlier problems, two new elements were added: ic-based program that was unlikely to receive many re- support for the newly-established NCPD and for the devel- quests for service. opment of an IEC program by that agency It was later Quantitatively, the project achieved most of its construc- amended to use surplus funds to finance development of tion and training targets. However, organizational support surgical facilities for sterilization. Efforts beyond IEC to for effective integration of FP and MCH activities was nev- promote demand for smaller families and FP were not be- lieved to be politically feasible at the time. This project (completed in December, 1990), like the first, was successful 5. However, this figure probably includes substantial funds for MCH and inofareas it cnstru and training c os were other health programs. This is certainly the case insofar as the Bank loans are concerned. 53 concerned, but made only halting progress with respect to ly on IEC to one coordinating all agencies involved in its IEC and other components. population work-donors, NGOs, other ministries, and By 1988, when the third population project was ap- District Population and Family Planning Committees- proved, further expansion of the health delivery system no and becoming the principal executing agency for the third longer appeared necessary, the political climate for family and fourth population projects. This agency has had signif- planning had substantially improved, and signs of in- icant teething problems and has grown only slowly in effec- creased demand were beconming apparent. The designers of tiveness. After a short residence in the Office of the Vice this project appear to have taken advantage of these chang- President, it was moved to the Ministry of Home Affairs, es by focusing on three areas: strengthening of NCPD (con- from which position it had difficulty promoting the in- struction of headquarters and district office buildings, staff volvement of other ministries and retaining good staff. Its training, and MIS and research and evaluation activities), IEC and promotional activities remain timid and it has not the establishment of a multisectoral IEC program (funds for assumed a strong leadership role in coordinating donors IEC activities in other agenciles, as well as NCPD), and sup- and NGOs or pressing other ministries to become involved port for NGOs to enhance their IEC and service delivery ac- in the population program. It has, however, sponsored a tivities. The only funds made available to the MOH were very useful series of conferences (see footnote 4) and devel- for the conversion of its Heal]th Education Unit into a Media oped a population policy statement enacted by Parliament. Production Centre to produce IEC materials for all users. The judgement is still to be made, therefore, on whether it The fourth project, initiated just two years later because of was a good idea to have promoted the development of this a critical shortfall in the availability of funds from other do- new organization, rather than continuing to influence the nors for purchase of contraceptives, is basically an amend- MOH to move more forcefully into the population field. ment or add-on to the third population project. Other program interventions included successful efforts In addition to these projects, the Bank involved itself in to persuade the MOH to liberalize guidelines for providing policy dialogue, sector work, and efforts to influence the contraceptives, to integrate FP into the mainstream of evolution of the program in a variety of ways. In the early MOH activities, and to offer sterilization services. The Bank days, these efforts were mostly aimed at raising the sense of also played a critical role in the financing of contraceptives urgency with which the GOK approached its population when funding was reduced by other donors. This was done problems. The general economic reviews produced by the despite restrictive guidelines about funding recurrent costs, Bank in 1969 and 1979 included sections on the negative by arguing that these expenditures are equivalent to invest- consequences for economic growth of rapid population ments in creating demand for a product which will eventu- growth and increasing population density on arable land. ally be purchased without subsidies. Finally, the Bank So too did the Population Sector Review of 1974 and the Eco- played a significant role in encouraging the GOK to make nomic Report of 1980, entitled Population and Development in greater use of NGOs in the FP program, in part by earmark- Kenya, a particularly good analytical document. In addi- ing project funds to NCPD for their use.6 tion, senior Bank officials from the President on down lost It is interesting to note that nearly all these interventions few opportunities to raise the issue with their counterparts. were undertaken by staff stationed at headquarters, albeit In the words of more than one observer, the Bank made a with rather liberal budgets for supervision and travel.7 It is significant nuisance of itself during this period. not clear how much more might have been accomplished There was, however, very little if any dialogue about had staff been stationed in the field. Matters might have program content before 1980. The first significant interven- otherwise progressed more quickly. It is also interesting to tion on that level involved support for the establishment of note that funding of NGOs, contraceptives, and other re- NCPD. This came, among other ways, in the form of a con- current costs was accomplished in Kenya at a time when it dition for release of the second tranche of the Second Struc- was commonly believed that it was difficult if not impossi- tural Adjustment Loan. This loan became effective August ble for the Bank to do so elsewhere. 27, 1982, the same day the second population project was signed. It is not clear whether such pressure was necessary to ensure the establishment: of this organization, since it was proposed and supported by important elements with- in the Government. Its creation, however, was opposed by the MOH, and as a consequE!nce, dialogue on other matters 6. Interestingly, one commentator pointed out that since the Government has traditionally provided funds to NGOs to help finance their health activ- of program content with the MOH remained negligible for ities and, in contrast to many other governments, has been willing to borrow several more years. at least on IDA terms for this purpose, it was not very difficult to make the The Bank was also influenitial in the Govemment's deci- case for extension to family planning. 7. Before the appraisal of the third project, the Bank appointed a doctor to sions to expand the role of N;CPD, from one focusing strict- serve in Nairobi, but with responsibilities for all of East and South Africa. 54 Implications and Lessons er, will not manifest itself spontaneously. For some time to come, free and convenient access to high quality services This case provides a good illustration of the virtue of pa- and supplies, plus considerable encouragement, will be re- tience and perseverance in the population field. It has taken quired to affect actions. Thus, while fundamental policy 20 years to reach the takeoff stage in Kenya; it could well changes are not required in the near future, funding must take that long in other countries in the early stages of the continue to expand, quality of services must improve, and demographic transition. As such, it should provide a good most important, promotional efforts like IEC and out- antidote to two self-defeating tendencies: to expect too reach-very weak aspects of the Government program so much too quickly and, when disillusionment sets in, to ex- far-must be significantly improved. pect that nothing will work. It also suggests that some re- However, given existing family size preferences, even if thinking about the way project goals and targets are the gap between desired and actual family size were totally specified and evaluated is in order in the population field. closed, the TFR would decline only to the range of 4-5 chil- Could anything have been done to speed up the process? dren per family This implies a population growth rate still Probably not too much. While each of the Bank projects in- substantially above 2 percent per year. To reduce the TFR cluded co-financiers, cooperation with other donors left further will require continued changes in social and eco- much to be desired at times. Significantly, more efforts nomic conditions of the type that have led to the decline so could have been made to correct program weaknesses-es- far, and/or selective interventions into those conditions to pecially in IEC, outreach, monitoring, evaluation, and re- encourage changes in desired family size norms that are search--and to attempt in other ways to influence demand out of phase with normal development processes. Promis- for smaller families. Better donor coordination on these and ing possibilities include programs aimed to improve the le- related issues would have helped. The Bank pressed the gal, economic, and educational status of women, induce Government fairly strongly on many of these issues- more rapid declines in infant and child mortality rates, and perhaps too strongly at times-and seems to have taken ad- improve old-age security options. vantage of opportunities to progress as they emerged. As Considerable ignorance remains, however, about what an example, it actively encouraged and funded NGO activ- has induced the initial changes in reproductive behavior, let ities, especially in areas in which the Government seemed alone what might be accomplished by deliberate policy unwilling or unable to act. Attempts to proceed more rap- changes. It is plausible that the pro-natalist family system idly might have been counter-productive. As indicated with the high value it places on large families has been above, the evidence with which to judge whether support weakened by the social, economic, and policy changes that for the development of NCPD was a wise move on the have taken place, but this is only conjecture. We do not whole is still inadequate. know which components have been the most significant Regarding the future, is the program now finally on the and how amenable they are to being influenced by policy A right track? There is good reason to believe that progress significant research effort to improve understanding of can continue for some time longer without significant what has happened so far, plus pilot projects to try out changes in policy directions. According to a 1989 survey of alternative interventions for the future are urgently needed married women of childbearing age, 32 percent indicated if one is to be ready for the time when diminishing returns that they wanted no more children but were not using con- to the present approach set in. Trying to understand the traceptJ.on. An additional 28 percent wanted to postpone basis for the incipient changes in Kenya (as well as in the next birth by at least two years, even though they were Botswana and Zimbabwe) may help in developing pro- not using contraception. Thus, 60 percent of these women grams for other Sub-Saharan countries where change is not can be classified as in need of family planning, a potentially yet under way. very large market for services. Most of this "need," howev- 55 7|, Senegal Senegal was selected for case study review because it is new. It is a traditional value. Further, 85 percent of those illustrative of a number of small Sub-Saharan countries who said that they wanted no more children were concen- with high fertility in which thiere is little indication of a sig- trated among women who already had five or more chil- nificant demand for family planning, relatively slow dren, with 73 percent having six or more. progress in social and economic development, and anly re- There has been a significant increase in knowledge about cent beginnings of official support for organized family contraception. In 1978, only 23 percent of women had heard planning services. It is also a country in which the Bank has of a modern contraceptive. By 1986, this percentage had in- had some influence on developing a policy favorable to creased to 69, although there is some evidence that the family planning. knowledge is superficial. This increase is probably the re- sult of the considerable coverage of the mass media in Sene- Demographic and Socioeconomic Trends gal and the frequent interaction between urban and rural populations in this small country. On the basis of its 1986 Dermographic and Health Survey, Given current socioeconomic levels and trends, repro- Seneg,- l's total fertility rate for 1983-86 was estimated to be ductive behavior is unlikely to change very rapidly in Sene- 6.37. This represents a decline from an estimated 7.55 for gal. Per capita gross national product in 1988 dollars was the 4-7 year period preceding the survey and from 7.1 in the $650. While similar figures are recorded for the Philippines, 1976 survey. Part of the decline in general fertility is due to Zimbabwe, and Egypt, Senegal appears much poorer. Its a rising age at marriage. per capita GNP declined 0.8 percent per year between 1965 Whether the estimated trend represents a real decline in and 1988, in contrast to the modest growth experienced by marital fertility is an open qu,estion. It is likely that part of these other countries. Obstacles to economic growth are the decline is attributable to a displacement of births back- substantial. The country has few natural resources. In par- ward in time from recent to earlier periods. In any case, ticular, rainfall is unreliable, only 20 percent of the land is there is little evidence of an increase in contraceptive prev- arable (and nearly all of that is in use), and deforestation alence that could have produced a decline in marital fertil- and desertification are palpable threats. Infrastructure is ity. The proportion of couples using modern contraception poor and less-than-optimal use is being made of these lim- increased only slightly, from 0.6 percent in 1978 to 2.4 in ited resources because of misguided economic policies. 1986. While an additional 7.4 percent were using traditional Educational levels are still low in Senegal, although im- methods, this mainly involved prolonged postpartum ab- proving slowly. In 1986, the DHS found that 77 percent of stinence, which has been com:mon for many years. women of childbearing years had no formal education, 14 While Senegalese women still want large numbers of percent had attended primary school, and only 9 percent children, there are some small signs of change. The mean had attended secondary school. These educational attain- desired number of children did decline from 8.8 in a 1978 ment levels are considerably below those of Botswana, survey to 7.2 in 1986. Further, in 1986, 19 percent of the Kenya, and Zimbabwe, the three Sub-Saharan countries women in marital unions were fertile, wanted no more chil- with significant declines in marital fertility and increases in dren, and were not using contraception. An additional 26 contraceptive prevalence. Infant and child mortality rates percent wanted to postpone al birth for at least two years. have also improved, but remain substantially higher than However, the idea that spacing births is desirable is not those in these three countries. 56 While 40 percent of Senegal's population lives in urban Family Health Project. Eventually, Condition Feminine, areas, the difference in modern contraceptive prevalence which later became the Ministry of Social Development rates is not great-6.7 percent in urban and 0.3 percent in (MSD), was assigned responsibility for family planning IEC rural areas. The rate of urbanization is modest by African and the MOH responsibility for clinical services. In 1990, standards-A percent compared to over 8 percent in Kenya the MSD was dissolved and the MOH, was assigned sole and Botswana and over 6 percent in Nigeria and Zimba- responsibility for all aspects of the program. bwe. These figures do not provide a very substantial basis After the legalization of contraceptives in 1980, USAID for the diffusion of new types of reproductive behavior. and UNFPA negotiated for family health projects to include Life is especially hard for Senegalese women. Strong tra- family planning. While these initiatives encountered oppo- ditional forces channel women into agriculture and/or sell- sition and bureaucratic obstacles, they have gradually in- ing in the informal sector, where they put in long hours. For creased their service points and the range of information, the most part, they are illiterate and uneducated. Islamic training, and service activities. Following an earlier, partial- traditions pose further restrictions on women, granting ly successful project, USAID designed a follow-through them status and security based on the number of children project to its initial family health project. This, too, encoun- they have, limiting wives' inheritance to one-eighth of their tered opposition and implementation problems. However, husbands' estate, and forbidding them employment in cer- with patience, compromises were worked out. This second tain jolbs. In its recent draft document outlining plans for a USAID project has enabled the Government and several National Program in Family Planning, the Government noes to make some progress, although not to meet initial outlined the socio-cultural constraints it perceives as affect- goals. While the proportion of all eligible women being ing the provision of family planning services in Senegal. served is still quite small, infrastructure has been created to These factors include pro-natalist traditions; fatalistic atti- encourage and accommodate additional use. This is espe- tudes and the view that children are the gift of God; false cially notable in the government sector, where there was es- rumors about family planning, which build upon irrational sentially no service 10 years ago. fears; and a tradition of males disassociating themselves During the latter half of the 1980s, CONAPOP'S secretar- from contraceptive use decisions. iat, the Directorate of Human Resources in the Ministry of There are a few positive signs. The concept of deliberate- Planning, undertook a process of discussion and consen- ly limiting births for spacing purposes is widely accepted sus-building which culminated in the publication of a na- and Islamic leaders have been playing a neutral or modest- tional population policy in 1988. The preparation of this ly supportive role. However, these seem weak influences policy statement, long overdue and much needed, was in- compared to the many pro-natalist cultural and socioeco- stigated and actively supported by the World Bank. This nomic factors in operation. statement was followed by a more operational document entitled the Population Action and Investment Plan (PAIP), Population Policies and Family officially adopted by CONAPOP in February, 1991. A major Planning Programs recommendation of the PAIP was that the Government es- tablish within the MOH a National Program in Family This pro-natalist heritage was reinforced by the French Planning (NPFP), which was done in April, 1991.2 These ac- colonial administration which made sale and distribution tions appear to have legitimized open support for family of modem contraceptives illegal, a prohibition that ceased planning among Government officials. Together with the to be enforced in the 1970s and was repealed in 1980. A slow development of a family planning service delivery turning point was Senegal's first-ever census, in 1976, system, they provide the basis for a moderate degree of op- which indicated a population size and growth rate much timism about future developments. larger than expected. These results led President Senghor to The formation of a program in the Senegalese context is warn of serious difficulties in providing basic services if the very significant. Prior to the establishment of the NPFP, all population growth rate were not curtailed. In 1978, the population activities were in the form of "projects," signify- Government created the Direction de la Condition feminine and assigned it 'tutelle" (responsibility) for family plan- 1. CONAPOP consists of representatives from various ministries, with the ning. In 1979, the National Population Commission (CON- Directorate of Human Resources in the Ministry of Plan serving as its secre- APOP) was established to coordinate various ministries' tariat. Its most significant action has been to agree on a population policy pa- activities with respect to demographic issues. However, it per prepared by the Directorate. 2. It is interesting to contrast these events with those in Kenya. There, the has never developed into an effective organization despite policy statement was drafted first and the process of consensus-building substantial donor efforts.' Turf battles ensued between around it took place thereafter. We are not in a position to say which process Condition Feminine and the Ministry of Health after the is better or more efficient; both appear to have worked. It does mean, how- ever, that Senegal is not as far behind Kenya as the difference in dates of their latter was made the executing agency for a USAID-funded policy statements might suggest. 57 ing an entity created by a donor which begins and ends spent many months mediating differences between the two with outside funding. The word "program" implies accep- rival ministries and eventually got them to agree to a sepa- tance of the activity within the Senegalese system, and rate project management unit with representatives from while the funding for the program might still come from both ministries. The result was that the project did proceed projects funded by donors, the sense of ownership has to implementation. This same field staff continued on an al- shifted from one that is external to internal. most daily basis to follow up during implementation, to good effect. Foreign Assistance and World Bank In 1985-86, the Bank undertook a general population Involvement sector review which concluded that USAID and UNFPA were doing as much as could usefully be done at the time Since independence in 1960, Senegal has received large in the FP field. It recommended that the Bank, instead of amounts of donor assistance. As of 1987, the per capita trying to supplement these activities (as it tried to do in amount of foreign aid was abcut US$80, almost double that the rural health project), concentrate instead on helping of other Sahelian countries. About 40 percent of this aid the Government develop a comprehensive population pol- comes from France and 8.5 percent from the World Bank. icy. This recommendation was accepted and eventually During the period 1960-87, the Bank committed approxi- implemented by making the development of such a policy mately $800 million to Senegal through 61 loans, 80 percent statement a condition for release of the second tranche of of which was on IDA terms. Most of these commitments the third structural adjustment loan. While news of this were allocated to transportation, industry, and tourism. approach provoked statements of resentment in other Lending to education constituted only one percent of the African countries, no significant negative reactions appear total, probably because this has been a focus of French as- to have developed in Senegal. The Government did at first sistance. After a number of years of disappointing results, express skepticism because it did not understand the pro- in part because of poor economic policies, the Bank began cess, purpose, and content of such a statement. However, to focus on structural adjustnient loans. Since 1985, these these doubts were overcome during the course of discus- loans have amounted to 40 percent of total lending. sions, and the Government, utilizing a Special Project So far, the Bank has had one project in the PHN area, a Preparation Facility grant and a multi-sectoral team, rural health project signed in 1982 which focused on the produced the statement, which eventually was published provision of buildings and equipment for expansion of ba- in 1988. sic health services. This project called for the provision of Subsequent events have proven that the policy and the FP along with MCH services in health clinics, and a stan- World Bank's role in its development have been welcomed dard package of equipment for this purpose was provided. by Senegal. This document and the process of consensus- However, this stipulation was ignored in the training pro- building that led up to it appear to have allowed and en- gram and not mentioned in either the supervision reports couraged activities in this field that were more difficult to or the PCR. Failure to implement this element, as well as initiate before. As indicated above, it provided the impetus most other software components of this project, is attribut- necessary for the Government to give family planning the able to lack of interest, insufficient budgetary allocations by status of a program and created a framework under which the Minister of Finance, rivalry and turf battles between the the many donors wishing to fund discrete population Ministry of Health and what later became the Ministry of projects may now do so in a more coordinated and rational Social Development, and poor relations between the Minis- manner. Another consequence has been the development of ter of Health and the Bank at that time. a Human Resources Project for Senegal, approved by the Of relevance here is the possibility that the Bank's oper- Board in April, 1991. A condition of negotiation was liber- ating style may also have played a role. The FP provision alization of restrictions on provision of FP services. As a was not a part of the original (,ovemment submission, but result, nurses can now provide all services except steriliza- was introduced at the behest of Bank staff on the basis of tion and traditional birth attendants can distribute pills. correspondence and discussions during brief visits. It was These are radical changes from earlier, when only doctors not developed in a collegial, problem-solving atmosphere could provide such services. A condition of approval was which might have led the Government to assume owner- official adoption of the NPFP. In addition to expansion of ship of the idea. The Bank also did less than it could have MCH and FP services, a few demand-side steps have been during implementation to follow up and encourage the added: an IEC program focused on youth, a WID compo- Government to proceed with this and other software ele- nent which includes a functional literacy campaign, and re- ments. It is interesting to contrast this experience with that search into legal issues like polygamy and personal of USAID, which had similar problems with its first family taxation. More central issues of female education have been health project. In this case, however, USAID's field staff left to a proposed education project. 58 Did tlhe Bank make the right decision in the 1980s to not appear willing at present to operate at this level; and leave assistance for FP up to USAID and UNFPA and en- that the risks involved can be minimized by starting on a courage policy development instead? Probably yes, on two small scale and slowly escalating. grounds. First, much of the early effort to establish FP activ- Without an intimate knowledge of field conditions, a ities required intimate contact and flexibility, which could judgement cannot be made with confidence between these be unde:rtaken effectively by these agencies' field staff, but two positions. Two things are clear, however. First, while it not by the Bank from Washington. Second, the Bank, with is true that an incremental approach to implementation its broacl policy agenda and access to senior policy circles, would reduce risks, this approach requires a different style was in a good position to discuss and encourage the devel- of operations than is typical for the Bank. It requires much opment of a population policy Thus, a good de facto divi- more intimate contact with day-to-day field operations, sion of labor appears to have developed between the much more technical support, and much greater willing- various donors in this field during this period. ness to make significant mid-course corrections. In the last few years, this division of labor has been Second, if one looks strictly at Senegal's demographic changed by the more pro-active role played by the Bank in and socioeconomic characteristics, it must be concluded developing the Human Resources Development (HRD) that the prognosis is for slow and reluctant progress in re- Project. Some donors have raised questions about the ap- ducing fertility rates in this country. It may be that latent propriateness of encouraging the Govemment to borrow demand exists, as appears to be the case in Bangladesh and for activities which other donors could fund on a grant ba- Kenya. For some reason, however, this is not being picked sis. Others have raised questions about whether this up in surveys, so that positive steps on the policy and ser- project, added on top of ongoing and planned USAID and vice delivery side will quickly begin affecting prevalence. UNFPA projects, will exceed the absorptive capacity limits Much will depend on whether the momentum on policy of of the Government. The project aims to extend family plan- recent months is sustained, and more important, how well ning and other services to the 600 health posts, the lowest these policy decisions are implemented. So far, there has level in Senegal's health delivery system. While agreeing been little public support by high-level government offi- on the importance of this extension at some point, critics cials, an element that seems to have been crucial in Kenya. point ot[t that to date, USAID has restricted itself to operat- In addition, the Govemment's implementation capacity ing at the next highest level, which consists of 44 health seems very limited. The HRD project provides a good op- centers, because of concerns that it is already straining the portunity to continue progress, but it is a complex, multi- Government's capacity to adequately supervise and pro- sectoral project that would be difficult to implement in any vide supporting infrastructure, transportation, and IEC country. services. Both points strongly suggest that the Bank consider The Bank's position is that it wishes to take advantage of ways to more intensively and intimately assist the Govern- the Government's current willingness to extend services to ment to implement this project than its typical supervision the periphery; that the urgency of the problem warrants procedures allow. If the urgency of the problem warrants taking more risks than is implied in USAID's approach; taking additional risks in this case, it also warrants a more that it is not displacing grant funds, since other donors do intensive effort to ensure a successful outcome. 59 8. Major Issues for the World Bank Neglect of Population Issues in Typically, the Bank's major investments in a broad range Non-Population Sectors of development sectors are made and justified without ref- erence to population issues. Had these issues been taken While there is evidence that family planning programs into account, their impact on demographic developments can contribute to fertility decline even where development could have been greater. This is certainly the case in educa- levels are low, there is no question that development, espe- tion and health, where there is good evidence that special dially in such sectors as education and health, has a facili- emphases on female education and on maternal and child tating effect in reducing the demand for children and health services have strong demographic effects. It is likely increasing the demand for family planning services. Other to be the case in a number of other sectors, for example, in aspects of development supported by the Bank (e.g., trans- the transport and communications sector, where a greater portation, communications, and electric power) are also im- focus on opening up isolated areas to external influences portant because they increase the circulation of information could be helpful; in agricultural and industrial projects that and new ideas and because they improve the capacity of could be designed to increase the number of jobs available the family planning program. to organize, provide services, to women; and in projects and programs concerned with and spread its message. social security, which might build incentives for smaller For example, there is evidence that in Indonesia a set of families into the rate and benefit schedules. dynamic social and economic programs supported strongly Recommendations for such selective interventions in the by the Bank had a profound effect on the aspirations of development process on behalf of population have been young persons for themselves and their children. These in made by every external advisory group to review the turn led to later age of marriage, a shift in desire for quality Bank's population activities since its inception. Their ne- over quantity of offspring, and a greater interest in family glect is probably due to a combination of factors: skepticism planning. The Bank could have done more to encourage about the argument for selective interventions, doubts progress in Indonesia's lagging health sector. This is, some- about what can be done, failure to establish significant re- what belatedly, a major thrust of the fifth Indonesia popu- search programs to overcome these doubts,1 and staff spe- lation loan, which was just inaugurated. cialization and compartmentalization.2 Underlying many The Bank could also have done more in both India and of these factors may be the fact that population is treated by Bangladesh to encourage greater investments in female ed- the Bank as a sector, with responsibility for it assigned to a ucation and in maternal and child health services, where specific admninistrative unit. Implicitly, this leaves other low levels and slow rates of improvement are inhibiting sectoral units free from any responsibility. However, popu- progress with family planning programs and threatening lation is not a sector any more than economic development future progress. Here too, recent initiatives are encouraging. Kenya is another case in which rapid progress in the ed- 1. For example, by setting up pilot projects to test the effects of different ucation and health sectors, strongly supported by the Bank, interventions. contributed to the recent, unexpected increase in family 2. Even within the population sector, there is compartmentalization. Most planning practice and fertility decline. Much less develop- staff assigned to this field are either demographers, family planning special- Y ists, or generalists responsible for loan processing. Social scientists concerned ment progress in Senegal uncloubtedly is a barrier to fertil- with how development variables influence-and might be modified to influ- ity decline there. ence-fertility decisions are rare. 60 or poverty alleviation. All three are more in the nature of Dialogue, Policy Promotion, and Sector Work strategic objectives that should be the responsibility of all sectors. The Bank is uniquely qualified among donors to encour- A step in the right direction was taken in 1980 when the age political commitment through a continuing dialogue Bank officially started lending for health and established a on population issues with top government leaders. Its fre- department whose aim was to integrate population, health, quent contact with such leaders about the overall lending and nutrition. The situation improved further after the 1987 program and about macro-economic issues, plus the high reorganization with the inclusion of education in a newly- regard placed on its analysis of economic issues, ensures established sector called Population and Human Re- that this is the case. The Bank appears to have had a signif- sources.3 A more important step was the establishment of icant supporting influence in the development of national the Women in Development and Safe Motherhood programs population policies in Indonesia, Kenya, Senegal, and which come closer to doing much of what is required to in- Bangladesh. It has had little influence in India, where a fluence desired family size without special pleading as is strong anti-natalist population policy was already in place, necessary in other fields. Promidsing initiatives can be found and in some Latin American and Middle Eastern countries, inniresigumbrfroecs,4 but conwhere ideological or religious opposition to such policies in an increasing number of projects,4 but considerably more can and should be done in a variety of fields. are still strong. canaythindgsh d be donpe inr ahe varietof fi velds. moreforAs illustrated in the case of Senegal and Kenya, the Bank Everything is in place for the Bank to move more force- can sometimes go beyond discussion by making fully anrd imaginatively in this direction. Health and educa- development of a policy statement a lending condition. In tion projects should be asked to routinely include dvlpeto oiysaenn edn odto.I tionmprojects shouldy be to routinely includnceonageofma ge both cases, the Bank undertook this approach in conjunc- components likely to have an influence on age of marriage tion with encouragement of an important group within the and desired number of children. Projects in other fields government, which felt it could benefit from external sup- shoulc[ be asked to routinely consider locational and gen- port. However, such situations are likely to be rare. der issues in their design. Pilot projects would help speed Moreover, as the recent efforts at population consensus- the accumulation of experience about what works. Eco- building in Africa demonstrate, it is possible to get govern- nomic and sector reports should be required to routinely ments to develop population policy statements that are consider possible demographic impacts of policy develop- based on the felt needs of their populations without using ments, just as they are asked to consider environmental and lending conditionality poverty impacts. Efforts in these directions would be en- In Latin America, Bank representatives have been reluc- couraged by assigning responsibility for promoting, moni- tant to stress these issues because of concerns that it might toring, and exchanging information on these initiatives to a jeopardize their influence in other areas. There, progress specific office.5 has been made in recent years by emphasizing the incorpo- ration of family planning services into the MCH aspects of health projects without explicit demographic justification. This is a desirable approach where necessary, although Bank staff should continue explaining the economic and so- 3. A logical extension of these attempts to integrate various social sectors cal valf slower popu latingr tha every appro- througF, administrative re-grouping is attempts to develop 'human resource cial value of slower population growth at every appropri- projects". The danger, however, is that such projects will be "Christmas ate opportunity. trees:", collections of unintegrated projects in various social sectors which Generally, however, there has been increasing acceptance would inake their implementation extremely difficult. A more practical ap- proach is to introduce population concerns into all relevant social sector of the Bank's efforts to promote the idea that policies to re- projects. For example, all education projects should attempt to increase pro- duce excessively rapid population growth are an important portion and duration of female enrollment, consider changes in curricula part of an effective development program. At a minimum, that might encourage acceptance of a small family norm and promote MCH p consider ways that teachers might be involved in family planning and MCH it has served to encourage and legitimize the worldwide outreach with parents of their pupils, etc. movement to increase government support for explicit 4. Solme of these developments may occur as a natural result of expansion anti-natalist policies and family planning programs. of capacity. In education, for example, the first priority is generally to expand the nurmber of spaces without regard to gender. Even if boys fill the spaces On the other hand, the Bank does not appear to have a first, after a point, further expansion will automatically focus on girls. In la- comparative advantage over other donors when it comes to bor ma:rkets, once unemployment falls to a certain level, the number of jobs influencing the development of programs to implement na- available to women will naturally increase. Such mechanisms cannot be re- lied upon as policy measures, however, because of equity considerations and tional policies. In part, this results from the generally lower the slowv rate at which they operate. staff inputs in the field compared to other donors. It also re- 5. The WID Office is one possibility, but its mandate would have to be sults from a failure at times to undertake the quantity and broadenied to include efforts to alter the perceived benefits and costs of chil- dren. The overlap with its current functions is sizeable, but leaves out a character of sector work in this field that would give Bank number of promising possibilities. staff convincing arguments for alternative approaches. 61 Consider, for example, what appears to be a central element Project-related Issues in all successful programs; that is, programs founded on com- munity-based outreach (as opposed to programs that are pri- Linkages to health and emphasis on hardware. Nearly all marily clinic-based). The esta,blishment of such programs Bank-funded projects with the word "population" in their requires an administrative structure that facilitates two-way titles include sizeable health components. More than two- communication and mobilizes actions initiated at the village thirds of all projects with the word "health" in their titles level-a very difficult task requiring intimate knowledge of lo- include population or family planning activities, and near- cal conditions and personalities. Amongst our sample, only ly all of these projects include sizeable hardware (construc- the Indonesia program has developed a truly effective system tion, fumiture, and equipment) expenditures. Indeed, in of this type, and it did so, not by creating a new administrative the four case study countries with at least two population structure, but by making imaginative use of the existing struc- loans, the average proportion allocated to such "hardware" ture. While it is difficult for donors to influence such a process, elements ranged between 60 and 70 percent. With the ex- USAID was able to do so using a high-caliber group of resident ception of the Indonesia projects, most of this was for representatives and consultants. The Bank helped in impor- expansion of the health infrastructure. Contrary to expecta- tant ways to implement the program after the ideas were de- tions, there is no clear trend toward lower percentages veloped and tested by others. In. India, neither agency has been of hardware in these countries, except in the case of very successful in influencing the program, in large part be- Bangladesh (see Table 3.3 in statistical appendix). cause theGovemment has resisted such advice. However, con- There are three main reasons for these characteristics. sidering the level and location of staffing and the absence of First, it is often determined that the best way to deliver significant and relevant sector work prior to 1988, it must also family planning services is through the health system, be said that, in this instance, the Bank did not try very hard. which in many countries has had to be expanded before it The Bangladesh case is sormething of an exception. Here, can serve this purpose. This determination, while usually the Bank has played a leading role in helping develop details correct, needs to be qualified, as suggested later in this of the program. The difference is largely accounted for by the chapter. Second, pure population programs consist mostly fact that the Govemment encouraged the Bank to play this of software activities, requiring technical inputs and donor role; the Bank devoted unusually intensive staffing to this field presence rather than capital investments. If packaged program, both in Washington and in the field; and other do- by themselves, these projects would tend to be small; too nors provided funds for this purpose. small, in fact, to have much policy leverage. Third, within The Bank could have had more influence on both general the donor community, the Bank has specialized in provid- and operational policy through its sector reports, which are ing construction for sectors like population and health. respected for their generally high quality. Prior to the cre- This represents a de facto division of labor with other do- ation of the PHN Department in 1979, relatively little pop- nors, who are unwilling to provide bricks and mortar. ulation sector work was done, perhaps because of the Construction is also an activity that can utilize Bank expe- primary pressure to develop loan volume and because the rience in other sectors and requires less continuing local Population Projects Department was out of the Bank main- field representation. stream. During the 1980s, the volume of sector work has Support for such hardware elements is especially impor- grown to the point where there is now in many countries- tant in the early years of a program, when infrastructure for Indonesia and India are cases in point-a reasonably solid the delivery system needs to be established. However, it is basis on which to make policy recommendations and de- based on the implicit assumption that other necessary com- velop new projects. This trend needs to be encouraged and ponents will be provided by the government or other do- increasingly focused on technical and operational issues in nors. In Sub-Saharan Africa, where many new population which the Bank has less strength. loans are now being made, this assumption is not justified; The Bank's capacity for pnrviding assistance at a more projects need to provide the whole package of required in- detailed operational level-and hence its influence-would puts for satisfactory completion and operation if they are to also have been greater had a larger share of the time of tech- succeed. Also, there is a tendency in some countries to con- nically competent population and health staff been devoted tinue providing building funds for a somewhat longer pe- to supervision and field operations . In some cases, it would riod than is justified, with the relative neglect of software also have helped if more of these staff had been located in components. Indonesia and India may be cases in point, al- the field.6 though this is an arguable issue. These characteristics-the tendency to package popula- 6. In fact, field representation has increased during the last two years, a tion and health components in one project and the focus on welcome development which should be monitored to determine whether the hardware support-have all too frequently resulted in fam- expected benefits from this placement do in fact occur. ily planning components of these combined projects being 62 given low priority This lack of clear priority appears both These three concems will surface as major issues if the in the project design stage, where programmatic detail trends to increase project size and recurrent cost financing about service provision and demand creation is often limit- continue. It would be wise to deal with them in detail in fu- ed, and in supervision, where sizeable disbursements ture project documents. remain one of the main indicators of satisfactory imple- Neglect of monitoring, evaluation, and research components. mentation. One of the frustrating aspects of this study has been our in- Since free-standing population projects without signifi- ability to determine more clearly whether population cant health components are the exception and likely to re- projects are having the desired impact or not. This inability main so, special attention must be devoted to insuring that is due precisely to the neglect of monitoring, evaluation, the technical design and implementation of the population and research components in these projects. While a moni- components are given the priority they require. This may toring and evaluation component is required in all projects, require some change in evaluation criteria, a topic ad- they are typically not developed in advance and are gener- dressed in the last section of this chapter. ally neglected during supervision. However, without base- Tendency for project size to grow and to fund an increasing line and control group data (acquisition of which should share of recurrent costs. In earlier years, population projects really start before project implementation), proper assess- (and PHN projects generally) grew in size, largely because ment is impossible. of increases in expenditures on hardware. In later years, the While it is obviously very difficult to separate out the im- share accounted for by hardware has stabilized and in some pact of a given project from all the other factors influencing cases declined. However, projects have continued to grow the situation, at a minimum, an assessment can be made of in size because of increased funding of recurrent costs. the proximate effects of individual components. For exam- These trends can be seen in Table 3.3, but become more ev- ple, it may never be possible to know what the impact of an ident when proposed projects, are taken into account. The IEC program is on desired family size. However, a great most dramatic example of this is the Bangladesh projects, deal can be inferred about its usefulness simply by know- which have more than doubled in size (in terms of funding ing how many times a given message was transmitted, levels) every five years, mainly because additional slices of what the size and character of the audience was, and what recurrent program costs have been financed by each subse- the audience thought of the message. Even these simple in- quent project. dicators, however, are not routinely gathered and assessed. These trends are occurring for a number of reasons: Part of the problem is the absence of adequate capacity growth in capacity to effectively administer larger pro- for undertaking such work. Such capacity needs to be local grams; growth in demand for contraceptive supplies and and to be independent of the principal implementing agen- services; concern that other sources of external assistance cies with respect to both the government and the donor. will not fill the gap; and a lowering of the barriers against While some improvements have occurred during the last 15 recurrent cost financing in fields in which they can be justi- years, this capacity remains inadequate in all the case study fied as developmental or capital expenditures. They raise countries.7 An elementary indication is the inadequate sys- three potentially serious concerns. tem for measuring presumed program impact on contracep- tive prevalence and fertility change. This is being remedied * The rapidity of the increase, at least in the case of Bang- mainly, but only partially, through sample surveys; usually ladesh, raises questions about whether the absorptive conducted by outside agencies.8 At present, however, none capacity limits of the governments in this instance are of the countries systematically studies contraceptive accept- being overwhelmed. Signs of waste and inefficiency not- ability, continuation rates, or side effects with adequate sam- ed in field visits suggest that this may be the case. ples. Nor are there adequate studies of the impact of IEC * Population project documents seldom portray any seri- efforts in any of the case study countries. While there is a ous concern about cost-effectiveness or efficiency. This is large volume of so-called operations research, its quality is quite different from education project documents, for often dubious. Judging from results, the efforts that the example, where issues of internal and external efficiency Bank and other donors have expended to assist in the devel- are more commonly discussed. opment of such capacity has been inadequate. * Increasing recurrent cost funding, no matter how justi- fied on developmental grounds, still raises serious con- 7. Such capacity is best developed in India. There, the problem is more that cerns about longer-run financial sustainability of these such capacity has not been effectively tapped and utilized, especially at the programs. Most project documents in this field also lack state level where implementation occurs. 8. These ad hoc surveys provide adequate measurements at the national or a serious consideration of this issue and what to do regional level at periodic intervals, but do not provide reasonable estimates about it. for local areas necessary to manage the program. 63 Another part of the problem is lack of effective demand are obtaining free supplies from donors. In addition, do- for the output of such efforts. It does not matter enough nors prefer to provide commodities, or tied grants to pur- what an evaluation study concludes or whether the moni- chase them, rather than provide (untied) grants to toring and evaluation component is satisfactorily imple- governments to purchase the output of domestic plants. mented to pay close attention to these issues. The situation Population staff in the Bank have also not been interested would improve if the Board required that evaluations of because of the availability of free supplies. However, if the past efforts be used to explaiin and justify projects proposed Bank is increasingly being called upon to fund the purchase to it.9 It would also help if senior management were to make of contraceptive supplies in hard currency, these positions it clear that a proposed project has not been properly de- should be reconsidered. signed-and therefore is not ready for Board presentation- Such a reconsideration is occurring in the Bangladesh until it includes a convincing plan for establishing a moni- program, where the Bank has agreed to fund the building toring and evaluation capacity and then for utilizing that ca- of a contraceptive production facility and provide a loan to pacity to answer questions about the project. the Government for the purchase of its output if a good fea- A larger role in commissioning research from local insti- sibility study for such an operation can be developed. The tutions and providing them with assistance in designing UNFPA is assisting in the attempt to produce such a study. and undertaking the work vvould also help. There are a va- Similar explorations should occur elsewhere. riety of questions of broad--as well as more narrow-rele- Project staffing requirements. These case studies suggest vance that could usefully be studied. In Bangladesh and that population projects, if undertaken properly, are more Kenya, what accounts for the initial success of the program staff-intensive than projects in most other sectors. Proper despite unfavorable culturail, social, and economic condi- design requires attention to processes and procedures that tions? If the answer is something like latent demand, where can be taken a bit more for granted in many other fields. does it come from and how can it be stimulated? Is this a Supervision invariably involves considerable technical as- case in which supply creates its own demand or will supply- sistance working with field-level problems-a time-con- oriented programs run into diminishing returns, as we sus- suming task if done correctly. Given this, one should expect pect, long before a satisfactory fertility rate has been to find that staff coefficients devoted to these projects are achieved? Can a similar phenomenon be counted on to start greater than for other sectors. the fertiiity transition in Senegal? Would a pilot project sim- However, as Figure 8.1 (which covers all PHN projects) ilar to Matlab be useful there to demonstrate the existence of demonstrates, this was the case in the early years, but over latent demand, and what it takes to reach it? In the past, time, these coefficients have declined to the point at which such efforts have been insufjficient to generate much interest they are now no different from the Bank average. It is not in the monftoring, evaluation, and research components. surprising that these coefficients have declined from the Lack of funding of contraceptive production facilities. If the early, formative days of this program, but it is surprising rise in demand for contraceptives is creating a financial they have fallen all the way to the Bank average. If the pic- problem that donors shouli consider filling, the scope for ture emerging from these case studies is correct, this sug- development of local production facilities should also be in- gests that population (and health and nutrition) projects are creasing. To date, the Bank has funded only one such facility, being forced to fit into a Procrustean bed determined by the a plant constructed in Indonesia in 1978-79. What has inhib- needs of other sectors. The consequence must be poorer- ited more activity in this field? For a time, concerns that de- quality projects and weaker implementation than could mand was insufficient for a plant of an economic scale may (and should) be the case. have explained lack of interest. In the larger countries how- The high staff coefficients of earlier years resulted ever-Indonesia, India, Bangladesh, Brazil, and Mexico- from the earmarking of staff time specifically for popula- this can no longer be the case. tion-related activities.10 This approach may no longer be Interviews within the Bank suggest the following rea- appro ' ~~~~~~~~appropriate in a more decentralized Bank. Nor would it be sons. Industry staff are not interested because the projects appropriate to mandate specific coefficients for these would be too small and because, at least in poorer coun- projects, since their requirements differ greatly. At a mini- tries, production would have to be subsidized or purchased mum, however, managers should understand that this sec- by a donor. Governments are not interested so long as they tor is likely to require higher coefficients and be willing to - ~~~~~~~~~~~~~~provide them when a good case can be made. 9. Virtually any probing question would require a review of past experi- ence. For example: why do you believe this project will take five years when the last two took seven; what evidence do you have the proposed IEC ap- 10. It may also have been the result of the small size of the program and its proach will work; why are you not proposing to focus on on-the-job training lack of integration with related sectors which resulted in staff inputs being in this project when the Audit considered it an error of omission in the pre- used less efficiently than is possible today. More recent declines may be the vious project? result of growth in lending proceeding ahead of recruitment. 64 Figure 8.1: Staff-Weeks per Project, PHN vs. All Sectors Lending Supervision 320 40 300- i1 3 \i5 N 280 Aa 35 Suptvii 7260 6 T 7 a. 220 2 0. 2 200 . .\s S I ~~~~~~~~~~~~~~~~~15 160 1 1........... ~"140 NI 120 ~~~~~~~~~~~~~75 7677 7879 80 8182 8384 858687 8889 90 100 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 .....PHN Allisectors Fiscal Year of Board Approval Source: World Bank data. Source: World Bank data. Donor Coordination and Involvement vertical program because they would get lost if they also of Non-Govemmental Agencies became just one more component of primary health care or of a MCH program, and because ministries of health were The division of labor that implicitly or explicitly devel- generally weak agencies. Conventional wisdom today is ops between the agencies assisting a given country is an im- nearly the reverse. The current thinking is that family plan- portant determinant of the effectiveness of the assistance ning should be integrated and delivered with MCH servic- program in this field. The case studies demonstrate that a es, for the following reasons. (1) Family planning is less well-rounded assistance package should include software likely to get lost in an MCH program today because it is far and technical assistance as well as hardware, grants as well better accepted as a legitimate intervention for both health as loans, detailed knowledge of local conditions, intimate and demographic reasons. (2) The time in a woman's life daily contact with counterparts, inputs into a variety of sec- when MCH is most pertinent is also the time when adop- tors, and periodic access to high-level decision-makers. No tion or resumption of contraceptive use is highly relevant. donor is in a position to provide all these elements; indeed, (3) Reducing child mortality is a facilitating condition for some may be best provided by NGOs. This is why the decreasing the demand for more children; similarly, in- mechanisms that have evolved for donor coordination and creased spacing between children is a facilitating condition involvenment of NGOs in Bangladesh are so attractive. In- for improving maternal and child health. (4) Good quality deed, the next step in this evolution may be the develop- family planning services require medical backup and su- ment of projects that are thought of as consortium projects, pervision. (5) Many countries which will not accept donor not Bank or UNFPA projects."1 This might be a useful direc- support for what they perceive to be population control tion in which to move in a number of African countries programs will accept support for family health and welfare where a wide variety of inputs is needed and the govern- programs with family planning components. ments are not in a position to provide effective leadership In all the case study countries, the Bank could do more to for the donor community. foster stronger MCH services, both for health and for fami- ly planning purposes. Fortunately, the Bank is now moving Country Organization of Population in this direction. The recently approved fifth population and Family Planning Activities loan in Indonesia, the Child Survival and Safe Motherhood Project in India, and the Fourth Population and Health In the 1960s, conventional wisdom argued that family Project in Bangladesh are cases in point. In Africa, seven planning services should be delivered through their own countries have initiated joint population/health loans in re- cent years. Additional countries in both Latin America and 11. There will, of course, still be the need for one agency to take the lead, Africa have family planning components in MCH pro- but it need :not be the largest contributor. grams supported by loans to the health sector. 65 While it is clear that healt:h ministries have an important for the country, and budget and plan accordingly. (2) If dur- role to play in delivering family planning and MCH servic- ing that period it is reasonable to have, for example, four es, there are a number of reasons for not giving them sole or projects, these should be thought of as necessary steps even principal responsibility for either family planning or along the way whose goals and monitorable targets are for other aspects of what should constitute a full-blown specified in terms of these steps, and not exclusively (as so population program. (1) Exclusive focus on the link be- often has been the case in the past) in terms of the long-rn tween MCH and FP will inevitably neglect young couples objectives of the program.12 To do otherwise encourages and men. (2) Ministries of htealth are generally among the frustration and disillusionment when in fact considerable weaker ministries. Moreover, they are often led by mem- progress is being made. (3) The program will undoubtedly bers of the medical profession who are more interested in require more than these projects alone to succeed, and the curative than in preventive care or public health programs. Bank may have to consider assisting the govemment with (3) As a consequence, unless special care is taken, this ap- some of these elements. Such elements may include sector proach tends to lead to neglect of other delivery channels, work, pilot projects and research, training programs, and for example, NGOs, private medical facilities, pharmacies, funding of recurrent cost items such as contraceptives that social marketing arrangements, and family planning pro- do not fit neatly into the time frame and scope of these grams in (or which might be developed in) other agencies, projects. (4) Progress toward achieving the program goals such as the military and large corporations. (4) For similar probably has to be monitored in different ways from reasons, IEC programs run through ministries of health are progress in achieving the goals of individual projects. Spe- often weak. (5) When structured this way, other minis- cial provision should be made for this. tries-for example, education, information, planning, agri- culture, and religion-tend not to participate in the The Extent and Allocation of Bank Resources program. All five of these problems have arisen in both In- dia and Bangladesh, where the ministries of health and Since 1980, 1 percent of the Bank's loan portfolio has family welfare have primary responsibility for their popu- been devoted to population projects (defined as projects lation programs. with population in their title)13. This amount constitutes The effort to correct these shortcomings has sometimes about 10 percent of the funds provided by bilateral, multi- led the Bank and other donors to support the establishment lateral, and private donors to the developing world for of high-level govemmental coordinating and facilitating population activities. Four countries (India, Indonesia, agencies along the lines of the highly successful BKKBN in Bangladesh, and Kenya) have received 79 percent of these Indonesia. Unfortunately, this is the only truly successful funds. Only 10 countries, constituting 32 percent of the de- example of such an agency It has worked there because of veloping world's population, have had two or more Bank- strong support from central authorities, astute and charis- assisted population projects. In 1988, the Bank was financ- matic leadership, and highly competent management- ing ongoing projects in 17 countries. In contrast, UNFPA conditions that are infrequently found together. In the end, and USAID are active in a much larger group of countries. decisions about administrative arrangements for various aspectsiof abcountry'sipoation prograngements wl havtou In 1988, the UNFPA had activities or projects that provided aspects of a country's population program will have to be more than US$1.0 million dollars per year to each of 31 based on the special circumstances of individual countries. countries, and USAID had the same in 39 countries. These data have led Bank critics to argue that the Bank has devoted too little effort to this field. Projections of re- Kenya exemplifies the value of patient building of source requirements based on numbers of couples project- consensu, ied to be using contraceptives in the future have been used cosnss inomain a. inrsnitr ove alos to argue that the Bank needs to do much more in the fu- 20 years before significant reproductive change becomes evident. A privately-run pilot project (Chogoria) in Kenya, which demonstrated the presence of latent demand and the 12. In countries just beginning to build a population program, a necessary first step is to create a delivery system and the institutional capacity to oper- fact that it can be turned into active demand by a good ser- ate it. This in itself is likely to be sufficiently difficult and time-consuming vice system, shows what can be done in the interim to de- such that several projects could usefully be devoted to this exclusively. The termine whether one is on the right track. Bank has helped effectively with the physical aspects of delivery systems, but has not been so effective with institutional aspects. (In earlier projects, the Several implications for the Bank and other donors fol- Banl sometimes obscured the situation by stating project objectives in terms low from this point. (1) If best judgement about a country is of fertility reduction, the implicit assumption being that everything else was that it will take (for example) 20 years of effort before sig- in place or would be put in place by others.) 13. This percentage is understated to the extent that population compo- nificant fertility reductions and health improvements can nents of health projects are left out, but it is overstated to the extent that be observed, the Bank should think in terms of a program health components of population projects are not excluded. 66 ture.14 On the basis of the case studies and general informa- grant funds, which could have been used flexibly, as was tion about Bank population activities in other countries, can the case in Bangladesh. Nor is there evidence that con- these judgements be supported? straints on progress resulted from lack of local funds or op- Before answering this question, it is important to note erating budgets, largely because donors generally provided that the Bank's role cannot be adequately judged by consid- funds for these purposes when serious shortages emerged. ering the extent of funding it has provided, at least not in This could become a problem in the future, however, if do- the aggregate. There are cases in which the Bank has had nors do not continue filling what may be a growing gap be- significant influence without providing funds directly for tween resource needs and govemments' ability to fund. population (Senegal). There are also cases in which the Additional funds could have been-and could still be- Bank has had little influence despite having invested con- used by these countries for demand-creating activities in siderable sums of money (India). Money is an especially non-population sectors. Also, there is some evidence that misleading indicator in the early stages of program devel- additional staff inputs, both at headquarters and in the opment. This aspect is relevant to many cases in Africa, for field, could have been usefully employed in several of the example.'5 A better proxy, at least for level of effort, is staff case study countries. Such inputs would have allowed and inputs, though this, too, is not a good indicator of influence. encouraged more effective supervision and the develop- The best approach is to derive judgements from reviews of ment of more sector and analytical work. However, there is individual country experiences, as this study tries to do. a limit on the extent to which external staff and technical as- Strictly speaking, however, the insights obtained thereby sistants can substitute for local capacity, especially where are limited to the case studies selected. are limllted o the ase stuies seected.governments are unwilling or unable to move faster. Moreover, the adequacy of the Bank's role cannot be as- g rents are anwillin oruabet move faster. sessed independently of what other donors are doing. The Of grae imotnetavnrasdfnigion reale q iondepenisnto whe athe total contribuonno of all doinor Ttries with existing Bank-supported projects is a shift in the is adequatein addiswheti the for of loans composition of activities, away from large-scale construc- IS adequate. In addition, Bank funds are in the form of loans tinpoet n oad mr rgamtcapoc or credits for which there is a limited market, whereas near- tion projects and towards a more programmatic approach ly all of other donors' funds are in the form of grants, often that includes the help of sectors in addition to population for program activities as distinct from projects, for which and health. It would also be useful to consider shifting the demand is much greater. composition of Bank and other donor funds between coun- Within case study countries, we have found no evidence tries, as recommended in the recent PHR paper. This would that progress has been constrained by lack of external funds serve to concentrate Bank financing on better-established for family planning programs per se. Additional Bank assis- programs and grant funds on programs in their formative tance in this field would not have been useful. It would in- stages. stead have overwhelmed these countries' already hard- If there is any scope for expanding the Bank's population pressed. absorptive capacity. If any additional external and related health programs, it is to countries without cur- funds at all were needed, the most useful would have been rently active Bank-supported population projects. While not all these countries need Bank inputs, either because they do not have a population problem or because other 14. The Population Crisis Committee argues (in The World Bank's Role in donors are providing adequate inputs, there is surely scope Global Population Efforts: An Agenda for Effective Action, 1989) that Bank activ- for useful expansion. However, to make this case adequate- ities in the population field have been constrained by a variety of factors. It ly requires an assessment of needs that lies beyond the recommends changing these so that the Bank can increase its level of lending from the 1989-90 average of $200-225 million to $1 billion a year. At this scope of this study. point, it vwould be contributing 13% of estimated global needs of $7.5 billion Thus, while expansion of funding for hardware in coun- a year. The U.S. Senate Committee on Appropriations (Report 101-519, Octo- ber 10, 1990, p. 47) expressed concem last year about the 'minor role" the tries in which the Bank has been active for many years is no Bank has played in international family planning assistance and asserted longer needed-and in some cases, should be reduced- that, "It is incumbent on the World Bank, with its enormous resources, to additional funds are needed in three areas: quite possibly make this a first priority, in dollars, as well as rhetoric." A PHR paper dated March 8, 1991, Strengthening the Bank's Population Work in the Nineties, argues for new countries; definitely for expansion of population- that because of successful efforts to increase demand and because of stagna- related activities in non-population sectors; and for rele- tion and planned declines in levels of bilateral assistance (especially USAID's vant software elements of population and population-relat- planned reduction in contraceptive funding), "the Bank will have to play a larger role in both mobilizing additional grant fund and in direct financing ed projects. These changes would require some net than it has in the past." (p. 17). increases in funding, though probably not as sizeable as 15. An especially good example is the African population initiative, a low- critics might like. Of greater significance, they imply some budget but staff- intensive activity that has been highly successful in break- ing the logjam of inertia and resistance to population activities in Africa, and changes in direction which could substantially improve the which now seems to be resulting in proposals for projects-for other donors, effectiveness of the Bank's efforts in this field. as well as for the Bank. 67 Implications for Evaluating and Staffing Bank inputs rather than outputs, and more important, on the spe- Population Activities cific inputs included in the project, rather than on whatever is needed to advance the sector goals. This review suggests that the population field is unique It is especially important in the population sector to cor- in a number of respects. A larger-than-usual proportion of rect these biases. In evaluating population projects, more non-project activities, some of which may be at least as im- weight should be given to proximate output measures, fo- portant as individual projects, seem to be required. Such ac- cusing on those believed to be the most important for tivities include advocacy efforts (such as the African achieving the sector goals-even if the associated inputs do Population Agenda Initiative), technical assistance for pol- not require the most resources. In evaluating Bank efforts in icy development, analysis of current population issues; and a given country, more attention should be paid to non- active cooperation with other donors. Within population project activities and efforts in other sectors. In evaluating projects, software components are often more important in Bank efforts not specifically related to a given country, reaching program goals than they are in other sectors. Such more weight should be given to the appropriateness of the activities and components are time- consuming and staff- content, relative to needs and what other donors are doing, intensive, but do not require large financial resources. Fur- than to the extent of the resource transfer. Overall, more at- thermore, population activities and projects tend to involve tention needs to be devoted to possible errors of omission- longer lead times and looser connections between initial in- what could have been done that would have made a larger puts and ultimate impacts than do projects in other sectors. difference-than to errors of commission. This is a sector that is still in its formative stages, where ex- These characteristics also have implications for staffing, perimentation and the search for new approaches remains especially important. While it: is convenient to refer to the moto hc aebe one u ndfeetcnet broeciadly ragmpopuantionW actiit:ies asnveaisector,r it i a above. If the Bank wishes to improve its effectiveness in this true that selective interventions from non-population sec- area, it must allow and encourage the use of more technical truetha seectve iterentonsfro nonpoplaton ec- staff for both non-pwroject and project work and should tors could be equally important at times in the achievement rstffr boct nonreojet and po e wokad shoul of population goals. consider locating more of this staff in the field in special These characteristics have a number of implications for circumstances. This is especially important in countries evaluation. The tvpical approach is to focus on projects without adequate indigenous capacity or which are lacking rather than non-project activities; to ask whether the project assistance from other donors. That the Bank has done as did what the SAR and President's Report said it would do; well as it has so far is a tribute to extra efforts expended by and within projects, to focus on the more expensive and a small core of dedicated staff-not a situation that should quantifiable inputs. While this is important for accountabil- be relied on for the long run, regular operation of an ity purposes, it leads to an excessive focus on projects, on important program. 68 Statistical Appendix Table 1.1: Demographic and Socioeconomic Indicators, Selected Countries and Years A. Basic Indicators Life Illiteracy Expectancy Rate Adult Urban Population GNP per Capita at Birth Female Mortality per 1000 Population Mid-1'88 1988 1965-88 Under 5 Maternal % (mnil) (US$) (% per yr) 1988 1985 1985 1985 1985 Indonesia 174.3 440 4.3 61 35 128.3 8.0 22.2 India 815.6 340 1.8 58 71 76.2 5.0 23.4 Bangladesh 108.9 170 0.4 51 78 175.7 6.0 10.4 Brazil 144.4 2160 3.6 65 24 127.3 1.5 67.5 Colombia 31.7 1180 2.4 68 13 112.9 1.3 64.2 Mexico 83.7 1760 2.3 69 12 59.3 0.9 66.4 Kenya 22.4 370 1.9 59 51 47.4 5.1 16.1 Senegal 7.0 650 -0.8 48 81 98.1 5.3 34.9 B. rime Series 1960) 1965 1970 1975 1980 1985 1988 1989 Total Fertility Rate Indonesia 5.4 5.5 5.5 5.0 4.3 3.7 India 6.6 6.2 5.8 5.3 5.0 4.6 4.0 Bangladesh 7.0 6.8 7.2 6.6 6.1 5.7 5.0 4.9 Brazil 6.2 5.6 4.9 4.4 4.0 3.6 3.3 Colombia 6.8 6.3 5.3 4.0 3.5 3.3 3.0 Mexico 6.8 6.7 6.5 5.5 4.5 3.8 3.4 Kenya 7.9 8.0 8.0 8.0 7.9 7.6 6.7 Senegal 6.4 6.4 6.5 6.6 6.6 6.6 Female Population Using ContracePtives (%) Indonesia 11.0 48.0 India 10.0 19.0 22.0 39.0 Bangladesh 3.0 7.7 12.0 25.0 32.0 Brazil 32.0 56.0 66.0 39.0 Colombia 33.0 49.0 65.0 Mexico 12.0 53.0 Kenya 1.0 7.0 7.0 27.0 Senegal 0.5 2.4 8.0 Infant (0-1) Mortality Rate (per 1000) Indonesia 137.8 127.6 118.0 108.6 99.0 80.6 India 163.4 149.8 139.0 129.6 114.0 103.4 93.0 Bangladesh 154.8 144.0 140.0 138.2 131.6 122.6 118.0 Brazil 114.2 103.6 94.6 83.8 74.2 66.2 Colombia 103.0 96.0 79.2 66.1 58.2 48.9 Mexico 90.8 81.8 73.0 63.6 55.8 49.4 Kenya 122.8 112.0 102.0 92.0 83.2 75.2 Senegal 179.2 171.2 164.4 157.2 146.8 133.6 70 Table 1.1 (continued) B.1rime Series 1960 1965 1970 1975 1980 1985 1988 1989 Crude Birth Rate (per 1000) Indonesia 43.9 42.7 41.9 38.4 34.5 30.5 India 47.6 44.8 37.8 36.0 33.7 32.0 30.5 Bangladesh 46.7 47.2 48.1 47.7 45.8 42.3 40.0 Brazil 42.6 38.7 34.8 32.7 31.2 29.3 Colombia 46.8 44.7 36.9 31.3 29.7 26.9 Mexico 45.7 44.9 43.4 37.7 32.8 29.9 Kenya 52.8 52.4 52.6 53.3 53.8 52.8 Senegal 47.4 46.9 46.6 46.4 46.1 45.7 Crude Death Rate (per 1000) Indonesia 22.6 20.2 18.1 15.4 12.1 9.8 India 23.5 20.3 16.5 15.1 12.6 11.2 10.2 Bangladesh 22.4 21.4 20.9 19.7 18.1 16.2 Brazil 12.8 11.4 10.2 9.2 8.6 8.1 Colombia 16.1 14.0 10.5 9.0 8.0 7.1 Mexico 12.0 10.6 9.4 7.9 6.7 6.0 Kenya 22.2 20.1 18.1 16.2 14.4 12.1 Senegal 24.1 22.8 21.7 20.8 19.8 18.4 Population Growth Rate (%) Indonesia 2.2 2.4 2.3 2.2 2.0 1.9 India 2.3 2.1 2.1 2.1 2.1 2.0 Bangladesh 2.6 2.7 2.7 2.8 2.6 2.5 Brazil 2.8 2.5 2.4 2.3 2.1 2.0 Colombia 3.3 2.4 1.9 1.9 1.9 1.8 Mexico 3.2 3.3 3.0 2.4 2.2 2.1 Kenya 3.2 3.4 3.7 3.9 4.1 4.1 Senegal 2.4 2.3 2.5 2.8 2.8 2.9 Physicians per 10,000 Population Indonesia 0.37 0.37 India 2.05 2.04 3.71 Bangladesh 1.18 1.18 1.49 Brazil 4.93 6.25 Colombia 4.45 Mexico 6.44 6.67 Kenya 1.25 1.27 0.99 Senegal 0.60 0.60 Hospital 13eds per 10,000 Population Indonesia 6.55 8.18 5.51 India 6.04 5.88 7.70 7.69 Bangladesh 1.55 2.11 2.78 Brazil 36.97 49.99 Colombia 22.25 17.18 Mexico 11.86 12.51 Kenya 12.64 12.50 16.67 Senegal 12.23 12.50 71 Table 1.1 (continued) B. Time Series 1960 1965 1970 1975 1980 1985 1988 1989 Health Expenditures (as % of GDP) Indonesia 1.4 India 2.3 1.9 2.0 Bangladesh 1.8 Brazil 3.8 4.2 Colombia 4.1 4.2 Mexico 2.9 2.9 Kenya 3.7 1.4 0.2 Senegal 1.7 1.6 Education Expenditures (as % of GDP) Indonesia 2.3 India 2.6 2.8 2.8 Bangladesh 1.1 Brazil 2.6 3.6 Colombia 6.4 3.0 Mexico 2.7 2.9 Kenya 7.4 6.9 1.2 Senegal 2.8 0.3 Gross Enrollment Ratio, Primary, Female Indonesia 73.0 78.0 100.0 115.0 India 56.0 62.0 67.0 Bangladesh 35.0 51.0 46.0 50.0 Brazil 87.0 97.0 Colombia 110.0 120.0 130.0 119.0 Mexico 101.0 106.0 119.0 114.0 Kenya 48.0 87.0 104.0 91.0 Senegal 32.0 34.0 36.0 45.0 Gross Enrollment Ratio, Secondary, Female Indonesia 11.0 15.0 23.0 India 15.0 16.0 22.0 Bangladesh 3.0 11.0 9.0 10.0 Brazil 26.0 28.0 36.0 40.6 Colombia 24.0 39.0 45.0 51.0 Mexico 17.0 28.0 45.0 54.0 Kenya 5.0 9.0 15.0 15.0 Senegal 6.0 7.0 9.0 Female Labor Participation Rate Indonesia 22.7 22.8 23.2 23.8 24.3 24.4 India 24.8 23.2 21.7 21.0 20.5 20.4 Bangladesh 3.4 3.5 3.7 4.0 4.3 4.4 Brazil 14.3 16.9 19.6 19.9 20.0 20.0 Colombia 12.7 13.3 13.9 14.3 14.4 14.4 Mexico 10.1 13.5 17.3 17.9 18.4 18.6 Kenya 36.9 36.0 35.1 33.2 31.7 31.3 Senegal 39.0 38.5 38.1 35.9 34.1 33.7 Sources: World Bank data modified by information frnm country sources where available. 72 Table 2.1: Donor Expenditures for Population Programs in Selected Countries, 1982-89 Country 1982 1983 1984 1985 1986 1987 1988 1989 Average Indonesia (US$ 1000) 27,307 15,515 15,450 8,526 18,894 23,803 7,792 13632 16,365 Bilateral (%) 80 67 64 18 68 47 23 74 55 Mul]tilateral (%) 10 15 15 48 11 13 6 0 15 Private (%) 10 18 21 34 21 40 71 26 30 India (US$ 1000) 20,113 43,704 45,575 43,062 23,415 23,159 22,365 32,401 31,724 Bilateral (%) 30 60 39 61 36 53 19 52 44 Multilateral (%) 47 31 51 28 40 31 50 31 39 Private (%) 23 9 10 11 24 16 31 18 18 Bangladesh (US$ 1000) 36,148 35,168 40,166 51,518 43,311 46,302 56,176 49,923 44,839 Bilateral (%) 80 75 76 80 72 66 67 78 74 Multilateral (%) 12 12 13 8 7 5 5 9 9 Private (N) 8 13 11 12 21 29 28 13 17 Brazil (US$ 1000) 6,987 4,644 7,379 7,199 7,666 6,728 15,770 11,973 8,543 Bilateral (%) 5 0 0 0 0 3 0 6 2 Multilateral (%) 4 4 5 28 100 1 39 26 26 Private (%) 91 96 95 72 0 96 61 68 72 Colombia (US$ 1000) 5,854 3,835 4,564 4,367 3,571 3,502 7,687 6,498 4,985 Bilateral (%) 0 0 2 3 14 24 20 2 8 Multilateral (%) 10 33 25 28 10 6 5 17 17 Pri vate(%) 90 67 72 70 76 70 75 81 75 Mexico (US$ 1000) 11,168 4,520 8,058 9,551 5,555 6,502 11,100 10,629 8,385 Bilateral (%) NA 1 3 9 0 3 5 13 4 Multilateral (%) 15 39 34 32 18 i5 21 17 24 Private (%) 85 60 63 60 82 82 74 70 72 Kenya (US$ 1000) 6,957 5,675 7,662 9,059 4,908 9,536 18,554 20,277 10,329 Bilateral (%) 68 60 67 68 33 39 35 56 53 Multilateral (%) 8 8 4 5 15 11 13 16 10 Private (%) 23 33 29 26 52 50 52 27 37 Senegai (US$ 1000) 729 968 2,045 11,142 3,274 6,752 2,350 4,284 3,943 Bilateral (%) 0 0 33 85 54 66 14 70 40 Multilateral (%) 67 65 38 8 14 9 23 4 29 Private (%) 33 35 29 7 32 25 63 25 31 Note: Figures are based on commitments. The result is that they shift considerably in years when large commitments are made, for example, in 1986, when the Bank's third population loan was committed to Bangladesh. Source: UNFPA, Global Population Assistance Report 1982-1989, New York City 73 X4 Table 3.1: Summary Data for Population and Population-Related Projects in Case Study Countries, FY7O-FY90 Total Terms Project Project Ratings3 Approval Completion Bank (IBRDI Cost Financing Plan1 (%) Type of Institutional Country/Project Name Year Year Funding IDA) (US$m) Bank Co-Financing Local Report2 Overall Sustainability Development Indonesia Population FY72 FY80 13.2 IDA 33.0 40.0 40.0 20.0 AUDIT S Second Population FY77 FY82 24.5 IBRD 60.0 40.8 59.2 AUDIT S Third Population FY80 FY85 35.0 IBRD 72.6 48.2 51.8 PCR S Fourth Population FY85 4A.0 IBRD 94.4 48.7 5;.3 FP & Safe Motherhood Project FY91 104.0 IBRD 148.4 70.0 30.0 India Population FY72 FY80 21.2 IDA 31.8 66.7 33.3 AUDIT S Second Population FY80 FY89 46.0 IDA 96.0 47.9 52.1 AUDIT U Third Population FY84 70.0 IDA 123.5 56.7 43.3 West Bengal 4th Population FY86 51.5 IDA 89.0 57.9 42.1 Fifth (Bombay & Madras) Population FY88 57.0 IDA 78.2 72.8 27.2 Sixth Population FY89 124.6 IBRD 182.0 68.5 31.5 Population VII FY90 96.7 IDA 141.5 68.3 31.7 Bangladesh Population FY75 FY83 15.0 IDA 45.7 32.8 55.5 11.6 AUDIT U Population & Family Health FY79 FY87 32.0 IDA 110.0 29.1 60.9 10.0 PCR S Likely Partial Third Population & Family Health FY86 78.0 IDA 213.9 36.5 46.5 17.0 Fourth Population & Health FY91 180.0 IDA 601.4 29.9 42.7 27.4 Kenya Population FY74 FY80 12.0 IDA 15.4 78.0 22.0 AUDIT S Integrated Rural Health & FP FY82 23.0 IDA 61.1 37.7 40.6 21.7 Third Population FY88 12.2 IDA 28.3 43.1 47.3 9.5 Population IV FY90 35.0 IDA 41.3 84.7 5.3 9.9 1. The following co-financing agencies participated in the above projects: Bangladesh I - ADAB, CIDA, Germany, NORAD, SIDA & the United Kingdom; Bangladesh 11 - ADAB, CIDA, Germany, the Netherlands, NORAD, SIDA & the United Kingdom; Bangladesh III - ADAB, CIDA, Germany, the Netherlands, NORAD & the Uniited Kingdom; India I - SIDA; Indonesia I - UNFPA; Kenya II - DANIDA, Germany, NORAD, SIDA, UNFPA & USAID; KENYA III - NORAD & the United Kingdom; Kenya IV - USAID. 2. All completed projects have PCRs, but not all completed projects have Audits. 3. The project ratings are drawn from Audits, where available; otherwise they are drawn from the PCR. S = Satisfactory; U = Unsatisfactory Table 3.2: Project Costs by Component in Population and Population-Related Projects for Case Study Countries Demand Project Admin. Total Foreign Supply of Generation Manpower Institutional and Not Project Exchange Project Services' and IEC Development Development Preparation Allocated Costs Component (percent) (percent) (percent) (percent) (percent) (percent) (US$mil) (percent) Indonesia Population 30.1 14.9 23.0 12.2 1.6 18.2 33.00 38.8 Second Population 35.9 15.6 33.6 14.9 - 0.0 44.30 18.3 Third Population 26.2 21.9 24.0 28.0 - 0.0 72.60 25.1 Fourth Population 37.4 20.0 13.1 11.8 - 17.7 94.40 43.9 India Population 79.3 - - 4.2 0.5 16.0 31.80 5.0 Second Population 64.5 6.3 5.3 0.8 1.3 21.9 96.00 17.9 Third Population 65.3 5.2 - 2.2 6.2 21.2 123.50 10.0 West Bengal 4th Population 53.9 5.6 10.6 1.3 4.7 23.9 89.90 8.0 Fifth (Bombay & Madras) Population 58.9 - 4.2 14.2 - 22.8 78.21 5.4 Sixth Population 27.3 0.0 9.7 37.0 - 25.9 182.00 7.2 Population VII 33.9 24.1 28.1 10.4 - 3.5 141.5 7.9 Bangladesh Population 6.0 12.7 39.1 3.9 0.6 37.7 45.70 52.6 Population & Family Health 65.6 10.5 20.8 1.9 1.2 0.0 110.00 33.1 Third Population & Family Health 81.7 4.2 - 1.3 0.8 12.0 213.85 31.5 Kenya Population 70.0 0.0 23.1 - 0.1 6.8 15.37 31.9 Integrated Rural Health & FP 40.6 19.8 16.8 - - 22.8 61.05 35.2 Third Population 60.1 8.6 0.0 14.7 - 16.5 28.34 32.5 Population IV 43.5 17.8 0.6 27.6 - 10.5 41.27 62.3 1. The supply of services includes the provision of health and family planning services (which all projects provided), as well as nutrition activities (provided in India I and Indonesia IV), and disease control or sanitation activities (provided in India III and India IV). Table 3.3: Appraisal Project Costs and Bank Financing, by Category of Expenditure, for Population Projects in Case Study Countries Total Project Civil Furniture TA & Operating Special Bank Civil Furniture TA and Operating Special Project Cost Works Equipment Training Costs Activities Allocated Financing Works Equipment Training Costs Activities Unallocated (US$m) (%) (%) (%) (%) (%) (%) (US$m) (%) (%) (%) (%) (%) (%) Indonesia Population 33.00 27.6 14.8 16.7 22.7 0.0 18.2 13.20 34.5 18.6 20.8 9.1 0.0 17.0 Second Population 60.00 16.7 10.5 8.7 38.0 0.0 26.2 24.50 31.4 30.2 24.9 0.0 0.0 13.5 Third Population 72.60 46.4 9.9 15.4 0.3 0.0 28.0 35.00 48.0 18.3 25.1 0.0 0.0 8.6 Fourth Population 94.40 28.3 39.6 32.1 - 0.0 0.0 46.00 50.4 23.9 25.7 0.0 0.0 0.0 India Population 31.80 39.6 14.7 8.4 14.6 6.5 16.1 21.20 39.6 14.7 8.4 14.6 6.5 16.1 Second Population 96.00 31.4 8.5 0.0 36.7 1.6 21.9 46.00 37.0 11.7 0.0 42.6 2.2 6.5 Third Population 123.50 43.6 11.9 0.0 23.2 0.0 21.2 70.00 57.1 18.6 0.0 18.6 0.0 5.7 West Bengal 4th Population 89.00 44.3 6.9 0.0 24.7 0.0 24.2 51.50 58.1 8.3 0.0 22.3 0.0 11.3 Fifth (Bombay & Madras) Population 78.21 10.8 25.2 6.1 57.9 0.0 0.0 56.96 12.6 25.6 8.3 53.4 0.0 0.0 Sixth Population 182.00 42.8 21.2 11.8 24.2 0.0 0.0 124.60 46.9 23.2 12.9 17.0 0.0 0.0 Population VII 141.50 37.3 16.7 26.0 11.0 9.0 0.0 96.68 42.2 17.1 23.5 9.2 7.9 0.0 Bangladesh Population 45.71 30.2 13.6 1.5 16.1 0.8 37.8 15.00 27.9 17.6 2.9 12.2 2.0 37.5 Population & Family Health 110.00 42.1 15.1 2.0 22.1 1.8 16.8 31.99 58.1 17.2 0.3 2.3 4.6 17.5 Third Population & Family Health 213.85 21.1 28.0 9.4 24.1 17.4 0.0 78.00 52.8 40.1 0.7 0.0 6.4 0.0 Kenya Population 15.38 89.3 8.9 1.8 - 0.0 0.0 12.00 66.7 5.4 1.7 0.0 0.0 26.3 Integrated Rural Health & FP 61.05 30.2 20.5 1.5 17.6 20.9 9.3 23.00 25.8 12.7 2.7 35.4 20.9 2.5 Third Population 28.30 19.8 20.5 24.7 18.7 0.0 16.3 12.20 20.5 34.4 30.3 0.0 0.0 14.8 Population IV 41.30 21.5 42.6 19.1 6.1 0.0 10.7 35.00 22.6 45.7 20.3 0.0 0.0 11.4 Table 41.1: Bank Lending by Sector, Case Study Countries, 1970-90 1970-79 1980-90 1970-90 US$milion Percent LIS$million Percent US$mfflion Percent Indonesia Education 186 6 1,349 11 1,536 10 Pop., Health & Nutrition 51 2 224 2 275 2 (Population) (38) (1) (81) (1) (119) (1) Agriculture & Rural Dev. 1,087 34 2,706 22 3,793 24 Industry 247 8 237 2 483 3 Energy & Power 632 20 2,758 22 3,390 22 Telecommunications 13 0 365 3 377 2 Transportation 519 16 1,470 12 1,989 13 Urban Dev., Water, Sewage 182 6 1,115 9 1,297 8 Other' 267 8 2,190 18 2,457 16 Total 3183 100 12,413 100 15,596 100 India Education 12 0 540 2 552 2 Pop., Health & Nutrition 21 0 573 2 594 2 (Population) (21) (0.3) (349) (1) (416) (1) Agriculture & Rural Dev. 2,385 33 6,915 27 9,300 28 Industry 693 10 2,745 11 3,438 10 Energy & Power 1,397 19 9,153 35 10,550 32 Telecommunications 358 5 659 3 1,017 3 Transportation 593 8 1,925 7 2,518 8 Urban Dev., Water, Sewage 548 8 1,804 7 2,352 7 Other' 1,240 17 1,565 0 2,805 8 Total 7,247 100 25,879 100 33,126 100 Bangladesh Education 58 9 333 12 391 12 Pop., Health & Nutrition 47 7 78 3 125 4 (Population) (47) (7) (78) (3) (125) (4) Agriculture & Rural Dev. 246 39 770 28 1,016 30 Industry 62 10 84 3 146 4 Energy & Power 28 4 901 33 929 28 Telecommunications 20 3 35 1 55 2 Transportation 125 20 424 16 549 16 Urban Dev., Water, Sewage 42 7 98 4 140 4 Total 628 100 2,723 100 3,351 100 Brazil Education 64 2 239 2 302 2 Pop., Health & Nutrition 19 0 605 5 624 4 (Population) 0 (0) (71) (1) (71) (0.4) Agriculture & Rural Dev. 589 15 4,591 34 5,180 30 Industry 776 20 562 4 1338 8 Energy & Power 859 22 2640 20 3499 20 Transportation 970 24 1,567 12 2,537 15 Urban Dev., Water, Sewage 574 14 2,158 16 2,732 16 Otherl 110 3 1,011 8 1,121 6 Tota:l 3,961 100 13,372 100 17,333 100 77 Table 4.1 (continued) 1970-79 1980-90 1970-90 US$million Percent US$million Percent US$million Percent Colombia Education 28 2 115 3 143 2 Pop., Health & Nutrition 25 2 61 1 86 1 Agriculture & Rural Dev 208 13 670 16 878 15 Industry 0 0 90 2 90 2 Energy & Power 440 27 1,656 39 2,096 35 Telecommunications 90 5 44 1 134 2 Transportation 208 13 602 14 810 14 Urban Dev., Water, Sewage 258 16 362 8 619 10 Otherl 384 23 692 16 1,076 18 Total 1,640 100 4,290 100 5,930 100 Mexico Education 0 0 251 2 251 1 Pop., Health & Nutrition 0 0 0 0 0 0 Agriculture & Rural Dev. 1,297 44 2,941 21 4,238 25 Industry 295 10 713 5 1,008 6 Energy & Power 250 8 910 6 1,160 7 Transportation 452 15 1,327 9 1,779 10 Urban Dev., Water, Sewage 147 5 1,859 13 2,005 12 Other1 521 18 6,301 44 6,822 40 Total 2961 100 14,302 100 17,263 100 Kenya Education 39 4 78 5 117 4 Pop., Health & Nutrition 12 1 70 4 82 3 (Population) (12) (1) (70) (4) (82) (3) Agriculture & Rural Dev. 298 30 346 21 643 24 Industry 0 0 166 10 166 6 Energy & Power 115 12 216 13 331 13 Telecommunications 34 3 77 5 111 4 Transportation 244 25 176 11 420 16 Urban Dev., Water, Sewage 159 16 100 6 259 10 Otherl 92 9 415 25 507 19 Total 993 100 1,644 100 2,637 100 Senegal Education 1 0 8 2 8 1 Pop., Health & Nutrition 1 0 41 10 41 7 (Population) (0) (0) (15) (4) (15) (2) Agriculture & Rural Dev. 68 35 146 35 214 35 Energy & Power 38 20 10 2 48 8 Telecommunications 0 0 15 4 15 2 Transportation 6 3 22 5 28 5 Urban Dev., Water, Sewage 68 35 86 20 155 25 Otherl 11 5 96 23 107 17 Total 193 100 423 100 616 100 Public sector management, development finance companies, non-project lending, technical assistance and unallocated. 78 Table 5.1: Civil Works, Furniture, and Equipment in Bank Population Projects, Appraisal Estimates Percent of Total Percent Excluding Unallocated Civil Furniture Civil Furniture Project Works & Equipment Total Works & Equipment Total Indonesia Population 33.7 18.1 51.9 41.6 22.4 63.9 Second Population 22.6 14.2 36.8 36.3 34.9 71.2 Third Population 64.4 13.8 78.2 52.5 20.0 72.5 Fourth Population 28.3 39.6 67.9 50.4 23.9 74.3 India Population 47.2 17.5 64.8 47.2 17.5 64.8 Second Population 40.1 10.9 51.1 39.5 12.6 52.1 Third Population 55.4 15.1 70.5 60.6 19.7 80.3 West Bengal 4th Population 58.4 9.0 67.4 65.4 9.4 74.8 Fifth (Bombay & Madras) Population 10.8 25.2 36.0 12.6 25.6 38.2 Sixth Population 42.8 21.2 64.0 46.9 23.2 70.1 Population VII 37.3 16.7 54.0 42.2 17.1 59.3 Bangladesh Population 48.6 21.8 70.4 44.6 28.1 72.7 Population & Family Health 50.7 18.2 68.8 70.5 20.8 91.3 Third Population & Family Health 21.1 28.0 49.1 52.8 40.1 92.8 Kenya Population 89.3 8.9 98.2 90.4 7.3 97.7 Integrated Rural Health & FP 33.2 22.6 55.9 26.5 13.1 39.5 Third Population 23.6 24.5 48.1 24.0 40.4 64.4 Population IV 24.1 47.7 71.8 25.5 51.6 77.1 79 Table 6.1: Commitments fer International Population Assistance by World Bank and Other Donors, 1952-89 (millions of current and 1985 US dollars) 1952-59 1960-69 1970-79 1980-84 1985-89 Total World Bank Current $ - - 196.6 270.7 413.8 881.1 1985 $ - - 273.4 263.8 323.2 860.4 Other Donors Current $ 6.7 305.5 2517.3 2156.0 2935.1 7920.6 1985 $ 23.1 954.7 4146.0 2108.7 2341.4 9573.9 Total Current $ 6.7 305.5 2713.9 2426.7 3348.9 8801.7 1985 $ 23.1 954.7 4419.4 2372.5 2664.6 10434.3 World Bank as % of Total Current $ - - 7.2 11.2 12.4 10.0 1985 $ - - 6.2 11.1 12.1 8.2 Source: Table I in United Nations Population Fund, Global Population Assistance Report, 1982-89. New York: United Nations Population Fund, 1991. 80 References References Arthur, W. B. and G. McNicoll. 1978. "An Analytical Survey ture." In Etienne Van de Walle and J.A. Ebigbola, of Population and Development in Bangladesh." Pop- eds., The Cultural Roots of African Fertility Regimes. ulation and Development Review 4(1): 23-80. Proceedings of the conference, University of Pennsyl- Caldwell, John C. and Pat Caldwell. 1990. "High Fertility in vania, February-March, 1987. Sub-Saharan Africa." Scientific American, May 1990, Kenya, National Council for Population and Development pp. 118-125. and Ministry of Home Affairs and National Heritage. Cleland, J. and C. Wilson. 1987. "Demand Theories of the Demographic and Health Survey, 1989. Nairobi. Fertility Transition: An Iconoclastic View." Population Mauldin, W. Parker and Robert Lapham. 1984. "Family Studies 41(1): 5-30. Planning Program Effort and Birthrate Decline in De- Demeny, P. 1975. "Observations on Population Policy and veloping Countries." International Family Planning Population Program in Bangladesh." Population and Perspectives 10(4): 109-118. Development Review 1(2): 307-321. Paul, Samuel. 1982. Managing Development Programs: The Dow, Thomas E. and Linda Werner. 1982. "Modem Transi- Lessons of Success. Boulder, Colorado: Westview Press. tional and Traditional and Contraceptive Patterns Phillips, J. F. 1987. "Translating Pilot Project Success into among Kenyan Women." Studies in Family Planning National Policy Development: Two Projects in Bang- 13(1): 12-23. ladesh." Frank, Odile. 1987. "The Demand for Fertility Control in Asia-Pacific Population Journal 2(4): 3-28. Sub-Saharan Africa." Studies in Family Planning 18(4): Population Crisis Committee. 1989. The World Bank's Role in 181-201. Global Population Efforts: An Agenda for Effective Ac- Frank, Odile and G. McNicoll. 1987. "An Interpretation of tion. Washington, D. C. Fertility and Population Policy in Kenya." Population Sinding, Steven W. 1991. Strengthening the Bank's Population and Development Review 13(2): 209-243. Work in the Nineties. Policy Research Working Paper Freedman, R. and D. Freedman. 1986. "Adding Demand- No. 802. World Bank, Population and Human Re- SiJde Variables to Study the Intersection between De- sources Department, Washington, D.C. rnand and Supply in Bangladesh." PHN Technical Singh, Susheela, and Deirda Wolf. 1991. "Estimating Abor- Note 86-28, World Bank, Population, Health and Nu- tion Levels in Brazil, Colombia, Peru." International trition Department, Washington, D.C. Family Planning Perspectives, 17(1): 8-13. Freedman R., S. E. Khoo, and B. Supraptilah. 1981. "Use of UNFPA. 1979. Report of Kenya Needs Assessment Mission. Modern Contraceptives in Indonesia: A Challenge to New York. the Conventional Wisdom." International Family Plan- - . 1989. Global Population Assistance Report 1982-1988. ning Perspectives 7(1): 3-15. New York. Hammerslough, Charles R. 1992. "Proximity to Contracep- U.S. Congress. Senate. 1990. Committee on Appropriations tive Services and Fertility Transition in Rural Kenya." Report No. 101-519. Washington, D.C. International Family Planning Perspectives 18(2). World Bank. 1980. Population and Development in Kenya. Jain, Anrudh. 1985. "The Impact of Development and Pop- Washington, D.C. ulation Policies on Fertility in India." Studies in Fami- . 1990. World Development Report, 1990. Oxford: Ox- ly Planning 16(4): 181-198. ford University Press. Kamuzora, C. L. 1987. "Survival Strategy: The Historical and Economic Roots of an African High Fertility Cul- 83 ANNEX 1 Annex 1 The World Bank and Bangladesh's Population Program ANNEX 1 Executive Summary Introduction until after 1988, when services were made widely avail- able. Improvements in IUD services had a similar im- Population policies and programs have been a central pact on use. priority of the Government of Bangladesh (GOB) since In- * Differentials in contraception use between educated and dependence in 1971. This report examines recent demo- uneducated, poor and prosperous, and older versus graphic changes in Bangladesh with particular attention to younger women are all narrowing, suggesting that the the role of population programs and policies, and the con- program is reaching disadvantaged and more difficult- tribution of the World Bank to this effort. The Bank has to-reach groups. supported three population projects over the 1973 to 1990 * Contraceptive use correlates with availability of services period. A fourth Bank project is currently under and outreach activities: with distance between workers' preparation. Although the strategies pursued by the and clients' homes, with frequency of contacts, and with Bangladesh program have been the subject of considerable the range and quality of services provided. Where good debate, there is now incontrovertible evidence that contra- services and information are routinely available, preva- ceptive use has increased, that fertility has begun to de- lence has reached 30 percent-and in special circum- cline, and that Bank financed elements of the Bangladesh stances 50 percent-even in the absence of concomitant population program have contributed to the observed changes in social and economic conditions. trend. Lessons from the Bank's success in Bangladesh are relevant to population lending in unfavorable institutional These findings should not be interpreted as meaning that settings elsewhere. social and economic change is not also important, but only that significant progress on the demographic front - while more difficult - can still be made, even in the absence of such changes. Latent demand for contraceptives appears to Since 1970, contraceptive use has increased from 3 to 32 be substantial, despite pervasive traditionalism and adver- percent, and the total fertility rate has declined from over 7 sity. -.to about 5. Some social and economic changes that contribute to in- While- a quantitative assessment is not possible, findings creasing demand for contraceptives and for smaller fami- from recent surveys suggest that the family planning pro- lies have undoubtedly occurred. During the last two gram has played an important role. decades, the probability that a child will die before reaching age 5 has decreased from 25 percent to 20 percent. Since * Mosi of the contraceptive methods bring used are the 1974, female literacy has increased from about 13% to 23%; modern, effective methods provided by program sourc- GNP per capita has inched up at an annual average rate of es. Government service agencies, private voluntary 0.4 percent; and subtle changes in women's status and the agencies, and the social marketing project supply over perceived value of children are likely to have occurred. But 90 percent of the commodities in use. the pace and level of these changes are hardly sufficient to * The timing of increases in prevalence correspond to the explain the observed changes in reproductive behavior. timing of service intensification. This is particularly evi- The most important characteristic of demand for contra- dent for sterilization, which was not widely practiced ception in Bangladesh is its tendency to remain latent in the 87 ANNEX 1 absence of effective support, encouragement, and services. The structural bifurcation of the population and health In rural Bangladesh, women are restricted by the customs programs has had detrimental operational consequences. of purda to remain close to their homes, to subordinate The family planning effort inadequately addresses health personal preferences to familial interests, and to defer to issues. Health services tend to omit family planning. Inter- their husbands in all aspects of personal decision-making. divisional conflict is rampant and field operations are inef- These features prevent women from acting on their own. ficient. Problems are most evident at the periphery, where Even in these circumstances, outreach efforts by supportive integration, from the client's perspective, presumably mat- family planning workers can have a pronounced demo- ters the most. graphic effect. Other problems with the program include: Despite evidence that program efforts are reducing Lapses in service quality. In the absence of comprehensive fertility, the achievement of a replacement level of approxi- operational planning, large numbers of field workers mately 2.1 is not possible in the near future using only were rapidly recruited and posted, facilities were con- supply-side approaches. Even in special pilot project areas stred, and strvites were con- under near-optimal conditions where the contraceptive structed, and services were implemented. Implementa- prevalence rate has reached 55 percent, couples continue to tion preceded the development of training and desire at least two sons and a daughter. In the absence of supervisory systems. Service quality suffered as a conse- very significant social and economic change, existing strat- quence. egies and resources are insufficient to achieve what appears * Poor service utilization rates. Caseloads at rural health and to be the Govesament's goal. family planning clinics are low. Studies have shown that poor utilization is related to poor service quality and in- The Bangladesh Program * adequate supply and logistics. o Problems in the utilization of non-governmental agencies The crisis mentality of the period immediately after In- (NGOs). In recent years, NGOs have proliferated. There dependence din the expansion of the modest family is a need to coordinate public and private sector effort at dependence resulted th perphry withouto stfln initiative.mll planning program that existed during the Pakistan era the periphery, without stifling initiative. without testing or careful phasing-in of its various * Weak interministerial coordination. There are few inter- elements. In the process, several bureaucratic legacies from ministerial or multi-sectoral activities (apart from the the Pakistan and even from t:he British era became institu- work of a few NGOs). Resources are provided to eight tionalized. One of these legacies is a bureaucratic culture line ministries whose activities are supposed to be coor- dinated by the National Population Control Council favorimg a centralized, top-down approach -features that (PC,btta raiainlcsasceait are not conducive to implementing multi-sectoral pr- *(NPCC), but that organization lacks asecretariat. grams which by their nature have to be managed at the * Inadequate concern about finances and internal efficiency. gramsmwhichtby leveltheir naturhave tonuatona at thep- Perhaps because of the crisis mentality that has pervad- community level. Another was continuation of a separa- this program cost recovery or tion between population and health activities. While a cat- ednmuch ofithe plotis poram,acost recover no egorical family planning program was constituted as a long-term financial planning for sustainability have not wicalof the health ministry and labeled as "integrated" by been central preoccupations. External inputs, however, wing othhelhmnsranlaeeas"nertdby cannot continue doubling every five years. Increasingly, virtue of its location, integration has occurred to some ex- furt consin must ely fivementsingly, tent only at the top of the hierarchy in the Ministry of further expansion must rely on improvements in effi- Health and Family Welfare, and in principle, at the bottom ciency, a neglected issue. - where village level health and family planning workers Although these operational constraints persist, criticism are told to work as a team. At all levels in between, no in- of the program in Bangladesh must now be reappraised in tegration has occurred. light of the recent evidence of fertility decline. Key ele- The first of these legacies has resulted in a mechanistic ments of the Bangladesh strategy that appear to have con- management style. Rigid, centrally-imposed targets are tributed to the success achieved to date are the following: viewed by senior program managers as critical to maintain- ing minimal performance standards. Demographic projec- * Communication. Owing to extensive publicity, program tions are converted into contraceptive use objectives, and outreach, and mass communication, knowledge of con- national-level goals are parcelled out to workers uniformly, traception is virtually universal. without regard to their past performance, current work * Outreach. A cadre of approximately 19,500 female work- prospects, or to local conditions and needs. Unrealistically ers has been hired, trained, and equipped to deliver fam- high targets, prepared in this iFashion, undermine morale. A ily planning services to couples in their homes. closely-related problem is the tendency to focus objectives * Clinical back-up. A system of maternal and child health on targeted outputs rather than client needs. (MCH) clinics now blankets the country and provides 88 ANNEX 1 primary care facilities for intrauterine device insertion, projects. Fortunately, these inputs have been made avail- side effects treatment, and basic family planning ancil- able to this program, thanks in large part to mobilization of lary health care. By 1989, 2,716 of the 4,325 unions had donor resources for this purpose. clinics constructed and equipped to provide services. By Thus, the Bank's success in the Bangladesh population the end of 1995, all unions will be equipped with MCH sector derives, not only from its capacity to marshall re- clinics. sources, but also from its successful coordination and lead- * Accessible non-clinical supply sources. Nearly all pharma- ership. These achievements, however, derive less from ceutical outlets in Bangladesh are supplied with low- established mechanisms of the Bank than from a serendip- cost, subsidized contraceptives. By 1989, about 40 per- itous combination of co-financing mechanisms that have al- cent of the couples practicing contraception in Bang- lowed Bank staff to circumvent customary norms and ladesh were using condoms and pills supplied through procedures. Five key elements are involved: subsidized commercial sales. * Surgical contraception. Sterilization services are provided * Lending. Foreign financing accounted for 46 percent of atSurgnoal costrtoepatien.tsineverylizat subdisrvicte hosit i total program costs during the 1975-80 period and 67 Batngladesh to patients in every subdistrict hospital inpercent during the 1984-88 period. The World Bank has Bangladesh. ~~~~~~~~~~~been a major actor in this effort, contributing between 10 * Ancillary health services. Efforts to extend child immuni- bnd a ajrctor thi eort, contributing bwnd 10 zation services to rural households have been linked to family planning outreach activities. Successful health been used for construction and equipment that cannot outreach services have contributed to the credibility of be funded from other sources. Since 1975, there have family planning. been three population projects, and negotiations for a * Services provided by NGOs. Originally organized as a fourth have been completed. public sector program, the population program in Bang- Co-financing. In addition to its own resources, the Bank ladesh is increasingly a collaborative effort involving has helped mobilize grant funds from other donors. NGOs. Approximately 120 private voluntary organiza- Two-thirds of the foreign assistance provided by Bank tions are involved in the program, providing services to projects has come from co-financiers. The consequence 40. percent of the couples practicing a method has been a larger fraction of project funding for software * Implementation-based planning. Early negotiations be- components and a greater degree of flexibility to make tween the Bank and the Government of Bangladesh mid-course corrections than is typical in Bank projects. were Bank-directed and largely external to the GOB. * Coordination. This arrangement-lending in combina- W'ith each successive project, however, leadership has tion with grants-involves such a large fraction of total shlifted to the Government. Moreover, the planners have exteral resources that donors not formally involved become increasingly pragmatic. Priorities for operation- find it necessary to coordinate their activities with those al change are set on the basis of informed trial and expe- of the Bank in order to be effective. For this reason, the aieche atr sthn th basistof tin e fiat Tan exe negotiations for a new Bank project in Bangladesh have rience rather than by administrative fiat. Thi hllas become particularly evident in the planning of the major policy significance for the country, affecting the Fourth Project, which involved GOB-led work confer- strategies of the national program and all foreign do- enices, field investigations, and technical reports. nors, whether co-financiers or not. Over time formal procedures for consultation have evolved which appear The Role of the World Bank now to be working well. * Operational leadership. With the encouragement of the The comparative advantage of the Bank as a donor in the GOB, Bank staff have played a lead role, not only in do- population field is its capacity to marshall resources for nor coordination and mobilization of grant resources for large-scale infrastructure projects. In Bangladesh, this the program, but also in program planning and foreign strength has been complemented by two additional ele- aid and administration. In so doing, the Bank has made ment:s: an institutional development perspective involving possible the effective administration of a much larger software and technical assistance that has been crucial to program than the GOB could have administered on its the success of the more traditional hardware components, own. To play this role effectively, greater staff resources and an extensive consultation system involving all major - both in Dhaka and Washington - have had to be de- parties-donors, the Government, and the Bank, plus a voted to these projects. This has been made possible by number of NGOs and advisors-that has permitted this funding for this purpose from co-financiers. comrmunity to adapt strategies to needs as they unfold. This * Innovation and financial flexibility. The flexibility and inno- approach requires more Bank staff inputs in both the field vative character of the program has been enhanced, not and;in headquarters than has been typical of its population only by resident technical assistance, but by an "Innova- 89 ANNEX 1 tive Project Fund" established by grants from co-finan- pro-active role and be prepared to engage in strategic ciers and headed by a steering committee of local planning to solve problems as they arise. Mechanisms in experts. This fund has been used for critically-needed re- the Bangladesh projects that provided the Bank staff search on service quality, contraception innovations, and with such flexibility merit review for application else- other issues not anticipated in original Bank agreements. where. * Intensive staff inputs. The success of Bank efforts in Bang- Implications for Bank Operations in Similarly- ladesh relate directly to heavy staff inputs and dynamic Constrained Settings leadership on the ground. Arms-length project design and negotiation and intermittent supervision would not Elements of the Bank's Bangladesh experience that hold have worked in this case. The same principle will apply promise for application in other cases where the institu- in any setting where institutional capacities are weak. tional setting is constrained in similar ways include the * Co-financing, donor coordination, and the phased use of grant following: and loan funds. The capacity to program a package of loan and grant funds has been extremely useful in Bang- Phased implementation.Ely Bank effotin panglades ladesh. Where the need for population assistance coin- were inordinately complex. The national program was cides with limited capacity to develop a project, financed before fundamental structural problems in the consideration should be given to forming a consortium program were resolved. Premature implementation of consbefo n should be usortium the first project on a large scale institutionalized organi- even before lending begins. Grants could be used to ini- tiate activities on a pilot basis, fund resident staff, and zational problems that subsequent projects have at- create a framework for later appraisal. The program tempted to solve. By starting on a smaller scale and would then be scaled up and involve increasing compo- more simply, such problems would have been reduced nents for IDA lending. Established Bank lending proce- and the second project would have found it easier to ap- dures and project cycles would thus be invoked after a ply the lessons learned from the first. u f project was developed and tested. As a program ma- * The role of pilots. While numerous pilot studies of indi- tures, loans would play a larger role and grants would vidual components were specified in appraisals, they ac- be phased out. complished much less than they might have because of flawed research design and i:mplementation. For maxi- More generally, the Bangladesh case demonstrates the mum benefits, pilots should try out the whole proposed need for the Bank to adjust its operating style to the setting. project design in one or two districts prior to finalizing Weak indigenous capacity for management, technical lead- the design. That would mean, for example, that a pilot ership, and coordination require correspondingly creative trial of the fifth project should be included in the fourth donor mechanisms for supervision, technical support, and project. In countries just starting with a series of Bank interagency liaison. While the Bank could have provided projects, it might mean that the first project itself should more advisory inputs on the technical level, the character of be a pilot for the second. its operations in other respects is laudatory and resulted in - Flexibility. In settings where demand is fragile and ad- the mobilization of grant funds - especially successful in ministrative capacity limited, donors must take a more Bangladesh - that would not otherwise have occurred. 90 ANNEX 1 1 . Introduction In the chronicle of concerns about development con- survey responses evinced substantial unmet demand for straints in Bangladesh, progress achieved in the population contraception. Others argued that supply-side strategies sector has received less attention than its due. Surveys con- were unlikely to succeed unless there were prior changes in ducted over the past two decades show that contraceptive underlying economic and social determinants of reproduc- use prevalence has increased from 3 to 32 percent, and sug- tive motives.2 Amid considerable controversy, without first gest that the total fertility rate has declined from over 7 to resolving critical issues in the debate, a large and complex about 5. 'Child mortality has also declined, particularly program was launched throughout the country. Con- among infants. Changes are modest in comparison to de- straints, difficulties, and skepticism about prospects for mographic trends observed elsewhere in Asia, but surpris- success, so prominent in the literature about Bangladesh at ing in light of arguments that have been marshalled to the time of its Independence, must now be reassessed in explain why Bangladesh fertility and mortality rates are light of evidence that fertility has begun to decline. high and constrained to remain so. Bangladesh thus repre- The critical contribution of the Bank to the Bangladesh sents a demographic anomaly. Recent trends are inconsis- population program derives from the timing, co-financing tent with conventional theory on how demographic change and donor coordination mechanisms, scale, and complexity relates to societal conditions. of the four population projects. This :report examines the contribution of the World Bank Projects are timed to coincide with the planning cycle of to the Bangladesh population program, a program that has the GOB. Planning cycles, in turn, correspond to cycles of had extensive support from the World Bank since Indepen- organizational change. The coincidence of project planning dence. Negotiations for the first project commenced in with national program planning places the Bank in a 1972, at a time when population lending was new to the unique position to collaborate in the policy development Bank. This project, planned by a small technical staff in the process, and to collaborate with the Government in plan- Planning Commission and in the Bank, set the stage for a ning at a time when the GOB traditionally reviews pro- pattern of active involvement and leadership that has char- grams and undertakes new commitments. acterized the Bank's role to date. Since then, two additional The first Bangladesh population project was the first projects have been implemented and a fourth is about to Bank project in the social sectors to be "co-financed," an ar- begin. Each has been timed to coincide with five-year plans rangement whereby donor grant funds are administered by commencing in 1975 and extending through 1994. the Bank. Australia, Canada, the Federal Republic of Ger- The First Population Project was prepared before an in- many (FRG), the Netherlands, Norway, Sweden, and the ternational consensus had emerged on directions for popu- United Kingdom contributed to the First and Second lation policy that are appropriate for settings such as Projects. Sweden withdrew from the Third Project but may Bangladesh. Some analysts advocated family planning ser- rejoin the Fourth Project. Japan has joined the Fourth vices as the central thrust of population policy, arguing that Project.3 1. In 1989, two national sample surveys were conducted that produced 2. See for example, Demeny 1975. nearly identical estimates of contraceptive prevalence (Huq and Cleland, 3. In this report, "co-financiers" refers to formal participants in the projects 1990, and Mitra, 1989). The Fourth Five Year Plan uses a figure of 35 percent. and "donor consortium" refers to co-financiers and other donors partidpat- It should 'be noted that traditional methods are included in these estimates. ing in the project preparatory and monitoring process. 91 ANNEX 1 Because the combined resources of the project co-finan- cause the growth in salary contributions from the Third to ciers are so large, and coordination among major donors is the Fourth Projects reflects inclusion of encadred health ser- vital to the program, the entire consortium of population vice staff rather than an incremental commitment of re- donors participates in strategic Bank project meetings, in- sources.7 cluding donors not formally involved in project funding.4 In contrast to other programs, the share of external funds Meetings address a range of issues and needs extending far contributed by co-financiers has been consistently greater beyond the scope of lending, to include such diverse issues than that of IDA. Over the four projects, this share has var- as research, technical assistance requirements, service qual- ied from 56 to 63 percent. ity, logistics, and the like. Mechanisms for the Bangladesh The allocation of funding from loans versus grants has World Bank Projects thus extend well beyond the tradition- changed with time. As Table 1 shows, the proportionate al role of the Bank in lending, to include the administration distribution of project resources was similar for IDA and of foreign aid, the coordination of donors, and technical as- grant components. By the Second Project, however, IDA sistance. funding was concentrated on hardware, while grant sup- Taken together, the four projects comprise the largest sin- port was directed to software components. By the Third gle source of external support to the Bangladesh popula- Project, construction was entirely IDA funded and salary tion program in the post-Independence era. Table 1 at the support was entirely grant-supported. This pattern contin- end of this case study presents appraisal data for the three ued into the Fourth Project. Bangladesh population projects and preliminary budget As the program expanded and prevalence increased, the data for the Fourth Project.5 T'he most striking feature is the component of the program devoted to contraceptive com- approximate doubling in size of each successive project. modities, related pharmaceuticals, and basic equipment in- About half of these funds were allocated to hardware (civil creased dramatically, from $6.2 million in the First Project, works and equipment) and about half to software (training, to $59.9 million in the Third Project. By the Fourth Project, staff salaries, and special activities), of which salaries were commodity and equipment costs were the largest compo- by far the largest component. The First Project allocated nent of overall project costs. This growing commitment to 37.8 percent to contingencies, reflecting the uncertainty and commodity support was borne by the co-financiers. institutional constraints prevailing in the immediate post- The achievements of these three projects, and plans for independence period. Capacities to absorb funds and im- the fourth may contribute insights into World Bank deliber- plement programs improved with time, disbursement lev- ations on strategies appropriate for traditional societies els increased, and the unallocated component declined to where demographic changes are not expected to spontane- 16.8 percent in the Second Project and to zero in the Third ously arise. The sheer size of the Bank's commitment to the and Fourth Projects.6 Bangladesh population program heightens the need to de- The bottom row of Table 1 shows the percentage distri- rive lessons from this experience. bution of sources of funding. As can be seen, the Govern- While the declines in fertility and mortality that Bang- ment has progressively increased its share, from 10-12 ladesh has experienced so far do not represent a demo- percent in the first two projects to 28 percent in the fourth. graphic transformation of the country, they are substantial However, the budgetary increase, from US$5.3 million to when weighed in light of social and economic barriers to US$214 million in the third, and US$165 million in the the modernization of reproductive behavior in Bangladesh. fourth, is considerably less than these figures imply, be- More important, operational features of the Bank's Bang- ladesh projects may be relevant to lending strategies else- where. References to the economic, social, and 4. These include: United States Agency for International Development demographic problems of the world's least developed (USAID), one of the largest bilateral donors, but not a co-financier; the Asian countries often focus on the Bangladesh case. Of the Development Bank (ADP), which is increasingly active in population lend- w ing and relates its programs to components of the program that are also IDA- funded; the United Nations agencies, which are implementing some compo- the largest recipient of foreign assistance. Given the exten- nents of the project; and the United Nations Population Fund (UNFPA), the sive investment in the Bangladesh population program, United Nations Children's Fund (UNICEF), and the World Health Organiza- and emerging evidence of success, it is appropriate to re- tion (WHO,) which administer technical assistance contracts for the Bank. 5. The budget for the Fourth Banglladesh Population and Family Health view the role of the Bangladesh family planning program Project has not been finalized. A pre-appraisal budget of $571.4 million has been proposed, and will not change appreciably. 6. Problems with the capacity of the program to absorb funds continued 7. See Karim et al. 1985. The actual contribution of the Government of into the Third Project owing to a variety of administrative problems. Dis- Bangladesh to population, including contributions external to projects, is not bursement involves a myriad of steps and procedures, complex and archaic documented. Also, there is no single source of information about total secto- accounting mechanisms, and frequent procedural lapses. Resident Mission ral budgets, including external resources, other than projects of the Bank. For in Bangladesh (RMB) support to the Project often takes the form of monitor- this reason, there is considerable confusion about the actual costs of the ing the flow of funds, resolving delays, and arranging for disbursement. Bangladesh population program. (People's Republic of Bangladesh 1985.) 92 ANNEX 1 in recent demographic developments, take stock of con- of the Bank's mechanisms and the congruence of project tinuing difficulties and constraints, and assess implications plans with Bangladesh's institutional capacities. The insti- of the Bangladesh experience for programs elsewhere. tutional style of the Bank and its relationship to the donor This report examines evidence as to whether the overall community are reviewed for lessons from this experience strategy has been sound-first, by reviewing demographic for population sector lending in other, similarly con- trends for evidence that intended outcomes were achieved strained institutional contexts. The report concludes with and by reviewing the congruence of societal conditions recommendations for Bank operations that emerge from with programmatic aims. Next, the report examines the im- the Bangladesh experience. plementation of the projects in terms of the appropriateness 93 ANNEX 1 2 h ne Demographic Situation in Bangladesh Until recently, the broad outlines of demographic For the historical period for which reliable records are dynamics in Bangladesh were typical of South Asia in available, fertility has been extremely high in Bengal.10 Al- general. Mortality and fertility were extremely high until, though seasonality is pronounced, annual total fertility with the introduction of basic preventive health measures, rates remained above 7.0 for the first seven decades of this mortality declined and rapidl population growth ensued. century." Fertility decline is thus a relatively recent phe- Although fertility rates declined in South India and Sri nomenon, probably commencing in the late 1970s, and Lanka in the 1960s and 1970s, rates in North India, gradually accelerating in the 1980s. Pakistan, Nepal and Bangladesh were uniformly high. The relative role of proximate determinants of this trend Fertility changes in Bangladesh in the 1980s have intro- has been examined in the course of several national sur- duced a new element of diversity into the demography of veys over the past 20 years and summarized in two recent South Asia. GOB-sponsored studies.12 Survey results corresponding to Mortality in Bangladesh declined gradually during the the earliest and latest estimates from this series of studies first half of the century and raipidly in the 1950s and 1960s. are presented in Table 2 at the end of this case study The Progress was interrupted by the 1969-70 liberation war and famine in 1974-75. A breakdown in public sector or- ganizational problems in the immediate post-indepen- 8. At least one study has shown that parents perceive survival prospects dence period disrupted hea:lth services and delayed the of children as improving, but this finding may relate more to special health introduction of child survival programs. By the late 1970s, services in the study area than to the naware that fewer children are dy1g however, economic condition-s began to improve, the po- and attribute this to better health, even if mortality rates are unchanged. Per- litical climate stabilized, and mortality declines that had ceptions of mortality and their influence on reproductive aspirations merit been interrupted since the late 1960s resumed. further research in Bangladesh. 9. That female life expectancy is lower than male life expectancy is due to While the progress in mo:rtality reduction achieved in several factors. For reasons that are not well understood, infectious disease the 1980s is significant, as yet there is no convincing morbidity is generally higher among women than men. Maternal mortality, evidence that rural Banglacdeshis perceive the survival at about 6 per thousand live births, is 70 times rates reported from developed countries. Most important, femnale child mortality rates are substantially prospects of today's children as markedly improved. higher than corresponding rates for males, a relationship that is apparently Infant mortality ranged between 130 to 160 in the 1950s related to traditional male gender preferences and to selective neglect of girls and 1960s, and remained above 120 in the 1980s. In the in the allocation of familial resources. (United Nations Economic and Social Commidssion for Asia and the Pacific 1981.) As a consequence, female life ex- 1960s, roughly one quarter of all children died in child- pectancy in Bangladesh is one of the lowest in Asia (D'Souza and Chen 1980). hood versus one in five by the 1980s. Modest changes of 10. UNESCAP 1981 and 1984. this magnitude are unlikely to have altered parental 11. Long-term fertility trends are reviewed in the report of the Committee on Population and Demography, National Academy of Sciences (1981). An perceptions of child survival or to have induced recent overview of natural fertility dynamics in rural Bangladesh appears ini Menk- changes in reproductive behavior.8 Life expectancy re- en and Phillips 1990. mains low in Bangladesh-54 and 49 years among males 12. An overview of these surveys appears in Duza 1990a and fertility and females, respectively.9 ~~~~~~trends are analyzed in Kabir and Rob 1990. The most recent survey, pub- and females, respectively.9 lished after the Kabir and Rob review, is a study of the Bangladesh National Institute for Population Research and Training (Huq and Cleland 1990). 94 ANNEX 1 table employs the method proposed by Bongaarts (1978) to are likely to be offsetting, with lactation continuing to rep- decompose fertility into proximate determinants.13 resent a major fertility depressant.15 As Table 2 shows, an increase in the age of marriage has Estimates from six national surveys show a consistent in- reduced fertility among very young women. The fertility- crease in use of contraception, beginning at very low levels reducing effects of this trend have been offset somewhat by in the post-Independence period, and continuously in- declining rates of widowhood and increased rates of remar- creasing to nearly a third of all couples by 1989 (Figure riage among divorced women. Nuptiality changes have 2.1).16 Most of this increase has been in use of modern thus reduced fertility, but changes are offsetting and the methods. According to the 1989 Bangladesh Fertility Sur- overall effects are modest (Cm declined from .92 to .86). vey (BFS), 10 percent of all couples have accepted steriliza- Nuptiality changes account for a decline of .3 births in the tion, a figure representing about one fourth of all modern TFR over the 1975-89 period. contraception users. Various estimates of the fertility-reducing effects of abor- Fertility aspirations have declined, possibly helping to tion have been marred by estimation problems, incomplete increase contraceptive practice. The mean desired family data, and other difficulties. Some analysts estimate that size declined from four children in 1975 to 2.9 in 1989. Even about 17 percent of all pregnancies end in abortion. A more telling is the fall, from 28 to 9 percent, in the propor- rough estimate of the fertility effect of abortion can be ob- tion of women giving non-numerical, fatalistic responses to tained by comparing direct estimates of the 1975 Bang- the question on desired fertility. Several special studies, ladesh Fertility Survey (BFS) fertility (TFR=6.7) with where services are intensely purveyed, suggest that preva- Bongaaits estimates implied by a null effect of abortion lence rates exceeding 50 percent may be achieved if existing (TFR=7.2), for which the value of Ca (abortion) is .93. demand is met.17 The relative slopes of trends in contracep- Roughly comparable estimates emerge from special studies tive prevalence have been estimated in a recent United Na- of abortion in Bangladesh. For lack of data on this issue, Ca tions-sponsored review of the demographic situation in in Table 2 is assumed to be constant. South Asia and are reproduced in Figure 2.2 below. If cur- Prolonged post-partum infecundity that arises from pro- rent trends continue, prevalence will reach 50 percent by longed breastfeeding is the most important fertility-inhibit- the year 2000, a level that will surpass prevalence in the ing variable among the proximate fertility determinants in large states of North India and substantially exceed preva- Bangladesh.14 In some settings, the modernization of repro- lence in Pakistan and Nepal.'8 ductive behavior has lead to reduced breastfeeding and Figure 2.1: Trends in Contraceptive Prevalence Among shorter durations of post-partum amenorrhea. Although Currently Married Women in Bangladesh, 1969-89 various studies of reproductive behavior have shown that CoMrraceptive Llse Fliev.lence breastfeeding durations have remained stable over the last 35_______ 15 years and exceed 28 months on the average in both ur- ban and rural areas, lactational protection may have dimin- Traditional ished, as suggested by comparisons of 1975 and 1989 30 surveys showing a decline in the length of postpartum amenorrhea from 18 to 12 months. A trend toward less in- 25 tense breastfeeding could explain this, partially offsetting the fertility-depressing effects of contraception. The C. in- 20 dex in Table 2 is based on estimated rather than reported duration, since recall of amenorrhea durations may be sub- 15 Modern ject to bias. In fact, it seems possible that the role of abortion 15 has increased (implying a Ca decline) and post-partum in- fecundity has decreased (implying a Ci increase). Effects 10 _ ~~~~~~~~~~~~~~~~~5 13. Empirical investigation has shown that virtually all areal variance in total fertility rates can be explained by variance in the fertility-reducing ef- 0 I . i....L. L.I_ I I fects of contraception, exposure to intercourse, abortion, and post-partum in- 1970 1 9i7i 1980 1985 I 9 Jo fecundity (Bongaarts and Potter 1983). Bongaarts developed indices for the proportionate reduction in fertility that can be attributed to each of these fac- tors. In the Bongaarts model, the total fertility rate is given by the product of the total fecundity rate and four indices: contraception (C ), marriage (C ), 15. Huq and Cleland 1990. abortion (C ), post partum infecundity (C.). Following hisCframework, thense 16. See Mitra 1986; Mitra and Kamal 1985; and Mitra 1989. indices have been computed for Bangladelsh and reported in Table 2. 17. See Koenig et al. 1987; Phillips et al. 1988. 14. Huffman et al. 1980. 18. Mauldin 1990. 95 ANNEX 1 Figure 2.2: Contraceptive Prevalence, South Asia, Although contraception is the predominant factor in the 1970-2000 Bangladesh fertility decline, prevalence rates of 50 by the Percent of MWRA year 2000 would achieve a fertility decline falling far above Sa the current target of a net reproduction rate (NRR) of 1. Moreover, there is reason to doubt that the slope of the Fig- 40 _ BadnjglaLdes)h / ure 1 trend will be accelerated in the future.20 Caution is 40- India (LNI) * * Pakistan / , - suggested by the experience of special projects in Bang- * c* Nepal ladesh. No intensive service research project has produced 30 - , . levels of prevalence commensurate with the requirements 20 - - y - . .-... of replacement fertility. Producing replacement fertility re- 20 -*, ^ . . quires a prevalence rate of 70 percent-a level of contracep- 10 - ----......-. tive use as yet unsubstantiated by demographic or 1 0 , , -* ..---'operations research. Known strategies can r-educe total fer- , / ., :, . . . tility to about 4, but most couples continue to desire two 0 | , . * o * . | ., .. , ! E .... sons and a daughter. It is not clear whether future declines t970 1975 1980 11985 1990 1995 2000 in reproductive aspirations are in the offing or whether Year such changes are even required to sustain further increases in contraceptive use.21 The dynamics of contraceptive use Table 2 shows the demographic implications of in- thus merit careful field study and continuous review. That creased contraceptive use in Bangladesh.19 Prevalence in- current programs are succeeding does not ensure that fu- creased from 7.7 in 1975 to 31.4 in 1989. Because the ture aims will be attained with existing strategies and efficiency of contraception also improved, owing to the in- known resources. creasing proportion of all users who are sterilized or using Successive Bank projects have aimed to foster increases an intra-uterine device, contraceptive effectiveness in- in contraceptive use and reduce fertility. The precise impact creased. Taken together, the increases in prevalence and ef- of strategies will never be known. Although targets have ficiency reduce Cc from .93 to .71. This implies that the been overly ambitious, demographic trends are proceeding predominant factor in the Bangladesh fertility decline has in the desired direction and contraception has been the been contraception: Of the TFR decline of about 1.8 births leading cause. This lends support to the view that the pro- over the 1975 to 1989 period., 1.5 births are attributable to gram has had a role in the coincidence of the observed increased practice of contraception and 0.3 births are attrib- trend with the intensification of services, and evidence that utable to nuptiality changes. Thus, to understand fertility most users depend upon program sources for supplies. changes in Bangladesh requires an explanation of trends in Demographic research thus lends support to the view that contraceptive use. the broad thrust of the Bank's strategy in Bangladesh has been sound. 20. See for example, Khuda and Howlader 1985. 19. Ministry of Health and Population Control 1979; Huq and Cleland 21. In East Asia, contraceptive use continued to increase even in the ah- 1990. A second national contraceptive survey in 1989 produced similar re- sence of prior changes in reproductive aspirations (See for example, Chang sults (Mitra 1989). et al. 1987). 96 ANNEX I 3,. The Environmental, Social, and Economic Context The January 1990, population of Bangladesh was esti- worked as well as their proponents had hoped. There is mated to be 113 million, growing at a rate of 2.3 percent. general agreement, however, that population dynamics Dispersed in an area of only 144 thousand square kilome- and societal well-being are inextricably linked in Bang- ters, the Bangladesh population is 90 percent rural. Its den- ladesh, and that reducing rates of population growth repre- sity, 819 per square kilometer, is the highest of any country sents a critical policy issue, since reproductive change is in the world. Arable land is only a tenth of a hectare per unlikely to occur spontaneously. capita and over a half of it is under water during the mon- Of the Bangladeshi social characteristics with ramifica- soon season. Under even the most optimistic assumptions tions for organizing human services, the most striking is of dramatic fertility decline, Bangladesh will have a popu- the "diffuse social structure" of rural Bangladesh-a char- lation exceeding 400 million in the next century. It is diffi- acteristic of social organization that affects capacities to cult to construct a scenario of economic development or structure formal organizations that deliver effective com- population redistribution that can sustain this anticipated munity- based services. Located on small clusters of artifi- population growth. cially-elevated land, household groupings are fragmented into isolated groupings known as Baris. Baris lack clear A Diffulse Society leadership and traditional linkages between baris are frag- mented and diffuse. As a consequence, village government Much has been written about the societal ramifications of does not exist, . This lack of structure permeates all social poverty, risk, and adversity in Bangladesh, and how tradi- systemS in Bangladesh.24 Models for community-based ser- tional social institutions perpetuate high fertility.22 Skepti- vice systems that have been developed effectively in East cism about the role of family planning programs, and the Asia do not transfer well to Bangladesh. approach used by the Bank in its population lending, is There is little doubt that the geography of East Bengal grounded in research showing that Bangladesh is an inaus- has historically impeded efforts to impose external rule and picious setting for family planning programs.23 Extreme severely constrained contemporary efforts to develop par- poverty traditionalism, and environmental adversity are ticipatory social programs at the periphery. Without village rooted in constraints to increasing agricultural productivi- leadership to relate to, development efforts tend to be ty, enhancing women's status, improving health condi- externally imposed and artificially contrived, lacking an in- tions, and reducing fertility To some observers, economic digenous organizational base with links to the political sys- and social problems attest to the need for an effective fam- tem or the bureaucracy. Although community-based ily planning program in Bangladesh. To others, such prob- programs often succeed on a small scale when charismatic lems explain why family planning programs have not leaders can exercise influence, they fail when activities are extended to scale. 22. See Cain 1981 and 1983. See also the critical commentary on this per- spective by Robinson (1986) and Cain's reply (1986). According to Robinson, the climate of risk may not have the constraining effects posited by Cain. 23. See for example, Demeny 1975. 24. See Arthur and McNicoll 1978. 97 ANNEX 1 Weak traditional village organizational structures may protection of personal status. Status defines access to re- explain why rural life is so rif e with conflict, and why secu- sources and the parameters of personal influence over the rity concerns are so salient to rural Bangladeshi families. As use of resources; it is defined less by organizational rules one study has noted: than by patronage and the power of lateral networks. Sta- tus therefore becomes a necessary obsession: every nuance "Economic pow er, icl poweand ptie arexall of title, formality, and activity is mediated by its pursuit; toe. ay large e enthconvrtib.e toe othce or exam- and yet, status is never entirely secure, always vulnerable pie, by using wealth to b:ribe the police or by using to encroachment by rival networks or patrons. Personal force to extract money or steal land. However, few gain like advancement in the village is viewed as a zero power-holders specialize in all these simultaneously; gai eadnementhin the age is ed as gaiero sum game-the good things in life are limited, and gainers rather they maximize the benefits of their special skills represent a threat to those left behind.26 The diffuse society, and opportunities, and have close ties with others pervasive poverty, and the concept of limited good under- who have different abilities." mine organizational capabilities. Bangladesh Rural Advancement Committee, 1983 Thus, to advance in status or acquire wealth, the villager Low Levels of Educational Attainment must form alliances, join networks, and cement lateral rela- tionships among peers either to foster individual gain at Just as illiteracy and conservatism impede social change, the expense of others, or to protect vital interests from limited educational opportunities hamper the process of change of any kind. organizational development or renewal. Educational stan- Patronage is also importarLt in the allocation of resources dards are low, and staff capabilities to train cadres of sym- and insurance against risk. Although patronage has been pathetic and creative village workers are limited in every traditionally defined by lancd tenure relationships, the vil- sense: both trainers and trainees have limited competence. lage patronage system may be changing with the increasing Deficiencies at senior levels sustain technical deficiencies so displacement of peasants from land. Landlessness is not that the obstacles to staff development are substantial, per- only becoming more common, but trading, petty business, vasive, and synergistic. and other non-agricultural sources of income are becoming Moreover, the traditional values of villagers are shared, increasingly important components of the village economy, as well, by organizational personnel whose perception of and increasingly important sources of economic mobility the world and their job is unenlightened by exposure to a and patronage. Although land retains its value as the ulti- modern economy, to a strong educational system, or to mate source of security, patronage and networks built outside ideas. Workers thus share the conservatism of around trade relations are increasingly important sources society at large: women are expected to attend to their of prestige, influence, and political power.25 household duties, and are reluctant to leave their village. The diffuse social order profoundly affects the organiz- Not surprisingly, female domiciliary workers tend to work ing capabilities of formal bureaucracies in Bangladesh. For- near their homes, if at all. Exchanges between clients and mal organizational structures embrace the strict application service providers are consequently infrequent, and the of bureaucratic rules, encouraging hierarchical thinking, quality and quantity of outreach activity is impaired. Male mechanistic management, a:nd inaction. Informal bureau- workers are reluctant to deal with family planning; justifi- cratic culture introduces dynamic components, directed ably perhaps, because men cannot readily talk with wom- however, at survival, not performance. Lateral networks en, and most certainly cannot discuss something as are formed, dissolved, and reformed for personal gain, sensitive as family planning. with lines of accountability to peers and patrons in a man- ner that insulates individuals from the influence of formal Low Levels of Development organizational structure. The pursuit of personal gain blends formal organizational contacts with extra-organiza- Although per capita income in Bangladesh is among the tional networks and alliances. Individual prerogatives con- lowest of any country in the world, painfully little progress dition the setting of goals, the perception of achievement, has been registered in improving the economic status of the and the allocation of resources. Formal organization, like rural poor over the past three decades. Wage labor is pur- the village social order, is typically diffuse. sued by increasing numbers of rural Bangladeshis, but real The fragmentation of formal organizations is compound- wages have declined.27 Educational levels, while already ed by strong cultural values supporting the pursuit and low, have not improved, and literacy among women - at 26. See for example, Foster 1967. 25. See for example, Khan 1977. 27. See World Bank 1981. 98 ANNEX 1 20 percent - is the lowest in Asia and substantially below pursue any opportunity for wage labor or petty trading, literacy levels for men. even if the gains are modest.30 Under the difficult economic conditions confronting the Strong patronage systems and networks formed around typical Bangladeshi family, children have economic value, economic activities insulate personnel from pressures to not only for their productive activities in the household perform their organizational duties. In any case, supervi- economy, but also for their insurance value to parents. Sons sors also have goals external to organizational objectives, are thus valuable to parents, and strong gender preferences so that applying sanctions is rarely considered, and in any are consistently reflected in survey data.28 case, would be ineffectual. Control mechanisms are there- fore virtually nonexistent in public sector organizations Status of Women because organizational duties threaten vital personal in- terests. Hiring, training, and supervising a work force The life cycle of women is a sequence of dependencies on must confront an increasingly unfavorable climate for men: father, husband, and sons. Dowries absolve fathers of motivating workers to work and disciplining those who future obligations to daughters; but with marriage, wom- do not. en's economic and social roles are circumscribed by the tra- Environmental constraints. The environmental setting ditions of purdah. Maternal dependency on sons is poses obvious challenges to social outreach programs that common, owing to large spousal age differentials and high depend heavily on organized activity at the periphery. mortality, and the absence of extra-familial sources of secu- Most of the country's 66,000 villages are inaccessible by rity to disadvantaged women. A woman facing adversity road. Although river transport is extensive, scheduled lacks a broader collective to turn to, in part because poverty launch transport that could be used for field management is so pervasive, and in part because social structure miti- is virtually nonexistent. Diffuse management systems, gates against the distribution of risk. Lateral kinship links seemingly lacking in basic control mechanisms, perfor- are weak, so that protection from adversity derives from mance standards, or rigor, undoubtedly reflect fundamen- vertical lineages, particularly sons. In the absence of social tal ecological constraints to effective communication. security, a family without sons is vulnerable to encroach- ment ont vital interests. The Demographic Role of a Supply-side Approach in Bangladesh Economic Deterioration Given the comprehensive system of constraints on the The cleterioration of the rural economy over recent de- family planning program in Bangladesh, powerful argu- cades has undermined the capacity of the public sector to ments have been marshalled to explain its difficulties. So- develop and sustain service programs. As their real wages cial, economic, and institutional conditions consistently have declined, civil servants have spent more and more mitigate against success. Although survey data from the time in supplemental activities such as trading, farming, 1960s and 1970s suggested a considerable gap between re- and conmmerce during normal work time. The situation has productive intentions and fertility regulation behavior, progressed to the point where these supplemental activities the demographic relevance of family planning services often provide more income than the civil service job itself. was much debated until projects subjected perspectives in The result is weakening supervision and motivation at all the debate to formal tests in field experiments.31 Although levels, down to and including outreach workers.29 much debated in the past, there is now little reason to At the same time, social institutions, grounded in the dominant economic role of agriculture, have been altered bymit t declineofi rola ownegrsip.uSysturems ofn palteric 30. While economic conditions in rural areas have not improved, these by the dlecline of land ownership. Systems of patriarchy changes have been in some sense modernizing -forcing men to travel for and power, grounded in traditional land tenurial relation- trading, inducing families to diversify their economic pursuits, enhancing ships, have been uprooted as the proportion of landed the value of education, while undermining the traditional economic func- households has declined. The traditional elite retain much tions of the extended family and kindred relationships. (See Khan 1977.) Some social theorists have hypothesized that growing poverty can lead to of their influence, but traders, middlemen, and non-tradi- the modernization of reproductive behavior (e.g., Freedman et al. 1981). This tional entrepreneurs have become increasingly important issue has not been investigated in Bangladesh, and merits investigation, however. actors in the rural economy. This trend is complicating 31. Several projects have had careful research and monitoring, and at least econom [c roles even for the subsistence farmer, who must one study has incorporated an experimental design and demographic mon- itoring. See the reviews by Chowdhury and Huda 1990; Alauddin and Khan 1983; Choudhuri and Akhter 1990; Phillips 1987. Studies suggest that simple distribution of contraception fails, however, because many couples are am- 28. See Cain et al. 1979; Cain 1981, 1983, and 1986. bivalent about contraception, and a support system is required that address- 29. See Koenig and Simmons 1989. es a range of family planning needs (Rahman et al. 1980). 99 ANNEX 1 doubt that there is substantial demand for family plan- * Adapting strategies and organizational designs to social ning. Supply-side approaches can have net demographic conditions contributes to success. Traditional gender effects in rural Bangladesh.32 roles require women to remain close to home and to First, there is substantial demand for family planning in retain familial duties, even if they are hired into full-time rural Bangladesh that remains latent in the absence of an ef- jobs. Men will travel in their work roles, but other fective supply of contraceptive services. When localities are economic roles compete for their time. A few full-time isolated from information and services, the prevalence of and well-paid male workers are more effective than a contraceptive use remains low. If services are delivered to couples in their homes, bas, or hamlets, contraceptive use olarge male staff; a large part-time female work force is ' ' ' ~~~~~~~~more effective than a small staff of full-time female rapidly increases. This suggests that women demand workers. Sociologically appropriate designs for family planning, but lack sufficient resources, motivation, supers. maagement approl,iand ifraon ar or social support to act on these desires. Pervasive ambiva- supesion, management ct,n f maton lence about contraception, relluctance to contracept, and so- suggeste byetsu cial pressures mitigate against the success of family * All successful projects have mechanisms for local flexi planning.~ ~ ~ ~ ~ ~ ~ ~ ~~ ~l sucesfu proect havenen mechanismsiv forice loalrexi planning. But when convenient contraceptive services are bility and resource allocation. Staffing patterns reflect offered to rural women, a third to a half of all couples will task requirements. Replicating the success of special use a method. projects in the public sector will require meaningful de- Second, there is no single "best design" for a family plan- ning program in Bangladesh. Some successful projects have pacities to adapt strategies to local conditions and needs. used an integrated approach. emphasizing health services, while others have emphasized the provision of contracep- The themes of effective outreach, service quality, strong tion. Still others have worked through development pro- supervision, and decentralization recur in case studies of grams, with family planning consigned to an ancillary success in Bangladesh. The demographic significance of service role. Some successful projects function as women's these elements of success has been demonstrated in careful- programs. Organizational philosophies and designs of suc- ly controlled studies, lending support to the view that pro- cessful projects thus differ. gram activities have contributed to contraceptive trends. Third, although there is no single best design, there are common elements of successful projects:33 Conclusion • Frequent contact between outreach workers and clients Conventional interpretations of the determinants of fer- increases contraceptive use. Establishing this requires a t d manaemen sytem hat s oientd totheneedof rral tility decline center on the role of economic and social bet- terment in reshaping reproductive aspirations. Little that women for regular encouinters with service providers, has occurred in Bangladesh leads one to conclude that such Workers need to understand basic technical tasks, have improvements are occurring. No analyst of the institutional basic management infonrmation, supervisory support, context has concluded that economic conditions are im- and reliable logistics. proving rapidly in ways that are favorable to the modern- * The quality of services matters. In Bangladesh, the avail- abihequality of follow-up multiplersIn methadesh, and ancil ization of reproductive behavior. Yet fertility has begun to ability of follow-up, multiple methods, and ancgllary decline. health services provided by trained and caring workers The critical characteristic of Bangladesh society that ex- are elements of successful family planning services. Al- plains this apparent anomaly is the phenomenon that is though acceptor targets are emphasized by the national termed "latent demand." Powerful traditional forres con- service system, targets do not seem to matter if workers tinue to structure high fertility. Other emerging forces pro- are supported and supervised. duce counteracting effects, even though these trends are * Establishing a strong supervisory system has been a pri- not appropriately characterized as development. Since ority of every successful project. Supervision systems women are isolated, impoverished, and confined, changes differ, but all successful projects have clear lines of au- in demand have emerged without affecting contraceptive thority delegated to supervisors, decentralized person- behavior. The determinants of these changes are not well nel decision-making, systems for management control, understood, but relate to the declining economic value of and salaries commensurate with the task. children with secular changes in the rural economy, the pressures on families to seek wage income, pressures that counteract traditional restrictions on the role of women, 32. See for example, Phillips et al. 1982 and 1988. 33. See for example, Nag 1990; Nag and Duza 1988; Alauddin and Khan and other factors. 1983. 100 ANNEX 1 Family planning program communication activities Under these circumstances, the basic thrust of the Bank's undoubtedly nurture emerging demand, but outreach ser- strategy in Bangladesh-to support the provision of conve- vices have had the most important effects. If trained and nient contraceptive care to the rural poor-has been sound. well-supervised workers visit rural households on a There is substantial demand for family planning, despite all regular basis, and offer a range of farrmily planning services, of the constraints that have been noted. There is a need to contraceptive use increases and fertility declines. Even assign priority to meeting that demand by intensifying out- when such services are readily available, however, demand reach and improving service intensity and quality. Even is fragile, and the need for intensive outreach and ancillary without a favorable climate, much has been achieved with health services continues to be acute. As adoption rates supply-side approaches, and more can be achieved in the increase, contraceptive failure rates and discontinuation future. Field research suggests that an optimal supply sys- rates increase as well. Turbulent use dynamics betray an tem produces a prevalence rate of about 50 percent. Even underlying ambivalence about contraception that is sus- with an imperfect system in place, however, a substantial tained by social conditions, even as reproductive behavior demand for services has been met, contraceptive use has in- is modernizing. creased, and fertility has declined. 101 ANNEX 1 4. 1The Niational Family Planniing Program If family planning prograrn effort is traced to its origins the district level, effectively insulating the routine activities in the Pakistan era, the Bangladesh program is one of the of the bureaucracy from village life. A position in the public oldest programs of its kind in the world. Numerous re- service was prestigious; salaries were high in comparison views, appraisals, and consultant missions have examined to wage income outside the public sector. the structure of the Bangladesh population program, its In the post-colonial era, the role of the civil service field operations, and the effectiveness of various operation- changed dramatically. The mandate of the civil service was al components.34 extended to include village-based programs, and its size Although management reviews aim to diagnose opera- was greatly expanded in health, development, agriculture, tional problems and recommend changes, Bank assess- and other sectors. Although sustained by post-colonial re- ments have had considerably more impact on program gimes, bureaucratic traditions developed during the Raj design than studies sponsored by the GOB directly or by were inappropriate for social action programs requiring other donors, because Bank assessments often correspond outreach, decentralization, and flexibility Family planning to project appraisals. This chapter reviews the institutional programs, in particular, require active organizations at the context of the Bank's population projects in Bangladesh, the periphery- village-based services that influence reproduc- history of the program, and its major institutional achieve- tive behavior of couples through the provision of informa- ments. tion and contraceptive care. The dysfunctional preoccupation with demographic goals, targets, adminis- The Institutional Context trative orders, and rigid rules, so often noted in analyses of the Bangladesh program, have historic roots that are not The Colonial Legacy readily dislodged with exhortations for change. Programs are consequently overly-centralized, decision-making British India established traditions of administration and tends to be autocratic, and operational planning is based governance that affect the bureaucratic culture of contem- upon fiat rather than trial and learning.35 Sectoral minis- porary GOB programs. The colonial civil service, oriented tries lack mechanisms for interagency coordination, so toward revenue collection and maintaining public order, there is little capacity to implement multi-sectoral pro- instituted narrow spans of authority, top-down decision- grams or initiatives involving the private sector. Structural making, and extreme centralization. Positions in the British change is promulgated without attending to the operation- civil service were reserved for a select few and located at al implications of orders. In the population sector, weak administrative capacities have led to logisticical and distribution problems that 34. Among the more comprehensivie reviews are the studies by Chauls et al. 1984; P and M Consultants 1977; anid Korten 1975. The Bank's population projects and certain internal Bank evaiuations have addressed organizational 35. Some analysts have attributed considerable significance to the role of structure and impact. GOB commiss:ioned management studies more typi- research in fostering program innovation in East Asia. Conversely, failure to cally focus on the effectiveness of cornponent strategies (e.g., Mahbud et al. apply research to policy deliberations has detracted from the success of 1990). South Asian programs. 102 ANNEX 1 artificially constrain the availability of supplies. Supervi- * The quality of services was poor, undermining the credibility sion is weak, outreach is sporadic, and other organizational of family planning. Incentives for clients, pressures on problems artificially restrict access to methods. Priorities staff, and rapid expansion of the program resulted in and programs are shifted centrally without adequate atten- poor training and lapses in service quality Rumors and tion to operational planning and capacities to actually im- problems with the intrauterine device were particularly plement activities in the field. In its brief history, emphasis damaging.37 in the population program has been shifted from popula- * Fissures between health and population specialists arose when tion control, to population welfare, and then to MCH-based family planning was placed in a separate organization. Orig- integrated services. In the First and Second Five Year Plans, inally intended to enhance the prestige of population, five reorganizations were ordered. While these shifts led to and avoid associating family planning with the organi- new components, procedures, and implementing mecha- zational weaknesses of the Ministry of Health, the deci- nisms, the intended changes in structure, coordination, and sion failed to achieve its operational aims. organizational design have not been undertaken as Organizational rivalries launched at that time have nev- planned, creating structural confusion. er been effectively resolved. In the face of such constraints, the Bank has had a unique * Dysfunctional organizational traditions continued in the role to play. Its projects and mechanisms can foster major Bangladesh period. Structural arrangements established change in the way the program is structured and the man- in the Pakistan program have been reorganized in sever- ner in which it is implemented-changes that responsible al subsequent plans of the GOB, but certain elements of Government officers sometimes seek to undertake, but are the Pakistan program remain. First, the program is pre- powerless to pursue without the institutional backing of dominantly a public sector effort. Although private the Bank. agency contributions are growing in importance, this is a relatively recent development. Second, policies and Legacies of the Pakistan Period priorities are centrally planned for the country as a whole, and units responsible for operational planning The predominant thrust of population policy in are isolated from units charged with implementation. Bangladesh has been clinical family planning services aug- Large-scale change is ordered without information flow- mented with village outreach and mass communication ac- ing upward from the field. Unanticipated problems tivities--a policy focus developed in the Pakistan era in arise, because decisions are centralized, and there is little three phases: 1) A private, non-governmental agency phase capacity to adjust strategies to local needs. Third, major (1953-59) was implemented by the Family Planning Associ- changes in the program's focus, content, and structure ation of Pakistan, but subsidized by government clinical are promulgated without field trial or phased imple- services in large towns. This program failed to have demo- mentation. In the view of some respected observers, this graphic effects, but its pilot projects and training programs represents the most serious single deficiency of the developed experience in family planning that was applied Bangladesh program.38 to subsequent efforts. 2) An integrated health service phase * Political support was not translated into political action. As (1960-65) established the Government as a primary service the East Asian experience demonstrates, lethargy in the provider, and created, for the first time, an outreach pro- civil service can be overcome if political support is gram. This phase also failed because the staff was poorly strong and if commitment at the top translates into coor- trained, services were narrowly focused on family planning dinated action at various levels of government. Ever to the exclusion of health, and resources were inadequate. since Independence, and even before in the Pakistan era, 3) An intensive family planning service phase was national leaders have provided strong support to the launched in 1965. Clinical services, communication pro- grams, and outreach were expanded and intensified. An 36. A useful review of the history of family planning efforts in East Paki- autonomous Family Planning Board was created as an stan and its impact on the Bangladesh program appears in Duza 1985. Early agency that was independent of sectoral ministries, includ- concerns about the consequences of rapid population growth are summa- ing the Ministry of Health. Although intended to accelerate rized in an influential article by Khan (1973). 37. The program had no demographic impact but did have some effect on program effort, the program achieved little more than pro- contraceptive knowledge, although it is unclear whether stated knowledge moting public awareness of population issues and increas- in surveys corresponded to well defined understanding of how contracep- ing basic knowledge of contraception.36 tion is practiced (Sirageldin et al. 1975). 38. See for example, Freedman 1987. Various units were established for re- Although the Pakistan program developed models for search, such as the East Pakistan Research and Evaluation Centre, the Central clinical services, communication, and outreach, elements of Evaluation Unity, and the Academy for Rural Development at Comilla. the Pakistan legacy weakened family planning effort in the Nonetheless, research was viewed as a means of assessing impact rather than a resource for guiding operational planning. The important rural experi- Bangladesh era: ments of the Comilla academy had no appreciable effect on policy. 103 ANNEX I national family planning program. Presidential speech- mandate to start afresh. Senior civil servants, with consider- es to the parliament, cabinet, and the public have ex- able support from the donor community, argued that the ur- tended clear support to the program, and this has gency of the situation required rapid implementation of the undoubtedly provided impetus to activities on the existing program. The latter view prevailed, with the result ground. What has been missing, however, is a political- that workers from the old program were kept on and the or- ly-supported chain of command that reaches from the ganizational culture in which they worked survived. President's office to the village to provide a backbone for multi-sectoral programs that are otherwise spread over The First Three Five-year Plans and the World Bank several ministries. Key actors in the program lack mech- anisms for marshalling political leadership for village- By financing a large, complex multi-sectoral and based development activities that villagers relate to as a centrally-planned program in the First Project, the Bank group. Strong support at the top has yet to be translated could be faulted for subsidizing a program that would most into a grass roots movement with links to national par- certainly be difficult for the new Government to implement. ties, development programs, or local government. Cen- But it is appropriate to weigh such criticism in light of pre- tralized multi-sectoral programs that fail to empower vailing opinion in the early 1970s and the context in which local development are unlikely to establish the links be- negotiations were undertaken. Because Bangladesh was tween population and development that are used to jus- viewed as a crisis zone, most external observers and Gov- tify their existence as World Bank population project- emient planners believed that there was no time for a funded initiatives.39 phased-in approach. Anything less than a comprehensive and ambitious program was unthinkable in light of the ob- The Immediate Post-war Context vious urgency attached to the program. Many competing strategies were debated at the time-family planning Preparation of the First Fives Year Plan was initiated short- outreach versus health integrated with family planning ver- ly after cessation of the war with Pakistan and in the midst sus demand-generating development approaches. In the of a famine. A major international relief action was under- view of government planners and their colleagues at the way, and an atmosphere of crisis permeated all national Bank, the most cautious approach under the circumstances planning and donor negotiations at that time. The devasta- was to try all three strategies simultaneously: (1) categorical tion from this war was particularly debilitating to the health family planning administered by a population wing of the and social service sectors of the Government. The bureau- health ministry, (2) integrated health and family planning cracy had collapsed, universities and training institutes services in maternal and child health program, and (3) were decimated, and many of the critically-needed health multi-sectoral programs for demand generation in the fields facilities were destroyed. Basic communication was disrupt- of education, agriculture, rural development, labor, and ed, further straining capacities to organize effective Govern- communication. Funds were channeled to various minis- ment. The nation's only political party was new to the task tries for population activities, although budgets were never of governing, and its extensive grass roots organization was functionally integrated. Creating village-based multi- oriented more to the war effort than to development. Mech- sectoral development and health activities would have re- anisms for coordinating complex tasks at the periphery- quired careful field trial, extensive institutional where family planning services would have their effects- development, a phased-in approach permitting lessons to simply did not exist. NeverthLeless, a sense of urgency pre- be learned and problems to be solved at each stage, and vailed, leading to deliberations on creating and financing a meaningful decentralization to the sub-district level and be- large and complex program.40) low. A decade or more would have been required, Much debated at the time was the future role of staff from beginning with a small, highly focused program in a few lo- the Pakistan program. Senior planners pressed for funda- calities and gradual expansion of program coverage and mental strategic and organizational change, arguing that scope with time. past efforts had failed and the revolutionary regime had a The First Five-year Plan 39. An insightful critique of the Bangladesh multi-sectoral strategy GOB interest in population planning was manifest in its appears in Demeny 1975. 40. An influential conference in the immediate post-Independence period First Five-year Plan (1973-78), and the new Government focused on projections and consequences issues (Bangladesh, People's Re- took several actions to establish a program even before the public of, 1972). Although drastic action was called for, there was little atten- First Plan was developed. A policy conference in 197241 led tion to what that entailed in practical operational terms. Discussion was dominated by debate on alternative administrative structure of the program and whether family planning should be integrated into health, rather than on designs for field trials that would resclve controversy. 41. See Bangladesh, People's Republic of, 1972. 104 ANNEX 1 to a decision to place population within the health ministry two wings have concerns about their relative bureaucratic under an arrangement termed "functional integration." rank, tenure, and authority.48 This policy disbanded the discredited Pakistan Family The dysfunctional structure has origins that can be Planning Board by creating a population wing of the health traced to policies of the Pakistan era, and even earlier to a ministry and consolidating activities in a directorate of the bureaucratic culture extending into the Raj. The present ad- population wing. In practice, functional integration has ministrative reality, however, is that lines of authority are meant separating health and family planning into separate confused, supervision is weak, and morale is low. Official and distinct structures, but placing the two structures exhortations to correct operational problems have had little under a single minister. impact on underlying systemic problems. Although there have been shifts in focus, the Bangladesh The First Population Project. The First Bangladesh Project prograrn has retained the basic structure established in the established the Bank as the most influential donor in the First Five-year Plan.42 The Family Planning Directorate is population sector in Bangladesh. Innovative procedures divided, into five sub-units responsible for administration, and mechanisms were outside of the mold of Bank lending training, communication, planning, and service delivery. operations in other sectors and atypical of the Bank's ap- The sub-directorate for service delivery has a chain of com- proach to population lending elsewhere. Various prece- mand extending to the periphery. dents set in the First Project continue to guide strategic The organizational structure. The most serious structural plans of the Bank in Bangladesh. problem associated with the program relates to the Large-scale projects. The First Project budget was $47.5 integration of health with family planning and the respec- million, nearly two thirds of which derived from grants tive roles, responsibilities, and authority of officers in delin- from Canada, Sweden, Germany, Australia, Great Britain, eated wings of the Ministry of Health and Family Welfare Norway, and the Netherlands. The large scale of the project (MOHFW).43 The organizational structure of the set a pattern for lending that continues, since nearly all as- Bangladesh health and population program is complex, pects of the population plans of Bangladesh have been at although the fundamental structural characteristic is its least partially funded under Bank projects. Unlike Bank delineation of health and family planning into separate, commitments elsewhere, the First Project focused on soft- vertical, wings, joined at the top by the Office of the ware. Eight ministries were involved, covering aspects as Secretary, MOHFW, and at the bottom, by multi-purpose diverse as communication, clinical services, education, wo- village workers, but bifurcated at levels in between extend- men's affairs, mothers' clubs, rural cooperatives, agricul- ing from units in Dhaka, to districts,44 sub-districts,45 tural extension, and other issues. unions46 and wards.47 At least some of the pressure for a large-scale project in Efforts to streamline this structure have been character- Bangladesh relates to "lending pressure" from within the ized as "integration." In fact, considerable overlap in health Bank, deriving from its traditional focus on large loans for and faimily planning functions exists, since maternal and development infrastructure. More important, however, child health services are delivered in the family planning were antecedent events setting the context for the loan, dis- wing, and family planning medical officers are assigned on cussed above. A Bangladesh economic crisis led to a foreign deputation from the health wing. Various attempts to create exchange crisis, reinforcing interest at the Bank in large in- a unified chain of command have failed, not because health frastructure projects. Institutional interests aside, there was and family planning are separate, but because staff in the clearly a need for infrastructure: hospitals were in disrepair or destroyed by the war, convenient clinical facilities did 42. There are no mechanisms for coordinating the work of NGOs. not exist, and training centers were inadequate. Although 43. Ch oudhuri and Akhter 1990. criticized at times for a "bricks-and-mortar" bias, the Bank 44. Districts or "zillas" are 64 geographic units grouped into four regional identified priority issues in its commitment to hardware, divisions. Divisions have little administrative authority, and divisional direc- and balanced that commitment with co-financed support to torates are located in Dhaka. In each district there are separate offices for the Civil Surgeon, responsible for the health wing, and the Deputy Director for software. Family l'lanning, responsible for population activities. Each district has Complexity. Eight sectoral ministries were to implement about 6 subdistricts. 45. Thie subdistrict is the basic unit of civil administration for the police, components of the program, offically described as "multi- various development activities, and the civil service administration. The sectoral." Components were specified for women's pro- population of subdistricts typically range between 200,000 and 300,000. grams, mothers' clubs, agricultural outreach, community 46. Th ere are typically about 10 unions in an upazilla, each with an elected programs, communication and broadcasting, and other council. 47. There are three wards to each union, and each ward has a uniform activities collectively designed to foster demand for staffing pattern for family planning workers. The Health Division also has staff posted at this level, but different areal units are delineated, "mouza." Work areas and worker density differ for the divisions. Nonetheless, central orders have been issued for workers at this level to operate as a team. 48. BRAC 1990. 105 ANNEX 1 contraception and enhance the availability of services. and vocational training programs were specified in the Opinions vary as to whether activity at the periphery actu- First Plan and were expanded over subsequent plans ally represented a multi-sectoral approach, or is more ap- with funding from the Bank. Scaling-up was considered propriately characterized as a complex, interagency family desirable because multi-sectoral activities generated po- planning service system. Most consistent with the concept litical support for the program in several ministries. of "multi-sectoral approaches" is the work of a few private Whether the approaches used were sufficiently efficient voluntary agencies that combine family planning with vil- or effective to justify their institutionalization was never lage development activities. T'hese are activities external to seriously appraised. the official program, however, and have yet to constitute a * Operational optimism. Beginning with the First Project, substantial proportion of the overall volume of services in there has been a tendency for appraisals and project Bangladesh. The Bangladesh approach to multi-sectoral agreements to overestimate the capacity of the GOB to population activities differs substantially from the design implement stated operational goals. According to the of such approaches in East Asia, where development com- First Project audit, implementation was delayed consid- ponents are closely linked vvith family planning service erably, even for components as straightforward as con- components, and entire multi-sectoral systems are coordi- struction: nated by a single agency Multi-sectoral coordination has 1. "Disbursements lagged substantially behind schedule, been weak in Bangladesh for a variety of reasons: and by December, 1979, the first closing date, only 35 * Weak coordinating mechanisms. The NPCC was charged percent of the total amount had been disbursed. The to- with the task of coordinating the program, but it lacked tal amount was disbursed three and one half years later. a secretariat,49 and its principal contribution was to le- Start up project activities were delayed by over a year by gitimize policy rather than to coordinate operations. delays in declaring the credit effective, and further de- Lacking a secretariat and clear lines of authority over layed in formalizing donor commitments. The construc- sectoral ministries, the NPCC has never functioned ef- tion component was further delayed due to technical, fectively as a coordination agency.5° Its first policy dec- organizational, procedural, and contractual problems. laration, in 1976, nonetheless legitimated the broad Once completed, many difficulties were encountered in outlines of policy action. A consequence of the weak co- making the facilities operational, largely due to design ordinating role of the NPCC is that each "multi-sectoral" and quality control deficiencies. Nearly all the sites vis- scheme is administered by a separate ministry. In theory, ited by the audit mission required major repairs and these efforts created a multi-sectoral program, but from maintenance, while much of the equipment provided the perspective of Bangladeshi villagers, little about this under the project was either not installed or not func- program was likely to be perceived as a package, since tional at the time." (World Bank, 1986) activities were parcelled out to eight different ministries, 2. Difficulties of the First Project reflect a pattern of project each with serious implemnentation problems of their execution. Ambitious goals are set, implementation lags, own. problem- solving follows, and remedial steps are under- * Weak research and evaluation. Despite difficulties in taken that eventually move the program forward. achieving coordination, there have been several First 3. Operational optimism derives, in part, from the tenden- Project pilot efforts of sectoral ministries that may have cy of the Bangladesh Government to set unrealistic de- had some impact, although evidence of this is anecdotal. mographic objectives and assume implementation There has been little systematic accumulation of knowl- capacity is adequate to achieve them. A series of such edge about what works and what does not. targeting exercises has been undertaken, each in turn * Premature expansion. Pilots involving community providing guidelines for a forthcoming plan and re- development, women's cooperatives, mothers' clubs, sponding to the failure of previous targets to achieve stated aims. Demographic goals are converted into con- traceptive use targets, under an assumed method mix.51 49. The NPC has been renamed the National Population Control Council This target is compared to current prevalence, and objec- (NPCC). tives are dvlpdfor the national pormand divid- 50. Analysts of the role of multisectoral programs note that it is difficult to developed program make any definitive statement about their impact (Duza et al. 1989). Some ed by cadre so that each individual worker has a evaluation studies show that prevalence in special project areas is higher monthly performance goal for new acceptors of various than national level prevalence rates, but since comparison areas are lacking, contraceptive methods. This scheme for converting and areal variance in prevalence is pronounced, no conclusion can be reached regarding the efficacy of various schemes. The role of multisectoral programs nonetheless enhances the standing of the MOHFP vis a vis the ministries that it funds to carry out these activities. Even though the MOHFP has little coordinating authority to exercise over this activity, it strongly sup- ports the multisectoral components of the program. 51. See procedures proposed by Bongaarts and Stover (1986). 106 ANNEX 1 demographic objectives into task objectives has been as- oped. The "union" is the primary level of government in sociated with serious dysfunctions: Bangladesh, typically comprising from 15 to 20 villages and First, national-level demographic targets are unrealis- a population of 20-25,000. A key component of the First tic. The First Five-year Plan specified cautious demo- Plan was a pilot of the construction of union-level clinics graphic targets, but when their implications for future for basic maternal and child health services. Each team of growth was cause for alarm, they were revised in 1976. field workers is led by a male supervisor at the union level. Working from the premise that sustaining population Until 1985, each supervisor had three female outreach growth below the level of 120 million in the year 2000 workers for village outreach activities.54 Clinics are staffed was a matter of economic necessity, it was concluded by a male medical assistant and a female paramedic, re- that replacement fertility was required by 1985, imply- sponsible for health and family planning, respectively. ing a decline in the crude birth rate from 40 to 17 in The Second Five-year Plan proposed little in the popula- nine years-an unprecedented change in reproductive tion sector that was fundamentally new. Rather, the Bank's behavior with no empirical justification. Successive Second Project a. ed to consolidate progress, expand pro- plans have noted the "failure" of the previous plan, gram coverage, and sustain key initiatives of the First and ai new set of ambitious and unattainable goals. Project-the construction of health and training facilities, While targets have been revised, operational implica- the training of technical staff, and the expansion and refine- tions are not thoroughly examined.52 t Second, performance targets for workers are inflexible ment of communication activities. and unrealistic. If prevalence in a locality is low, tar- Despite its rather modest operational goals, the Second gets can be readily met. As prevalence increases, how- Plan called for an astonishingly ambitious target of replace- ever, meeting fixed targets is increasingly difficult. The ment fertility in five years. Targets were revised again in introduction of new, reversible-method programs is 1982, setting a pattern that continues: an order is issued de- often resisted, because workers believe that successful claring an ambitious target, followed by a new order revis- introduction of methods other than sterilization will ing it when evidence indicates that it cannot be attained. undermine opportunities to meet sterilization targets. In the course of the First and Second Five-year Plans, Rather than motivating workers, targets are detrimen- there were five reorganization orders addressed to imple- tal to morale. menting integration. All were met with opposition from the Third, targets lead to inappropriate emphasis on staff of the health ministry. None of them addressed the methods rather than client needs. In an ideal program, fundamental problems of a bifurcated structure, and disin- couples select methods from a range of available op- tegration was institutionalized in the process. tions. Targets can lead to efforts to recruit acceptors of The Bank's Second Project continued to support salaries particular methods rather than to care for their clients' and software components of the program, and greatly more general contraceptive needs. This may have led expanded the construction program for rural clinics. The to undue emphasis on sterilization, and a neglect of total budget was US$110 million, of which construction spacing as an element in contraceptive care. was roughly half of the estimated cost. Of the US$10 * Operational leadership. The Bank was the first major do- million, US$32 million was financed by IDA, and the nor to assist the Bangladesh program. A pattern of lead- remainder was co-financed. Since the IDA funding was ership and assertiveness was established in the First structured as the resource of last resort-to be used only if Project that continues to date. Procedures were devel- oped for contracting the construction of clinics, supply- grant funds could not be absorbed -athe Bank's primary ing clinics with equipment and drugs, training workers, role in Bangladesh Project II was to administer foreign aid setting goals, and monitoring performance. Basic sub- for the consortium. systems developed in the First Project continue to the The Third Plan represented a further expansion of the in- present. Numerous other examples of operational devel- itiatives launched during the previous decade. It called for opmnents can be cited, each supported by the resident a near-doubling of female village workers and continued mission or Bank consultants to the project. Most impor- construction programs of previous plans. Targets were tant among the various new initiatives were commit- revised again, to reflect the goal of replacement fertility by ments to establish service outreach and hire women to the year 2000. provide village services aimed at increasing program coverage. A focus on the periphery.53 The First Plan established a pat- tern for activities at the periphery that is still being devel- 54. Unfortunately, this design has separated field teams from the clinical program. Current plans call for expanding the paramedical team, and as- signing field supervisory duties to them. In the Third Plan, nearly 10,000 ad- 52. See Khuda and Howlader 1985. ditional female village workers were hired, increasing the union level field 53. The need for this focus has been reviewed by Heaver (1988). staff to 5 per supervisor. 107 ANNEX 1 In response to concerns about the state of child health, * The coordination of non-governmental organizations. Re- new goals were spelled out for maternal and child health strictions on the role of NGOs have been relaxed in re- activities. Female outreach workers, previously restricted cent years, and NGOs have rapidly proliferated. There is to family planning duties, were given duties in disease con- now a recognized need for mechanisms to coordinate trol, immunization, and ante-natal care. Apart from the public and private sector effort at the periphery, without added emphasis on health, there were initiatives designed stifling initiative or re-imposing operational constraints. to improve management and management information, * The pace of implementation. The implementation of the NGO utilization, and research coordination. program is weakened by operational deficiencies in its Apart from these new cormponents, the most salient fea- strategic planning system and management sub-sys- ture of the Third Project was its emphasis on sustaining in- tems for supervision, training, personnel management, itiatives from the First and Second Projects. The total cost logistics, and MIS. Performance targets continue to be was US$213.8 million, of which US$78.0 million was IDA unrealistic, inflexible, and dysfunctional. Rewards for funding, US$21.8 GOB counterpart funding, and US$114.0 achievement are lacking; sanctions for poor perfor- million was grant funding from Australia, Canada, mance are rarely imposed. Germany, the Netherlands;, Norway, and the United KiGdom.Iny budgNetaryrterms, eorwach,succes e projectedou- Various initiatives are proposed in the Fourth Project to blegdomthe n budgetaof theprecseding pretive.project dou- improve managerial capacities, both centrally and in the field. Most important among them are efforts to improve The Current Program: ThLe Fourth Five-year Plan supervision and strengthen management information ca- and the Forthcoming Fourth Population Project pabilities. Specifics are not yet developed, but mechanisms for field diagnosis of problems and communication of op- The Bangladesh program has been beset with operation- erational deficiencies to senior officers were initiated in the al deficiencies that are noted in numerous reports55 and Third Project, and will be refined and developed in the World Bank appraisals for the first three projects. The Third Fourth. and Fourth Projects address.key problems: The Elements of Program Success in Bangladesh * Bureaucratic triage. A possible solution to the structural problems of the health ministry is simply to allow As yet, the precise mechanisms through which program dysfunctional cadres to retire over time without replace- activities translate into reproductive change are incom- ment. In keeping with this strategy, retiring male field pletely understood, but it seems likely that at least six of the workers in the health division will be replaced with elements of the Bangladesh strategy have been crucial to women who will be trained to provide both health and the success achieved to date:56 family planning care. As these new workers are phased in and the previous cadres retire, prospects for meaning- Communication ful integration of family planning with primary health care will improve. Owing to extensive publicity, program outreach, and * Improving training and continuing education. In past mass communication, knowledge of contraception is virtu- projects, training has been limited to service providers, ally universal. Young women learn of contraception before and until recently, mid-level administrators and political marriage. Family planning is openly discussed in the pub- leaders were not provided with basic training. Efforts to lic media, undermining traditional conservatism about in- improve training are incorporated in the Fourth Five- ter-personal communication on this issue. According to Year Plan. Management training continues to be weak, survey research, about 6 percent of all married women had but systematic efforts to diagnose management prob- heard about contraception prior to the Pakistan Third Plan lems and take corrective action are the focus of forth- in the mid-1960s. By 1969, this had increased to 52 percent, coming training plans. although few women knew where to obtain services. By the * Expanding coverage and outreach. Female outreach work- end of the 1970s, 94.8 percent of all women knew of at least ers were first hired in the course of the First Project, and one method;57 by 1983, 98.6 percent knew of at least one their numbers were increased from 13,500 to 23,500 in method.58 Most women know about several modern meth- the Third Project. To further increase coverage, a new ods, where to obtain them, and what costs are involved. cadre of volunteer workers is proposed. 56. See Robinson 1985. 57. National Institute of Population Research and Training, Government of Bangladesh 1981. 55. See for example, Duza et al. 1989. 58. Mitra and Kamal 1985. 108 ANNEX 1 Outreach tributed to the increase in contraceptive use. Owing to the large number of clinics, even with low case loads averaging A cadre of approximately 19,500 female workers has about 30 clients a day, the volume of care is substantial. By been hired, trained, and equipped to deliver family plan- 1989, 2,716 of the 4,325 unions had clinics constructed and ning services to couples in their homes. To address the so- equipped to provide services. By the end of 1995, all unions cial isolation of village women, outreach workers are will be equipped with MCH clinics. instructed to visit households and provide women with basic family planning services in their homes. Although Accessible Non-clinical Supply Sources problems have been noted with this outreach system,59 sur- vey responses suggest that 40 percent of all rural women Nearly all pharmaceutical outlets in Bangladesh are are visit:ed by female village workers at least once quarter- supplied with low- cost, subsidized contraceptives. Contra- ly.60 Contraceptive use correlates with distance from a ceptives are widely advertised and conveniently available worker's home to a client's home, with the frequency of en- at low cost in every locality of Bangladesh. Recent decisions counters with workers, and with the range of services that to encourage private voluntary agencies to expand services workers can provide.61 Various studies suggest efforts to in rural areas have increased service accessibility. By 1989, make services accessible in Bangladesh are critically impor- about 40 percent of the couples practicing contraception in tant to the impact of the program. Because of the social Bangladesh were using condoms and pills supplied restrictions that purdah imposes on rural women, most through subsidized commercial sales (Mitra and Associates women are not free to travel to fixed service points, even for 1990). The success of subsidized commercial sales of con- primary health care. In the absence of outreach, demand for traceptives is due to the efforts of a privately-owned and family planning can be substantial, and contraceptive use USAID-subsidized "Social Marketing Project" (SMP). minimal, even if clinical services are readily available. For While not involved directly, the SMP has enjoyed full en- this reason, outreach has played a particularly crucial role couragement and support from the GOB for licensing, im- in the Bangladesh family planning program. portation, and other legal needs. This collaborative effort of the GOB with the SMP is the most successful project of its Clinical Back-up kind in the world. Maternal and child health clinics now blanket the coun- Surgical Contraception try and serve as primary care facilities for intrauterine device insertion, side effects treatment, and basic family Sterilization services are offered in every subdistrict hos- planning ancillary health care. Basic ambulatory services pital in Bangladesh by medical officers trained to provide are also offered. Staffed by a female paramedic and a med- vasectomies and tubectomies to clients free of charge.64 ical assistant, clinics are now located in 2,716 of the 4,325 unions62 in Bangladesh. If the conservation goals of the Ancillary Health Services Fourth Population Project are met, more than half of all couples will live within five miles of a clinic by 1995. At Efforts to extend child immunization services to rural each clinic, long-acting contraceptives are offered and an- households have been linked to family planning outreach cillary health care is provided at no cost. Although there are activities. Successful health outreach services have contrib- extensjive operational deficiencies of the clinical program, uted to the credibility of family planning. Clinical support and considerable room for improvement in the quality of for family planning is offered by paramedics who also pro- care,63 there is little doubt that the clinical system has con- vide basic maternal and child heath services. Much needs to be done to improve the quality of such services. None- theless, basic health care facilities for women and children 59. See BRAC 1990. are closely linked to the availability of family planning. 60. By 1983, about a third of the contraceptive pills and condoms in use wereprovided by outreach workers (Mitra and Kamal 1985). By 1989 the role of outreach had increased considerably: of the supplies that women dis- cussed wvith interviewers, 42, 20, and 28 percent of the pills, injectables, and 63. For example, regulations stipulate that paramedics reside in the clinics condoms currently used were provided by female outreach workers (Huq they serve -a regulation that is often broken owing to security concerns or and Cleland 1990). supervisory lapses (Mahbud et al. 1990). There is considerable need to im- 61. Assessment of social and programmatic determinants of areal variance prove paramedical training and medical back-up, logistics support, and su- in contraceptive use has shown that the quality and intensity of outreach is a pervision (see BRAC 1990). key determinant of contraceptive use. (See Rahman 1984 and 1986; Phillips 64. Field workers are given monthly targets for recruiting sterilization cli- et al. 1986.) ents. Whether targets for specific methods are helpful or detrimental is the 62. A union is a unit of local government with an elected council serving a subject of considerable debate (see Huq and Ahmad 1989). There is little population of 20,000 to 30,000 people. doubt, however, that priority is placed on long acting contraception. 109 ANNEX 1 NGO Services nisms for insuring that something would actually be done to offer the services that national leaders were promoting. Originally organized as a public sector program, the In light of these constraints, it seems surprising, in retro- population program in Bangladesh is increasingly a collab- spect, that the Bank would devise an ambitious and com- orative effort involving NGOs. By 1990,120 NGOs were in- plex, multi-sectoral population project. Population Project I volved in providing family planning services. Although was an exceedingly high-risk venture, developed by a estimates are undoubtedly subject to error, it is estimated small group of planners in Dhaka and counterparts in that approximately 20 percent of all modern method users Washington in a matter of a few months. The institutions are supplied by NGOs.65 The proliferation of NGOs, and supported in the project provided the terms of reference for services provided by them, has been facilitated by the deci- all subsequent Bank assistance to Bangladesh in the popu- sion to permit NGOs to provide services in rural areas, and lation sector. to receive foreign aid funds for such activities. Even more surprising than the ambitiousness of the strategy pursued is the fact that it seems to have worked, Conclusion despite severe societal and institutional constraints. The outreach component, in particular, has fulfilled a critical It was against a background of considerable institutional need. Clinical services are also important, however, partic- weakness that the First Population Project was launched in ularly for sterilization and the IUD. Bangladesh. The economy was in disarray, disrupting oper- Despite considerable operational difficulty in imple- ations in the civil service, ancd capacities to mobilize politi- menting the Bangladesh projects, project size has been in- cal support for the program at the periphery were weak. creasing rapidly-in part because the volume of services The most basic technical tasks to be performed by the new provided has been expanding, but in part because each new Bangladesh family planning program faced formidable or- project sustains previous components as new commitments ganizational constraints-services were in demand by indi- are added. A critical issue for the future will be the sustain- vidual couples, but social support for contraception was ability of co-financier participation, expansion of GOB weak at best. Implementing a service system was further funding, and cost recovery. Attention in the past has fo- impaired by weak administrative capacities in the public cused on the question of whether family planning can suc- sector. Social and bureaucratic forces undermine mecha- ceed in Bangladesh and strategies for implementing a large-scale program that reaches every town and hamlet. In the future, deliberations will shift to strategies for improv- 65. USAID 1990. At present, USAII) is the only donor assisting NGOs in ing program efficiency and operational effectiveness. family planning in Bangladesh. 110 ANNEX 1 5* hne Bangladesh Population Projects and the Bank's Institutional Style The GOB depends heavily on external assistance to its Canada, Germany, Norway, and the Netherlands. Sweden population program. Of the resources committed to the participated in the First and Second Projects; Japan may population sector in the 1975-80 period, 46 percent was for- join the Fourth. From its inception, this arrangement has eign-financed. By the 1984-88 period, this had increased to posed administrative challenges related to the diversity of 67 percent, a level of dependence that is likely to persist in donor mechanisms, priorities, and commitments to be the coming decade. The World Bank has been a major actor accommodated. Each co-financier, in turn, adrminisisk: 6f in this effort. Originally involved in global population pro- bilateral aid program in Bangladesh, and some of this assis- grams in the early 1970s, the Bank's commitment to the tance is in the population and health sector. In addition to Bangladesh population program commenced soon after ne- the co-financiers, USAID has a program in population and gotiatiorns were launched in 1972-a time when population health, with annual budgets of approximately US$25 mil- lending was new to the Bank and an emerging priority lion. UNFPA also has a major program, although the level issue. of assistance is somewhat less than the USAID commit- ment. Major support for child health programs is rendered Lendirng Mechanisms by UNICEF. Although UNFPA and USAID are not formal parties to lending agreements, they participate in the The Bank's lending mechanisms, which were developed project development and assessment process. ADB, WHO, for large-scale loans for infrastructure, employ procedures and most small foundations active in the population field and staffing patterns that are better-suited to hardware and have projects or activities in Bangladesh. ADB in particular capital-intensive lending programs than to software-inten- must liaise closely with the Bank because it has a program sive, grant-supported projects in population.66 Nonethe- of lending for infrastructure in health that augments exist- less, mechanisms of the Bangladesh projects address key ing Bank programs. Not surprisingly, the diversity of do- limitations of Bank procedures through collaboration with nors with disparate interests and philosophies has posed donors, and mechanisms that provide the Bangladesh diplomatic and administrative challenges for the Bank. At Government and donors alike with key advantages. issue has been a dispute between donors over priorities for child health components of the Third Project. Structural Co-financing problems in the MOHFW exacerbate such problems, be- cause responding to external pressure faces fundamental Of the funds committed to Bangladesh projects, over 60 operational obstacles. percent have been grant funds from Australia, Britain, The administrative mechanisms associated with co-fi- nancing are complex, involving accounts in Washington, revolving accounts in Dhaka, and mechanisms for monitor- 66. See for example, Crane and Finkle 1981; Wolfson 1983. ing the flow of resources from various sources to the 111 ANNEX 1 relevant sub-component of the project that a particular do- issues, however, and welcome efforts from the Bank to keep nor is funding. This machinery alone requires extensive ad- them informed about the broad range of exchanges be- ministrative support, although staff costs required for tween the RMB and the Government. Useful and compre- administrative operations are covered by the short-term in- hensive aides memoires accompany each Bank mission to terest that accrues on funds awaiting disbursement. Apart Dhaka, and these are now shared more widely than has from minor disruptions unrelated to the Bank, the adminis- been the case in the past. Aides memoires are also prepared trative support of the consortium arrangement works re- for each meeting between the RMB staff and GOB officers, markably well. In the view of some senior GOB officers, but circulation is apparently internal to the Bank. Without this administrative capacity of the Bank is vital to the pro- further encumbering communication tasks, there could be gram at its current stage of development. The role that has greater information sharing.67 Such mechanisms should be evolved for the Bank in the population sector is somewhat viewed in the manner of fine tuning an arrangement that is atypical of its role in ot:her development sectors, as already functioning, however. In the absence of the Bank described below: and its consortium arrangements, communication among donors would be fragmentary at best.68 The Bank as a Coordinating Agency Liaison mechanisms of the Fourth Project. In the Fourth Project, innovative liaison mechanisms have been devel- The co-financing arrangement for the Bangladesh oped that have improved coordination, without the consor- projects represents a critically-needed coordination mecha- tium as a pressure group. In addition to the regular nism. Although participating donors strongly support the exchanges that occur between GOB officers and the resi- Bangladesh program individually, serious inefficiencies dent mission staff, there are donor consortium meetings in would arise if each were separately negotiating bilateral Dhaka involving resident cofinancier officers. The most im- projects with the GOB. It is clear that no alternative to the portant improvement in this process has involved Bank as- Bank's mechanism for donor coordination is acceptable to sistance to GOB-sponsored work conferences and planning the GOB at this time. Senior GOB officers welcome the role sessions addressed to strategic planning and donor coordi- of the Bank as a coordinator and view alternative mecha- nation. Building on these local working sessions, there have nisms as unwieldy at best, and quite possibly, threatening been meetings in Europe, sponsored by the Bank, and at- tio he resource needs of the program. The Bank, as a lend- tended by representatives of the Government, the co-finan- ing agency, is viewed less as a donor than as a collaborating ciers, and the Bank. Bank-sponsored meetings involve agency The notion that any donor could somehow replace information exchange on progress, plans, and coordination the Bank as a coordinator of donors is unacceptable, since issues. All such mechanisms are in addition to the routine donor coordination could evolve into a mechanism for ex- supervision and appraisal missions of the Bank, and most ternal direction and control of the program. This is a rea- participants agree that this has improved the technical sonable concern given the fa,ct that nearly all resources for content of projects and collegiality among participating the program derive from external assistance. The Bank's agencies. style of relating to the GOB as a borrower makes its role as a donor coordinator acceptable as a neutral partner in deci- Mechanisms for Enhancing Project Flexibility sion-making. If the Bank did not exercise a coordinating role, some other less effective and less acceptable Resident Mission Staff mechanism would emerge. The combination of the resident mission presence and Initiative for Improving Donor Coordination the co-financier resources has given the Bank an element of flexibility in Bangladesh that it lacks elsewhere. The follow- With the growing complexity of projects, there has been ing examples of the role of the Bank in this context are in- increasing attention to improving coordination. structive: Liaison in Dhaka. Project consultants and technical advi- sors are contributed to the project by co-financiers. Advi- sors develop a role within project subcomponents, but do 67. Caution must be exercised, however, lest the sharing of information not always have direct links to the RMB. At times, dialogue undermine the climate of trust and collegiality between the Bank and the between RMB staff and GOB officers does not adequately GOB. A monthly liaison note, circulated for discussion, might improve do- nor communication. address technical issues. Mechanisms have improved in re- 68. Somewhat surprisingly, the home offices of donor agencies often fail to cent years, however, owing to the increased frequency of communicate decisions to their Dhaka counterparts. Lines of communica- technical meetings and exchanges. Advisors assigned to a tion flow from Washington to various home offices of the cofinanciers, where the follow-up of information can sometimes stop. Similar mechanisms could project component typically have broader interest in policy be developed for dissemination of information in Dhaka. 112 ANNEX I The compensation controversy. The policy of compensating training, service quality, logistics, and other elements of the clients arnd referral agents for sterilization has always been program were undertaken in the course of the first two controve:rsial. Intended as a means of covering costs, pay- projects. Where institutional capacities are weak, the role of ment of any kind could become an incentive. Although resident technical support is critical. such disbursements were not funded by the Bank, publicity A support team in Washington. Just as the resident mission about ethrical issues threatened European and North Amer- presence has enabled the Bank to respond to changing ican donor commitment to the Bangladesh program. Al- needs, a project team in Washington has provided back- though Government officers were aware of the controversy, stopping. In the view of senior officers in the GOB, the role they had no appraisal of the extent to which controversy of the Bank in coordinating the donor consortium has threatened long-term donor relations. Using the "innova- maintained a vitally-needed grant component of the tive project mechanism" and co-financier funding, a com- project. Without the consortium providing grants, IDA prehensive study was commissioned that resolved major funding could be controversial in the Finance Ministry and controversies and led to needed procedural change. The the scale of operations would have to be greatly reduced. diplomatic role of the Bank staff was critical to this initia- The Bank's role in coordinating this arrangement depends tive: the diplomatic role of the RMB clarified the need for a upon a small team in Washington. In the view of some co- study; liaison with the donor community from Washington financier representatives, this team has sustained donor translated political pressure on an emotional issue into an support for Bangladesh activities at times when funding objective scientific inquiry; and the study results preserved could have lapsed. the fabric of donor support and introduced appropriate Several examples can be cited of how critical events re- changes in the program.69 quired donor liaison, consultants, or backstopping. Seem- The quality of surgical contraception. In the 1980s, a series of ingly routine administrative operations have required carefully-conducted studies showed that the rates of steril- support. In one such example, the flow of funds from ization-related complications and deaths were unaccept- Canada for salaries, commodities, and other critical compo- ably high. Responding to the urgent need to re-train nents of the Third Project was interrupted by cash flow surgical staff, revise procedures, and monitor change re- problems in Ottawa. However, accounts in Washington are quired recruitment of an expert team of surgeons, logistical managed so that advances paid by one donor can cover support, and a sustained regimen of technical assistance. such cash flow needs. By agreement with the various do- Because the Bank's staff could respond to these needs, tech- nors concerned, resources continued to flow to Bangladesh nical improvements were instituted, and epidemiological without interruption. surveillance was undertaken demonstrating a marked The innovative projects find and the role of research. While reduction in morbidity and the elimination of excess the first project contained a sizeable research component, mortalit,7.70 many unanticipated needs and ideas for new approaches The health and family welfare centres. Convenient clinical arose during implementation. This led program designers referral 'points have been critical to family planning pro- to search for a more flexible mechanism to provide finances grams around the world. The First Population Project pro- for these purposes. The Innovative Projects Fund was the vided for pilot construction of rural clinics, but no response. Funded by IDA and administered by a committee mechanism for its expansion when the pilot succeeded and of Bangladeshi policymakers and population experts, in the need for clinics was demonstrated. In the course of the conjunction with IDA, it has been used to support - project, however, Bangladesh devalued its currency, creat- among other things - trials of innovative delivery sys- ing a situation whereby agreements and priorities could be tems, acceptability of new contraceptives, the evaluation of re-examined. Funds for the construction of 80 clinics were the effectiveness of a sterilization compensation scheme, identified in the course of negotiations-funds that would and other innovative work of NGOs that held promise for otherwise have been lost to the project. Because resident the national program. It has recently also been used to fund staff in D)haka were aware of needs and opportunities, and part of the census. could undertake sustained negotiations, resources commit- In recent years, the fund has not always been fully uti- ted to the project could be efficiently utilized. lized and some projects would have benefitted from more In numerous cases, conditions changed and initial thorough preparation. Somewhat closer supervision could project agreements were insufficient for emerging circum- take care of this problem. It remains a very valuable tool for stances. Critically-needed shifts in construction programs, the program. The contribution of co-financing to improving managerial ca- pacities in the GOB. Key program management dysfutnc- 69. See Cleland and Mauldin 1990 and the studies on which their assess- ptios we the said der frogm oaniationa tditions ment was based (Kamal et al. 1988; Khuda et al. 1988; Huq and Ahmad 1989). hons, we have said, derive from organizational traditions 70. See Khan and Mia 1984. that are difficult to dislodge. Central among them is the ten- 113 ANNEX 1 dency of senior administrative officers to issue directives complexity of the project and its co-financing, opening the without information on the operational implications of the process to participation of all donors and agencies can be order or feedback on its implementation. The Bank's Third fatal to closing an agreement. For this reason, projects have Project incorporated a mechanism aimed at resolving inter- been developed in Washington and modified in the ap- nal communication problemns in the MOHFW. The "Man- praisal process. This essential procedural strategy is criti- agement Development Unit" is funded by co-financiers, cized because projects prepared in this fashion are removed but functions as a management consulting team reporting from reality, over-designed, and impervious to change. to senior officers of the Ministry. At the request of the GOB, In the Fourth Project preparation, steps have been taken its management support functions are being expanded in to resolve this problem by sponsoring work conferences, re- the Fourth Project. vising pre-appraisal drafts in response to conference out- Ultimately, the problems that the Bank faces in coordi- comes, and vetoing preliminary plans to donors and the rtating donors relate to problems that the Bangladesh Gov- GOB well in advance of the formal Bank appraisal. Al- ernment must eventually solve, since donor consortium though this process has been labor-intensive, it has been and coordination mechanisms are intrinsically GOB re- characterized as a breakthrough in donor coordination, sponsibilities. In recognition of this need, important mech- greatly strengthening GOB participation in project prepara- anisms of the Fourth Project aim to develop GOB planning, tion and enhancing donor interest in funding project donor coordination, and implementation capacities. At components. least one project cycle will be required before such mecha- A strength of the Bank operation in Bangladesh is the ex- nisms are fully developed, but most important among them tent to which its representatives are viewed by GOB officers is a Project Finance Cell responsible for monitoring the flow as trusted colleagues in the project development and imple- of resources and the disposition of accounts.71 Improving mentation process. There is continuous and frequent dia- accounting operations will address donor concerns about logue between the RMB and MOHFW officers. Strong accountability to their governments for grant resources rapport and mutual trust between resident Bank officers committed to the GOB through the Bank project mecha- and GOB counterparts has been critical to the preparation nism. The relative advantage of joining the consortium ver- of the four population projects. Given the two-year project sus administering projects relates directly to the capacity of development cycle of the Bank and the complexity of the the Bank to insure that implementation proceeds as projects, close working relationships are essential. Any con- planned and resources are committed efficiently. tentious issue, whether large or small, can retard the pro- cess of project development and jeopardize the flow of Project Preparation Cycles critically- needed resources. Bank officers are pressed to maintain the pace of project development, and GOB officers The long project planning cycle and the complexity of share their concern that lapses in the preparation of projects project components has led to various problems. Projects can produce major disruptions. Thus, there are powerful are enormously important to the program, setting in mo- pressures on Bank staff and their GOB counterparts to work tion changes that can have major programmatic effects over closely together as members of a common team. There is lit- a five-year period and obstructing opportunities for change tle doubt that this effective partnership has marshalled re- in the interim. Once the process of project formulation be- sources for the Bangladesh program that it would not gins, there is considerable pressure on the GOB to specify otherwise command. components. This can lead to resistance at the Bank to any While this critically-important role of the Bank is ac- change that would slow down the project cycle. Once an ar- knowledged by co-financier representatives, the closeness rangement is reached, it is followed by five years of inflex- of the collaboration is criticized by some as unbalanced and ibility in the GOB until the cycle is repeated. Given the disproportionate to the actual diplomatic requirements of the project development task. Co-financiers feel excluded from this dialogue, and unaware of critical exchanges until 71. Much remains to be done in developing this unit. Basic modern man- it is too late in the project cycle to contribute substantively agement methods are yet to be developed, however, so that donors rely on to what is going on. To the credit of the RMB, much has the Bank rather than the Government for basic information on the disposi- b . M tion of resources. Spread sheets, computerized accounting MIS, and other been done to improve donor communcation. Meetings of management tools, so widely used in businesses in Bangladesh, should be in- the donor consortium are more frequent now than in past stituted in the Project Finance Cell. Not only would this improve financial project development cycles. Nevertheless, while communi- control and rates of disbursement, butt the donor coordinating capacities of the GOB would be enhanced. 114 ANNEX 1 cation has undoubtedly improved, co-financier criticism of overwhelm implementation capacity, with little thought the Bank communication style in Dhaka persists. being given to internal efficiency or long-run financial sustainability. The Continuing Operational Problems of Bank Projects Operational Mechanisms The Scale of Operations Apart from the scale issue, there are continuing criti- cisms of Bank mechanisms and suggestions for operational Because of the demographic circumstances of Bang- improvements. ladesh and its recent history of war and calamity, its popu- Technical leadership. The recent reorganization of the Bank lation program was viewed as an emergency action at the reduced the concentration of population expertise in the time of its creation. The language of crisis and the logic of Population, Health and Nutrition Division and increased impatience pervade the early planning literature of the the technical population staff in operating divisions. Reor- Bangladesh program. Bank officers, in response to the Plan- ganization had no effect on resident mission staffing, and ning Commission requests, opted for a large, centrally- has not contributed needed additional staff expertise to planned program. There was no blueprint for action in such population and health project planning for Bangladesh. In situations-no success stories to build on, no consensus in the view of some external observers, reorganization may the policy literature to guide strategic planning. Owing to have even weakened technical support for the population the climate of crisis, there was pressure on the Bank to un- projects somewhat. What is needed, in this view, is a team dertake something large and complex and to include nearly of officers such as the present support team for Bangladesh, anything that could plausibly be argued to have potential. who are knowledgeable about population issues and pro- A multi- sectoral program was designed that was integrated grams and available to work full-time on the project, deal- with health and categorical family planning, both supply- ing with administrative problems that inevitably arise in oriented and demand generation-oriented, focused on coordinating a large and complex undertaking in a weak hardware as well as software-as if no focus in fact was re- institutional setting. Demographic research expertise is a quired because everything that was needed would be tried. less pressing need than the program monitoring and super- There was remarkably little deliberation about what was vising roles of administrators with expertise in population unknown-how best to provide services, whether services project management. When specific demographic analyses under any conceivable scheme would work, how to struc- are required, it is important to have technical staff to turn ture tasks at the periphery, and how to organize a bureau- to, but these skills are best acquired from a research team cracy to ensure that tasks were actually implemented. that is specialized substantively in population studies rath- There was little deliberation, however, about what was er than regionally specialized in operational units. But it unknown: about how best to provide services, whether ser- had no effect on resident mission staffing or the Bank's con- vices under any conceivable scheme would work, how to tribution to planning of population and health projects at structure tasks at the periphery, and how to organize a bu- the technical level. In addition to the team of officers work- reaucracy to ensure that tasks were actually implemented. ing on administrative problems and overall planning-ab- Nor was a well-articulated strategy adopted to overcome solutely necessary in this weak institutional setting-more such knowledge gaps, for example, by starting small, scal- technical inputs, best acquired from a research team spe- ing up, and revising plans on the basis of lessons learned. cializing on this country rather than being regionally ori- Such a c autious, phased-in approach had no constituency ented, would have been very useful. in Bangladesh and little in the Bank or amongst other do- The greatest need is for expertise that can help resolve nors. The style of project development and negotiation was strategic planning issues in Bangladesh, reviewing the rel- simple: a few able and energetic senior staff from the Plan- ative merits of competing options for improving program ning Commission interacted with a few like-minded col- performance, and guiding the design of a project based on leagues from the Bank and developed a program to serve what has worked or not worked in previous projects. Deal- 64,000 villages. Donors were eager to contribute. The polit- ing with complex issues in interagency coordination re- ical chemistry was right for bold action. Trial, experimenta- quires close collaboration with institutions affected and a tion, phased development of focused initiatives-the continuous presence in Dhaka. For reorganization to make elements of sound strategic planning-were not part of this a difference to population lending, resources would need to chemistry. While more technical inputs and feedback from be committed to fielding a larger operations staff, including the field have gone into the designing of later projects, administrative and technical assistance personnel. Techni- much in this culture of urgency remains. Project size has at cal assistance, pilot testing, and feasibility research would least doubled every five years, constantly threatening to need to be expanded. As things stand now, field staff is 115 ANNEX 1 bogged down with administrative tasks and has little time idence that workers are not performing or that clinics are for technical issues of strate,gic planning. Such issues are underutilized is a growing source of concern. Even the suc- addressed by other donors and their consultants,72 but few cess of the program is, in some sense, a source of concern. of the co-financiers believe this is an adequate substitute for The successful expansion of services has accelerated the Bank inputs. cost of commodities, drugs, and other recurring costs.75 The role of co-financed advisory support. When institutional The role of research in strategic planning. Until recently, the support to a large and complex program is an issue, a tech- role of research in strategic planning has been informal, and nical assistance approach is required. This in turn requires much of what researchers do has had little effect on the de- a resident team of experts--something co-financiers are sign of five-year plans or Bank projects. Informal effects de- providing through their bilateral aid programs. As the rive from the impact of small-scale studies on official number of technical staff grows, and their cohesion as a opinion of what works. In recognition of the need to en- group grows, they have becoime critical of the Bank for do- hance the role of research, recent work seminars focus on ing less than they believe is needed to provide coordination policy planning and research issues simultaneously. The and leadership in articulating problems as they see them to Fourth Project incorporates mechanisms for translating the GOB and seeking satisfactory resolutions.73 Advisors research into action and formalizes innovations of the Third typically have an office in their counterpart institution, but Project.76 maintain close links with their respective embassies and Research activities and findings are nonetheless diffuse donor representatives. Their technical contribution is much and difficult to digest. Although problem-solving studies needed, but their link to the mechanisms of the Bank's are extremely useful, linking decision-making mechanisms projects could be strengthened. There is need for improved to experimental studies might contribute the Appraisal technical coordination among advisors, possibly involving Process for the Fifth Project.77 contiguous office space at the RMB or at least a framework This would involve: for technical meetings and exchange. Informal exchanges that occur between advisors and the RMB could be struc- * confronting each project with a set of critical operations tured into a more formal mechanism for technical back- to be resolved (implementing integration, improving stopping. This would obviate the need for increasing Bank supervision, coordinating outreach with NGOs, staff while creating greater cohesion between co-financier improving service quality, etc.); advisors and RMB. To avoid the potential pitfall of creating * designating a district where these problems are to be an isolated pressure group, liaison mechanisms should in- resolved, tested, and researched; clude GOB technical counterparts.74 * requiring that each project prepare Fifth Five-year Plan Costs and efficiency. Even more fundamental problems for recommendations in time for the appraisal; the consortium mechanism rmay emerge in the coming de- alowing each project to work with separate resources cade. With the dramatic increase in the size of successive and institutions; but project budgets, there is growing concern that the Bank's specifying a coordination and steering committee mech- approach is tantamount to an uncritical underwriting of the anism. Bangladesh public sector. Just as there is a need to scale up A common framework, developed in this fashion, would what works, there is a need to phase out what does not. Ev- greatly facilitate the utilization of research by creating a family of studies dealing with common and difficult-to- 72. USAID, for example, has a staff cf 20 and a sizeable budget for consult- resolve systemic issues. ants for a much smaller and more focused project. 73. The contrasting institutional roles of lending and donor agencies may account for some of this criticism. Because lending involves resources that are owned by the borrower, accountability for the pace of implementation is less strictly enforced by Bank officers than is the case for grant officers award- 75. The Fourth Project calls for studies of cost recovery for the first time, ing foreign aid (See, for example, Wclfson, 1983; Crane and Finkle, 1981). plus increased participation of the GOB in covering recurring costs from the Cofinancing involves the diplomatically challenging task of merging project revenue budget. monitoring organizational cultures of the Bank and the cofinanciers. Perhaps 76. See for example, MOHFP 1989a and 1989b. These reports and proceed- reflecting the complexity of this task, representatives from the resident mis- ings of work conferences illustrate the increasingly sophisticated participa- sions of cofinanciers express concern that they cannot account to -their home tion and planning role of the MOHFP in Bank projects. (Duza 1990a and offices for the state of implementation of project components that they are 1990b.) See also, case studies by Yunus et al. 1984; and Koblinsky et al. 1989, paying for, and cannot count on the RMB to press the GOB to implement ac- addressed to the need to translate research into strategies for the Third tivities on schedule. Project. 74. At present, liaison involves RMB staff and senior officers who simply 77. These include the German Technical Cooperation Project in do not have time for more meetings of the sort proposed. A technical coun- Munshiganj Division, projects of the Intemational Centre for Diarrhoeal terpart is a high ranking officer involved in implementation of a project com- Disease Research in Matlab, the MCH-FP Extension Project, special studies ponent. These are formally designated directors of subcomponents of the of the Bangladesh Rural Action Committee, the Pathfinder Fund, and Project. possibly others. 116 ANNEX 1 Prospects for Replicating the Bangladesh Approach support from the Bank itself for these activities. Incen- in Otheir Population Projects tives, as defined by routine performance evaluation The flexibility that derives from co-financing has led to criteria, encourage staff to minimize supervisory inputs, extensive staff inputs and adaptive planning. Numerous to de-emphasize liaison, and to structure agreements so examples of crises could be cited that could have threat- that little administrative support will be required for ened Bank population projects. Staffing and flexible their effective execution. Effectiveness of a loan, in turn, funding arrangements permitted Bank officers in the field is judged more by the volume of lending and gross indi- and in Washington to deal with problems in ways that sus- cators of impact than by evidence that institutional ca- tained implementation and improved project performance. pacities are being developed. That things were somewhat Such problems are not unique to Bangladesh. Success in different in the Bangladesh program is a tribute to the many countries where lending is under consideration will specific task managers involved and to donors willing to depend upon intensive staff inputs, creative financing ar- provide grant funds for the purpose. Short of a very sub- rangements, and extensive institutional development. stantial change in the Bank's operational culture, these At present, a Bank officer responding to the need for two conditions would have to be met in other countries technical support, donor liaison, or adaptive project plan- to generate similar results. ning and implementation will find little institutional 117 ANNEX 1 6. Conclusions and Recommendations Developing and expanding family planning services has in population should have an unusually intensive staffing been the principal thrust of the Bank's support to the pop- commitment for the project preparation period. Consider- ulation program in Bangladesh. Despite considerable initial ation should be given to establishing mechanisms for pre- controversy about the wisdom of supply-side priorities in project technical assistance and pilot studies, not governed such settings, there is now clear evidence that the program by the Bank's project cycle, but instead designed to inform has worked. Much needs to be done to clarify the role of the project planning process. First projects should be small, programs and policies vis-a-vis the role of societal and eco- with an emphasis on phasing in regions or activities, in nomic trends, but there is strong evidence from special ways that inform the generation of subsequent projects. projects to suggest that contraceptive services are critical to First projects should have high supervision ratios, and a crystallizing and sustaining the modernization of repro- strong emphasis on donor coordination. Since lending is not ductive behavior in Banglacdesh. Prevalence has been in- required for preliminary pilot activities, first projects creasing, particularly for sterilization, and fertility has should be financed largely, if not entirely, by grants. Loan begun to decline. This suggests that the overall approach to components should be small and software-oriented. population assistance that has been used in Bangladesh, The Bank has been a major factor in the development of emphasizing supply-side strategies, is sound and merits the Bangladesh program. As the largest single donor and consideration even in settings where social and economic leader of the consortium, its missions and resident staff conditions are unfavorable to family planning programs. continue to exert major influence over operational policies Developing a successful supply-side strategy is a com- and programs of other donors. This role is welcomed by plex undertaking, particula:rly in settings where family GOB officials, who view the Bank's role as indispensable in planning is neither widely practiced nor obviously in de- mobilizing resources, providing smaller donors with a mand. In such settings, there is considerable risk of strate- method of participating, and reducing administrative gic error. Starting initiatives on a small scale, learning from pressures on them. success, and scaling-up reduces the risk of failure. Large- The decision to post resident staff to Dhaka has been crit- scale central planning, so endemic to Bank initiatives in this ical to the success of the Bank's efforts in Bangladesh. In sector, can underwrite premature bureaucracies and insti- comparison to other sectors, or to Bank population projects tutionalize the elements of failure. While much about the elsewhere, staff time assigned to the Bangladesh projects is Bangladesh program can be characterized as a success, dys- unusually high. In the view of donors and GOB officers, functional elements have been externally funded that are however, this arrangement has been crucial to marshalling now extremely difficult to disband. the resources that were required to develop the program. Where programs are new, early projects define the terms In retrospect, decisions to develop a large, multi-sectoral of reference for all projects that follow. Bureaucratic ma- program in the First Population Project were premature. In chinery is subject to formal rules and procedures that the the absence of capacities to coordinate such a program cen- Bank and other donors cannot change. Because of the bu- trally, or mechanisms to implement complex programs at reaucratic momentum that a first project establishes, there the village level, the First Project should have been more fo- should be special mechanisrms associated with them-par- cused with subsequent projects designed to scale up or ticularly, if the Bank is planning a lead role in establishing phase in what was learned at earlier stages. Consideration a program, as it did in the case of Bangladesh. First projects could still be given to targeting the development of the 118 ANNEX 1 Fifth Project in one or two focal districts as a component of * A coordinated consortium approach can be useful to all the Fourth Project. When bureaucratic capacities are weak, parties concemed. Operational advantages of the Bank a phased-in approach is probably more workable than the are combined with advantages of bilateral aid. Where project planning and implementation approach of the there is a need for population assistance, combined with Bank. a limited capacity to develop a project, and an interest in While much has been done to improve mechanisms for Bank funding, consideration should be given to mecha- project planning, there is continuing criticism of the Bank's nisms for forming a consortium even before lending be- tendency to centrally plan activities for large bureaucracies gins. In such settings, grants would initiate activity on a before there is adequate knowledge about what works and pilot basis, fund resident advisory assistance, and create how to implement it. The dysfunctional structure and staff- a framework for an assessment mission. This arrange- ing pattern of the MOHFW that was created in the First ment would recognize the comparative advantages of Population Project continues to absorb considerable re- different donor mechanisms at different stages in a pro- sources in the forthcoming project. Changes that are sug- gram's development. Initial efforts should focus on gested by field research could improve efficiency and small-scale trials and grants rather than loans. Institu- performnance, but much of what is needed is neither politi- tionalizing a program should involve gradual scaling cally nor operationally feasible. up, using pilot projects to identify components for IDA There is a need for mechanisms that are external to lending. Established Bank lending procedures and project and lending cycles. Population assistance is needed project cycles would be invoked after a project was de- most in doubly constrained settings such as Bangladesh; veloped and tested. where societal constraints hamper demand for contracep- The Bangladesh program, and the Bank's contribution to tion and institutional constraints limit capacities to orga- this effort, have long been the subject of considerable dis- nize effective systems of supply. Much has been learned in cussion and debate. Much has been said about how condi- the Bangladesh case that should guide lending in such sit- tions constrain demand for contraception and how uations: institutional weaknesses constrain efforts to develop sys- tems for the effective supply of family planning services. * Pilot:s should be external to the lending cycle, but target- Crtiis and. skpiimaou.h.oeo tepormi ed on preparing the material for an initial appraisal mis- agltacdsm and skeptncsm about the role of the program pi sion. ~~~~~~~~~~~Bangladesh can now be confronted with the demographic slon. . . reality that fertility has begun to decline. For all of its com- - Deeoigapormi osrie etnsivle plexity and difficulties, the Bank's efforts in Bangladesh heavy Bank staff inputs. Where institutional capacities peiyaddfiute,teBn' fot nBnlds hreavyBank staflexinpits Whdleaerehip institqutioa cepaciti must now be labeled a success. Mitigating against this suc- success in Bangladesh relates directly to heavy Bank cess has been the tendency of projects to go to scale before success inpus Bangldynades relatershdire toheavyrank strategic planning is complete and operational implications sa Flexibiinputs andrdynamicalto leders.Thi e o e grund.ou are resolved. But outweighing these limitations, and con- Flexanibilitys criticBalk tofsucerss ThereinginDhaka arnutributing significantly to success, are a set of mechanisms exarmples of how Bank officers, working in Dhaka and in providing the Bank and the GOB expertise, flexibility, and Washington with flexible resources and mechanisms, leadership that have extended the role of the Bank beyond solved major problems that would otherwise have ledrsi tahvexendteroefteBnkbyd seolvrdizedm problctperformsthatnwouldc otherwis its traditional focus on lending into the domains of donor coordination and institutional development. 119 Table 1: Bangladesh World Bank Population and Family Health Project, Appraisal Budgets, by Component and Financing Source, 1975-94 > 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Population Project I Population Project 11 Population Project III Population Project IVd z GOB GOB GOB GOB X Component Cofin- Count- Cofin- Count- Cofin- Count- Cofin- Count- IDA aticed erpart IDA anced erpart IDA anced erpart IDA anced erpart Loan Grants Funds Total Loan Grants Funds Total Loan Grants Funds Total Loan Grants Funds Total Total Civil works 4.18 7.27 2.37 13.82 18.60 24.62 3.14 46.36 41.15 0.00 3.32 45.07 64.2 19.7 8.2 92.1 197.35 (percent) (27.8) (28.6) (44.6) (30.2) (58.1) (36.7) (28.5) (41.1) (52.8) (0.0) (10.8) (21.1) (35.7) (7.7) (5.0) (15.4) (20.35) Equipment/ Supplies 6.64 3.35 0.21 6.20 5.50 7.82 3.30 16.62 31.26 23.03 5.58 59.87 26.8 17.8 151 59.7 142.39 (percent) (17.6) (13.2) (3.9) (13.6) (17.2) (11.7) (30.0) (15.1) (40.1) (23.2) (15.3) (28.0) (14.9) (7.0) (9.2) (10.0) (14.69) Training 0.43 0.27 0.00 0.70 0.10 1.78 0.37 2.25 0.58 18.55 1.02 20.15 37.7 40.4 1.5 79.6 102.70 (percent) (2.9) (1.1) (0.0) (1.5) (0.3) (2.7) (3.4) (2.0) (0.7) (18.7) (2.8) (9.4) (20.9) (15.8) (0.9) (13.3) (10.59) Salariesa 1.83 5.52 0.02 7.37 0.73 21.05 2.50 24.28 0.00 35.86 15.71 51.57 5.5 32.3 134.1 171.9 255.12 (percent) (12.2) (21.7) (0.4) (16.1) (2.3) (31.4) (22.7) (22.1) (0.0) (36.1) (43.1) (24.1) (3.1) (12.7) (81.3) (28.7) (26.31) Special Activitiesb 0.30 0.05 0.00 0.35 1.46 0.54 0.00 2.00 5.01 21.96 10.22 37.19 45.8 144.8 6.1 196.7 236.24 (percent) (2.0) (0.2) (0.0) (0.8) (4.6) (0.8) (0.0) (1.8) (6A) (22.1) (28.0) (17.4) (25.4) (56.8) (3.7) (32.8) (24.36) Unallocatede 5.62 8.93 2.72 17.27 5.6 11.20 1.69 18.49 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 35.76 (percent) (37.5) (35.2) (51.1) (37.8) (17.3) (16.7) (15.4) (16.8) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (3.7) Total 15.00 25.39 5.32 45.71 31.99 67.01 11.00 110.00 78.00 99.40 36.45 213.85 180.0 255.0 165.0 600.00 969.56 (percent) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) Percent of Total (32.81) (55.55) (11.64) (100.0) (29.08) (60.92) (10.0) (100.0) (36.47) (46.48) (17.05) (100.0) (30.0) (42.5) (27.5) (100.0) a. Public sector salaries and related operating costs. b. Research, innovative activities, and special NGO projects (including salaries). c. Physical and price contingencies unallocated at appraisal, but consigned to specific components during implementation. d. Budget at preappraisal (subject to modification). ANNEX 1 Table 2: Estimates of the Proximate Determinants of Fertility in the Bangladesh Fertility Surveys of 1975 and 1989 Fertility Component Survey Year 1975a 1989b Total Fecundity ('IT) 15.3 15.3 Contraceptive Prevalanece (u) 7.7 31.4 Contraceptive Effectiveness (e) 0.779 0.849 Lactatiornal Infecundity (i) 17.5 16.6c Bongaarts indices: - Marriage ICIm 0.92 0.86d Abortion [Cla 0.93d 0.93 Postpartum Infecundity lC]1 0.56 0.57c Contraception iC]C 0.93 0.71 Total Fertility Rate (TFR) 6.7 4.9 a. Ministry of Health and Population Control Bangladesh Fertility Survey 1979; Kabir and Rob 1990 b. Huq and Cleland 1990. c. Assurmng estimated rather than observed durations of amenorrhea. d. Indirect estimate based on 1975 Bangladesh Fertility Survey data. 121 ANNEX 1 References . S. R. Khanam, and S. Nahar 1979. "Class, Patriar- chy, and Women's Work in Bangladesh." Population Alauddin, Md. and Y Khan 1983. "A Study of Five NGOs and Development Review 5(3): 405-438. in Population Activities in Bangladesh," University Chang, M-C., R. Freedman, and T-H. Sun. 1987. "Trends in of Dhaka, Institute of Social Welfare and Research, Fertility, Family Size Preferences, and Family Plan- Dhaka (mimeographed). ning Practice: Taiwan, 1961-85," Studies in Family Arthur, B.and G. McNicoll 1978. "An Analytical Survey of Planning, 18(6)1: 320-337. Population and Development in Bangladesh." Chauls, D., B. Ryder, and W. Zaman. 1984. "An Assessment Population and Development Review 4(1): 23-80. of Management of the Bangladesh Population Pro- Bangladesh, People's Republilc of. 1972. "Proceedings of the gram." Management Sciences for Health, Boston, National Seminar on Family Planning," Ministry of Massachusetts (unpublished report, January 1984). Health and Population Control, Dhaka. Choudhuri, S. R. and H. H. Akhter. 1990. "Supply Aspects . 1979. Bangladesh Fertility Survey: First Country Re- of Meeting Demand for Family Planning." In M.B. port - 1978, World Fertility Survey. Ministry of Health Duza, ed., South Asia Study of Population Policy and and Population Control, Dhaka. Programmes: Bangladesh. United Nations Population - 1985. Government budget documents. External Re- Fund, Dhaka. sources Division, Ministry of Finance. Dhaka (un- Chowdhury, A. M. R. and K. S. Huda. 1990. "Lessons from published). Successful Experiences," In M. B. Duza, ed., South . 1989a. Status Report on Bangladesh Third Population Asia Study of Population Policy and Programmes: Bang- and Health Project. Ministry of Health and Family ladesh. United Nations Population Fund, Dhaka. Welfare, Planning Section, Family Planning Wing, Cleland, J. and C. Wilson. 1987. "Demand Theories of the Management Information Unit, Dhaka Fertility Transition: An Iconoclastic View." Population (unpublished). Studies 41(1): 5-30. - 1989b. Proposed Programme for the Health and Family and W. P. Mauldin. 1990. The Promotion of Family Planning Sector in the Fourth Five Year Plan, 1990-95. Planning by Financial Payments: The Case of Bangladesh, Ministry of Health and Family Welfare, Dhaka. Research Division Working Paper No. 13. The Popu- Bangladesh, Rural Advancement Committee. 1983. The Net: lation Council, New York. Power Structure in Ten Villages, Dhaka: BRAC Printers Committee on Population and Demography 1981. Estima- - . 1990. "A Tale of Two Wings: Health and Family tion of Recent Trends in Fertility and Mortality in Planning Programmes in an Upazilla in Northern Bangladesh, National Academy of Sciences Report Bangladesh," Research and Evaluation Division, No. 5. Washington, D.C.: National Academy Press. Bangladesh Rural Adviancement Committee, Dhaka Crane, B. B. and J. L. Finkle. 1981. "Organizational Impedi- (unpublished). ments to Development Assistance: The World Bank's Bongaarts, J. 1978. "A Fra:mework for Analyzing the Population Program." World Politics, 33(4): 516-553. Proximate Determinants of Fertility."Population and Demeny, P. 1975. "Observations on Population Policy and Development Review 4(1): 105-132. Population Program in Bangladesh." Population and and R. Potter. 1983. Fertility, Biology, and Behavior: Development Review 1(2): 307-322. An Analysis of the Proximate Determinants. New York: D'Souza, S. and L. Chen. 1980. "Sex Differentials in Mortal- Academic Press. ity in Rural Bangladesh."Population and Development and J. Stover. 1986. The Population Council Target Set- Review 6(2): 257-270. ting Model: A User's Manual. Working Paper No. 130. Duza, M. B. 1985. "Fertility Regulation in Bangladesh: Pol- The Population Council, Center for Policy Studies, icy Evolution and Execution,"Cairo Demographic New York. Centre Working Paper No. 10. Cairo Demographic Cain, M. 1981. "Risk and Insurance: Perspectives on Fertil- Centre, Cairo (unpublished). ity and Agrarian Change in Rural India and . 1990a. "Overview of Findings." In M.B. Duza, ed., Bangladesh. "Population and Development Review 7(3): South Asia Study of Population Policy and Programmes: 435-474. Bangladesh, United Nations Population Fund, Dhaka. 1983. "Fertility as an Adjustment to Risk." . 1990b. "Government of Bangladesh-Donor Pre-ap- Population and Development Review 9(4): 688-702. praisal Workshop on the Fourth Health and Family - 1986. "Risk and Fertility: Reply to Robinson." Welfare Project, 1991-1995: Report on Key Issues and Population Studies 40(2): 299-304. Recommendations Emerging from the Workshop." Ministry of Health and Family Welfare Dhaka (un- published). 122 ANNEX 1 - et al. 1989. "Family Planning Programme Perfor- Khuda, Barkat-e- and S. R. Howlader. 1985. "Population mance in Bangladesh and Strategies for Improve- Policy under the Third Five Year Plan of Bangladesh: ment: Report of the Task Force," Ministry of Health How Realistic?" Paper presented at the biennial con- and Family Welfare, Dhaka (unpublished). ference of the Bangladesh Economics Association at Fostei-, G. M. 1967. "Peasant Society and the Image of Jahangirnagar University, Savar (December 17-20). Limited Good." In J.M. Potter, M. N. Diaz, and G. M. , R. Hasan, and S. N. Mitra 1988. Follow-up Survey of Foster, eds., Peasant Society. Boston: Little Brown, and Tubectomy and Matched Cases, National Institute for Company. Population Research and Training, Dhaka. Freedman, R. 1987. "The Contribution of Social Science Re- Koblinsky, M. A., S. J. G. Brechin, S. D. Clark, and M. Y search to Population Policy and Family Planning Pro- Hasan. 1989. "Helping Managers to Manage: Work gram Effectiveness." Studies in Family Planning 18(2): Schedules of Field Workers in Rural Bangladesh." 57-82. Studies in Family Planning 20(4): 225-234. -, Siew-Ean Khoo, and B. Supraptilah. 1981. "Contra- Koenig, M. A., J. F. Phillips, and R. Simmons. 1987. "Family ceptive Use in Indonesia: A Challenge to Conven- Size Preferences and Contraceptive Use in Matlab, tional Wisdom."International Family Planning Bangladesh: 1975-1984." Studies in Family Planning Perspectives 7(1):3-15. 18(3): 117-127. Heaver, Richard, 1988. Reaching People at the Periphery: Can - and R. Simmons. 1989. "Constraints on Supply and the World Bank's Population, Health, and Nutrition Demand for Family Planning: Evidence from Rural Operations do Better? Working Paper WPS81. Popula- Bangladesh." Paper presented at the IUSSP Seminar tion and Human Resources Department, The World on the Role of Family Planning Programmes as a Fer- Bank, Washington, D.C. tility Determinant, Tunis. Huffrnan, S. L., A. K. M. A. Chowdhury, J. Chakraborty, and Korten, David C. 1975. "Organization and Management of N. K. Simpson. 1980. "Breastfeeding Patterns in Rural the Division of Population Control and Family Plan- Bangladesh." American Journal of Clinical Nutrition 33: ning."The Ford Foundation, Dhaka (unpublished). 144-154. Mahbud, M. A., A. T. P. L. Abeykoon, A. K. M. Ahmed-ul- Huq, Md.Najmul and S. Ahmad. 1989. Study of Compensa- Ghani, A. Z. M. Mahbudur Rahman, and Md. Abdur tion Payments and Family Planning in Bangladesh: A Rab. 1990. "An Impact Evaluation of the Union Synthesis. National Institute of Population Research Health and Family Welfare Centres." PDEU Report and Training, Dhaka. No. 54, Population Development and Evaluation Kabix, M. and A. K. Ubaidur Rob. 1990. "Fertility and its Unit, Planning Commission, Government of the Peo- Proximate Determinants."In M.B. Duza, ed., South ple's Republic of Bangladesh, Dhaka (unpublished). Asia Study of Population Policy and Programmes: Mauldin, W. P. 1990. "Summary Comments in the July 18- Bangladesh. United Nations Population Fund, Dhaka. 19,1990 Workshop." South Asia, Study of Population Kamal, G. M., A. U. Ahmed, and M. Khan. 1988. Clinic-based Policies and Programmes, sponsored by UNFPA, Survey of Sterilization Clients. National Institute for (unpublished). Population Research and Training, Dhaka. Menken, J. and J. F. Phillips. 1990. "Population Change in a Karirn, Md. A., M. A. Rashid, M. M. Abdullah, and S. Kha- Rural Area of Bangladesh: 1967-1987."The Annals of tun. 1985. A Study of the Status of Foreign Assistance to the American Academy of Political and Social Science Bangladesh Population Control Programme, 1973-1982. 510: 87-101. Population Control Wing, Planning Cell, Ministry of Mitra, S.N. 1986. "The Bangladesh Contraceptive Preva- Health and Population Control, Government of the lence Survey, 1985: Preliminary Tabulations."Mitra People's Republic of Bangladesh. Dhaka. and Associates, Dhaka (unpublished notes). Khan, A. R. 1977. "Poverty and Landlessness in Rural . 1989. "The Bangladesh Contraceptive Prevalence Bangladesh." ILO Report No. 9, International Labour Survey, 1989: Preliminary Results." Mitra and Organization, Geneva. Associates, Dhaka. __- and A. Mia, eds., 1984. Contraceptive Practice in and G. M. Kamal. 1985. Bangladesh Contraceptive Bangladesh: Safety Issues, PIACT, Dhaka. Prevalence Survey - 1983. A final report submitted to Kharn, M. R. 1973. "Bangladesh Population During the First the Ministry of Health and Population Control, Peo- Five Year Plan, Period 1973-78: An Estimate."The ple's Republic of Bangladesh. Mitra and Associates, Bangladesh Economic Review 1(2): 186-198. Dhaka. 123 ANNEX 1 Nag, M. 1990. "Family Planning Success Stories in . 1986. Tradition, Development, and the Individual, Bangladesh and India." IPaper prepared for the World Asian Population Change Series No. 1. Australian Bank on family planning program effectiveness. The National University, Department of Demography, Population Council, New York (preliminary draft, Canberra. restricted). , W. H. Mosley, A. R. Khan, A. I. Chowdhury, and J. and M. B. Duza. 1988. "High Contraceptive Preva- Chakraborty. 1980. "Contraceptive Distribution in lence in an Unfavorable Social Setting: A Qualitative Bangladesh: Some Lessons Learned." Studies in Case Study in Matlab, Bangladesh." A report submit- Family Planning 11: 191-201. ted to the Swedish Institate of Development Studies, Robinson, W. 1985. "Some Reflections on Recent Attacks New York, The Population Council, (unpublished). on the Population Programme in Bangladesh." National Institute of Population Research and Training. Population Issues Research Center, Institute for 1981. Bangladesh Contraceptive Prevalence Survey - Policy Research and Evaluation, The Pennsylvania 1979. Ministry of Health and Population Control, State University, State College, Pennsylvania People's Republic of Bangladesh, Dhaka. (unpublished). Phillips, J. F. 1987. "Translating Pilot Project Success into . 1986. "High Fertility as Risk Insurance." Population National Policy Development: Two Projects in Studies 40(2): 289-298. Bangladesh." Asia-Pacific Population Journal 2(2): 3-28. Sirageldin, I., M. Hossain, and M. Cain. 1975. "Family Plan- , W. S. Stinson, S. Blhatia, M. Rahman, and J. ning in Bangladesh: An Empirical Investigation." The Chakraborty. 1982. "The Demographic Impact of the Bangladesh Development Studies 3(1): 15-24. Family Planning Health Services Project in Matlab, United Nations Economic and Social Commission for Asia Bangladesh," Studies in Family Planning, 13(5): 131- and the Pacific. 1981. Population of Bangladesh, 140. Country Monograph No. 8. Bangkok, , M. B. Hossain, R. Sirnmons, and M. A. Koenig. . 1984. "Study of Levels and Trends of Fertility in 1986. "Worker-Client Exchanges and the Dynamics of Bangladesh Using the Census Data." Asian Popula- Contraceptive Use in Rural Bangladesh." Paper tion Studies Series No. 62-J, Bangkok. presented at the annual meeting of the Population United States Agency for International Development. 1990. Association of America, San Francisco. USAID Population and Health Assistance Briefing Paper. - , R. Simmons, M. A. Koenig, and J. Chakraborty. Office of Health and Population, Dhaka. 1988. "The Determinants of Reproductive Change Wolfson, M. 1983. Profiles in Population Assistance: A in a Traditional Society: Evidence from Matlab, Comparative Review of the Principal Donor Agencies, Bangladesh." Studies in Family Planning 19(6): 313- OECD, Paris. 334. World Bank. 1975a. The Population Program of the Govern- P and M Consultants. 1977. "Management Study" Report ment of Bangladesh: A Sector Review, World Bank, Pop- of a group of consultants to the Ministry of Health ulation and Nutrition Projects Department, and Population Control, People's Republic of Washington, D.C. Bangladesh. Dhaka (unplublished). Yunus, M., J. F Phillips, M. Koblinsky, and R. Simmons. Rahman, M. 1984. "Determinants of Areal Variation in Con- 1984. "Strategies for Implementing Change in a Rural traceptive Practice in Bangladesh."A thesis presented Health and Family Planning Programme in for the degree of Doctor of Philosophy at the Bangladesh." Paper presented at the annual meeting Research School of Social Sciences, The Australian of the National Council for International Health, National University, Canberra. Washington, D.C. 124 Annex 2 Thte World Bank and Indonesia's Population Program ANNEX 2 Executive Summary Introduction tional infrastructure which facilitated the work of the na- tional family planning program. In this developmental In the 20 years since 1970, Indonesia's fertility rates have context, the BKKBN helped to legitimize small planned fallen by 40 percent, mainly as a result of the increasing use families and provided the contraceptive supplies and ser- of contraception. The Indonesian family planning program vices to make that possible. There is little doubt that both (BKKBN) has been immediately responsible for the increas- the favorable developmental trends and the effective fami- ing use of contraception by providing information, contra- ly planning program contributed to the fertility decline, al- ceptive supplies, and services. It has legitimated the though this cannot be quantified. concept of family planning in the thinking of Indonesians through. a remarkable, pervasive, community-based pro- Indonesia's Family Planning Program gram. As one of the intemational donors that has given the program financial support and technical assistance, the The BKKBN has created a program which operates effec- World Blank has played a significant, although limited, role tively at every level-the nation, the 27 provinces, the 301 in these developments. Support by the Bank and other do- districts, the subdistricts, and eventually at the village and nors also has helped to legitimize the family planning pro- hamlet level. Through this hierarchy, it has succeeded in gram which has been somewhat controversial, especially in reaching the mass of married couples in the several hun- its earlier years. The Bank has also had an effect on popula- dred thousand hamlets below the village level. BKKBN, tion trends through its major support of Indonesia's broad- formally a coordinating agency rather than a line agency, based development program. works through such such ministries as those for health, ed- ucation, home affairs, information, rligion, and the armed Causes of Indonesia's Fertility Decline services. However, BKKBN also has important implemen- tation capacities which enable it to carry out directly some Indonesia's total fertility rate declined from about 5.5 to important activities and o test, expand, and hasten new in- 3.3 between 1967-70 and 1984-87. An increase in the age at itiatives. While many family planning programs espouse marriage accounted for part of the decline, but the main the goal of country-wide, community-based activities, cause was a substantial increase in the use of contraception Indonesia is one of only a few developing countries in by couples of childbearing age, from less than 10 percent in which this is a reality. the 1960s to 48 percent by 1987. The other factors which The success of the family planning program is attribut- potentially might have affected the decline-abortion, able to: divorce, fecundity, delayed marriage-consummation, breastfeeding-appear in combination not to have had 1. Thecontinuing strong political supportof President Su- much effect on fertility. harto and, through him, of the whole political-adminis- Substantial social and economic change associated with trative system. Indoesias cnsidrabe deelopentcontbutd tothe 2. An administrative structure which facilitates communi- Indonesia's considerable development contributed to the cto n oiiaino cina h rs-ot ee fertility decline in two important ways: (a) by decreasing cation and mobilization of action at the grass-roots level the demand for children and increasing the demand for for a wide range of development activities, including contraception and (b) by greatly improving the general na- 127 ANNEX 2 family planning. This system is unusually effective for a program, but these generally have not been of central developing country. importance to the program's success. While the Bank 3. Stable and effective leadLership. The charismatic chair- has given only limited support to routine field service man of the BKKBN and most of his principal deputies operations, it did provide important initial support for have been in leadership positions for more than 15 salaries when field workers were first employed to ex- years. tend the outreach of health clinics. The Bank generally 4. Significant social and economic development, which has has not played a leading role among donors in such soft- affected attitudes about mnarriage, family life, and repro- ware aspects of program operations. duction; increased openness to new ideas; and created a 4. Through support for Indonesia's general development transportation and communication infrastructure effort. More indirect, but possibly as important as direct through which the program could function as it expand- program support, has been the Bank's large-scale sup- ed throughout the Indonesian islands. port (15 billion dollars in the last 20 years) for Indone- 5. Considerable financial and technical support from inter- sia's development efforts in education, health, national and bilateral donors. This support does not transportation and communications, electrification, ag- subvert indigenous Indonesian direction of the pro- riculture, and other support for the economy. Develop- gram, in part because the Indonesians themselves have ment in these other sectors is changing Indonesian paid an increasing share of program cost, reaching 70 society in ways which probably affect the population's percent by 1980. The BKKBN has been generally suc- interest in smaller families and family planning. cessful in gaining and coordinating donor assistance for its policies, in contrast to the domination by donor direc- The distinctive characteristics of the Bank's population itionolichreristincofts som other ouinatrie It hasoworkdred work in Indonesia has resulted in a de facto division of labor closely wiaracthedo in domevotheloping rie n . Itiats, with such other donors as USAID, UNFPA, and UNICEF in support of the BKKBN. The Bank has specialized in provid- The Contribution of the World Bank to Indonesia's ing essential "hardware"infrastructure: buildings and their tamily Planning Program furnishings, equipment, and vehicles. Two-thirds of the Bank loans have been for this purpose. The Wor- d Bank has contributed to the success of the Even during the first project, when there was more em- program in a number of ways: phasis than later on software and program operations, 53 percent of the credit went for buildings, vehicles, and 1. Through population-policy dialogues with top govern- equipment. Part of the explanation for the Bank's emphasis ment leaders, to whom its broad lending program in on hardware was that the Indonesians (like other borrow- other sectors has given it continuing access. Its general ers) preferred grants rather than loans for software compo- prestige, arising from support for the whole range of de- nents. Loans were more acceptable for capital costs and velopment activities, has given legitimation to the pop- other hardware. ulation sector. Especially after the first few years, new ideas for the pro- 2. Through loans for basic program infrastructure: build- gram operations generally were developed and tested by ings, equipment, and vehicles. Two-thirds of Bank loan BKKBN in collaboration with donors other than the Bank. money was for this purpose. This is regarded as the The Bank then sometimes provided the resources for large- Bank's distinctive and principal contribution to program scale extension of such already-tested program initiatives. operations. The buildings provided the physical base for It has been perceived as-and indeed was-the major pro- the program as it expanded throughout the islands. The vider of the physical infrastructure of the buildings and the Bank provided headquarters and supply warehouses, vehicles needed for program activities as they expanded not only in Jakarta and the 27 provincial capitals, but throughout the islands. also in the 301 district capitals. They also provided a net- The fact that, in the software area, the Bank played the work of important training facilities. As the major pro- less glamorous role of complementary extension of the vider of vehicles, the Bank made possible the mobility ideas developed by the BKKBN and other donors is not essential for field services, supplies, training, IEC work, necessarily to the Bank's discredit. It can be seen as a real- and administration. istic recognition of the most effective division of labor, giv- 3. Through specific operational program components: en the local staff and modus operandi of the various training, population education, a hospital post-partum organizations in Indonesia. program, research and evaluation, and information- The program emphases of the Bank were determined in education-communication (IEC). Most of these "soft- part by its operating style and experience. The compo- ware"components have mnade some contribution to the nents of the Bank's four five-year projects have been de- 128 ANNEX 2 veloped over a period of several years by missions is not perceived as having provided intellectual leadership. recurrently visiting Indonesia. Because the projects are fi- Bank staff have not played an important role in helping the nanced by loans rather than grants, their development al- BKKBN to develop the major ideas guiding the program af- ways involves the planning agency (BAPPENAS) as well ter the first few years. Such a view of the Bank's role in pop- as the BK1KBN. This further complicates project develop- ulation is in sharp contrast to its acknowledged important ment and implementation. intellectual leadership role in Indonesia's general develop- Unlike the other major donors, the Bank has not had a ment program and in macro-economic policy for Indonesia. significant resident field staff for population, except for lim- The Bank has population staff capable of exercising such ited periods. Traveling Bank missions from Washington, leadership, but, at least in Indonesia, the implicit or explicit however qualified, could not have the knowledge of Indo- choice was to let other donors who had competent resident nesia or the personal day-to-day relations with BKKBN staff lead the way in policy for software operations. staff necessary for cooperatively developing and testing While the perceptions about the Bank's role in popula- ideas in a very dynamic program. In contrast, USAID, in tion appear to be generally correct, they do not give the addition to a resident staff of three population officers, has Bank sufficient credit for its more limited but useful sup- had many resident expatriate and Indonesian consultants port of specific project components such as training, popu- working day-to-day with the BKKBN in operations re- lation education, and IEC activities. Indonesian officials cite search and in other ways. The Bank has fielded such resi- as an example of unusual Bank "flexibility,"its support for dent staff or consultants only in a very limited way field salaries of the newly recruited field workers when Once developed, the Bank's project components have there was no other support for this crucial first extension of been difficult to change, largely because such decisions had services away from limited fixed clinic sites. to be refEerred to Washington. In addition, because the The Bank staff was concerned about the management ca- projects were financed as loans rather than grants, time- pability of the BKKBN, especially during its first decade. consuming approvals were required from the Indonesian However, the third loan period Bank reports complimented government. By contrast, USAID reviewed and revised its the BKKBN on improved management and claimed some program annually and had the capacity to authorize new credit for it. Even then, however, the Bank was concerned expenditures locally on short notice. about too frequent administrative reorganizations. BKKBN In the "soft"aspects of social sectors, such as population management no doubt has improved, and the Bank and and health, flexibility is necessary both to adjust projects to other donors may have contributed to the improvement. changing conditions and to seize opportunities to test new However, the BKKBN deserves most of the credit for initiatives. The difficulty of programming such activities creating an indigenous management style and structure five years in advance is less relevant for hardware compo- which has been effective for many years, as judged by its re- nents. Further, the Bank has had successful experience with sults in the field. As for reorganizations, these seem to have hardware components for many sectors and countries. been necessary adjustments for a dynamic program, re- The Bank's undoubted capacity for analytical sector sponding to new ideas and to the problems of extension to work was rarely applied to population in Indonesia during the outer islands. the Bank's early projects. Even when such reports were lat- The criticisms of BKKBN management generally turn er done, they do not appear to have affected the design or out to refer to meeting dissonant administrative require- implementation of the Bank's first four loan projects very ments of the Bank and other very bureaucratic Indonesian much. In at least two instances in which Bank projects in- agencies. Criticisms have been less directed to management cluded potentially important operations research projects of the program in the field. For some time, the BKKBN has to test new ideas, the results were not very useful, in part been generally regarded by both Indonesian and expatriate because of inadequate collaboration with Indonesians in observers as considerably better managed than other Indo- the initial design. However, two recent reports are highly nesian human resource agencies or than other family plan- relevant to programmatic needs and are likely to influence ning programs in the region. the fifth population loan, now being negotiated. One deals The Bank's major contribution to the success of the gen- with the prospects for the program in the 1990s. The other eral Indonesian development program was not designed to is a much-needed economic analysis of the program. meet population policy objectives. However, it had this ef- The net result of the Bank's distinctive operating style fect. The major social and economic changes associated has beerL that, for most of the past 20 years, it has been per- with the development program are generally believed to ceived by the Indonesians and other donors as well quali- lead to lower fertility. For example, there have been signifi- fied to provide essential hardware infrastructure and to cant improvements in education, a sector in which the Bank support the large-scale extension of ideas tested and devel- has been active and one with recognized links to lower fer- oped by the BKKBN working with other donors. The Bank tility. The Bank has had a much smaller role in the health 129 ANNEX 2 sector, in which Indonesia lags. The health system is impor- in Indonesia's general development program has in- tant for population, because it is a major factor in the creased its credibility in supporting the population pro- delivery of family planning services and because of its ef- gram and in its continuing policy dialogue with the fect on reproductive motivation. The Bank might have highest levels of Indonesian leadership. done more in this sector, both from a welfare and a popula- While primary credit for achievements belongs to the In- tion perspective. donesians, the Bank and other donors have made signifi- cant contributions. Because of its operating style, the Bank's The Bank's Relation to Other Donors role in assisting program operations has been mainly in providing hardware and in implementing software projects The Bank's relations to the other donors were not very developed by others. The division of labor between the good for many years. When interviewed in the late 1970s, Bank and other donors in the past was reasonable and the other donors characterized visiting Bank missions as ar- worked rather well in terms of results. However, if the Bank rogant and poorly informed. A recurrent complaint was is to play a larger role in software projects, it would need to that Bank missions asked for information and reports but change its style of work. The decision not to have a resident did not reciprocate. Happily, when interviewed in 1990, do- staff for most of the period probably has limited the scope nor representatives unanimously reported a complete and effectiveness of the Bank's contribution to the opera- change to relationships of r,eciprocity and trust in recent tional program. years. Further, there was appreciation of joint or coordinat- ed activity Such coordinated activities may be of increasing Lessons importance as the financial needs of the program expand with growing numbers of both actual and potential contra- What might the Bank have done differently or better ceptive users. that could have increased its contribution to the popula- tion program? A Larger Perspective Especially in the first decade, a resident Bank presence The contributions of both the primary actor, the BKKBN, in the population field might have speeded up imple- and the supporting donors is best understood in larger per- mentation, made the Bank more flexible, and allowed it spective. The BKKBN is responsible for the supply aspects to contribute more to program development and soft- of most of the rise in contraceptive prevalence, which in ware components. turn accounts for most of the fertility decline. However, the * The Bank could have been more sensitive and coopera- BKKBN properly defines itself as an integral part of Indo- tive in its relations with other donors. While relations nesia's comprehensive development program. The sub- have improved over time, even today more regular re- stantial social and economic changes associated with that porting of the division of labor with other donors could development program probably help to account for a sub- still be helpful part of project completion reports, audits, stantial rise in age at marriage, which in turn accounts for and other documents as a basis for evaluating and di- perhaps 20-25 percent of the fertility decline. Beyond that, recting Bank activity. thes socal hangs hve hlpedto rodue cange in * Earlier and stronger initiatives in the health sector these social changes have helped to produce changes inwudhvmbe sfl family life and in reproductive values, which affect the de- would have been useful. mand for contraception and, thus, facilitate the work of the lMore generally, Indonesia would have been a good BKKBN. It does not denigrate the remarkable success of the place pophave and gier intors,tor ex ges BKKBN to put it in this broaler context. Similarly, the role between population and other sectors, for example edu- of the donors, including the Bank, is placed in proper per- cation, female employment and, of course, health. The spective if it is described as fa,cilitating but not directing the opportunity still exists. work of the distinctively Indonesian BKKBN. * Indonesia's truly successful outreach and community- In this broader perspective, the contribution of the based program needs far more study than it has so far World Bank has three aspects. First, the loans in the popu- gotten to identify the factors associated with that success World Bank has three aspects. First, the loans in the popu- that might be transferable. lation sector have strengthened the program in specific ways, especially in providing buildings and vehicles, but much less so in software components. Secondly, the much Issues for the Future larger portfolio of Bank loans for other development sec- tors has contributed to social changes, which tend to Indonesia faces major challenges if it is to reach its goal corshange thendsired nu r sochialdr chand, theh demnd for of reducing its total fertility rate from 3.4 to the replacement change the desired number of children and the demand for leeof21Thmuharrnmbrfcntcpivuss family planning services. Finally, the Bank's important role level of 2.1. The much larger number of contraceptive users 130 ANNEX 2 needed to reach the low fertility targets will require large but with less detailed project development and monitoring increases in program expenditures, with the amount de- by the Bank. If the Bank is to take on more software compo- pending on how rapidly the BKKBN succeeds in its pro- nents, it should consider a resident population staff in In- gram for much greater privatization of contraceptive donesia, unless it decides the BKKBN no longer needs the supply and greater cost recovery in public services. Such resident staff support it has received from USAID. major changes raise issues of reorganization of the Indonesia's broad development program, with contin- program, shifts to more permanent methods of contracep- ued Bank support, should facilitate continuing social and tion, and more reliance on self-motivation of contraceptive economic change, which in turn will facilitate the continu- users. ing success of the family planning program and fertility USAID, the other major source of external funds, plans decline. Such indirect support through other sectors to scale down its participation greatly by 1995. The future strongly complements but is no substitute for strng role of the Bank depends on the extent to which it would support for the BKKBN. The Indonesian population pro- continue what it has done in the past and the extent to gram has had a significant effect going beyond the facilitat- which it might assume some of the financial and program- ing indirect effects of progress in other sectors. The Bank matic contributions that have been carried by USAID. can substantially increase its support for the population Further, the way the Bank structures its assistance will sector without significantly increasing its total Indonesian depend on its assessment of the extent to which BKKBN loan portfolio. can manage its own program with continuing financial aid, 131 ANNEX 2 1 . Introduction Population policy has been an important element in In- The Bank may also have had an effect on fertility and donesia's development program almost from the beginning other demographic variables through its lending for of its New Order government under the leadership of Pres- projects in other sectors. In the 20 years under review there ident Suharto (1966). At that time, Indonesia's population has been considerable social and economic development, as of 107 million was growing at the rapid rate of 2.35 percent1 well as a mortality decline in Indonesia. Bank loans and per annum. Indonesian leaders developed a vigorous fam- projects in education, health, transportation and communi- ily planning program because they believed that rapid pop- cation, and more generally for economic development, may ulation growth would impede their plans for social and have contributed to social and economic changes, which in economic development and that the lower fertility required turn could lead to a decrease in the number of children de- to reduce population growth would also improve family sired, an increased demand for controlled fertility, and a welfare. It is significant that Professor Widjojo, an impor- fertility decline. The effects of the family planning program tant leader of Indonesia's development program, was the and other basic social and economic changes could have author of the seminal work on Indonesia's population.2 had synergistic effects. As a result of its development and population programs, The Indonesian case, unlike that of some other countries, Indonesia's fertility rates have declined substantially in the is one in which the Bank's activity in the population and last 20 years. A case study of this significant fertility decline other development sectors might have had an influence. offers an opportunity to evaluate the role of the World Bank There is something important to be explained, because con- in this historic event. The remarkable Indonesian family traceptive use did increase and fertility did decline. The planning program (BKKBN) is generally believed to be re- causes of these changes are in the interactive influence of sponsible for much of Indonesia's fertility decline, because the family planning program and general social and eco- of its role in increasing the us,e of contraception. It is plau- nomic development. The Bank's influence could be either sible that the World Bank may have contributed to this re- direct, through its support for the BKKBN, or more indirect, sult by providing funds for the family planning program through its support of the general development program. through loans, through technical assistance, and by its dia- Changes in other such demographic factors as age at mar- logue with Indonesian leaders about population policy and riage could affect fertility independently of any direct pro- related development issues. gram effort, but possibly as a result of social and economic change. A precise evaluation of the Bank's role is not possible be- cause of the complexity of the forces at work: changingy 1. United Nations, Department of International Economic and Social Af- . . fairs, World Population Prospects, 1988, Population Studies No. 106, (New marital and reproductive values and behavior under the in- York: United Nations, 1989). Since Indconesia, like most developing countries, fluence of a complex and dynamic family planning pro- does not have vital statistics adequate 1:o the purpose, such demographic sta- gram as well as social and economic development in a tistics as birth rates, death rates, rates of natural increase, total fertility rates, cu and life expectancy are necessarily estimates from census and sample sur- lturally and economically diverse country. Further, the veys. The word "about"or "approximately"should be understood as preced- Bank was only one of several major donors with diverse ing all such cited numbers. However, it is believed that the demographic data but overlapping programs. Neither the Bank's nor other and trends cited are correct in the general order of magnitude. 2. N. Widjojo, Population Trends in Indonesia (Ithaca, New York: Cornell donor interventons were made or could have been made in University Press, 1970). terms of an experimental design through which cause and 132 ANNEX 2 effect could be neatly assessed. The best that can be done is of the family planning program and of other social forces in to assess what are plausible or probable influences of the producing these changes is discussed. Only then is it mean- program and the Bank. ingful to discuss the role of the Bank, vis-a-vis other donors, This paper proceeds by first describing the fertility de- in the achievements of the family planning program. cline and its immediate demographic causes. Then, the role 133 ANNEX 2 2 . Indonesia's Declining Fertility and its Immediate Causes Since 1970, Indonesia's fe:rtility has fallen more rapidly as to limit the number of births. Increasing the interval be- than anyone had expected at the time, in view of its rela- tween births reduces annual fertility rates, in addition to tively low level of development, cultural diversity, large ar- the effect of the smaller total number of births per couple. eas of Muslim fundamentalism, and a legacy from the Part of the fertility decline is attributable to a steady but Sukarno period of a bloated bureaucracy, political instabil- gradual increase in the age of marriage, which decreased ity, and a poor health structure. the period of risk of pregnancy. The mean age at marriage The total fertility rate, estimated at 5.5 for 1967-70, had increased from 19.3 in 1971 to 21.1 in 1985.5 It is estimated fallen to 3.3 by 1984-87.3 This 40 percent decline, while not that 20-25 percent of the fertility decline between 1967-71 as large or as rapid as that in some other Asian countries, is and 1987 is attributable to changes in marital patterns.6 The substantial in terms of the initial expectations. increasing age at marriage is probably related to greater ed- The most important immediate (proximate) cause of the ucation, fewer arranged marriages, and greater indepen- fertility decline was an increase in the use of contraception dence of young people. Later age at marriage generally by married couples to limit the number of children and to must be a result of such social changes. China increased age increase the spaced interval between births. In the 1960s, at marriage by sanctions and vigorous action through its the contraceptive prevalence rate was probably less than 10 family planning program, but this was not the cause of the percent. By 1976, it had increeased to 19 percent, and by 1987 age-at-marriage increase in Indonesia. it was 48 percent.4 There was still significant regional and Other changes in the marriage institution7-declines in provincial variation in 1987 The rate was 51 percent in the incidence of divorce and widowhood-had the effect of Java-Bali, where the program began, and 41 percent in out- increasing fertility, thus partially offsetting the opposite ef- er Java-Bali II, where the program began later. Within the fect of the rising age of marriage. Java-Bali region, the lowest prevalence rate was 46 percent (West Java) and the rates were as high as 68 percent in Bali and Yogyakarta. 5. Terence H. Hull, "Fertility Decline in Indonesia: An Institutionalist In- The increasing use of contraception led to fertility de- terpretation,"Research Note No. 72, International Population Dynamics Pro- cline, not only by reducing the number of children each gram, Department of Demography, Australian National University, May couple had, but also by increasing the interval between 1987. 6. The 25 percent estimate for 1967-71 to 1976 is from T.H. Hull, V. Hull, births, as some couples used contraception to space as well and M. Singarimbun, "Indonesia's Family Planning Story: Success and Chal- lenge,'Population Bulletin, Vol.32, No.6, November 1977, p.45. The 20 percent estimate of the percentage of change between 1976 and 1987 attributable to 3. Central Bureau of Statistics, National Indonesia Contraceptive Prevalence contraception and changing marital pattems due to the latter is from M. Adi- Survey, 1987, Jakarta, January 1989, p. 53. This report is abbreviated as NICPS etomo, S. Kitting, and S. Taufik, Fertility Transition in Indonesia: Trends in Prox- in subsequent references. imate Deterninants (Jakarta: Lembaga Demografi, 1989). 4. Data for contraceptive prevalence, desired number of children, and oth- 7. The discussion of marriage, divorce, non-consummation, post-partum er aspects of reproductive behavior cited in this paper for 1976 are from Cen- abstinence, and arranged marriages draws on the article by Hull cited in tral Bureau of Statistics and World Fertility Survey, Indonesia Fertility footnote 5, as well as his "Marriage and Divorce Trends in Indonesia,"Re- Survey, 1976: Principal Report, Jakarta, 1978. This report is abbreviated as IFS search Note No.87, International Population Dynamics Program, Depart- in subsequent references. The comparable data for 1987 are from NICPS. ment of Demography, Australian National University, March 1988. 134 ANNEX 2 Traditional marital patterns of early marriage followed more children, more than 50 percent made such a report. In by a period of non-consummation and a practice of post- 1987,21 percent of NICPS respondents said they wanted no partum abstinence - common in some parts of Java - had more children but weren't using contraception. There are the effects of keeping fertility lower than it otherwise might no reliable data on the incidence of abortion in Indonesia, have been. These practices may help to account for the fact but insofar as it occurs because of unwanted or mistimed that fertility was not especially high in Java, even before the pregnancies, it would account for some part of the fertility rise of contraceptive prevalence. The apparent decline in decline. the last 20 years of periods of non-consummation and of Overall, then, it appears that the decline in fertility was postpartum abstinence8 should have had a pro-natalist ef- mainly due to the rise in contraceptive prevalence. The in- fect, further offsetting the effects of rising age at marriage. creasing age at marriage and a possible increase in abor- There is no firm evidence for Indonesia of a decline in the tions could account for some of the fertility decline, but this incidence or duration of breastfeeding which, in some de- was at least partially offset by the decreased incidence of di- veloping countries, has had pro-natalist effects offsetting vorce, widowhood, early non-consummation, and postpar- the early effects of adoption of contraception. Neither the tum abstinence. Thus, the major proximate cause of the 1976 IFES survey nor the 1987 NICPS survey indicates any fertility decline was the increased use of contraception, the change in breastfeeding patterns. However, some micro- factor most subject to the effects of the family planning pro- studies indicate a decline in some sub-populations.9 The gram. negative relation of education to mean duration of breast- An important primary cause of increased use of contra- feeding, amenorrhea, and postpartum abstinence10 sug- ception during the transition from high to low fertility is a gests that the antinatalist effects of breastfeeding should be decline in the number of children parents want. There has declining as the education level has been rising. been such a decline in Indonesia. Between 1976 and 1987 A rising incidence of abortion is usually common during the percentage of wives with any specific number of chil- the early period of rises in contraceptive practice. This is be- dren who said they wanted no more children increased and cause, during such a period, many couples have not yet the mean number of children wanted decreased as follows: adopted contraception or have used it ineffectively. As a re- sult, many have unwanted or mistimed pregnancies and Number of Living Percent Wanting Mean Number of are likely to tum to abortion. According to the 1987 NICPS, Children No More Children Children Wanted 6 percent of the births in the preceding five years occurred 1976 1987 1976 1987 to women who wanted no more children and an additional 18 percent occurred earlier than wanted. Further, among 0 4 5 2.9 2.5 those with three or more children, substantial numbers said 1 9 12 3.2 2.7 they would have preferred fewer children if they could 2 29 42 3.5 2.9 start fannily-building over again. Among those with five or 4 or more 57 81 5.7 4.1 Total 39 51 4.2 3.2 8. Geoffrey McNicoll and Masri Singarimbun, Fertility Decline in Indone- Source: 1976 IFS and 1987 NICPS. sia: Analysis and Interpretation, Report No. 20, Committee on Population and Demography, National Academy of Sciences (Washington, D.C.: Nation- Changes in such a fundamental value as the number of al Academry Press, 1983); and Terence H. Hull, "Marriage and Divorce Trends in Indonesia,"Research Note No. 87, International Population Dynamics Pro- children wanted usually arise in large part as a result of gram, Department of Demography, Australian National University, March changes in the family and in the society at large. However, 1988.aswwilsethBKB'inesvinomtoaned 9. GeoffTey McNicoll and Masri Singarimbun, Fertility Decline in Indonesia: as we will see, the BKKBN's intensive information and ed- Analysis and Interpretation, Report No. 20, Committee on Population and De- ucation program probably reinforced the effects on de- mography, National Academy of Sciences (Washington, D.C.: National mand for children of considerable social and economic Academy Press, 1983). 10. Indicated in both IFS and NICPS. change. 135 ANNEX 2 3 i @ Social Changes as Probable Causes of Rising Use of Contraception Later age at marriage, a decline in the number of chil- cy in linking the younger generation to the world commu- dren desired, and the increasing use of contraception to nication system and to modern technology through the have smaller planned families in Indonesia are plausibly re- mass media, written instructions, advertising, and other sults of profound social and economic changes, the power- messages. ful family planning program, and the interaction between Education was perceived, with some justification, by the social changes and the family planning program in both parents and young people as providing entry to better Indonesia. jobs in the government and in the growing private sector. The dynamic social and economic development pro- Most secondary school graduates, indeed, did get civil ser- grams introduced by the New Order under President Su- vice jobs. harto in the late 1960s must have had profound effects over The expansion of education was a major thrust of the time on family life, on the life view of young married cou- government development program. Under the INPRES ples, on the aspirations of parents for their children, and on program initiated in 1974, there was a 250 percent increase the children's own aspirations. in the education share of the second Repelita Program bud- The success of the Indonesian development program (in- get (1974-78). This program resulted in its first five years in cluding the family planning program) has depended to a over 145,000 new primary schools and the rehabilitation of substantial degree on the stable and effective administra- 56,000 others. Primary school enrollment increased from 13 tive system developed under the New Order. Vertical lines to 27 million between 1973-74 and 1984-85. This education of authority were strengthened to insure that orders and re- expansion brought changes to tens of thousands of villages. sponses flow up and down between various levels. Military It was accompanied by continuing emphasis on the growth personnel have been effectively used in civil positions of new opportunities for the younger generation, both in within the administrative system. Of critical importance the mass media and in communications to the villages has been the ability of this system to mobilize attention, ac- through the political system. tion, and response at the village level. While their political While the increase in educational achievement was sub- systems differ in many respects, Indonesia and China are stantial, a sober view needs to consider also that the quality similar in their effective linkages between the national gov- of education still leaves much to be desired. There is little ernment and the village base. In this respect, they differ support for recurrent essential supplies and equipment. from India. The success of the Indonesian family planning Much learning has been of the rote variety, and textbooks program cannot be understood without reference to the ad- often are lacking. The mobilization of parents to support ministrative-political system within which it operates. schools is commendable, but it often has the effect of the en- A primary development program for a massive increase richment of programs in the urban and wealthier areas. in educational levels succeeded in making primary educa- While the specific ways in which education influences re- tion virtually universal, in greatly increasing secondary productive behavior are not certain, it is clear that educa- school enrollment, and in greater equity of education for tional attainment has been related in Indonesia, as boys and girls." More years in school exposed the new gen- eration to new ideas, in part through the school itself, but 11. The discussion on education in this paper draws on an internal Ubrld also through the life-long effect of greater functional litera- Bank report and on interviews with specialists at the Bank on the Indonesian education sector. 136 ANNEX 2 elsewhere, to wanting fewer children, to greater knowledge torized vehicles, radios, blue jeans, commercially-prepared and use of contraception, and to actually having fewer chil- foodstuffs, and other popular goods of the world consumer dren. Longer periods in school probably had the effect, markets. While all Indonesians could not afford all the new also, of making young people more independent of their goods, there is little doubt that even the poor aspired to families. have them. At the same time as educational levels rose, there were T. Hull, a long-time, perceptive student of Indonesian major increases in mass media communications via radio, society, writes that,15 "Even as some people remain poor, television, printed media, and government-organized in- the majority of the society is becoming oriented to higher formation and education programs at the village level. forms of consumption. A wide variety of consumer goods These affected both the educated, younger generation and and services is promoted through advertising and exem- the less literate older generation, linking them not only to plary use, and are distributed through the far-flung archi- the system of ideas involved in government programs, but pelago by a growing network of traders and retailers. Most also to the world-wide communication system, whose mes- of the new consumer goods and the middle class styles are sages transcend the local and the national. Between 1978 outside the reach of the mass of the Indonesians ... Full par- and 1984 alone, the proportion of the population who ticipation in the new consumerism is not necessary for this watched television in the preceding week increased from 20 innovative institution to have a strong influence on peo- to 58 percent, while the proportion who listened to the ra- ple's decision-making. More important is the strength and dio increased from 50 to 75 percent.'2 immediacy of the images of a new life which are created A parallel trend was the development of a more exten- and desired, if not immediately for themselves, then in the sive and improved road network and of buses, trucks, and future for their children. In this way mass consumption... .is small motorized vehicles, which greatly increased the cir- clearly related to the development of elements of ideology." culation of people, goods, and ideas. The network was es- "There is no traditional village in Indonesia. Change has pecially dense in Java-Bali, where 62 percent of the swept the bulk of the population along quickly, but even Indonesian population lives. There was a considerable in- the backwaters are caught up in new patterns of consump- crease of migration and urbanization. Especially important tion, trade, education, and migration."16 was circulatory migration, in which family members com- Improvements in health, in infant mortality, and in life muted to new employment opportunities, either on a daily expectancy are of special relevance to family decisions basis or for longer periods, while maintaining their links to about childbearing.17 Infant mortality is estimated to have their families in the villages. Circulatory migration helps to declined from 142 in 1971 to 112 by 1980 and to 71 by 1985. integrate villages into larger social, economic, and ideation- Child mortality declined from an estimated 134 in 1972-76 al networks.13 to 101 by 1982-87. Life expectancy is estimated to have in- Between 1972 and 1987 the number of licensed motor- creased from 43-45 in the 1960s to 56 by 1980-84, and to 60 ized vehicles increased as follows:14 by 1988. Lower mortality, especially of infants and children, can contribute to lower fertility through several mecha- 1972 1987 nisms. It is likely that the declines in infant and child mor- (thousands) tality were a result in part of the declines in fertility. Smaller Trucks 131 953 families and greater spacing between births increased fam- Buses 26 303 ily resources and attention for each child and increased the Motorcycles 615 5554 access of many infants to prolonged breastfeeding. Passenger cars 277 1170 Undergirding all of these changes were broad improve- Total 1049 7980 ments in the economy.18 In 1967, Indonesia was still very poor, with a per capita GNP of only US$50 and with a ma- The expansion of the Indonesian transportation and jority of the population classified as below the poverty line. communication network was associated with rapid growth in the availability of consumer goods: bicycles, small mo- 15. Terence H. Hull, 'Fertility Decline in Indonesia: An Institutionalist In- terpretation,"Research Note No. 72, International Population Dynamics Pro- 12. Central Bureau of Statistics, Welfare Indicators, Jakarta, 1988, Table 3.4. gram, Department of Demography, Australian National University, May 13. Graeme J. Hugo, "Population Movement, Economic Development 1987. and Social Change in Indonesia since 1971,"Paper prepared for the Annual 16. Personal communication. Meeting of the Population Association of America, Chicago, April 1987; and 17. Data from NICPS, p. 69; and from World Bank, Indonesia: Family Plan- Graeme J. Hugo, Terence H. Hull, Valerie J. Hull, and Gavin W. Jones, The De- ning Perspectives in the 1990s (Washington, D.C.: World Bank, 1990). mographic Dimension in Indonesian Development (Singapore: Oxford Universi- 18. This discussion draws on Graeme J. Hugo, Terence H. Hull, Valerie J. ty Press. 1987). Hull, and Gavin W. Jones, The Demographic Dimension in Indonesian Develop- 14. Central Bureau of Statistics, Environmental Statistics of Indonesia and ment (Singapore: Oxford University Press, 1987); and several internal World Statistical Pocket book for 1972/1973, Jakarta. Bank reports. 137 ANNEX 2 The New Order government made a commitment to broad- The scope of economic change should not be exaggerat- based economic development and, especially, to rural de- ed, since very considerable numbers of Indonesians still are velopment. With the oil boom, there was a surge in eco- just above the poverty level. Per capita GNP is still not nomic growth, with emphasis on productive use of labor high, but in the last two decades, enough Indonesians have and resources, and with relatively little in the way of direct improved their standard of living so there is some evident transfers, consumer subsidies, or public employment. Be- basis for aspirations for a better material standard of living. tween 1974 and 1978, econonmic growth was rapid, but the Associated with the social and economic changes re- effect on poverty reduction was modest due to slower ini- viewed have been fundamental changes in the outlook of tial growth in the rural sector. There was considerable re- the young adult generation and in its relation to the older duction in poverty beginning in 1978, as the Indonesian generation.20 Young people are more independent of fa- version of the Green Revolution dramatically increased rice milial direction. This is evident in later marriages, a sharp production and the government invested more heavily in decline in arranged marriages, and increased influence of rural development. peer groups. Greater education links the younger genera- Beginning in 1983, the sharp drop in oil income and tion to new ideas transcending the family and the commu- changes in the international economy led to some sharp, nity. The mass media brings images of romantic rather drastic shocks and retrenchment in some development ac- than traditional relations between husbands and wives tivities. However, Indonesia is credited by World Bank and constantly reinforces the consumption orientation of economists with mounting a successful macro-economic young adults and their children. The family is becoming adjustment program, so that per capita incomes increased more child-centered and less oriented toward service to between 1984 and 1987. In these new adjustment initiatives, parents. private investment was stressed and did increase substan- These basic ideational changes probably have contrib- tially. Non-oil sectors have grown considerably Agricul- uted to the openness of young couples to the family plan- ture, with some government protection, continued to do ning program's message that the small planned happy well. The manufacturing sector has grown rapidly. While family is good for them and for Indonesia. This message the resource-based plywood sector has been number one, has been widely and repetitively transmitted through the number two has been in texti:les and clothing. This is espe- mass media and the political and administrative system. It cially important for the population, since this sector pre- is so pervasive that it may well have contributed to, as dominantly employs women. In general, the government well as being strengthened by, the broader ideational acted to free up the private sector. It stimulated labor-inten- changes affecting the younger generation. sive, export-oriented production and helped to develop a The ongoing transformation of Indonesian society just supporting infrastructure. described is relevant to understanding the success of the It is estimated that as mucb as 50 percent of rural house- family planning program. The changes in education, holds in Java-Bali have significant income from the non-ag- health, transportation and communication networks, con- ricultural sector.19 Another indicator of rising standards of sumption levels and aspirations, standards of living, the living is an increase in the proportion of households with family, and the increasing linkage to the world system of electricity, from 6 percent in 1971 to 31 percent in 1985. ideas are in directions generally believed to be conducive According to an internal World Bank study, the net result to lower fertility.21 of economic changes is the following estimated decline in Based on the 1976 and 1987 studies, there is specific evi- the proportion of the population below the poverty level: dence for Indonesia that the number of children desired is negatively, and the use of contraception positively related to education.22 There is also evidence from the 1976 study Percent Below that the rapid rise in contraceptive use from 1970 to 1976 in- Time Poverty Line Early 1 960s 70-75 Late 1970s 40 20. This discussion draws on Terence H. Hull, "Fertility Decline in Indone- Late 1980s 22 sia: An Institutionalist Interpretation,"Research Note No. 72, International Population Dynamics Program, Department of Demography, Australian Na- tional University, May 1987; Nathan Keyfitsz, "An East Javanese Village in 1953 and 1985: Observations on Development,"Population and Development Such estimates depend heavily on the concepts used to Reviezo, Vol.11, No.4, December 1985, pp. 695-719; and interviews with Indo- define poverty and are affected by the relatively poor data nesian social scientists. base. However, it is likely that the basic trend is correct. 21. Ansley J. Coale, "The Demographic Transition Reconsidered,"lnterna- tional Union for the Scientific Study of Population, International Population Conference, Liege, 1973, Vol. 1, pp. 53-72; John C. Caldwell, "Toward a Restate- ment of Demographic Transition Theory,"Population and Development Review, Vol.2, Nos. 3-4, September-December 1976, pp. 321-366. 19. Verbal report from scholars at Gadja Mada University. 22. In IFS and NICPS. 138 ANNEX 2 volved relatively high use rates for those with a high stan- conventional wisdom has been that the very poor in LDCs dard of living, high ownership of consumer durables, high want to have a large number of children because for them, household income, and high occupational status. This sug- even young children are a net economic asset, apart from gests that the general social and economic trends previous- the security they offer in old age. However, a contrary argu- ly discussed were sufficiently favorable to a reduced ment is that there are situations in which the resources per demand for children and an increased use of contraception capita are so low that having additional children is not ratio- so that they should have facilitated the work of the family nal from the point of view of the individuals involved. It is planning program. possible that in rural Indonesia, where there often was ex- However, a simple explanation in terms of moderniza- treme pressure of population on very limited land, the fam- tion trends and higher living standards is not sufficient for ily planning program may have helped large numbers of Indonesia, because contraceptive use was also relatively couples to define their desperate situation as being a result high among the very poor, among those with low living of having too many children. Large numbers of poor villag- standards and low household income, and among those in ers, while ambivalent about such a definition, might under farming. For example, consider the following 1976 data for their circumstances be led to adopt it by the pressure of au- a standard of living index which includes components for thority figures in the local community. housing, lighting, water supply, and ownership of modern It seems very unlikely that the most disadvantaged cou- consumer durables:23 ples would have adopted contraception so quickly and in such large numbers without the influence of the family Standard of living index Percent Using Modern Contraception planning program. For them, the role of the program was Unadjusted Adjustedp probably particularly important in the early part of the 1970s, before the substantial effect of the social and eco- Very low 38 39 nomic changes could have their significant effects later in Low 31 35 Medium 31 36 the decade. 'edium-high 30 37 It does not detract from the appreciation of the achieve- High 42 55 ments of the family planning program to take into account a ~~~~~~~~~~~~~~~the changing social setting in which it operated. For those aAdjustecl for effects of region, number of living children, wife's age, wife's who were b off, mernization ma exain thoti educatior., and husband's occupation. who were better off, modemzation may explain the moti- vation to accept program services. In the case of the disad- This is illustrative of a curvilinear relationship for a num- vantaged-the mass of the Indonesian population-the ber of economic variables with relatively high rates for the program probably had to work much harder to crystallize most disadvantaged and the most advantaged groups. The latent demand. relatively high rate of use for the very poor with the lowest Apart from its effect on the motivation of potential ac- standard of living is remarkable. These were couples in ceptors, social change produced an infrastructure which fa- housing made of the poorest materials, without electricity, cilitated the work of the family planning program. For with water availabl only from a riverorpublicwell,and example, improvements in transportation and communica- with w r i e y mtions strengthened the program's capacity to interact with without any of the 13 modem consumer durables entering its staff, both face to face and through the exchange of mes- the index. sages, to transmit IEC messages to the public, and to man- It has been suggested24 that two different processes pro- age an efficient supply line. duce these patterns of contraceptive use. The conventional agean efficient25 line. Lapham and Mauldinm have shown that the strength of modernization effects of better education, higher occupa- family planning program effort depends on a country's tional status, and higher living standards produce the rising development level. Sixteen of 17 countries ranked low on levels of use as expected among those who are better off. At social setting (as of 1970) were ranked "weak"or 'very the same time, a Malthusian principle is at work for the very weak or none"on program effort (as of 1982). Fifteen of 18 poor. The hypothesis is that those who live in the most dire ranked "low-middle"on social setting were "weak"or povert,~ ~ ~ ~ ~ ~ ~ ~ ~~ane adoptdl cotrnpto whena they were linkeak"orh poverty adopt contraception when they are linked to the "very weak or none"on program effort. Indonesia's social outside world with higher living standards and are present- and economic development improved its capacity to mount ed with the idea and means for family limitation by a legit- an effective population program, but its program effort imate organization that reaches the village level. The ranked higher than that of other countries with similar de- velopment levels. 23. Ronald Freedman, Siew-Ean Khoo, and Bondan Supraptilah, "Modem Contraceptive Use in Indonesia: A Challenge to Conventional Wisdom," 25. Robert J. Lapham and W. Parker Mauldin, "Contraceptive Prevalence: World Fertility Survey, Scientific Reports, No. 20, March 1981. The Influence of Organized Family Planning Programs,"Studies in Family 24. ibii. Planning, Vol 16, No. 2, May/June 1985, Table 4. 139 ANNEX 2 4. The RIole of the National Family Planning Program The Indonesian family pla[nning program (BKKBN) has sources were more common, 78 percent still reported such been immediately responsible for the rise in contraceptive program-based sources in 1987. prevalence and, therefore, for much of the fertility decline, The number and variety of service points has expanded at least in the sense that contraceptive information and sup- rapidly. During its first phase (1970-73), a clinic-based fam- plies have come to the increasing millions of users through ily planning program worked through the Ministry of its network of services. BKK:BN clearly deserves credit for Health's maternal and child health program. In this initial what happened on the supply side. To what extent it is also period, 2,200 clinics were built and 6,600 field workers were responsible for creation of demand-the motivation for recruited. In subsequent stages, the outreach of the pro- contraception-is a more difficult question, to be discussed gram was greatly expanded to provide community-based later. service points, first in Java-Bali and then in successive phas- Prior to the inception of the BKKBN in 1970, there was es to the outer islands. By 1988, there were 63,000 village little use or knowledge about modern contraception. Up to and more than 190,000 sub-village posts (VCDC) which are 1967, the subject was taboo in the mass media and not men- operated by village family planning volunteers organized tioned by public officials. In a 1968 KAP survey in Jakarta, and supervised by family planning workers. This means three-quarters of the responcdents claimed not to know of that there are VCDC posts in virtually every village and in any method of fertility regulation.26 There was then a pro- a large number of hamlets in addition-a remarkable out- found change, as family planning became an important is- reach network. In addition, as of 1988, there were about sue for the New Order Government, with strong leadership 225,000 Posyandus scheduled to serve children under five by President Suharto. By 1976, three-quarters of the mar- in villages once a month, through a mobile team which ried women under 50 surve,yed in Jakarta said that they brings not only clinical family planning services, but also had heard of at least one modern method, and the percent- MCH services-immunization, nutrition, and oral rehydra- age was similar for Java-Bali as a whole. By 1987, 94 percent tion therapy It should not be surprising that all these sup- indicated knowledge of at least one modem method. plementary services are not always available every month. The rapid rise in the use of modem contraceptives from It was estimated in 1990 that about 85 percent of the Po- very low levels to 26 percent (in Java-Bali) by 1976 and 51 syandus were in place and that they were covering 50 per- percent by 1987 (48 percent for all of Indonesia) occurred cent of the targeted children.27 through the rapidly expanding information and service An important additional instrument of the program is network coordinated by the BKKBN. The source of contra- the acceptor groups that are intended to reinforce efforts to ceptive services in 1987 was still overwhelmingly (87 per- recruit users and maintain practice. An estimated 240,000 cent) reported as facilities organized and coordinated by such groups in 1986 may have included three to six million the BKKBN: family planning clinics, hospitals, health cen- individuals. ters, and various other public services at the village and BKKBN is a coordinating agency, not a line ministry. It hamlet level. Even in urban areas, where private sector works to a substantial degree by organizing information and supply services through the Ministry of Health and 26. Haryono Suyono, The Adoption .of Innovation in a Developing Country: The Case of Family Planning in Indonesia, (Chicago: University of Chicago, Community and Family Study Center, 1974). 27. Estimates by UNICEF personnel in February 1990 interview. 140 ANNEX 2 through the Home Ministry's civil administration at the "The government's strategy in recruiting implement- provincial, district, subdistrict, and village levels. It also in- ers has been to work downward in the administrative volves such other ministries as Education, Religion, De- hierarchy and then outward. Senior officials began by fense, and Information. Because it is not a line ministry, putting pressure for results on those below them, who BKKBN has had the flexibility needed for its bold policy of did the same with their subordinates. Officials at each decentralization and for varying the program to meet the level then worked outward to involve others, starting circumstances of different provinces. It has been able to use with their wives and close acquaintances. Since in innovative approaches crossing bureaucratic boundaries. most cases clients are recruited in their village of resi- While the BKKBN is primarily a coordinating agency, it dence, the activities of village heads were particularly also has important direct implementation capacities, for ex- important... Sometimes these potential program im- ample, by using its field workers to organize local volun- plementers co-operated with little urging, but in other teer groups, through its considerable IEC activities, its cases regional government leaders spoke of ordering massive training activities, and in the important area of or obliging subordinates... contraceptive supply. "The core strategy for recruiting clients has been a In its coordinating activity, the BKKBN staff works very combination of individual persuasion and community closely with the government officials at every level, as well influence. Our village studies suggest the following as with relevant officials from other ministries. characterization of the style of field operations in the Free-.standing, coordinating agencies such as the BKKBN Indonesian family planning program in the 1970s and have failed in a number of other countries because they did early 1980 ... If an intended program client ... .was unre- not have the power or skill to win the cooperation of line sponsive to persuasion or accepted but later dropped agencies. The success of the BKKBN in mobilizing action, out of the program, the village head, other administra- despite the rigidity and sometimes the opposition of line tors, their wives, or members of acceptors' groups agencies, is due in no small part to the high level of political were likely to stop by to talk about family planning. commitment of President Suharto to the family planning Local religious leaders may have argued ... that fertili- program. This is evident not only in his frequent public ty control was not only not against religion, but... .was statements, but also in strong financial support from the a religious obligation. The greater the number of local central Government. Success in meeting family planning program implementers, the greater were the opportu- targets is one important way in which officials at every ad- nities for them to exert an influence on unresponsive ministrative level have their work appraised. Failure to members of the community. And the larger the num- achieve targets at one level is quickly followed by an inqui- ber of acceptors, the smaller was the room of move- ry from a higher level and by steps to correct the situation. ment for holdout. Particularly strong influence could A m.ajor source of strength of the BKKBN has been the be exerted on poorer members of the community, unusual continuity of its able leadership. The charismatic who, moreover, were highly susceptible to influence chairman and his principal deputies have led the program from motivators." for more than 15 years. Such unusual continuity might Such a comprehensive community-based program is the have had a stultifying effect, but the program continues to goal of many other developing countries and even exists on be innovative. Further, there is a consensus among observ- paper in many places where the reality on the ground is dif- ers that the leadership has been politically adept in the dif- ferent from program rhetoric. However, successful imple- ficult task of working through line agencies jealous of their mentation of such a program is rare. China may be the only authority and prerogatives. The BKKBN has also been skill- other country to be as successful as Indonesia in such a ful in getting strong financial support and technical assis- strategy. tance from international donor agencies. Donors have been Given the ethnic and socioeconomic diversity of Indone- happy to share the credit that comes from supporting a suc- sia and the thousands of villages and sub-villages of vary- cessful program. ing size and population to be covered, it should not be The BKKBN has mobilized local governments and local surprising that the degree of implementation varies consid- groups on an awesome scale to bring informal community erably. Warwick29 has done an important study of how the pressure on couples to practice contraception. It has also variation in program inputs at the village level is related to made contraceptive supplies so universally available at the local level that supply is not the major issue for the decision 28. Donald P. Warwick, "The Indonesian Family Planning Program: Gov- to use or not to use them typical in many other developing emment Influence and Client Choice,"Population and Development Review, countries. Vol. 12, No. 3, September 1986, pp. 470-471. Key aspects of the operation are summarized by War- 29. Donald P. Warwick, "The Indonesian Family Planning Program: Gov- emment Influence and Client Choice," Population and Development Review, wick2t8 on the basis of study of a good sample of villages: Vol. 12, No. 3, September 1986, p. 477. 141 ANNEX 2 contraceptive prevalence rates. He summarizes his findings reorganizations of the BKKBN, which troubled Bank staff, as follows: who preferred a more stable organization. The BKKBN has been unusually successful in working "ithe best qu antitative predictors of success in the with Muslim religious leaders, in contrast to programs in family planning program were the presence and activ- most other Muslim countries. This has been facilitated by ity level of implementers, acceptors' groups, and trained bttn tan Indonesian tradition of working out problems through otherai birthr attndants andilabiliteral medintrat w. frequent conferences and discussions in which confronta- other programs and avail.ability of medical backup. tion is avoided and accommodation and consensus are These data strongly support the strategy of the Indo- stressed. Muslim leaders were asked for advice and co-opt- nesian family planning programn which puts heavy srse.Msi edr eeakdfravc n oot nemphsian fam pluaning progrntsam, whpyic usheavy ed to a common enterprise. The effectiveness of this ap- emphasis on persuading clients and supplying servic- proach is evident in the relatively high acceptance rates in areas of Java known to be more orthodox in their Muslim The fact that such prograrm inputs are strongly related to identification. It is also relevant that Muslim fundamental- the desired program outcome in a study independently col- ism is not strong in some parts of Indonesia. lecting data at the village level is good evidence that the A comparative view of the standing of the Indonesian program does make a substantial difference. Negating this family planning program can be derived from the Mauldin- proposition would require evidence that program inputs Lapham rankings30 for 1982 of the strength of family plan- are placed where demand is greatest, which might have led ning program effort for 100 developing countries. Indone- to higher contraceptive prevalence rates there even without sia ranked fifth among 100 countries. It ranked first for the the program. There is undoubtedly some bias in the place- service and service-related activities,"fourth with respect ment of medical facilities, because these follow the pro- to "policy and stage setting,"and third with respect to gram of the health department and are not controlled by "record-keeping and evaluation."Its overall ranking would BKKBN. A recent Bank study has indicated that the place- have been higher were it not for its ranking of 24th on ment of such health facilities affects the availability of such "availability and accessibility"of different fertility control effective, more permanent rmethods as the IUD, steriliza- services, resulting from the fact that safe sterilization and tion, and implants. This affects the mix of available contra- abortion services were not easily available. Indonesia was ceptives which, in turn, affects the prevalence rate. also one of only nine countries (among the 88 rated) whose The scale of the management, personnel, and logistical overall program effort score increased by at least 25 percent problems on such a large and growing program, operating between 1972 and 1982.31 Indonesia was the only country in so many diverse places, and dealing with sensitive is- classified as having "strong"program effort (as of 1982) sues, is impressive. An essential element is the steady and among the 18 countries ranked as "low-middle"on the de- appropriate flow of contraceptive supplies to service velopment level of its social setting (as of 1970). This sug- points, varying in their contraceptive mix. Further, tens of gests that Indonesia's program success exceeded thousands of workers of different types had to be selected expectations on the basis of its development level. Howev- and trained. During the Third Five-Year Plan alone, 323,000 er, the considerable social and economic change since 1970 people were trained for over 100 categories of work. In might have reduced this discrepancy. 1990, BKKBN had a staff of 42,000: about 1,000 at national BKKBN's performance was very favorably assessed as a headquarters, 25,000 field workers and supervisors, and model for other agencies in an internal Bank report on In- 16,000 at provincial and district offices. A sophisticated donesia's management development done in the mid- communication and managernent information system had 1980s. to be developed to send and receive messages to and from As a further indication of reputation among external ob- the thousands of service training, supply, and other facili- servers, the family planning program was cited as a "suc- ties. Large numbers of buildings had to be planned, built, cess,"exemplifying a high quality of strategic management, and maintained for clinics, su-pply depots, training schools, and BKKBN offices at various administrative levels. The evolution of the program from an initial clinic basis to outreach by field workers, and then a radical expansion through use of village volunteers, followed small-scale ex- 30. W. Parker Mauldin and Robert J. Lapham, "The Measurement of Fam- perimental trials of the new approaches. Since the small- ily Planning Inputs,"in Robert J. Lapham and George B. Simmons, eds., Or- scale trials were supervised by the program managers, their ganizing for Effective Family Planning Programs, (Washington, D.C.: National replication on a large scale did not face the problems asso- Academy Press, 1987), pp. 545-582. 31. W. Parker Mauldin and Robert J. Lapham, "Conditions of Fertility Die- ciated with pilot projects exter nal to the system. These ma- cline in LDCs: 1965-1980,"Paper prepared for Annual Meeting of the Popu- jor changes in the scale of activity were associated with lation Association of America, Minneapolis, May 1984. 142 ANNEX 2 in the highly regarded book by Samuel Paul32 on the man- 15-20 absolute percentage points resulted in gross overesti- agemeni- of development programs. mates of resources needed for maintenance and gross un- The success of the BKKBN in designing and managing derestimates of new users required to reach targeted levels. the complex and changing program is impressive both in The illusion that abnormally large reported prevalence in- absolute terms and in comparison with other countries. creases were real created a temporary euphoria that was However, a proper perspective also requires consideration misleading about what was possible in a short time period. of the limitations and problems of the program as well as Independent measurement of contraceptive use in good so- the achievements. cial surveys is more reliable for national and provincial es- First, the achievement of about a 50 percent contracep- timates, but it does not provide estimates needed for tive prevalence rate in about 20 years is notable, but did not assessing and modifying performance in local areas. The go as far or as rapidly as the rise in Thailand, China, Korea, accuracy of the program's system of local estimates is a con- and Taiwan, China. To be sure, those countries had some fa- tinuing problem. Changes made since a Bank internal re- cilitating cultural and socioeconomic factors not found in view of these problems in the mid-1980s are still producing Indonesia. The comparison underscores the fact that the prevalence rates acknowledged to be too high by the BKK- achievement of family planning programs is not simply de- BN official in charge of the system in February 1989. termined by their vigor, resources, and intelligent manage- Fourth, some critics of the Indonesian family planning ment. It is also affected by the social, economic, and cultural program allege that acceptance of family-planning services circumstances of the country. Unfortunately in Indonesia, sometimes results from strong social and administrative very large population increases going well into the next pressures that are morally coercive. It is probable that there century are almost inevitable, unless the already significant were such pressures in East Java in the 1970s. Acceptances pace of progress in family planning is accelerated. in "special"high-pressure drives near the end of the pro- Second, in the drive to meet quantitative targets in a rel- gram's fiscal year were shown to have lower continuation atively short time period, less-than-optimal attention has use rates than those in the normal program.35 This suggests been given to the quality of services, as indicated by the lev- that special drive acceptances involved a lower level of vol- el of understanding by acceptors of the methods offered untary commitment. Something of the same character may and accepted and by the availability of adequate medical have been involved more recently in "safari"drives, in backup for the problems of acceptors. A number of pro- which thousands were brought together for carnivax'- ii . gram cbservers have called for higher-quality services, events, including recruitment for family planning. The con- even at the cost of a possibly slower pace.33 These are prob- sequences of social pressure in the program are much less lems of which the BKKBN is aware. The Chairman of the in Indonesia than in India or China, because abortion and BKKBN indicated in 1986 that the program must do more sterilization are not part of the program. The line between to provide "... accurate, true, and honest information education and information, on the one hand, and occasion- [about contraception] without intimidation and false ex- ally heavy-handed persuasion by local groups and author- pectations."34 ities, on the other, is sometimes hard to draw. There has Third, the Indonesian program, like others which have a been a tension between social pressure to meet targets and strong l:arget-achievement orientation, faces the problem of encouragement of genuinely voluntary, community-based over-reporting of performance by local service points. The support for the program.36 Insofar as the program depends information system, which produced credible estimates of on external influence rather than personal choice, it does prevalence until at least 1976, subsequently greatly exag- not bode well for recent program initiatives for major shifts gerated. the rise of prevalence rates, especially between of program services to the private sector. It is also problem- 1982-83 and 1985-86. Exaggeration of prevalence rates by atic for intentions to scale down promotion of community socialization for family planning, on the assumption that self-motivation is enough because "small-family"values 32. Samuel Paul, Managing Development Programs: The Lessons of Success, have already been internalized. The BKKBN leadership has (Boulder, Colorado: Westview Press, 1982). expressed concern as to whether contraceptive prevalence 33. T. H. Hull and V. J. Hull, "Health Care and Birth Control in Indonesia: Links through Time,"paper presented at the Annual Meeting of the Popula- tion Association of America, San Francisco, 1986, mimeo, p. 15; cited in Don- ald P. Warwick, "The Indonesian Family Planning Program: Government Influence and Client Choice,"Population and Development Review, Vol. 12, No. 35. J. Sullivan, H. Suyono, W. Bahrawi, A. Hartoadi, "Contracepdve Use- 3, September 1986. Effectiveness in Mojokerto Regency, Indonesia: A Comparison of Regular 34. H. Suyono, "Success, Challenge and Future Prospect of National Fam- Program and Special Drive Acceptors,"Studies in Family Planning, Vol. 7, No. ily Planning Program in Indonesia,"National Family Planning Coordinating 7, July 1976, pp. 188-196. Board, 1986, mimeo, p. 7; cited in Donald P. Warwick, "The Indonesian Fam- 36. T. H. Hull and V J. Hull, "Health Care and Birth Control in Indonesia: ily Planning Program: Government Influence and Client Choice,"Population Links through Time,"paper presented at the Annual Meeting of the Popula- and Development Review, Vol. 12, No. 3, September 1986. tion Association of America, San Francisco, 1986. 143 ANNEX 2 rates would continue to rise if the government relaxed ef- crystallize latent demand for smaller numbers of children forts to mobilize community influence for acceptance. created by other social changes. However, the extent to Fifth, representatives of all the major donors believe the which it independently, substantially changed values about BKKBN is currently administratively top-heavy.37 While family size is an open question. It is likely that the overall insisting that administrative and technical staff at provin- increase in effective demand was a synergistic joint effect of cial and district headquarters are essential, BKKBN leader- the broad social changes and the strong family planning ship has been moving to reduce higher-level administrative program. Given what is at stake and the independent value staff levels. of both development and the service aspects of the pro- Despite these problems, there is no doubt that the family gram, it is prudent to consider both as essential parts of the planning program was successful on the supply side. On Indonesian population policy. the demand side, whether the program has persuaded The program very probably led to more rapid adoption many couples to want fewer children is more difficult to as- of contraception by those who wanted no more children. It sess. The program certainly conveyed such a message to probably also led many ambivalent couples to decide they millions through its community-based information and ed- wanted no more children. What is less certain is whether it ucation system. That message could have reinforced the ef- converted many who definitely wanted more children to fects of the social and economic forces which were working decide to have no more. in this direction. The role of the program was probably par- Assessing whether population programs reduce fertility ticularly important in the early 1970s, especially for poor, requires cross-national analyses. There is support for the rural couples, before the substantial effects of the social and proposition that they do in such studies.39 For Indonesia economic changes came into play later in the decade. specifically, it is possible to say in summary that it is very However, no one has yet ,developed a methodology for probable that the program contributed substantially to the disentangling the changes in the desired number of chil- fertility decline, because: dren which are induced by sDcial change, and those which * fertility fell mainly because of the increased use of con- arise from the direct communication of such ideas by an or- ganized program or in other ways. That such ideas have an the supply of contraceptive servicesp autonomous force and are not simply a reflexive result of t s o social change is part of the 'ideational hypothesis."38 It is * Indonesia's family planning effort and contraceptive veryalhangile ihart ofthe stideatincal hypothesis."3f It is prevalence were both higher than expected on the basis very plausible that the strong social pressure of the prftescieooioidctr-aoe n gram and Its extensive IEC anid service pr-ograms helped to o h oieooi niaosaoe n gram and its-extensive lEG and service programs helped to * it is unlikely that, without the program, the unexpected considerable adoption of family planning by poor, uneducated, and rural couples would have occurred so 37. As of January 1990, BKKBN hatd 42,180 employees: 25,026 were field dl workers and supervisors; 980 were adlministrative and other support staff at rapidly. national headquarters; and 16,174 were administrative and support staff at 27 provincial and 301 regency/ municipal offices. 38. John Cleland and Christopher VVilson, "Demand Theories of the Fertil- 39. Robert J. Lapham and W. Parker Maulding, "Family Planning Program ity Transition: An Iconoclastic View,"Plopulation Studies, Vol. 41, No. 1, March Effort and Birthrate Decline in Developing Countries,"International Family 1987, pp. 5-30. Planning Perspectives, Vol. 10, No. 4, December 1984, pp. 109-118. 144 ANNEX 2 5 , The Bank's Contribution to the Indonesian Program's Success Overview and Background The Bank's distinctive characteristics affected its role in population in Indonesia, as in many other countries: The World Bank plausibly can be credited with contrib- 1. The Bank-uniquely-was involved in the broad range uting to the success of the Indonesian family planning pro- of development activities which affect population gram. AL continuing series of Bank loans have supported trends, although these were not designed for that both specific family planning projects and Indonesia's gen- purpose. eral development program from the beginning of the New 2. Because it made very large loans for all major aspects of Order (,overnment. Three Bank projects have been com- development, it had access to the highest levels of lead- pleted, a fourth is almost completed,40 and the negotiations ership in the Government. are unclerway for the fifth. The loans for the first four 3. Its sector analysis and research had an excellent reputa- projects totalled US$122.4 million.41 The population sector tion, so it was potentially influential. loans have financed important specific aspects of the pro- 4. It was able to finance large projects, and the Bank's sys- gram. Especially after the first project, the major emphasis tem of accounting for time and overhead appears to fa- was on buildings, equipment, and vehicles, but there was vor larger rather than smaller loans. also more limited input into such software components as 5. The Bank's operational style in population in Indonesia training programs, population education, research and involved: evaluation, and field staff. More diffuse is the Bank's influ- * Loans for five-year periods, corresponding more or ence through policy dialogue with top Indonesian leaders less to the Indonesian plan periods (Repelita) and and the reported influence in the development of the pro- specifying projects and budgets for the whole gram's management structure. period. More indirect, but possibly as important as the Bank's * Project development, supervision, and review by specific work in the population sector, has been the Bank's staff members based in Washington who made re- large-scale support (almost US$15 billion) for Indonesia's current trips to Indonesia. Unlike other donors, the development efforts in education, health, transportation Bank did not have resident field representatives for and cornmunications, electrification, agriculture, and other population for most of the period covered. support for the economy. Development in these other sec- 6. The fact that the Bank made loans rather than grants tors has been changing Indonesian society in ways which probably helps to account for its emphasis on hardware probably affect the population's interest in smaller families (buildings, equipment, and vehicles) rather than soft- and famnily planning. ware (training, service, IEC, etc.). Third World borrow- ers generally prefer grant rather than loan funds to support software activities. Software grants are avail- 40. Information about the four Bank projects as reported by the Bank was able from other donors less prepared to fund capital drawn from the four appraisal reports, project completion reports for the three cormipleted projects, and audit reports for the first two projects. hardware projects. 41. Includes actual disbursement for the first three projects and the amount of the loan for the fourth. 145 ANNEX 2 These distinctive characteristics of the Bank both facili- leadership, who favored an aggressive population policy tated and limited what it could accomplish in various as- throughout this period. However, the Bank's support prob- pects of its work. ably helped to legitimate the BKKBN's strong policy within Indonesia, where there has always been some ideological The Bank and Policy Dialogue and bureaucratic opposition. The Bank has had much less influence over the last 20 Compared with other donors, the Bank potentially has years, as compared with USAID, with respect to more spe- an important advantage in policy dialogue about popula- cific operational policies through which the broad policies tion because Bank officials and staff are frequently negoti- are implemented. The first project (joint between IDA and ating with such key government officials as ministers of UNFPA) involved more software and less hardware (53 finance, planning officials, and even with presidents or percent) than later projects and influenced the early course prime ministers. This results partly from the fact that the of the program. However, during most of the 1970s, when Bank makes loans rather than grants, but mainly because the basic Indonesian program was further developed, US- its loan portfolio is so large. While countries naturally pre- AID had continuously resident in the country three officers fer grants rather than loans, especially for the social sectors, who spoke Indonesian, were in almost daily contact with this is offset, at least in part, by their interest in large loans their Indonesian counterparts, were often in the field, and in the economic sectors. helped the BKKBN to develop its changing operational Nevertheless, this potential comparative advantage of program. The USAID team worked closely with the BKK- the Bank has not been realized in many countries where BN in developing the important shift from clinic-based to there is disinterest in or opposition to population policy. village-based contraceptive supply, information, and moti- Several reviews of the Bank's; work in population have re- vation in 1975-76.42 The USAID mission had flexible funds ported that Bank economists in such countries are reluctant which could be committed on short notice without refer- to advocate or even discuss this issue. ence to Washington. In 1976, the BKKBN director for a ma- In indonesia, however, the Bank officials have worked jor province with a successful program told an interviewer with leaders who saw population issues as important for for the Berelson Committee that the Bank could not give development policy from the early days of the New Order. him the kind of flexible and timely support for new initia- After the announcement of a national family planning pol- tives that he got from USAID staff who knew his province icy in 1968 and even before the creation of the BKKBN in well. 1970, the Bank had a series of missions and officers discuss- By contrast, for most of the period since the first Bank ing family planning in the context of health with the Indo- population loan, there was no resident Bank field represen- nesians and with other donors. A joint initial mission with tative for population. There were such residents for two WHO and UNFPA occurred in November 1969. In May limited periods, but neither spoke Indonesian and neither 1970, President Suharto, already a warm advocate of family had supporting technical staff. Almost all of the Bank's planning, had further discussion with Robert McNamara, work in population was done by staff based in Washington President of the Bank, about a mission to develop a project who came to Indonesia for visits of 2-4 weeks a few times a with support from the Bank and other donors. In the course year, supplemented by occasional missions for planning of seven Bank missions between 1969 and 1971, the first new loans, reviewing loan completion, and other special project, eventually a joint one with UNFPA, was formulat- purposes. ed. Several Bank officials successfully encouraged a sub- Half a dozen international and bilateral donor represen- stantial expansion of the scale and geographical coverage tatives in Indonesia, interviewed in 1976 and 1978, were of the family planning program. Also, the Bank staff collab- unanimous in the view that the Bank's influence on the na- orated closely with government officials in selecting, bud- geting, and coordinating specific project components for 42. T. Reese, a USAID population officer, and Dr. Haryono Suyono, long- the First Project. Dr. Widjojo Nitisastro, an economist and time Chairman of the BKKBN, were co-authors of reports on this important leader of BAPPENAS and Coordinating Minister for Eco- phenomenon. The reports acknowledge the role in preparation of the reports nomic, Financial, and Industrial Affairs, at that time had a ofthe field notes of another USAID officer. This kind of immediate participa- tion in program policy development is inconsistent with the Bank's operat- strong interest in this issue. He was an important figure in ing style. It can well be argued that the Bank should not be so involved, but, these developments. absent such participation, the Bank is less likely to play an important, inno- The Bank, indeed, has had frequent interaction with key vative role in the development of program policy. The relevant reports are: The Bank, indeed, has hadfrequent interaction with key H. Suyon, S. H. Pandi, I. B. Astawa, Moeljono, and T Reese, "Village Family Indonesian officials, a by-procluct of the almost US$ 15 bil- Planning-The Indonesia Model, Institutionalizing Contraceptive Practi- lion in financing of 221 loans to Indonesia from 1968 ce,"National Family Planning Coordinating Board, Technical Report Series, through 1989. Such special leverage probably was not re- Monograph No. 13, Jakarta, July 1976; and H. Suyono and T Reese, "Inte- grating Village Family Planning and Primacy Health Services-The Indone- quired to discuss population issues with the Indonesian sian Perspective,"Jakarta: BKKBN, 1976. 146 ANNEX 2 ture of the program was limited by its failure to have a res- periods. Traveling Bank missions from Washington, how- ident staff with some authority to change projects. While ever qualified, could not have the knowledge of Indonesia the BKKBN staff, interviewed at that time, were naturally or the relations with BKKBN staff necessary for coopera- restrained in any criticism, the project completion report for tively developing and testing ideas in a very dynamic pro- the second population loan indicates that the BKKBN had gram. USAID, for example, in addition to a resident staff of suggested that Bank projects would go more smoothly if three population officers, has had many resident expatriate there were such representation. and Indonesian consultants working day-to-day with the When this issue was raised with the director of the BKKBN in operations research and in other ways. The Bank Bank's Indonesia office by a Berelson Committee member has done this only in a very limited way. in the rnid-1970s, he indicated that the kind of detailed The Bank has been less flexible than other donors, partly work done by USAID's mission staff was technical assis- because its typically infrastructure components did not re- tance which was not appropriate for the Bank. He indicated quire the flexibility needed for software components. Its that, if such assistance were needed, it should be provided projects, designed for five-year periods, have taken two by creating a special agency for this purpose as part of a years or more to develop. Once developed, the project com- loan project. ponents have been difficult to change, largely because such While part of the explanation for the Bank's emphasis on decisions were referred to Washington. In addition, be- hardwa:re is the fact that they worked through loans rather cause the projects were financed as loans rather than grants, than grants and their extensive experience with hardware, time-consuming approvals were required from the Indone- it was a]so a result of the BKKBN perception that the Bank's sian government. By contrast, USAID reviewed and re- operatirig style did not permit the year-to-year flexibility vised its program annually and had the capacity to desirable for many software activities. authorize new expenditures locally on short notice. An In- Since a good case can be made for a division of labor donesian official who has observed the whole human re- among donor agencies, it can be argued that little would source field for many years said in an interview, "We say have been gained if the Bank had replicated USAID's resi- that, if you want to develop ideas for a new social institu- dent staff and done more on programmatic software. How- tion, don't go to the Bank." ever, this means accepting the fact that, without a resident The net result was that the Bank has been perceived by staff the Bank had much less influence on program deci- the Indonesians and other donors primarily as well-quali- sions than it might have had. Since the important macro- fied to provide essential hardware infrastructure and to level political decisions about population policy were made support the large-scale extension of ideas first tested and relatively early, it is the policy decisions at the changing developed by others. The Bank is not perceived as having prograra level which determined its course for most of the provided intellectual leadership. Bank staff have not period under review. played an important role in helping the BKKBN to develop the major ideas guiding the program since the first project. Support for the Family Planning Program Such a view of the Bank's role in population is in sharp con- trast to its acknowledged important intellectual leadership Most of the Bank's effort in population has been in direct role in Indonesia's general development program and in suppori: of the family planning program. macro-economic policy for Indonesia. The Bank has popu- With respect to specific project components, the distinc- lation staff capable of exercising such leadership, but, at tive cha.racteristics of the Bank's population work in Indo- least in Indonesia, the implicit or explicit choice was to let nesia has resulted in a de facto division of labor with such other donors who had competent resident staff lead the other donors as USAID, UNFPA, and UNICEF. This has de- way in policy for software operations. termined the balance among the specific components of the While these perceptions about the Bank's role in popula- Bank's projects. The Bank has specialized in providing tion appear to be generally correct, the Bank deserves credit hardware: infrastructure, buildings and their furnishings, for its support - in a more limited way - of such specific equipment, and vehicles. New ideas for the program gener- program components as training, population education, ally we:re developed and tested by BKKBN in collaboration and IEC activities. with other donors who concentrated on the software of pro- It is not possible to assess in any precise way the effect of gram design and implementation. The Bank sometimes the Bank's loans and activities in the population field on the provided the resources for large-scale extension of already- rise of contraceptive prevalence and the decline of fertility, tested program initiatives, but did not lead in software which are the ultimate marks of program success, because components. the program is large and complex. It involves multiple In- Unlilce the other major donors, the Bank has not had a donesian agencies. There are multiple donors with overlap- significant resident staff for population, except for limited ping agendas and activities. The program is dynamic, 147 ANNEX 2 changing in many respects over the years. Such change and capitals. This was supplemented by a network of important flexibility have contributed to its success. For this changing, training facilities. An extensive building program during complex situation, it is not feasible (especially after the fact) the third project provided infrastructure for decentraliza- to establish direct links, for example, between the Bank's tion of program management. loans for building training schools and contraceptive prev- While there were delays and shortfalls on quality, the alence. building programs were reasonably well executed. An In- Nevertheless, the Bank can reasonably be credited with a donesian official with extensive experience with building significant role in BKKBN's success insofar as: programs indicated recently that Bank standards and pro- It is plausible that these elements have contributed to cedures resulted in buildings substantially better than oth- program success in changling reproductive behavior and er government buildings, especially outside of Jakarta. norms. Buildings are in use and in general, funds are available Unfortunately, the evidence that can be cited from sys- for their maintenance. Buildings have been used for signif- tematic evaluation of specific project components is rather icant program purposes: training that appears to be impor- thin. Project completion reports rarely provide evidence of tant for the program went on in the training centers built; systematic evaluation in terrms of the component's continu- the program has been administered from BKKBN offices in ing operation and impact. NMost of the evidence provided 27 provinces and 301 district population and family plan- has to do with pro forma completion of project components ning centers; warehouses have been used for storage of during the project period. Evaluation, according to prede- contraceptives and supplies; inpatient annexes to health termined criteria, was called for in various audits and centers were being utilized. project reports. However, this was not done systematically It is plausible that having buildings, furniture, and over the years. equipment to carry on such activities helped the program. Beyond that, BKKBN officials believe that the buildings Assessing Specific Components of have had an important value in symbolizing the impor- the Bank's Projects tance the Government and important international agen- cies attach to family planning/MCH goals. Hardware Vehicles In terms of civil works, furniture, equipment, and vehi- cles account for about two-thirds of the expenditure of This major component of all the projects provided the Bank loans in the first four projects.43 Even during the first mobility essential for field services, supplies, training, IEC project, when there was more emphasis than later on soft- work, and administration. It facilitated reaching the popu- ware and program operations, 53 percent of the credit went lation at the grass roots, especially in less accessible and less for buildings, vehicles, and eqluipment. Part of the explana- densely-settled areas. There is some evidence of higher ac- tion for the Bank's emphasis on hardware was that the In- ceptance rates associated with mobile clinic services and donesians (like other borrowers) preferred grants rather with increasing the mobility of field workers by providing than loans for software components. Loans were more ac- them with motorcycles. ceptable for capital costs and other hardware. These are types of assistance with which the Bank has long experi- Training ence in many sectors and countries. A number of Bank au- dits and reviews indicate that these were relatively easy for Training of large numbers of family planning officials the Bank as compared with software elements. Further, and various kinds of workers at various levels presumably hardware project components can absorb relatively large contributed to the effectiveness of BKKBN's thousands of investments which fit nicely with established Bank proce- workers at every level. Sixteen training centers were built dures. Approximately 40 percent of all expenditures during in Java-Bali during the first project period with Bank sup- the four population projects iwas for buildings and 26 per- port, and were provided with vehicles and other equip- cent was for equipment, furn:iture, and vehicles. The build- ment. In these centers, a yearly average of 8,100 officials ings provided the physical base for the program as it were trained in 1970-76, 10,000 in 1976-79, and 15,000 expanded throughout the islands. The Bank provided during 1978-83. Under the second project, 10 provincial headquarters and supply warehouses not only in Jakarta training centers were built in the Outer Islands-I area. and the 27 provincial capitals, but also in the 301 district These trained over 42,000 participants between 1979 and 1986-a shortfall from the 75,000 targeted in the project 43. Actual expenditures for projects one through three, plus appraisal es- esign. Under the Third Project, additional training centers timates for project four. were built in the Outer Islands-II area, and there was sup- 148 ANNEX 2 port for a wide range of training activities throughout the to create population centers at 10 or more provincial uni- country. The reported numbers trained included: versities exceeds the supply of trained and capable leaders. The new BKKBN research centers supported by the Supervisors of family planning workers 2,529 Bank, while doing some useful work, are not yet operating Doctors 27 at a very high level, according to both Indonesian and for- Subdlistrict and health center midwives 2,494 eign observers. They are handicapped by a shortage of ex- Sucistrionan helth tenantsretrminedwi 21,49 peri enced staff for a very ambitious program that is not Traditional birth attendants (retrained) 21,500 sufcinl fousd ThX97KPsre NCS a Prim-ary health care nurses 1,186 sufficiently focussed. The 1987 KAP survey (NICPS) has Nursina shealthacarenurses 1,186 been very useful and another is planned for 1991. Continu- Nur sing school teachers 790ato-tesuisrendrwy Paraimedical personnel (for IUD insertion) 6,613 ation-rate studies are under way. Parmedical pron(Such micro-studies as those being done at Gadja Mada Such a volume and range of training, indicative of the University provide insights into the changing Indonesian complexity of the program, was made possible by the de- family structures and reproductive behavior. However, velopment of the training infrastructure in all three of the there apparently is no provision in Indonesia for surveys on initial Bank projects. the family to provide national and regional parameters on such issues as changing authority and dependency rela- Population Education tions between the younger adults and their parents and how these are related to changing reproductive behavior. Support for large-scale population education programs This is an important research challenge in Indonesia, rele- has proc[uced materials and curricula and trained trainers, vant to population policy because of the inter-action of so- as well as thousands of teachers. An evaluation of this com- cial and economic change, vigorous development and ponent indicated that large numbers were exposed to it, population programs, and regional cultural variations. The learned a reasonable amount of the material taught, and re- Bank should have special interest and expertise in such tained it at least until an examination. It is plausible-but studies because of its involvement in both the population by no means certain-that this contributed to ideational and broad development programs. changes in the younger population about the desirability of As previously indicated, the BKKBN system for estimat- lower fertility, smaller families, and family planning. ing contraceptive prevalence, initiated with strong USAID The Bank support for population education included support early in the program, served the program very well buildings and their furnishings, audio-visual equipment, for a time in quickly providing data at every level down to consultants, training materials, fellowships, vehicles, and the village. and it was not far discrepant from the 1976 IFS. staff support. During the third project period, 117,000 cop- However, in the last decade, that system produced large ies of population-education instructional materials were overestimates of prevalence, with serious consequences for produced and were being used in primary and secondary planning and policy: The recent revision of the system still schools throughout the country. The materials were again produced a probable overestimate of prevalence of approx- reproduced and circulated under the fourth project. imately 10 percent in 1989. So far, the program has done quite well without the ben- Research and Evaluation efit of many sound research studies on the relationship be- tween specific program inputs, socioeconomic The Bank (and, for that matter, the BKKBN and other do- development, and reproductive behavior. Research on nors) has not been very successful in helping to improve what does and does not work in program operations has the indigenous infrastructure for research and evaluation not yielded the kinds of reports that should be possible, activity. As recently as the current Fourth Project, important given the level of program success. In April 1985 the ap- goals in this area were the creation of three research centers praisal report for the fourth Bank project indicated that the in the BKKBN, a regular series of KAP and continuation- BKKBN still had only limited capacity for undertaking op- rate studies every two years, and the strengthening of 10 erational research. There has been only modest improve- provincial university centers. ment since then. The Demographic Institute at the University of Indone- However, the successful expansions of the program from sia and the Population Studies Center at Gadja Mada clinic basis to outreach by field workers, and then to large- University are the premier national Indonesian academic scale use of village volunteers were based on experimental centers in the population field. The latter is well-known for pilot projects by BKKBN with the collaboration of USAID. its sociological-anthropological micro-studies. Qualified In- These trials did not result in scientific reports. However, donesian and foreign observers are agreed that the attempt because of their supervision by program managers, the large-scale expansion following testing did not face the 149 ANNEX 2 problems associated with pi]ot projects external to the op- The Ministry of Population and Environment erating program.44 The Bank's undoubted capacity for analytical sector The Bank has supported several kinds of activity in train- studies was not much utilized in population in Indonesia ing, research, population policy development, and other ac- during the Bank's early projects. Even when such reports tivities of the Ministry of Population and Environment. were later done, they do not appear to have affected the de- This Ministry is charged with encouraging and coordinat- sign of the Bank's four loan projects very much. In the two ing a broad range of population activities in various minis- instances in which Bank projects included potentially im- tries. While the mandate of the Ministry is important, for a portant operations research projects to test new ideas, the variety of reasons, the achievements are still mainly poten- results were not very useful, in part because of inadequate tials for the future. A recent Bank sector report45 character- collaboration with Indonesians in the initial design. How- izes the population section of the Ministry as "... a small ever, two recent reports are highly relevant to program office supported by consultants, and it has a limited capac- needs and are likely to influence the fifth population loan ity to fulfill its functions." now being negotiated. One deals with the prospects for the Other donors have always recognized the Bank's special program in the 1990s. The other is a much-needed econom- contribution in hardware infrastructure. However, for ic analysis of the program. many years, other donors as well as some Bank staff saw the Bank as being at a disadvantage in software compo- Support for Field Staff nents. Uneasiness about the software components was ex- pressed as follows in the audit of the second Bank project: The Bank helped to finance field staff salaries during the "Some concern has been expressed by certain donor first project and later, for the special outreach effort for agencies that the Bank is expanding into areas such as trans-migration resettlement areas. With these exceptions, research and evaluation in Indonesia, where the other the Bank generally has left the direct support of the largely agencies have already spent considerable resources recurrent costs of field staff to the Indonesian Government. and in which they are better equipped to monitor, giv- The Bank's initial support of field staff salaries was recently en their representation in the field. Given resource cited by BKKBN officials as especially helpful at a crucial constraints, countries may be better off obtaining stage of the program. resources for certain software elements from else- where, especially when these can be financed from grants and/or when these software components re- quire close supervision which the Bank is not The Bank has provided stupport in several projects for equipped to provide." The ank as povidd suportin svera proectsfor However, in recent years, this evaluation has been mod- IEC equipment, consultants, other training costs, field pro- Hoer,ircntyasthsvluinhsbenmd /ECequipment, consultans, other trainingcosts,fielfied to some extent. For example, appreciation of the role duction costs, operating costs for media centers, and vehi- of the Bank in implementing on a large scale projects devel- cles for IEC work. While there are qualitative reports of oped by others is illustrated in a recent UNFPA evaluation wide-spread exposure to posters, films, pamphlets, and report.46 UNFPA supported the development of six media other IEC activities, there are few data on the sizes of audi- projection centers in Java-Bali. The Bank then responded ences and none on possible effects on reproductive behav- favorably to a BKKBN request to replicate them in all other ior or values. provinces. The UNFPA initiated an IEC project covering the cost of developing training materials and of educating mas- Hospital Postpartum Program ter trainers, with the understanding that the costs of train- ing the supervisors and field workers would be covered by This component helped to increase the number of hospi- the Bank. The UNFPA also reported cooperative large-scale tals providing postpartum farnily planning services as part assistance from the Bank in supporting the institutionaliza- of the international programs coordinated by the Popula- tion of population education in both the formal and infor- tion Council. A Bank mission reported in 1977 that 25 per- mal sectors. cent of abortion and obstetric cases became new family planning acceptors. 45. World Bank, Indonesia: Family Planning Perspectives in the 1990s (Wash- 44. This aspect of the program is given particular emphasis as an impor- ington, D.C.: World Bank, 1990). tant basis for its success by Samuel Paul in Managing Development Programs: 46. United Nations Population Fund, Report on the Evaluation of Indonesia The Lessons of Success, (Boulder, Colorado: Westview Press, 1982). Country Programme, New York: United Nations, November 1988. 150 ANNEX 2 Similarly, UNICEF staff recently reported47 that their point given exceptional stress during its briefing on work in developing the program for Posyandu posts was Indonesia in 1976. greatly facilitated by the Bank's support to BKKBN for rap- This litany of complaints about management occurred id early expansion of these posts to the 25,000 mark before just when the BKKBN was having great success in provid- its subsequent expansion to 225,000 in over 60,000 villages. ing services and information to rapidly growing numbers In summary, the preceding review indicates reasonable of contraceptive users. Was the Bank failing to understand but variable achievement in implementing specific project what was, apparently, a very successful indigenous man- components, which plausibly could contribute to program agement style because it had a preconceived, non- success. Given the considerable success of the program, it is Indonesian view of what management should be? In docu- likely that the specific Bank inputs, taken together, made a ments covering the third and fourth loan periods, there is a significant contribution to program success. The major greater readiness to acknowledge management strengths at Bank investment in buildings and vehicles, providing es- BKKBN, and indeed, to claim credit for them. sential program infrastructure, could not have been provid- Understandably, the BKKBN has had many management ed by other donors. Most of the software components were problems. Initially, the inexperienced BKKBN staff had to useful, although in this area the Bank was a follower rather deal with the politics and arcane procedures of multiple In- than a leader. While its software functions could have been donesian agencies, as well as those of international foreign carried out by other donors, it is unlikely that they could donors. The fact that the Bank provided loans rather than have absorbed the costs of doing that. Besides, the Bank's grants increased the problem, because loans required spe- important role in hardware components and policy dia- cial procedures with BAPPENAS and the Ministry of Fi- logue might have had less credibility without some in- nance. Significant recurrent problems for the Bank and volvement in funding the program's actual operations. BKKBN were delays by the GOI in releasing required coun- terpart funds. The Bank itself was also initially inexperi- Developing Management Capacity enced in dealing with the multiple Indonesian agencies and the other donors. During the first Bank project, the Bank's The Bank staff have frequently claimed in their reports resident population officer was reported to have spent most that it has helped to develop BKKBN's management capac- of his time on relations with multiple agencies, dealing not ity, presumably through its very methodical appraisal, re- with substantive issues, but with internal Bank and GOI porting, supervision, and review procedures. procedures and paperwork. BKKBN not infrequently suf- A number of knowledgeable observers, interviewed in fered from the inconsistencies between the Bank's proce- the 1970s and more recently, claim that the Bank's require- dures and those of the rest of the Indonesian govemment. ments are onerous in terms of reporting and planning and During the third Bank project, the Bank staff spent 140- related documents, together with the time required from 150 weeks for supervision.48 BKKBN apparently could not BKKBN staff to work with numerous supervisory missions remedy an exceedingly slow rate of disbursement of project and to correspond with Washington. funds. It was necessary for Bank missions to persuade BAP- Another major donor also has reporting requirements PENAS and the Ministry of Finance to release the necessary that are no less time-consuming and detailed than those of counterpart budget allocations. An able staff member in the Bank. However, BKKBN's ability to deal with these are residence might have been more effective and, in the end, facilitated by the resident staff of the other agency. less expensive than so many missions from Washington. A very knowledgeable Indonesian government official BKKBN's management problems, as described in Bank indicated in an interview that administrative rigidities reports, seem to have occurred much less with the agencies were as often those of the Indonesian Government as of the and personnel it was supervising and coordinating to carry Bank. HIe indicated further that the administrative prob- out the program than with the Indonesian and donor agen- lems of working with the Bank arose mainly from having to cies from which it obtained its resources. correspond with Washington rather than working with a The Bank's audit of the second project stated that, "Over- resident representative for population. all the BKKBN was a well-managed organization and, in During the First Project, the Bank was constantly con- general, it implemented the project well. However, its effec- cerned about the failures of BKKBN and GOI management. tiveness has, in the audit's view, been impaired from time The audit report for Project One cites 12 of 13 supervision to time by extensive reorganizations, and this continues to reports as indicating that BKKBN management was an im- be a cause of concern."Similar concerns with the changes in plementation problem. The Berelson Committee heard this 48. This included some time spent in supervising several other Bank projects and in preparing for a later project - time which could not be sepa- 47. In interviews in Jakarta, February 1990. rated out. 151 ANNEX 2 BKKBN organization appear in later Bank reports. There to spend out their annual budgets, BKKBN does so regular- does not appear to have been adequate recognition by the ly because its data-based MIS system provides accurate Bank that changes are to be expected-indeed, may be de- month-by-month accounts of its financial status in relation sirable-in a program whose character necessarily changes to its detailed forward plans. Several donor representatives with rapid technical, social, and economic change and with also regarded it as better managed than family planning changes in the public's knowledge and acceptance of the programs in other Southeast and South Asian countries. program's basic ideas. No doubt, the interaction over many years on many The Bank often has tried to micro-manage and micro-su- projects of the BKKBN with the Bank and other donors has pervise from a distance. The Appraisal Report on Project contributed to the development of better management Four has an astonishing list of "agreements and recommen- structures and skills. But the Indonesians themselves de- dations"on which assurances were obtained as part of loan serve major credit in the managerial area on the basis of negotiations. These involve requirements for detailed ac- successful program results from the early years, the adap- tions and reports with specified early deadlines. Some of tation of the program to the Indonesian environment on a these seem clearly unattainable in the time specified and, regional basis, and such specific administrative indicators indeed, this has proved to be the case. After 20 years of suc- as its ability to spend out its budget according to plan. cessful experience, the BKKBN might well regard such re- BKKBN now faces new managerial challenges. It is seri- quirements as unnecessary and condescending. ously trying to shift as much of the program as possible to The contributions of the Bank and of other donors to de- the private sector (K-B-Mandiri) and to increase cost recov- veloping management skills should be evaluated in the ery in the public sector to contain the mounting costs of the perspective of the primary role of the Indonesian govern- increasing number of acceptors. Such a transition will re- ment and leadership. The BKKBN has been imaginative quire both experimentation and substantial reorganization. and effective in molding the program to fit Indonesian in- USAID, which is heavily involved in the privatization pro- stitutions. The considerable linancial, technical, and intel- gram, has indicated that it intends to radically reduce its lectual resources provided by international and bilateral general participation in the program after 1995. The Bank donors, no doubt, have been important. However, the may have to decide whether Indonesia will need assistance Indonesian leadership has been unusually successful in ne- with the new management problems and whether the Bank gotiating for and utilizing these resources in the Indonesian is capable of providing it. context. Some other Asian countries which have had simi- Is Indonesia ready to make its own decisions and man- lar resources available to them have not had similar suc- age its own program? The Bank and other donors have cess. Is is indicative of the BKKBN's attention to managing been making grants and loans for almost 20 years and the donor relations that it has prepared manuals for its staff, three principal donors all claim to have helped to build pro- laying out the operating style and the detailed require- gram management infrastructure. And, after all, the pro- ments of each donor. gram is eminently successful. If, as is likely, continuing The independent role of the Indonesians in managing financial assistance is needed, especially in foreign ex- their own program may be indicated by the rising share of change, perhaps such loans could be made with less super- program spending financed by the Government of Indone- vision, but with final accountability according to broad sia during the first three completed projects: guidelines. This could be done by allowing more flexibility * Repelita 1 (1969-70 - 1973-74) 40 % on project loans or by providing sector or program loans. * Repelita 11 (1974-75 - 1978-79) 52 % * Repelita III (1979-80 - 1982-83) 72 % Relations with Other Donors A Bank audit suggested that the major Indonesian con- When the Bank was developing its first project, many tribution to program funding reduced its need "to accom- other donors were already active in Indonesia (USAID, modate to widely divergent views of population experts or UNICEF, WHO, IPPF, the Population Council, the Ford to the substantial political sensitivities of donors."Such Foundation, and the Swedish and Japanese Governments). problems were reported as occurring in another Asian The entry of the Bank and the UNFPA onto the scene, with country where the Bank and other donors provided a large major funding for a joint project, involved a readjustment share of the program funding. which was described in the first project completion report The BKKBN is regarded by knowledgeable Indonesian as "... long and somewhat contentious, with each agency and foreign observers as being much more effectively man- attempting to strengthen its own institutional position, aged than other Indonesian human resource agencies and while simultaneously expecting co-operation from the oth- departments. While other Inclonesian agencies are unable ers."Resentments resulted from the perception that the 152 ANNEX 2 Bank saw itself as the coordinating agency. Indeed, the Possible Bank Influence on Fertility through Bank did become the executing agency for the joint Bank/ Assistance in Non-population Sectors UNFPA project, with several of the other agencies imple- menting specific aspects of the work. Discussions of the Bank's role in population almost always Resentments developed during this initial period per- call attention to the potential leverage from the Bank's invest- sisted for many years, exacerbated by the fact that the ments in other social and economic sectors. While the idea of Bank usually had no resident representative who could linking population and other sectors is plausible, there is little develop informal personal ties with the resident repre- evidence that this is done very often, either in Indonesia or sentatives of the other donor agencies. Representatives elsewhere. For example, while the Ministry of Health (MOH) of the other donor agencies told a member of the Berel- provides a substantial portion of the family planning services son Committee in 1976 that the Bank was "standoff-is- in Indonesia, neither the Bank loans for health nor those for h"and 'arrogant."They complained that they often were population provide direct support of family planning activi- interviewed by visiting Bank mission staff who didn't ties by the MOH. Similarly, while the Bank projects have pro- know ]'ndonesia well, but that they almost never got vided support for the population education program, this has feedback on the Bank's reports, ideas, or intentions until been done through the BKKBN without direct support to the after decisions were made that often affected them. An Ministry of Education. While such direct linkages with other outside mission evaluating USAID in 1979 reported that sectors are probably desirable, they would be unlikely to have "There is little contact with the World Bank, because it as powerful effects on reproductive motivation and behavior does nta arsetenasafas general progress in the social and economic sectors. doeexperts n it havebargsident toechicalostaff orr ation. The World Bank's loans and projects in such other sectors as The experts it brings in for specific loan arrangements education, agriculture, health, transportation and communi- do consult with USAID staff, but there is little opportu- cation, and, more generally, for economic development have nity for sustained contact once they depart." contributed to the major social and economic changes of the In recent years, relations with other donors have been last 20 years in Indonesia. As previously indicated, these in very much improved because of a determined effort by a turn could have played a role in the decrease in the number of relevant Bank staff member to keep other major donors children wanted, the increased use of contraception, and the informed about what happens during and following vis- fertility decline. iting missions. Donor representatives with many years The Bank's spending for the Indonesian development pro- of experience reported a cooperative change for the gram has been substantial for a broad range of development better in relations with the Bank. There has been cooper- programs. The scope of the Bank's loan program is indicated ation in funding specific projects. There have been help- by the following distribution of Bank commitments to Indone- ful exchanges of information and cooperation in sia for fiscal years 1968-89, by sector: evaluation of program components of mutual interest. Also, in recent years, Bank sector reports in population Sectors US$ millions and in such related fields as health have won the respect Agriculture and rural development 3,794 of staff of other donor agencies, both in Indonesia and in Education 1,536 the central offices of the agencies. References to such Health and nutrition 156 reports have recently appeared in publications of other Industry 540 agencies. Population 119 These recent developments indicate that, with suffi- Power/energy 3,305 cient sustained effort by a motivated staff member, do- Small-scale enterprise 798 nor relations can be improved even without a resident Telecommunications 27 population representative. Tourism 16 Transportation 2,016 More formal and comprehensive joint planning and Urban development 960 budgeting by the donors, even if it were feasible, would Water supply and sewerage 147 not necessarily be desirable. An integrated program de- Other economica 1,372 sign in this sensitive area by a consortium of outsiders Total 14,786 would be resented and probably rejected by the 'Development finance companies, trade policy, private sector development. Indonesians, who take justifiable pride in the successful program they have designed and executed. The BKKBN These data provide concrete evidence of the Bank's major coordinates the donors to suit Indonesian needs and support for a wide array of crucial development sectors. sensibilities. Most population specialists would agree that substantial progress in these sectors taken together should facilitate 153 ANNEX 2 social and familial changes generally associated with fertility central govemment expenditures on health fell by more than decline. Chapter 2 indicates that substantial social and eco- 40 percent between 1982-83 and 1987-88. These were far nomic changes have indeed occurred. There is little doubt greater than the declines for the family planning program. that the Bank has contributed to the Government's efforts in The Ministry of Health recognizes the difficult problems in these sectors. the health system and is attempting to address them. The fact that the population sector accounts for less than The rather poor status of the health system suggests that 1 percent of the Bank's total lending program suggests that, the mortality declines in Indonesia may have been associated if a larger population lending program were deemed desir- with rising standards of living as much as with improvement able, it could be done with little increase in the total of de- in health services. As previously indicated, the family plan- velopment lending to Indonesia. ning program's success in reducing fertility probably has fa- Specific comments may be useful on the education and cilitated the decline in infant and child mortality. health sectors - the two generally believed to have the The health system is important for fertility decline, both closest connection to changes in reproductive behavior. because lower infant mortality affects reproductive intentions The Bank has made 29 education loans since 1965. The and because the health system provides much of the essential proportion of Bank lending for education has been larger family planning services. A recent internal Bank report found for Indonesia than for any otlher country. (The substantial that BKKBN bears about half of family planning delivery progress in education has been described in Chapter 2.) Ed- costs: the Ministry of Health, 40 percent: and the community, ucation is the social factor with the clearest evidence of ef- the remaining 10 percent. The massive Posyandu system de- fects on reproductive behavior, both in general and for pends to a considerable extent on the system of health clinics Indonesia specifically. While much of the Bank's education for the medical aspects of its monthly services in several hun- lending has been for higher education, the lending for pri- dred thousand hamlets. Further, there is general agreement mary education (US$61 million), secondary education by BKKBN and the donors that the future success of the pro- (US$293 million), and teacher training for primary and sec- gram will depend to a considerable extent on a shift toward ondary schools (US$275 million) was not inconsiderable. such more permanent contraceptive methods as Norplant, re- Emphasis on textbook produiction and teacher training quiring more clinical medical services. More clinical services aimed to improve the quality of education. Emphasis on would also be required if there was an increase in voluntary technical education was especially relevant to elevating sterilization, which is the most used method in the United economic aspirations of young people and their parents. States. There is similarly a requirement of better health servic- The Bank's strong support for higher education (US$596 es for the goal of improving the quality of family planning million) gave Indonesia greater latitude to invest its own services, especially with reference to side effects. funds in primary and secondary education. All of this suggests that the Bank might usefully have con- The Bank's efforts in health are more recent than in other sidered both a more integrated approach to population and sectors. In general, Bank health projects have not yet dealt health and more substantial strategic assistance to upgrading with the central problems in Indonesia's health system. Ac- the health system as a whole. This would have served both cording to a recent internal health sector report by the Bank, health and population objectives. It is pertinent that in cur- Indonesia's health programs dci not compare favorably with rent negotiations for the fifth population project, a major ma- those of other countries in Southeast Asia and, in some re- ternal and child health component is contemplated involving spects, to those of even poorer countries elsewhere. Indonesia a large loan element going directly to the Ministry of Health, spends much less of its GNP or of its budget on health than with expectation of a major additional component in the next other countries in the region. It spends 2.5 percent of its bud- health loan. get on health as compared to an average of 5 percent in the re- More generally, since population trends are influenced by gion. In comparative terms, there is a low density of health intersectoral forces, the Bank could exercise greater leader- services and low utilization of both the inpatient and outpa- ship in developing an explicit intersectoral strategy for the tient services available, perhaps in part because of the low population field. density. The density of services and their utilization is lower The benefits flow not only from the other development sec- in poorer areas, raising substantial questions of equity. Skilled tors to population, but in the reverse direction as well. A Bank personnel are disproportionately in better-off places and at staff member has recently estimated49 a rate of return of 12 leading hospitals located there. There is a stress on curative percent per annum for expenditures on family planning, services as compared with preventive and child-centered ser- based on savings in the health and education sector resulting vices. Hospital beds per capita are an extremely low 0.6 per from births averted. thousand, compared with 0.9 in low-income countries and 1.3 in the lower middle-income countries with which Indone- 49. World Bank, Indonesia: Family Planning Perspectices in the 1990s (Wash- sia should be compared. During the fiscal crisis of the 1980s, ington, D.C.: World Bank, 1990). 154 ANNEX 2 The Role of the Bank in the Next Decade adopt innovative strategies and targeting for these groups. Strong public program efforts should be maintained at the The Indonesian population and development programs same time that the private sector is encouraged to become still face a formidable task in further reducing fertility and a more significant partner in family planning." bringing population growth rates to the desired zero- To achieve the targeted reduction of total fertility rates growth level. Despite the considerable success in reducing from 3.41 in 1988 to 2.88 in 1994 is estimated to require an fertility through increased use of contraception, the annual increase in the number of contraceptive users from 14.2 to rate of increase of the population only fell from 2.35 percent 19.4 millions-a 27 percent increase. The Bank report esti- in 1965--70 to about 2.00 in 1985-88.50 The rate of popula- mates that, to achieve the contraceptive prevalence rates for tion growth did not decrease more rapidly, mainly because 1994, the BKKBN budget would have to grow by 13 to 16 the decline in fertility was offset to a considerable extent by percent per annum. Projections for longer periods into the a decline in mortality Despite its effect on retarding the future, with much larger increases in numbers of contra- growth rate decline, the mortality decline is, of course, val- ceptive users, involve very large budget increases, if it is as- ued not only for the intrinsic value of saving lives, but also sumed that most services will continue in the public sector because lower child mortality facilitates further fertility de- with little cost reimbursement. For this reason, the BKKBN cline. has launched a major long-term drive to increase the pro- Indoniesia still has a very young population, a result of portion of services provided by the private sector (the K-B- formerly high (and presently still moderately high) fertility. Mandiri program) and for greater cost reimbursement in This young age structure is partly responsible for the con- the public sector. USAID is devoting a major share of its tinued relatively high crude birth rate and rate of natural current Indonesia budget funds to this purpose, but intends increase. Even when total fertility rates reach the replace- to phase out most of its program support by 1995. The Bank ment level, the population will continue to grow for de- is supporting this privatization initiative, although it is less cades because of the large number of couples of sanguine than USAID about how rapidly privatization can childbearing age produced by earlier high fertility rates. take place. Furthei, while the decrease in total fertility rates to 3.3 is a Estimates of probable trends in contraceptive practice considerable achievement, reaching replacement-level fer- depend on estimates of demand as they appear in different tility of about 2.1 will take considerable time, even assum- strata of the population whose composition is changing. ing continued good progress. The demand, in turn, depends not only on the program, but Therefore, large increases in population are almost inev- also on the extent to which continuing social and economic itable. The size of projected population increases depends development will affect the demand for children and con- on the assumptions made about fertility and mortality, but traception. However, given the central importance the GOI all credible projections indicate large population increases. attaches to reducing population growth, it is likely that it For example, recent projections by the United Nations Pop- will continue to support the program strongly, until it is as- ulation Division51 indicate a population of 216 to 253 mil- sured that autonomous actions by individuals will meet the lion by 2010 and 228 to 305 million by 2025, depending on need. Presumably the Bank ought to take a similar ap- the assumptions made. Actual zero population growth may proach while holding a watching brief on the development not be attained before the end of the twenty-first century, of strongly internalized small, planned family norms and even if replacement-level fertility is reached by 2010. the availability of services in the private sector. For the immediate future, a recent Bank sector report52 If USAID does radically reduce its participation, the presents an excellent but sobering picture of the challenge Bank may need to consider larger loans, unless privatiza- for Indonesia in the population field in the 1990's: "Further tion is much more rapid than expected or other major do- declines in fertility and in the population growth rate will nors appear on the scene. Large building programs are no take considerably more effort, because the program will longer needed, although there is a major need to replace ob- need to reach unprecedented numbers of contraceptive solescent vehicles. As USAID phases out its program, there acceptors and motivate couples with high-fertility will be major unmet needs in the software areas it has been attitudes, less likely to know about family planning and to covering. Programs to develop structures appropriate for seek services (e.g., remote areas, urban poor). The detailed the transition to privatization and self-motivation require programming of family planning operations will need to considerable software elements. As already indicated, the Bank is not presently well-equipped to field such software projects effectively, if they require the kind of in-country 50. United Nations, Department of International Economic and Social Af- continuing technical assistance USAID has given, with its fairs, World Population Prospects, 1988, Population Studies No. 106, (New considerable staff and consultant group. If the Bank does York: Uniited Nations, 1989). 51. Ibid. decide to move into such software areas, it should consider 52. World Bank, Indonesia; Family Planning Perspectives in the 1990s (Wash- having field staff resident in Indonesia and giving them ington, D.C.: World Bank, 1990). some latitude for changing projects and re-allocating funds. 155 ANNEX 2 6. Lessoins from the Bank's Indonesian Experience In current negotiations for a fifth loan, a major expan- in the field were not sympathetic to the idea of reducing re- sion in software areas is being discussed. It is significant sources in such areas. They insisted that, even where the that this comes with a plan to make the Bank's relations rates were at such high levels, the staff needed to work to with BKKBN "more flexible" by having an annual review maintain the motivation for practice and also to work in of BKKBN-prepared plans for the next year, with the other aspects of family welfare defined as part of BKKBN's Bank's supervisory missions having authority to make general mission. reasonable changes in the field including budgetary real- Eventually, in the life of any family planning program, locations within broad categories. If this innovation devel- there should be a time when the program can function in- ops, it will be important to know whether such flexible dependently, at first in terms of program design and imple- annual reviews are feasible without a resident representa- mentation, and eventually, in financial terms as well. The tive and with the present very limited authorized time for strength of the BKKBN suggests that the Indonesian pro- supervisory missions. gram may be one the Bank should study closely from this However, a different course of action is possible, on the point of view. assumption that after 20 years BKKBN is competent to de- What factors account for the success of the Indonesian vise and manage its own programs without detailed over- program? To what extent are these factors ones to which the sight and consultation. On that assumption, a sector loan Bank made a contribution? To what extent is that experi- would be appropriate, involving design and implementa- ence relevant elsewhere? What could the Bank have done tion of the program by the Indonesians within broad guide- better or differently which might have made its contribu- lines and with provision for accountability for expenditures tion greater? and program results. Consider first the favorable factors: In interviews in Indonesia in February 1990, it appeared Strong, continuing, and reliable political support for the family that BKKBN was ready for sornething like a sector loan, but planning program. that BAPPENAS preferred the discipline and checks of * The Bank's influence through policy dialogue was sig- something closer to present procedures. nificant, especially in the early years, but also later in In several provinces of Indonesia, where prevalence lev- providing continuing legitimation for a program has al- els are at about 70 percent, there are many subdistricts in ways been somewhat controversial. which contraceptive prevalence is virtually at saturation * Providing such support was not difficult, because the levels. These provide an opportunity to experiment with Indonesian leadership was always favorable. changes in the deployment of staff and resources in such * The Bank has the potential for such broad policy sup- maturing programs. A recent iinternal economic analysis of port everywhere it functions, but its influence is condi- the program suggests that In areas in which the program is tioned by the extent of indigenous leadership support. well-established and not too dependent on outreach activi- * However, the Bank had relatively little influence in ty, as in Yogyakarta, savings may occur by reducing the working with the BKKBN to develop specific BKKBN's field staff and limiting their activity to IEC. In in- operational policies through which the broad policies terviews in 1990, BKKBN staff at central headquarters and were implemented. 156 ANNEX 2 Skillful, charismatic, and stable leadership. development program alone. Population should be seen While the Bank's financial support and appreciation as an integral part of a general development effort, with strengthened the hands of the BKKBN leaders, the Bank reciprocal effects involving relatively small direct costs probably had little influence on their selection and suc- for the population sector. cessful career advancement. The leadership knew how A pervasive network of community-based volunteer groups and to develop and implement a program suitable for the In- workers providing information, services, and most important, so- donesian social environment. cial support and pressure for the legitimacy and desirability of * This was a fortunate given indigenous asset. small, planned families. * It is not transferable. Advice to choose and support good * The BKKBN has been exceptionally successful in build- leadership is gratuitous. ing this network. A good general administrative structure with the capacity for * The Bank and other donors have supported this effort in communication to and from the local community level and for various ways. mobilizing local community involvement. * However, the success of this massive effort rests mainly * The BKKBN used and skillfully adapted the general ad- on the consensus-building meetings and group activi- ministrative-political system to its special needs. ties, the central Govemment's ability to mobilize action * This important institutional asset was not developed by at the village base effectively, and on the BKKBN's high- the B3ank or any other donor. priority effort toward building this important network * It is basically a result of Indonesia's distinctive social, and maintaining its vitality over many years. political, and cultural history. * Within Indonesia, the BKKBN is a leading model for de- * It is not easily transferable or created de novo elsewhere. velopment efforts utilizing this community-based, vol- A strong and successful development program which produces untary-effort approach. both social and economic changes favorable tofertility decline and * The basic idea of using voluntary groups in community- an infrastructure which facilitates family planning program based service and social support networks is relevant for operaticns. almost all countries. Each country must build such net- " Indonesia has had such a successful development pro- works to fit its own situation, but it may require social gram and its leaders appreciate its links to the popula- innovations. It will be difficult where govemments do tion program. not have strong capacities for communication inter- * The Bank has contributed substantially to the broad de- changes with the villages and for mobilizing community velopment program in a range of sectors generally be- involvement. However, since program success will be lieved to be relevant to lower fertility and mortality. difficult without such networks, how to implement i Among sectors believed most closely linked to popula- these ideas deserves close attention almost everywhere. tiorn, the Bank did much in education but considerably The BKKBN as an effective, free-standing coordinating agency. less in health. Such sectors as transportation and com- * The ability of the BKKBN to move regular ministries to munication, not generally thought of as population-rel- action has depended on its very strong political support. evant, may have had considerable effect indirectly, both * Its effectiveness has also depended on its capacity for di- on ideas and attitudes and in facilitating the program rect implementation when needed. operations. * The Bank and other donors supported the creation of the * Supporting development broadly is something the Bank BKKBN and contributed substantially to its develop- works at wherever it operates. How best to do this tran- ment. scends population issues. * Such an agency deserves serious consideration in other * By reducing the numbers of children to be served, the countries, as an altemative to placing this function with Indonesian family planning program is saving resources typically weak ministries of health. However, agencies in the education and health sectors, so there are recipro- on the BKKBN model will be unable to elicit action from cal benefits in the linkage of population and other devel- the agencies they are supposed to co-ordinate without opment sectors. Further, smaller planned families very strong political support. improve matemal and child health. Strong support for the BKKBN by both multilateral and bilateral * While the interdependence of population trends and de- donors. velopment usually gets nominal acceptance, citation of * While primary credit for program success belongs to the the Indonesian example may be useful in policy dia- BKKBN, the financial, technical, and intellectual support logue elsewhere. Specifically, it may be useful to empha- of the donors was considerable. size that relatively small marginal expenditures for the * The Bank's distinctive contribution was in hardware ele- population sector helped move Indonesia's fertility de- ments, where it has a comparative advantage: cline more rapidly than expected from the general 157 ANNEX 2 1. The network of buildings and equipment and the ve- However, in places like Sub-Saharan Africa, the Bank's hicles provided under Bank loans were essential in- role will be very limited if it does not have the capacity frastructure for the program. The Bank specialized in to assist in the software aspects of IEC, creation of effec- providing these elements successfully. tive demand, and service delivery. These are particularly 2. The Bank has long experience with hardware ele- important where the existing infrastructure is poor and ments in many sectors and countries. general development levels are low. 3. The hardware elements may seem prosaic, but they * The Bank could have been more sensitive and coopera- are important, providing other donors are support- tive in its relation to other donors. The benefits of a bet- ing the program software elements which the build- ter relationship have become evident in recent years. ings and vehicles must serve. More regular reporting of the division of labor and rela- 4. These are desirable and. deliverable elements in most tionships with other donors would have been a helpful other countries, although the infrastructure neces- part of project completion reports, audits, and other doc- sary for good building and transportation compo- uments. This is still desirable. nents will vary considerably * The Bank could have taken earlier and stronger initia- The Bank's contribution in the software area consisted of tives in the health sector, which has both direct and indi- implementing ideas developed by others to a significant rect effects on the population sector. Emphasis on the extent. intersectoral linkage of health and population might 1. A de facto division of labor existed between the Bank have strengthened the case for improving the health and other donors, in which the Bank implicitly ac- system, especially in maternal and child health. This op- cepted their leadership in advising on program de- portunity still exists, since the Indonesians are well sign. aware of the linkage and of the inadequacies of their 2. Implicit recognition by the Bank of this division of health system. labor is to its credit. Especially in recent years, this * More generally, Indonesia is a place in which the Bank, has involved co-operation with other donors in eval- cooperatively with the Indonesians, might have devel- uation and implementation of software project ele- oped and tested strategies for intersectoral connections ments. between population and other sectors. The Bank is cer- 3. Such a division of labor was reasonable as long as tainly the best-qualified among donors to take a lead in the Bank did not wish to allocate the resident staff this direction. The BKKBN's efforts in linking income- and consultants needed to develop and test software generation and other development activities to the pop- ideas in the field. However, this meant that the Bank ulation program have not been systematically evaluat- did not exercise leadership or gain experience in pro- ed. The Community Incentive scheme under the second gram software areas which are of increasing impor- Bank project did not produce useful results for various tance. reasons. However, the Bank has not given this general 4. If such a division of labor is recognized and fostered, problem the sustained attention it merits, either in Indo- it is probably transferable to other countries, provid- nesia or elsewhere. ing other donors are available for the complementa- The Bank might have provided more intellectual leader- ry roles. ship in analyzing: What might the Bank have clone better or differently that * The interconnections between development, the popula- could have increased its contribution to the population tion program, and the socio-cultural structure in Indone- program? sia. Indonesia is a good example of these inter-sectoral * Especially during the first decade, a resident Bank pres- and inter-institutional connections which are at the ence probably would have speeded and improved even heart of the Bank's general mission. the hardware components. A resident field representa- * The elements making for success in a program which is tive for population could have made the Bank's projects eminent in its ability to field a truly successful commu- more flexible and efficient by facilitating changes in nity-based program. There is enough variability be- project components as needed. Further, it might have tween areas in the country in program achievements to led the Bank to a larger and. more innovative role in the make a study of determinants worthwhile. related software areas. However, substantial gain from a To carry out or to commission and monitor such studies resident staff would have required giving that staff lati- requires commitment of staff working on Indonesia and tude in making changes in projects and in reallocating other selected countries for some period of time. The PHN funds. That the Bank did not do more in the software ar- staff allocation is too small to add this to its other responsi- eas in Indonesia probably was not important for the In- bilities. The knowledge base in human resource areas is not donesian program, because other donors met that need. as well-developed as in areas with greater hardware 158 ANNEX 2 components. Social and economic history is being made in is needed, investment in this area can be considered as rel- the hurnan resource areas in places like Indonesia. The evant for the indirect population component of the general Bank should consider more than average allocation of ana- development programs. lytical staff time to the PHN area. If additional justification 159 I I The World Bank Headquarters European Office Tokyo Office 1818 H Street. N.W. 66 avenue d'l6na Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome Chiyoda-ku, Tokyo 100, fapan Telephone: (202) 477-1234 Telephone; (1) 40.69.30.00 Facsimile: (202) 477-6391 Facsimile: (1) 40f69.30t66 Telephone: (3) 3214-5001 Telex: WUI 64145 WORLDBANK Telex: 640651 Facsimile: (3) 3214-3657 RCA 248423 WORLDBK Telex: 26838 Cable Address: INTBAFRAD WASHINGTONDC Cover illustration by Bill Fraser ISBN 0-8213-2081-5